EVALUATION AND CHARACTERISATION

OF BIOADHESIVE SUPPOSITORIES
FORMULATED USING COMMERCIAL
HYDROGENATED PALM KERNEL
STEARIN

LAU HUI LING

Thesis submitted to the University of Nottingham

for the degree of Doctor of Philosophy

AUGUST 2015
ABSTRACT
Rectal route of drug administration is particularly useful when patients cannot tolerate

orally yet are unable to receive parenteral injections. Furthermore, studies have shown

that it may be possible to circumvent first pass metabolism if absorption was localised

in lower rectum. This is theoretically achievable if suppositories were bioadhesive.

The aim of this thesis was to evaluate two types of commercial hydrogenated palm

kernel stearin (HPKS) namely ChocExa (CE) and Supersocolate SpecialTM (SS) as

base candidates for bioadhesive suppositories in comparison to cocoa butter (CB).

Robustness of these bases during suppository manufacturing was compared using both

DSC-simulated and extemporaneous methods. Diclofenac sodium (DcNa) which

undergoes extensive first pass metabolism was selected as model drug. Suppositories

containing 50 mg DcNa and 1-5 %w/w bioadhesive polymers manufactured using CB,

CE and SS as base were evaluated in terms of physical properties, drug release,

bioadhesive properties as well as stability under different storage conditions. The

bioadhesive polymers used were Carbopol® 974P NF (CBP), hydroxypropyl

methylcellulose 2910 (HPMC), poly(vinylpyrrolidone) K30 (PVP) and carboxymethyl

chitosan (CMCTS). Two self-fabricated methods using the texture analyser (tensile

and shear stress) were developed to study bioadhesion of suppositories against porcine

colon mucosa under simulated rectal conditions.

Physical characterisation found that CE and SS were comparable to CB in terms of

thermal profile, solid fat content (SFC), pH, viscosity and displacement values (DV)

but with added advantages of reduced polymorphism and less stringent manufacturing

parameters. Solidification of CB melt into suppositories was highly dependent on the

i
maximum heating temperature (Tmax) and cooling rate (Crate). HPKS on the other hand

were more robust, as long as it is completely molten HPKS would solidify into stable

β’ polymorph; while cooling rates did not affect crystallisation. All the bioadhesive

suppositories melt between 32.5-35.5 °C. Addition of CBP decreased rate and extent

of DcNa release in a concentration dependent manner, resulting in bi-exponential first-

order kinetics release pattern. The other bioadhesive polymers had minimal impact on

DcNa release. The tensile method to study bioadhesion found that bioadhesive

properties decreased in the order of PVP > CBP > CMCTS > HPMC while the shear

method PVP > CMCTS > CBP = HPMC. In both instances, HPMC showed poor

bioadhesion with limited benefit in development of bioadhesive suppositories.

Formulations containing 5 %w/w PVP and CMCTS were selected for subsequent

stability assessment based on considerations of complete DcNa release and good

bioadhesive properties. These suppositories required refrigeration as suppositories

stored for 200 days at room temperature (24.5 ± 2.5 °C; RH 58 ± 5 %) showed

graininess and loss of surface glossiness, increased melting point, possible

triacylglycerol (TAG) separation, higher SFC at 37 °C, prolonged softening times and

decreased amount of DcNa release. These changes were unfavourable for

suppositories and may lead to ineffective treatment. Generally, SS suppositories

subjected to accelerated ageing released DcNa more efficiently than CE.

Although both HPKS were suitable suppository base substitutes of CB, SS provided

superior stability in terms of resistance to depression of DcNa release. PVP on the

other hand conferred the best bioadhesive properties among all polymers evaluated.

Thus, SS suppositories incorporated with 50 mg DNa and 5 %w/w PVP may be a

potential candidate for further development.

ii
ACKNOWLEDGEMENTS
I owe a huge thanks to my supervisor, Dr. Tung Wai Hau for her guidance, patience,

support and continuous motivation. Her valuable comments and helpful suggestions

have assisted me in many ways to complete this thesis. I would also like to express my

deepest gratitude to my co-supervisor, Dr. Chin Chiew Foan for her generous support

and encouragement. Thank you to the School of Pharmacy and University of

Nottingham for the stipend and conference travel grant.

I would like to express my sincere appreciation to the academic staff in School of

Pharmacy, Prof. Andrew Morris, Assoc. Prof. Nashiru Billa, Assoc. Prof. Wong Kok

Thong and Assoc. Prof. Low Bee Yean for their kind guidance. A hearty thanks to the

academic staff in the Faculty of Engineering for their help with the instruments,

especially Assoc. Prof. Law Chung Lim, Dr. Hii Ching Lik and Ms. Lim Siew Shee. I

am also grateful to technical staff; Fareez, Nurzulaikha, Heri and Kak Wan (Faculty of

Engineering) as well as Mr. Wong, En. Wan, Kak Siti, Shankari, Nurul, Shamila,

Linda and Asma (Faculty of Science) for their kind assistance in lab. I truly am

thankful for the support from faculty administrative staff, Pn. Salma, Sabariah, Radha,

Fifi, Carol, Nini, Tila and Marcus. Thank you to all my lab mates for making the three

years more enjoyable. I am also grateful to Tissa for all his inspirational

encouragement.

Finally, I cannot thank my parents and siblings enough for their moral and financial

support in this pursuit; especially my sister Yi Ling who is a great source of emotional

support throughout this PhD. I am also indebted to my fiancé Heng Liang for keeping

me company through the late work nights. I would not have done it without all of you.

iii
To Mum, Dad, Ron, Mei and Heng Liang

iv
TABLE OF CONTENTS
Abstract.........................................................................................................................i

Acknowledgements.....................................................................................................iii

Table of Contents.........................................................................................................v

List of Figures...........................................................................................................xiii

List of Tables ...........................................................................................................xxii

List of Appendices....................................................................................................xxv

List of Abbreviations..............................................................................................xxix

Chapter 1 General Introduction

1.1 Introduction ................................................................................................. 2

1.1.1 Advantages of suppositories ....................................................................... 4

1.1.2 Patient acceptance and social stigma .......................................................... 6

1.1.3 Suppository base ......................................................................................... 7

1.1.3.1 Fatty bases ................................................................................................... 7

1.1.3.1.1 Cocoa butter (CB) ....................................................................................... 7

1.1.3.1.2 Palm kernel oil (PKO) and hydrogenated palm kernel stearin (HPKS) ... 10

1.1.3.2 Water soluble and water miscible bases ................................................... 10

1.1.3.3 Emulsifying bases ..................................................................................... 12

1.1.4 Ideal base characteristics ........................................................................... 12

1.1.5 Manufacturing of suppositories ................................................................ 12

1.1.5.1 Hand rolling .............................................................................................. 13

1.1.5.2 Cold compression method......................................................................... 13

1.1.5.3 Fusion moulding method .......................................................................... 13

v
1.2 Human rectum ........................................................................................... 14

1.2.1 Anatomy .................................................................................................... 14

1.2.2 Rectal mucus ............................................................................................. 15

1.2.3 Rectal absorption ...................................................................................... 15

1.2.3.1 Factors affecting rectal absorption ............................................................ 16

1.2.3.1.1 Physiological factors ................................................................................. 16

1.2.3.1.2 Physicochemical factors ............................................................................ 17

1.3 Bioadhesion ............................................................................................... 18

1.3.1 Theories and mechanisms of bioadhesion ................................................ 19

1.3.2 Factors affecting bioadhesion ................................................................... 20

1.4 Bioadhesive polymers ............................................................................... 22

1.4.1 Carbomers ................................................................................................. 23

1.4.2 Poly(vinylpyrrolidone) .............................................................................. 25

1.4.3 Chitosan and carboxymethyl chitosan (CMCTS) ..................................... 26

1.4.4 Hydroxypropyl methylcellulose (HPMC) ................................................. 26

1.5 Bioadhesive suppositories and its advantage ............................................ 27

1.6 Diclofenac sodium (DcNa) ....................................................................... 28

1.7 Aims and objectives .................................................................................. 29

Chapter 2 Preformulation studies

2.1 Introduction ............................................................................................... 32

2.2 Materials ................................................................................................... 34

2.3 Methods..................................................................................................... 34

2.3.1 Characterisation of suppository base ........................................................ 34

2.3.1.1 Thermal profile ......................................................................................... 34

vi
2.3.1.1.1 Thermogravimetric analysis ...................................................................... 34

2.3.1.1.2 Differential scanning calorimetry ............................................................. 35

2.3.1.2 Solid fat content (SFC) ............................................................................. 35

2.3.1.3 pH of molten base ..................................................................................... 35

2.3.1.4 Partition coefficient of DcNa in molten base/ distilled water ................... 36

2.3.1.5 Viscosity of molten base ........................................................................... 36

2.3.2 Optimisation of manufacturing parameters .............................................. 37

2.3.2.1 DSC–simulated suppository manufacturing ............................................. 37

2.3.2.1.1 Effects of Tmax on phase behaviour ........................................................... 37

2.3.2.1.2 Effects of Hrate and Crate on phase behaviour ............................................ 38

2.3.2.2 Extemporaneous manufacturing of suppositories at various Tmax ............ 38

2.3.2.3 Determination of displacement value (DV) .............................................. 39

2.4 Results and discussion .............................................................................. 40

2.4.1 Characterisation of suppository base ........................................................ 40

2.4.2 Optimisation of suppository manufacturing methods ............................... 49

2.5 Conclusion ................................................................................................ 63

Chapter 3 Formulation and assessment of suppository dosage form

3.1 Introduction ............................................................................................... 66

3.1.1 Routine analysis for manufactured suppositories ..................................... 66

3.2 Materials ................................................................................................... 68

3.3 Methods..................................................................................................... 69

3.3.1 Manufacturing of DcNa suppositories containing bioadhesive polymers 69

3.3.2 Dosage form analysis ................................................................................ 70

3.3.2.1 Visual inspection ....................................................................................... 70

vii
3.3.2.2 Weight variation ........................................................................................ 70

3.3.2.3 Melting point ............................................................................................. 71

3.3.2.4 Determination of DcNa content ................................................................ 71

3.3.2.5 Viscosity ................................................................................................... 72

3.3.2.6 Hardness .................................................................................................... 72

3.3.2.7 Softening time ........................................................................................... 73

3.3.2.8 Statistical analysis ..................................................................................... 74

3.4 Results and discussion .............................................................................. 74

3.4.1 Physical inspection .................................................................................... 74

3.4.2 Weight variation ........................................................................................ 75

3.4.3 Melting point ............................................................................................. 76

3.4.4 Determination of DcNa content ................................................................ 80

3.4.5 Viscosity ................................................................................................... 80

3.4.6 Hardness .................................................................................................... 82

3.4.7 Softening time ........................................................................................... 85

3.5 Conclusion ................................................................................................ 87

Chapter 4 In vitro drug release and release kinetics

4.1 Introduction ............................................................................................... 90

4.1.1 Drug release studies .................................................................................. 90

4.1.2 Statistical comparison and mathematical modelling ................................. 91

4.1.2.1 Exploratory data analysis .......................................................................... 91

4.1.2.2 Model-independent methods ..................................................................... 91

4.1.2.3 Model–dependent methods ....................................................................... 94

4.1.2.3.1 First order kinetics .................................................................................... 94

viii
4.1.2.3.2 Higuchi release .......................................................................................... 95

4.1.2.3.3 Korsmeyer – Peppas model....................................................................... 96

4.1.2.3.4 Weibull model ........................................................................................... 96

4.1.2.3.5 Bi-exponential first-order kinetic model ................................................... 97

4.1.3 Selection of the best fit model .................................................................. 98

4.2 Materials and methods .............................................................................. 99

4.2.1 Dissolution setup and evaluation of formulations .................................... 99

4.2.1.1 Experimental setup.................................................................................... 99

4.2.1.2 Evaluation of formulations ..................................................................... 100

4.2.1.2.1 The effects of drug loading on DcNa release .......................................... 100

4.2.1.2.2 The effects of bioadhesive polymers on DcNa release ........................... 101

4.2.2 Statistical and mathematical analysis of data.......................................... 101

4.2.2.1 Exploratory data analysis ........................................................................ 101

4.2.2.2 Model-independent method .................................................................... 101

4.2.2.3 Model dependent method ........................................................................ 101

4.3 Results and discussion ............................................................................ 102

4.3.1 Method development .............................................................................. 102

4.3.2 Exploratory data analysis ........................................................................ 103

4.3.2.1 The effects of drug loading on DcNa release .......................................... 103

4.3.3 Model- independent method ................................................................... 107

4.3.3.1 Effects of bioadhesive polymers on DcNa release.................................. 107

4.3.3.1.1 Fit factors (ƒ1 and ƒ2) .............................................................................. 107

4.3.3.1.2 Dissolution efficiency (DE) .................................................................... 111

4.3.3.1.3 Mean dissolution time (MDT) ................................................................ 112

4.3.4 Mathematical modelling ......................................................................... 115

ix
4.4 Conclusion .............................................................................................. 124

Chapter 5 Bioadhesion studies

5.1 Introduction ............................................................................................. 128

5.1.1 Methods to study bioadhesion ................................................................ 128

5.1.2 Detachment methods ............................................................................... 129

5.1.2.1 Tensile stress ........................................................................................... 129

5.1.2.2 Shear stress .............................................................................................. 130

5.1.3 Experimental design considerations........................................................ 131

5.2 Materials and methods ............................................................................ 133

5.2.1 Materials ................................................................................................. 133

5.2.2 Methods................................................................................................... 133

5.2.2.1 Preparation of sample discs..................................................................... 133

5.2.2.2 Preparation of large intestinal tissue ....................................................... 134

5.2.2.3 Preparation of simulated rectal mucus .................................................... 135

5.2.2.4 Experimental setup.................................................................................. 135

5.2.2.4.1 Tensile measurement .............................................................................. 135

5.2.2.4.2 Shear measurement ................................................................................. 137

5.2.2.5 Effects of instrument and test variables on the bioadhesive test............. 138

5.2.2.5.1 Tensile measurement .............................................................................. 138

5.2.2.5.2 Shear measurement ................................................................................. 139

5.2.2.6 Evaluation of the bioadhesive strengths in suppository formulations using

biological membranes ............................................................................. 139

5.2.2.7 Evaluation of synthetic regenerated cellulose membrane as an alternative

to biological membrane ........................................................................... 139

x
5.2.2.8 Data analysis ........................................................................................... 142

5.3 Results and discussion ............................................................................ 143

5.3.1 Preparation of large intestinal tissue ....................................................... 143

5.3.2 Effects of instrument and test variables on the bioadhesive test............. 143

5.3.2.1 Tensile measurement .............................................................................. 143

5.3.2.2 Shear measurement ................................................................................. 149

5.3.3 Evaluation of bioadhesive strength in suppository formulations using

biological and synthetic membranes ....................................................... 154

5.3.3.1 Tensile measurement .............................................................................. 154

5.3.3.2 Shear measurement ................................................................................. 156

5.3.4 Evaluation of synthetic regenerated cellulose membrane as an alternative

to biological membrane ........................................................................... 159

5.3.4.1 Tensile measurement .............................................................................. 159

5.3.4.2 Shear measurement ................................................................................. 161

5.4 Conclusion .............................................................................................. 162

Chapter 6 Stability studies

6.1 Introduction ............................................................................................. 165

6.1.1 Stability studies in suppositories ............................................................. 165

6.1.2 Chemical testing...................................................................................... 165

6.1.3 Physical testing ....................................................................................... 166

6.2 Materials and methods ............................................................................ 167

6.2.1 Materials ................................................................................................. 167

6.2.2 Methods................................................................................................... 167

6.2.2.1 Physical appearance ................................................................................ 168

xi
6.2.2.2 Thermal profile ....................................................................................... 168

6.2.2.3 Hardness .................................................................................................. 168

6.2.2.4 Softening time ......................................................................................... 168

6.2.2.5 DcNa release ........................................................................................... 169

6.2.2.6 Statistical analysis of data ....................................................................... 169

6.3 Results and discussion ............................................................................ 169

6.3.1 Physical appearance ................................................................................ 169

6.3.2 Thermal profile ....................................................................................... 171

6.3.3 Hardness .................................................................................................. 184

6.3.4 Softening time ......................................................................................... 185

6.3.5 DcNa release ........................................................................................... 187

6.4 Conclusion .............................................................................................. 193

Chapter 7 General Conclusion and future work

7.1 General conclusion .................................................................................. 195

7.2 Future work ............................................................................................. 199

7.2.1 Evaluation of bioadhesive suppository migration in the rectum ............ 200

7.2.2 Incorporation of DcNa-polymer granules ............................................... 200

7.2.3 Introduction of synthetic cellulose membrane as alternative to biological

membranes .............................................................................................. 201

7.2.4 In vivo bioavailability studies ................................................................. 201

Bibliography ...........................................................................................................203

Appendix..................................................................................................................224

xii
LIST OF FIGURES
Figure 1.1: Shapes and sizes of rectal, vaginal and urethral suppositories (University

of North Carolina Eshelman School of Pharmacy, 2015)............................... 2

Figure 1.2 : The general chemical structure of bioadhesive polymers used, (a)

Carbopol®(CBP); (b) poly(vinylpyrrolidone) (PVP); (c) N,O-carboxymethyl

chitosan (CMCTS) and (d) hydroxypropyl methylcellulose (HPMC). ........ 24

Figure 2.1: DSC thermograms showing the first and second heating profiles of

unadulterated (raw) bases (a) CB; (b) CE and (c) SS. Solid lines and dash

lines indicate the first and second heating respectively. Thermograms were

offset for clarity. Enthalpy of fusion was included on far right of each

thermogram. .................................................................................................. 45

Figure 2.2 : SFC (%) of unadulterated (raw) suppository bases (CB, CE and SS)

determined using (a) DSC method and (b) p-NMR method (MPOB p4.9 :

2004). ............................................................................................................ 47

Figure 2.3 : The viscosity of the bases (CB, CE and SS) under different shear rates

maintained at 37 ± 0.5 ºC. Mean ± SD, n=3. ................................................ 48

Figure 2.4 : Thermograms of the second heating cycle of CB samples. CB was

subjected to various Tmax (5 °C/min) before being cooled at 2 °C/min to 4 °C.

Subtle peaks are enlarged in the respective figure inset. Arrow denotes small

amounts of form 3A polymorph present as a shoulder. Asterisk denotes

residual form 4B polymorph from first heating. ........................................... 51

Figure 2.5 : Thermograms of the second heating cycle of SS samples. SS was

subjected to various Tmax heated at 5 °C/min before being cooled at 2 °C/min

to 4 °C. Asterisk denotes residual β polymorph from first heating. ............. 52

xiii
Figure 2.6 : Thermograms of the cooling cycle of CB samples at (a) 2 °C/min, (b)

0.5 °C/min and (c) 0.2 °C/min after undergoing first heating to Tmax of 37 °C.

Thermograms of the second heating cycle are shown in the corresponding

thermograms (d–f). First heating was conducted at 5 °C/min and subjected to

solidification at various cooling rates. Subtle peaks are shown in figure insets.

Solid arrows denote presence of form 2 while open arrows denote form 4A.

Asterisks denote exothermic event while hashes indicate presence of form

3A.................................................................................................................. 54

Figure 2.7 : Thermograms of the second heating cycle of the CB. First heating was

conducted to different Tmax 38, 39 and 40 °C (5 °C/min) and subjected to

solidification at either cooling rate of 2 °C/min (a–c) or 0.5 °C/min (d–f).

Smaller peaks are shown in the figure insets. Solid arrows denote presence

of form 2; while open arrows denote form 3A. Asterisks denote form 3B. . 56

Figure 2.8 : Thermograms of (i) cooling cycle and (ii) second heating thermogram

after first ‘heat–cool’ cycle of SS samples. The SS samples were subjected to

cooling rate of (a) 5 °C/min, (b) 2 °C/min, (c) 0.5 °C/min and (d) 0.2 °C/min

after first heating to Tmax of 42 °C. Both first and second heating cycles were

conducted at rate of 5 °C/min. Solid arrows denote presence of β’ form

(mixture of β’1 and β’2). ................................................................................ 57

Figure 2.9 : Thermograms of extemporaneously produced CB suppositories

manufactured from molten heated to various Tmax followed by solidification

at regulated lab conditions of 22.5 ± 1 °C; RH 63 ± 3 %. Smaller peaks are

shown in the enlarged insets. Hashes indicate the form 2 polymorph, while

solid and open arrows denote the presence of form 3 and form 4A

xiv
 polymorphs respectively. Asterisk shows the unmelted residual (raw) base.

...................................................................................................................... 59

Figure 2.10 : Thermograms of extemporaneously produced SS suppositories

manufactured from molten heated to various Tmax followed by solidification

at regulated lab conditions of 22.5 C ± 1 °C; RH 63 ± 3 %. Solid arrow

denotes the presence of small amounts of β’2 form, while open arrow denotes

β’1 form. Asterisk shows the unmelted residual base. .................................. 60

Figure 3.1: Direction of force exerted at (a) cylindrical circumference (barrel) (b)

pointed tip of suppository. ............................................................................ 72

Figure 3.2 : The experimental setup of the softening point apparatus. ........................ 73

Figure 3.3 : The common defects found on suppositories: (a) fissures or cracks; (b)

pitting; (c) fat-blooming; (d) particle sedimentation; (e) chipping. .............. 75

Figure 3.4 : The viscosity (cp) of suppositories made using (a) CB; (b) CE; and (c) SS;

each suppository contained 50 mg DcNa and 1-5 %w/w of bioadhesive

polymers (CBP, HPMC, PVP and CMCTS). Asterisks indicate formulations

which are significantly different from blank suppositories. ......................... 81

Figure 3.5: The hardness of suppositories produced using (a) CB; (b) CE and (c) SS;

each suppository contained 50 mg DcNa and 1-5 %w/w of bioadhesive

polymers (CBP, HPMC, PVP and CMCTS). Mean ± SD, n=6. Asterisks

indicate formulations which are significantly different from blank

suppositories. ................................................................................................ 83

Figure 3.6 : The softening time of suppository formulations produced using (a) CB;

(b) CE and (c) SS. Each suppository contained 50mg DcNa and 1-5 %w/w of

bioadhesive polymers (CBP, HPMC, PVP and CMCTS). Mean ± SD, n=3.

xv
 Asterisks indicate formulations which are significantly different from blank

suppositories. ................................................................................................ 86

Figure 4.1 : Cumulative percentage release of DcNa in (a) CB; (b) CE; (c) SS

suppositories containing 25, 50 and 75 mg of DcNa. Mean ± 2 SE, n=6. .. 104

Figure 4.2 : DcNa release profiles from suppositories made with different bases, each

containing 50 mg of DcNa. Mean ± 2 SE, n=6. .......................................... 106

Figure 4.3 : The DE of suppository formulations containing 50 mg DcNa and 1-

5%w/w bioadhesive polymers (CBP, HPMC, PVP, CMCTS) in (a) CB; (b)

CE; and (c) SS. Asterisks indicate formulations which are significantly

different from formulations without polymer (DcNa only). Mean ± SD, n=6.

.................................................................................................................... 112

Figure 4.4 : The MDT of suppository formulations containing 50 mg DcNa and 1-

5 %w/w bioadhesive polymers (CBP, HPMC, PVP, CMCTS) made from (a)

CB; (b) CE; and (c) SS. Asterisks indicate formulations which are

significantly different from formulations without polymer (DcNa only).

Mean ± SD, n=6. ......................................................................................... 113

Figure 5.1 : Experimental setup for testing tensile stress of bioadhesion using texture

analyser; (a) without temperature control (Thirawong et al., 2007; Tobyn et

al., 1995; Wong et al., 1999a) and; (b) with temperature control (Thirawong

et al., 2007). ................................................................................................ 129

Figure 5.2 : Experimental setups used to investigate the shear stress of the bioadhesion

joint between bioadhesive material and membrane; (a) modified Dia-Stron

rheometer equipped with pulley system (Mortazavi and Smart, 1995); (b)

dual modified tensiometer method (Leung and Robinson, 1988) and; (c)

vertical rod coupled to tensiometer (Jiménez-Castellanos et al., 1993). .... 131

xvi
Figure 5.3: Different segments of freshly excised porcine large intestinal tissues used

in this study, (a) intact large intestines with serosa, tunica muscularis and

submucosa layer; (b) split large intestines, with mucosa facing upwards. . 134

Figure 5.4 : The tensile bioadhesion study experimental setup using texture analyser

with heated copper membrane stage. .......................................................... 136

Figure 5.5 : The shear bioadhesion study experimental setup using texture analyser

with heated copper membrane stage. .......................................................... 137

Figure 5.6: A typical plot of force versus distance data for suppository sample disc

(CB + 50 mg DcNa + 1 %w/w PVP) tested with colon mucosa using the

tensile setup................................................................................................. 142

Figure 5.7: Effect of contact time on (a) Fmax and (b) Wad of SS discs containing 50

mg DcNa and 5 %w/w CBP against porcine colon mucosa using the tensile

setup. Mean ± SD, n=5-6. ........................................................................... 144

Figure 5.8: Effect of probe withdrawal speed on (a) Fmax and (b) Wad of SS discs

containing 50 mg DcNa and 5 %w/w CBP against porcine colon mucosa

using tensile setup. Mean ± SD, n=5-6. ...................................................... 145

Figure 5.9 : Effect of contact force on (a) Fmax and (b) Wad of SS discs containing 50

mg DcNa and 5 %w/w CBP against porcine colon mucosa using tensile

setup. Mean ± SD, n=5-6. ........................................................................... 146

Figure 5.10 : Effect of volume of SRM on (a) Fmax and (b) Wad of SS discs containing

50 mg DcNa and 5 %w/w CBP against the porcine colon mucosa using the

tensile setup. Mean ± SD, n=5-6................................................................. 147

Figure 5.11 : Effect of different segments (rectum and colon) of the porcine large

intestines on (a) Fmax and (b) Wad of SS discs containing 50 mg DcNa and 2–

xvii
 5 %w/w of CMCTS or CBP using the tensile setup. Values expressed as

mean ± SD, n=5-6. ...................................................................................... 148

Figure 5.12 : Effect of contact time on (a) Fmax and (b) Wad of SS discs containing 50

mg DcNa and 5 %w/w PVP against the porcine colon mucosa using the

shear setup. Mean ± SD, n=5-6................................................................... 149

Figure 5.13 : Effect of probe withdrawal speed on (a) Fmax and (b) Wad of SS discs

containing 50 mg DcNa and 5 %w/w PVP against porcine colon mucosa

using the shear setup. Mean ± SD, n=5-6. .................................................. 150

Figure 5.14 : Effect of contact force on (a) Fmax and (b) Wad of SS discs containing 50

mg DcNa and 5 %w/w PVP against the porcine colon mucosa using the

shear setup. Mean ± SD, n=5-6................................................................... 151

Figure 5.15 : Effect of volume of SRM on (a) Fmax and (b) Wad of SS discs containing

50 mg DcNa and 5 %w/w PVP against the porcine colon mucosa using the

shear setup. Mean ± SD, n=5-6................................................................... 152

Figure 5.16 : Effect of different segments (rectum and colon) of the porcine large

intestines on (a) Fmax and (b) Wad of SS discs containing 50 mg DcNa and 2–

5 %w/w of CBP or PVP using the shear setup. Mean ± SD, n=5-6. .......... 153

Figure 5.17 : The Fmax of (a) CB; (b) CE and (c) SS formulations containing 50 mg

DcNa and 1–5 %w/w of bioadhesive polymer (CBP, HPMC, PVP, CMCTS)

using tensile setup. Asterisks indicate Fmax values which are significantly

different from formulations without bioadhesive polymers. Mean ± SD,

n=5–6. ......................................................................................................... 155

Figure 5.18 : The Fmax of (a) CB; (b) CE and (c) SS formulations containing 50 mg

DcNa and 1–5 %w/w of bioadhesive polymer (CBP, HPMC, PVP, CMCTS)

using shear setup. Asterisks indicate Fmax values which are significantly

xviii
 different from formulations without bioadhesive polymers. Mean ± SD, n=5-

6. ................................................................................................................. 157

Figure 5.19 : The Fmax of the SS formulations containing 50 mg DcNa and 0-5 %w/w

of bioadhesive polymer (CBP, HPMC, PVP, CMCTS) tested with (a) colon

mucosa as biological membrane and (b) synthetic membrane using the

tensile experimental setup. Mean ± SD, n=5-6. ......................................... 160

Figure 5.20 : The Fmax of the SS formulations containing 50 mg DcNa and 0-5 %w/w

of bioadhesive polymer (CBP, HPMC, PVP, CMCTS) tested with (a) colon

mucosa as biological membrane and (b) synthetic membrane using the shear

experimental setup. Mean ± SD, n=5-6. ..................................................... 162

Figure 6.1 : The DSC thermogram of CB suppositories containing 50 mg DcNa and

5 %w/w PVP. Individual thermograms show the melting endotherm of

suppositories which were (i) freshly prepared; stored refrigerated at for (ii)

100 days and (iii) 200 days; stored at room temperature for (iv) 100 days and

(v) 200 days. Inset shows enlarged portions of the thermogram. ............... 172

Figure 6.2 : The DSC thermogram of CB suppositories containing 50 mg DcNa and

5 %w/w CMCTS. Individual thermograms show the melting endotherm of

suppositories which were (i) freshly prepared; stored refrigerated at for (ii)

100 days and (iii) 200 days; stored at room temperature for (iv) 100 days and

(v) 200 days. Inset shows enlarged portions of the thermogram. ............... 173

Figure 6.3 : The SFC of CB suppositories containing 50 mg DcNa and (a) 5 %w/w

PVP and (b) 5 %w/w CMCTS. ................................................................... 174

Figure 6.4 : The DSC thermogram of suppositories made using CE as suppository

base containing 5 %w/w PVP. Individual thermograms show the melting

endotherm of suppositories which were (i) freshly prepared; stored

xix
 refrigerated at for (ii) 100 days and (iii) 200 days; stored at room

temperature for (iv) 100 days and (v) 200 days. Inset shows enlarged

portions of the thermogram......................................................................... 177

Figure 6.5 : The DSC thermogram of CE suppositories containing 50 mg DcNa and

5 %w/w CMCTS. Individual thermograms show the melting endotherm of

suppositories which were (i) freshly prepared; stored refrigerated at for (ii)

100 days and (iii) 200 days; stored at room temperature for (iv) 100 days and

(v) 200 days. Inset shows enlarged portions of the thermogram. ............... 178

Figure 6.6 : The DSC thermogram of SS suppositories containing 50 mg DcNa and

5 %w/w PVP. Individual thermograms show the melting endotherm of

suppositories which were (i) freshly prepared; stored refrigerated for (ii) 100

days and (iii) 200 days; stored at room temperature for (iv) 100 days and (v)

200 days. Inset shows enlarged portions of the thermogram. ..................... 179

Figure 6.7 : The DSC thermogram of SS suppositories containing 50 mg DcNa and

5 %w/w CMCTS. Individual thermograms show the melting endotherm of

suppositories which were (i) freshly prepared; stored refrigerated for (ii) 100

days and (iii) 200 days; stored at room temperature for (iv) 100 days and (v)

200 days. Inset shows enlarged portions of the thermogram. ..................... 180

Figure 6.8 : The SFC of CE suppositories containing 50 mg DcNa and (a) 5 %w/w

PVP and (b) 5 %w/w CMCTS. ................................................................... 182

Figure 6.9 : The SFC of SS suppositories containing 50 mg DcNa and (a) 5 %w/w

PVP and (b) 5 %w/w CMCTS. ................................................................... 183

Figure 6.10 : The hardness of suppositories made using (a) CB; (b) CE and (c) SS

containing 50 mg DcNa and 5 %w/w PVP and 5 %w/w CMCTS after

various storage conditions up to 200 days. Asterisks indicate a significant

xx
 difference in hardness from ‘freshly prepared’ suppositories. Mean ± SD,

n=6. ............................................................................................................. 184

Figure 6.11 : The softening time of suppositories made using (a) CB; (b) CE and (c)

SS containing 50 mg DcNa and 5 %w/w PVP and 5 %w/w CMCTS under

different storage conditions. Asterisks indicate a significant difference in

softening time compared to ‘freshly prepared’ suppositories. Mean ± SD,

n=6. ............................................................................................................. 186

Figure 6.12: The cumulative DcNa release from (a) CB; (b) CE and (c) SS

suppositories containing 50 mg DcNa and 5 %w/w PVP under different

storage conditions and duration. Mean ± 2 SE, n=3. .................................. 188

Figure 6.13: The cumulative DcNa release from (a) CB; (b) CE and (c) SS

suppositories containing 50 mg DcNa and 5 %w/w CMCTS under different

storage conditions and duration. Mean ± 2 SE, n=3. ................................. 189

xxi
LIST OF TABLES
Table 1.1: The list of commercial suppositories for local action registered with the

Drug Control Authority in Malaysia. .............................................................. 3

Table 1.2 : The list of commercial suppositories for systemic action registered with the

Drug Control Authority in Malaysia. .............................................................. 4

Table 1.3 : The composition of the major fatty acids in CB and hydrogenated palm

kernel stearin (HPKS). .................................................................................... 9

Table 1.4: Summary of factors affecting rectal absorption of drugs (Allen et al., 2008).

...................................................................................................................... 16

Table 1.5: Factors affecting bioadhesion (Ahuja et al., 1997; Andrews et al., 2009).. 21

Table 1.6: Some of the bioadhesive polymers used in pharmaceutical dosage forms

(Irons and Robinson, 2003; Lehr and Bouwstra, 1992). ............................... 23

Table 2.1 : Reported polymorphic forms of CB and the variability in their

nomenclature. Table includes nomenclature used to describe CB polymorphs

in this thesis. ................................................................................................. 41

Table 2.2 : The reported polymorphic forms of lauric fats, PKO blends, HPKS and the

variability in their nomenclature. Table includes nomenclature used to

describe CE and SS polymorphs in this thesis. ............................................. 42

Table 2.3 : Qualitative assessment of CB and SS molten properties, ease of

manufacturing process and visual appearance of compounded suppositories

subjected to various Tmax, as determined by lab-scale extemporaneous

manufacturing process. ................................................................................ 62

Table 3.1 : The target weight, actual weight, melting point and DcNa content of CB

suppository formulations containing 50 mg DcNa and bioadhesive polymers.

Mean ± SD. ................................................................................................... 77

xxii
Table 3.2 : The target weight, actual weight, melting point and DcNa content of CE

suppository formulations containing 50 mg DcNa and bioadhesive polymers.

Mean ± SD. ................................................................................................... 78

Table 3.3 : The target weight, actual weight, melting point and DcNa content of SS

suppository formulations containing 50 mg DcNa and bioadhesive polymers.

Mean ± SD. ................................................................................................... 79

Table 4.1 : Summary of parameters used for in vitro drug release studies. ............... 100

Table 4.2 : Statistical comparison of formulations containing DcNa only (Rt) and

formulations with DcNa and bioadhesive polymers (Tt). Dissolution profiles

are considered similar when; 0 < ƒ1 < 15 and 50 < ƒ2 < 100. The highlighted

ƒ1 and ƒ2 values indicate similar dissolution profiles between Rt and Tt.... 109

Table 4.3 : The goodness of fit parameters obtained from equation fitting of drug

release data from all CB formulations. ....................................................... 116

Table 4.4 : The goodness of fit parameters obtained from equation fitting of drug

release data from all CE formulations. ....................................................... 117

Table 4.5 : The goodness of fit parameters obtained from equation fitting of drug

release data from all SS formulations. ........................................................ 118

Table 4.6 : The release constant (km) and release exponent (n) of formulations fitted to

Korsmeyer- Peppas model. ......................................................................... 120

Table 4.7 : The release parameters of formulations containing CBP fitted with Weibull

equation. ...................................................................................................... 122

Table 4.8 : The release parameters of formulations containing 1-5 %w/w CBP fitted

with bi-exponential first-order kinetic equation. All the r2 values for initial

and terminal phases were > 0.95 when goodness of fit of the data was

reviewed by linear regression. .................................................................... 123

xxiii
Table 5.1 : The fixed and variable parameters used for tensile force optimisation using

sample discs containing 50 mg DcNa and 5 %w/w CBP. .......................... 140

Table 5.2 : The fixed and variable parameters used for shear force optimisation using

sample discs containing 50 mg DcNa and 5 %w/w PVP. .......................... 141

Table 5.3 : The yield of biological membrane mucosa. ............................................. 143

Table 6.1: The physical appearance of PVP and CMCTS suppositories containing 50

mg DcNa after storage at various conditions up to 200 days. .................... 170

Table 6.2 : The melting points of various polymorphic forms of CB TAG. (Fatty acid

nomenclature under TAG column as follows: P = palmitic acid; O = oleic

acid; S = stearic acid) .................................................................................. 175

Table 6.3: The melting points of various polymorphic forms of saturated TAG. ..... 176

Table 6.4 : The DE (%) of suppositories made using CB, CE and SS containing 50 mg

DcNa and 5 %w/w bioadhesive polymer (PVP, CMCTS). Asterisks indicate

a significant difference in DE compared to ‘freshly prepared’ suppositories.

Mean ± SD, n=3. ......................................................................................... 191

Table 6.5 : The MDT (minutes) of suppositories made using CB, CE and SS

containing 50 mg DcNa and 5 %w/w bioadhesive polymer (PVP, CMCTS).

Asterisks indicate a significant difference in MDT compared to ‘freshly

prepared’ suppositories. Mean ± SD, n=3. ................................................. 192

xxiv
LIST OF APPENDICES
Appendix 1: Certificate of analysis of cocoa butter (CB). ......................................... 224

Appendix 2 : Certificate of analysis for Chocexa (CE). ............................................ 225

Appendix 3 : Certificate of analysis for Supersocolate SpecialTM (SS)..................... 227

Appendix 4 : Certificate of analysis for diclofenac sodium (DcNa).......................... 228

Appendix 5 : Certificate of analysis for Carbopol 974P NF (CBP)........................... 229

Appendix 6 : Certificate of analysis of hydroxypropyl methylcellulose 2910 (HPMC).

.................................................................................................................... 230

Appendix 7 : Certificate of analysis for carboxymethyl chitosan (CMCTS). ........... 231

Appendix 8 : The viscosity (cp) of CB suppositories measured at 50 rpm shear rate.

Outcomes were a result of ANOVA and post hoc Tukey’s HSD analysis.

Mean ± SD, n=3. ......................................................................................... 232

Appendix 9 : The viscosity (cp) of CE suppositories measured at 50 rpm shear rate.

Outcomes were a result of ANOVA and post hoc Tukey’s HSD analysis.

Mean ± SD, n=3. ......................................................................................... 233

Appendix 10 : The viscosity (cp) of SS suppositories measured at 50 rpm shear rate.

Outcomes were a result of ANOVA and post hoc Tukey’s HSD analysis.

Mean ± SD, n=3. ......................................................................................... 234

Appendix 11 : The hardness (N) of CB suppositories. Outcomes were a result of

ANOVA and post hoc Tukey’s HSD analysis. Mean ± SD, n=6. .............. 235

Appendix 12 : The hardness (N) of CE suppositories. Outcomes were a result of

ANOVA and post hoc Tukey’s HSD analysis. Mean ± SD, n=6. .............. 236

Appendix 13 : The hardness (N) of SS suppositories. Outcomes were a result of

ANOVA and post hoc Tukey’s HSD analysis. Mean ± SD, n=6. .............. 237

xxv
Appendix 14 : The softening time (min) of CB suppositories. Outcomes were a result

of ANOVA and post hoc Tukey’s HSD analysis. Mean ± SD, n=3. .......... 238

Appendix 15 : The softening time (min) of CE suppositories. Outcomes were a result

of ANOVA and post hoc Tukey’s HSD analysis. Mean ± SD, n=3. .......... 239

Appendix 16 : The softening time (min) of SS suppositories. Outcomes were a result

of ANOVA and post hoc Tukey’s HSD analysis. Mean ± SD, n=3. .......... 240

Appendix 17: Amount of DcNa (mg) released at each time interval in (a) CB; (b) CE;

and (c) SS suppositories containing 25, 50, 75 mg of DcNa. Mean ± 2 SE,

n=6. ............................................................................................................. 241

Appendix 18: Cumulative percentage DcNa release in a) CB; b) CE; c) SS

suppositories incorporated with 1-5 %w/w of CBP. Mean ± 2 SE, n=6. ... 242

Appendix 19 : Cumulative percentage of DcNa release in a) CB; b) CE; c) SS

suppositories incorporated with 1-5 %w/w of HPMC. Mean ± 2 SE, n= 6.

.................................................................................................................... 243

Appendix 20 : Cumulative percentage of DcNa release in a) CB; b) CE; c) SS

suppositories incorporated with 1-5%w/w PVP. Mean ± 2 SE, n=6. ......... 244

Appendix 21 : Cumulative percentage of DcNa release in a) CB; b) CE; c) SS

suppositories incorporated with 1-5%w/w CMCTS. Mean ± 2 SE, n=6. ... 245

Appendix 22 : The dissolution efficiency (DE) of CB suppositories. Outcomes were a

result of ANOVA and post hoc Tukey’s HSD analysis. Mean ± SD, n=6. 246

Appendix 23 : The dissolution efficiency (DE) of CE suppositories. Outcomes were a

result of ANOVA and post hoc Tukey’s HSD analysis. Mean ± SD, n=6. 247

Appendix 24 : The dissolution efficiency (DE) of SS suppositories. Outcomes were a

result of ANOVA and post hoc Tukey’s HSD analysis. Mean ± SD, n=6. 248

xxvi
Appendix 25 : The min dissolution time (MDT) of CB suppositories. Outcomes were

a result of ANOVA and post hoc Tukey’s HSD analysis. Mean ± SD, n=6.

.................................................................................................................... 249

Appendix 26 : The min dissolution time (MDT) of CE suppositories. Outcomes were a

result of ANOVA and post hoc Tukey’s HSD analysis. Mean ± SD, n=6. 250

Appendix 27 : The min dissolution time (MDT) of SS suppositories. Outcomes were a

result of ANOVA and post hoc Tukey’s HSD analysis. Mean ± SD, n=6. 251

Appendix 28 : The peak force of detachment (Fmax) of CB suppositories measured

using tensile setup against colon mucosa. Outcomes were a result of

ANOVA and post hoc Tukey’s HSD analysis. Mean ± SD, n=5-6. ......... 252

Appendix 29 : The peak force of detachment (Fmax) of CE suppositories measured

using tensile setup against colon mucosa. Outcomes were a result of

ANOVA and post hoc Tukey’s HSD analysis. Mean ± SD, n=5-6. .......... 253

Appendix 30 : The peak force of detachment (Fmax) of SS suppositories measured

using tensile setup against colon mucosa. Outcomes were a result of

ANOVA and post hoc Tukey’s HSD analysis. Mean ± SD, n=5-6. .......... 254

Appendix 31 : The peak force of detachment (Fmax) of CB suppositories measured

using shear setup against colon mucosa. Outcomes were a result of ANOVA

and post hoc Tukey’s HSD analysis. Mean ± SD, n=5-6. ......................... 255

Appendix 32 : The peak force of detachment (Fmax) of CE suppositories measured

using shear setup against colon mucosa. Outcomes were a result of ANOVA

and post hoc Tukey’s HSD analysis. Mean ± SD, n=5-6. ......................... 256

Appendix 33 : The peak force of detachment (Fmax) of SS suppositories measured

using shear setup against colon mucosa. Outcomes were a result of ANOVA

and post hoc Tukey’s HSD analysis. Mean ± SD, n=5-6. ......................... 257

xxvii
Appendix 34 : The peak force of detachment (Fmax) of SS suppositories measured

using tensile setup against colon mucosa and synthetic regenerated cellulose.

Outcomes were a result of ANOVA and post hoc Tukey’s HSD analysis.

Mean ± SD, n=5-6. ..................................................................................... 258

Appendix 35 : The peak force of detachment (Fmax) of SS suppositories measured

using shear setup against colon mucosa and synthetic regenerated cellulose.

Outcomes were a result of ANOVA and post hoc Tukey’s HSD analysis.

Mean ± SD, n=5-6. ..................................................................................... 259

Appendix 36 : Hardness values (N) of suppositories subjected to various storage

condition and duration. Outcomes were a result of ANOVA and post hoc

Tukey’s HSD analysis. Mean ± SD, n=6. .................................................. 260

Appendix 37 : Softening time (min) of suppositories subjected to various storage

condition and duration. Outcomes were a result of ANOVA and post hoc

Tukey’s HSD analysis. Mean ± SD, n=3. .................................................. 261

xxviii
LIST OF ABBREVIATIONS
ANOVA Analysis of variance

AIC Akaike information criterion

CB Cocoa butter

CBP Carbopol 974P NF

CBS Cocoa butter substitute

CE Chocexa

CMCTS Carboxymethyl chitosan

DcNa Diclofenac sodium

DE Dissolution efficiency

DV Displacement value

DSC Differential scanning calorimetry

ƒ1 Difference factor

ƒ2 Similarity factor

Fmax Peak force of detachment

FTIR Fourier transform infrared

GIT Gastrointestinal tract

HPKS Hydrogenated palm kernel stearin

HPMC Hydroxypropyl methylcellulose

HSD Honestly significant test

IM Intramuscular

IV Intravenous

KBr Potassium bromide

MDT Mean dissolution time

xxix
MPOB Malaysian Palm Oil Board

MSE Mean square error

MW Molecular weight

NSAID Nonsteroidal anti-inflammatory drug

PEG Polyethylene glycol

PKO Palm kernel oil

PKS Palm kernel stearin

p-NMR Pulsed nuclear magnetic resonance

PVP Poly(vinylpyrrolidone)

r2 Coefficient of determination

RH Relative humidity

Rt Reference formulation

SCF Simulated colonic fluid

SD Standard deviation

SE Standard error

SFC Solid fat content

SOS Stearic acid-oleic acid-stearic acid

SRM Simulated rectal mucus

SS Supersocolate SpecialTM

SSR Sum of squares residues

TAG Triacylglycerol

TGA Thermogravimetric analysis

Tt Test formulation

USP United States Pharmacopoeia

UV Ultraviolet

xxx
Wad Work of adhesion

%w/v Percentage weight per volume

%w/w Percentage weight per weight

xxxi
CHAPTER 1

GENERAL INTRODUCTION

1
1.1 Introduction

Suppositories are solid dosage forms intended for administration to the human body

via insertion into body orifices, mainly the rectum; where it softens, melts or dissolves

to release the incorporated medication which then exerts its therapeutic effects locally

or systemically. Suppositories can also be administered via the urethra or the vagina

(pessaries) (Allen et al., 2008).

Rectal suppositories are usually cylindrical with either one or two tapered ends,

shaped like a bullet or a torpedo. Length can be up to 32 mm and it usually weighs

about 1–2 g (Allen et al., 2005; Ansel, 1981). The various shapes and sizes of

suppositories are shown in Figure 1.1.

Figure 1.1: Shapes and sizes of rectal, vaginal and urethral suppositories (University

of North Carolina Eshelman School of Pharmacy, 2015).

Although less common nowadays, suppositories are a relatively old method of

administering medications to the human body; dating back to as far as the ancient

Egyptian civilization; as evidenced in the Ebers Papyrus scriptures from 1550 BC

(Bryan and Smith, 1930).

2
These days, suppositories are mainly employed as locally-acting laxatives (Table 1.1)

to promote defecation or to treat anorectal diseases such as haemorrhoids and

ulcerative colitis (Cooper and Gunn, 1987). Nonetheless, there is also a substantial

amount of commercial suppository formulations marketed for systemic delivery in

Malaysia (Table 1.2). Drugs prescribed as a suppository for systemic treatment

include analgesics, antibiotics, tranquilizers and antihistamines.

Table 1.1: The list of commercial suppositories for local action registered with the

Drug Control Authority in Malaysia.

Product Name Active Ingredient Strengths

Pentasa 5- aminosalicylic acid, mesalazine 1g

Salofalk 5- aminosalicylic acid, mesalazine 250, 500 mg

lignocaine, hydrocortisone acetate, zinc -

Xyloproct
oxide, aluminium subacetate

Bisacodyl bisacodyl 5, 10 mg

Pricolax bisacodyl 5, 10 mg

Dulcolax bisacodyl 5, 10 mg

hydrocortisone, framycetin sulphate, -

Proctosedyl
aesculin, cinchocaine

Liproct hydrocortisone acetate, zinc oxide, lidocaine -

zinc oxide, hydrocortisone acetate, -

Doproct
benzocaine

3
Table 1.2 : The list of commercial suppositories for systemic action registered with the

Drug Control Authority in Malaysia.

Product Name Active Ingredient Strengths

Primperan metoclopramide hydrochloride 10, 20 mg

Tramadol Stada tramadol hydrochloride 100 mg

Remedol paracetamol 125, 250 mg

Arfen paracetamol 125, 250 mg

Tempol paracetamol 125, 250 mg

Pritamol paracetamol 125, 250 mg

Poro paracetamol 125, 250 mg

Shoren diclofenac sodium 12.5, 25 mg

Dicloren diclofenac sodium 12.5 mg

Profenac diclofenac sodium 12.5 mg

Voltaren diclofenac sodium 12.5, 25, 50 mg

Almiral diclofenac sodium 100 mg

Diclogesic diclofenac sodium 12.5, 100 mg

Voren diclofenac sodium 12.5, 25, 50, 100 mg

Pritaren diclofenac sodium 12.5 mg

1.1.1 Advantages of suppositories

Oral drug delivery remains the most common route of drug administration and has the

highest rate of patient acceptance. However, the oral route of administering

medications may not always be the best option. For example, drugs which undergo

extensive first-pass metabolism or pre-systemic degradation such as lidocaine (De

Leede et al., 1983), diclofenac sodium (DcNa) (Menasse et al., 1978; Willis et al.,

4
1979) and salbutamol (Goldstein et al., 1987; Morgan et al., 1986). Conversely, drugs

administered to the lower rectum would largely be absorbed into the systemic

circulation, thus circumventing the hepatic first-pass metabolism to afford greater

bioavailability (Allen et al., 2008; Kokate et al., 2006; Watanabe, 2007). For drugs

with unpleasant taste and odour, such as cysteamine (a new treatment for nephropathic

cystinosis), suppositories may be an alternative route for drug administration (Buchan,

2011).

Rectal administration of drugs is also particularly useful when the oral route is

occluded or disrupted by nausea, vomiting or during acute convulsions (Allen et al.,

2005). Nausea and vomiting which limit oral intake of medications are common

symptoms in patients undergoing chemotherapy or in palliative care. Hence,

alternative routes such as rectal or transdermal drug delivery would be beneficial for

these patients (Allen et al., 2005; Davis et al., 2002; Warren, 1996).

Terminally ill and palliative care patients in the outpatient setting would often require

frequent administration of multiple analgesics. For these patients, the intravenous (IV)

and intramuscular (IM) routes are less practical as they may not always be under the

immediate care of qualified health care practitioners.

Clinically, there is also a great deal of unmet needs for alternative routes of

administering medications for patients who underwent total gastractomy, ileal

resectioning procedures, and patients inserted with nasogastric or nasojejunal tubes

where modified release oral formulations are very often rendered less effective. Hence,

5
non-peroral or transmucosa formulations such as rectal suppositories would be of

great value in the outpatient management of such patients.

Furthermore, various recent studies showed that the use of preoperative rectal

nonsteroidal anti-inflammatory drug (NSAID) have successfully delayed time before

the first request for anaesthesia; reduced the use of supplemental opioids and scored

lower on visual analogue scales in both major and minor surgeries in both adults and

paediatrics (Bahar et al., 2010; Fayaz et al., 2004).

1.1.2 Patient acceptance and social stigma

A study found that only 18 % of patients favoured suppositories over IM

postoperative analgesia (Carroll et al., 1996). Various studies showed that the main

reason for rejection of suppository was mainly due to the misconception where

patients regard its method of administration as a form of invasion or violation of

dignity, leading to subsequent humiliation (Colbert et al., 1998).

Conversely, Vyvyan and Hanafiah (1995) reported that 46 % of the middle aged

patients surveyed were receptive towards rectal drug administration; however 98 % of

these patients felt the need for discussion prior rectal administration. Meanwhile,

Bonner et al. (1996) found that only 15 % of patients aged between 15-91 years old

objected to suppository administration under anaesthesia, although 59 % of them

preferred to be informed preoperatively. Another study by Dodd et al. (2004) also

found high degree of acceptance towards suppositories in women for relief of

postnatal pain. Among younger children, Hinton et al. (2007) found that there was

considerable acceptance towards the use of suppositories as an alternative route to oral

6
dosage forms; and that there was also a 69.5 % caregiver acceptance of malarial

treatment via rectal route. These were encouraging findings which indicates

diminishing stigmatism towards the use of suppositories.

1.1.3 Suppository base

The type of suppository base used depends on the intended release profiles and nature

of the active drug. Suppository bases can be classified according to their physical

characteristics; fatty bases, water soluble bases and the emulsifying bases (Allen et al.,

2005; Cooper and Gunn, 1987).

1.1.3.1 Fatty bases

Suppositories made of fatty (oleaginous) bases must melt upon administration into the

rectum before the drug partitions into rectal fluids for absorption across rectal

membranes into the systemic circulation (Allen et al., 2008). This group of bases

include cocoa butter (CB), palm oil, palm kernel oil (PKO) and cottonseed oil or fat-

based glycerine compounds containing high molecular weight (MW) fatty acids such

as glyceryl monostearate (Allen et al., 2005). Fatty bases contain very little water and

has low tendency of hygroscopicity. Commercialised bases nowadays are usually a

combination of two or more fatty bases, for example Wecobee® which is derived from

fully hardened palm kernel and cottonseed oils (Stepan Specialty Product, 2014).

1.1.3.1.1 Cocoa butter (CB)

The traditional base, CB or theobroma oil is obtained from the roasted seeds of

theobroma cacao and presents itself as a yellowish-white solid at room temperature

7
which melts at 30–36 °C (Allen et al., 2005). CB is desirable due to its melting range,

non-irritant nature and miscibility with a wide range of medicaments. However, it

exhibits both rancidity and polymorphism on storage. The fatty acid composition of

CB is shown in Table 1.3.

It is generally accepted that CB exist in 4 different polymorphs, namely α, β, β’ and γ

forms although some literatures suggested up to a number of 6 distinct polymorphs

(Loisel et al., 1998; Marangoni and McGauley, 2003; van Langevelde et al., 2001;

Wille and Lutton, 1966). Each of the polymorphic forms exhibit different melting

ranges with β form being most stable. The presence of metastable polymorphs with

lower melting points are not conducive for suppositories, especially in warm tropical

climates (Allen et al., 2008). Other disadvantages of CB include its inability to

contract and detach from suppository moulds on cooling, thus necessitating the

lubrication of moulds with liquid paraffin to aid suppository removal (Cooper and

Gunn, 1987). There is also substantial batch to batch variation since CB is sourced

naturally and the fatty acid content was found to be affected by geographical origin of

the CB (Chaiseri and Dimick, 1989; Spangenberg and Dionisi, 2001).

These disadvantages of CB suppository bases prompted the development of newer

commercial bases with specific set of properties to overcome formulation difficulties.

8
Table 1.3 : The composition of the major fatty acids in CB and hydrogenated palm kernel stearin (HPKS).

Weight (%)
Fatty acid CB HPKS
(carbon no : Peyronel and
Spangenberg and Lonchampt and Toro-Vazquez et 2
double bonds) Rossell (1975) Siew (2001) Marangoni
Dionisi (2001)1 Hartel (2004) al. (2004)
(2014)
Lauric (12:0) 49.6 56.6 43.3
Myristic (14:0) 0.09 0.1 30.4 22.0 28.8
Palmitic (16:0) 25.1 26.8 25.8 11.5 7.9 12.6
Stearic (18:0) 37.4 35.6 34.5 2.8 8.6 14.2
Oleic (18:1) 33.0 33.5 34.9 2.4 0.2
Linoleic (18:2) 2.4 3.2 3.0
Arachidic (20:0) 1.1 0.9 1.0 0.3

1
Values quoted for sample CB-28, deodorized CB originated from Malaysia.
2
Values quoted for HPKS (Iodine value =1.8).
9
1.1.3.1.2 Palm kernel oil (PKO) and hydrogenated palm kernel stearin (HPKS)

Oil palm (Elaeis guineensis) is one of the richest vegetable oil plants and is widely

used in the food industry. PKO is produced as a by-product via extraction of the

residual kernels (Akinoso and Raji, 2011; Pantzaris and Ahmad, 2002; Zhou et al.,

2010).

PKO was found to contain 81.67 % of saturated fatty acids; mainly the short-chain

fatty acids, such as lauric (C12) and myristic (C14) acid. It has a slip melting point of

27–29 °C and iodine value of 16-20 (Goh, 1994). However, the slip melting point of

the PKO can be altered via hydrogenation or blending with other palm oil products

(Goh, 1994; Pantzaris and Ahmad, 2002). Further hydrogenation of PKO produces

hydrogenated palm kernel stearin (HPKS) with a melting point of 32.5-34.5 °C (Siew

and Ng, 2000; Siew, 2001). The common composition for HPKS is shown in Table

1.3.

Recently, Noordin and Chung (2007) developed two new suppository bases using

combinations of locally sources hydrogenated PKO, HPKS and hydrogenated palm

kernel olein with mixtures of stearic acid and glyceryl monostearate. The authors

found that the bioavailability of aspirin administered rectally in these bases were

superior to the equivalent dose administered orally.

1.1.3.2 Water soluble and water miscible bases

Unlike fatty bases, water soluble bases disintegrate and dissolve in rectal fluids upon

insertion into the human rectum (Allen et al., 2008). Since the rectum has very small

amount of fluid, complete vehicle dissolution can be difficult and water will be

10
attracted from rectal tissues towards the suppository via osmotic effect causing pain to

the site of administration. Polyethylene glycol (PEG), poloxamer and glycerinated

gelatin are among the common water soluble bases used to produce suppositories.

PEG are polymers made up of ethylene oxide and water, produced in various chain

lengths, MW and physical states. It is possible to formulate PEG bases with desired

consistency and characteristics by combining different grades of PEGs via fusion

process (Allen et al., 2008).

Meanwhile, glycerinated gelatin base could be easily prepared by dissolving 20 %

granular gelatin in 70 % of glycerine, added with 10 % of solution or suspension of

the desired drug. The resulting base is hygroscopic in nature and could potentially

irritate the rectal surface; thus requiring moistening by dipping into water prior to

insertion into the rectum (Allen et al., 2008).

Poloxamers on the other hand, are odourless, tasteless, water soluble, block co-

polymers which exhibit reverse-thermal gelation properties; they remain liquid at

room temperature and undergo phase transition to gel at body temperature (Choi et al.,

1998; Keny and Lourenco, 2010). Various studies investigated formulations of

thermogelling bioadhesive suppositories using poloxamer in attempts to eliminate

discomfort caused by insertion of a conventional solid suppository into the rectum as

well as to localise drug absorption within the lower rectum (Barakat, 2009; Choi et al.,

1998; Keny and Lourenco, 2010).

11
1.1.3.3 Emulsifying bases

These include mixtures of oleaginous bases and water miscible materials,

disintegrating agents, collagen and natural gums. When formulated as suppositories, it

disperses in rectal fluid to form oil-in-water emulsions due to its surface active

properties and spreads as a smooth layer over mucous membranes (Allen et al., 2008).

As an example, the Witepsol® series composed of triacylglycerols (TAG) of saturated

C12-18 fatty acids with varied portions of partial glycerides and fatty bases which

contain the TAG from palm, palm kernel and coconut oils with self-emulsifying

glyceryl monostearate and polyoxyl stearate (Cremer Oleo GmBH & Co. KG).

1.1.4 Ideal base characteristics

Allen et al. (2008) and Cooper and Gunn (1987) summarized that an ideal suppository

base should have the following qualities: (1) melt at body temperature or dissolve in

body fluids; (2) readily release medicaments; (3) physically and chemically stable; (4)

nontoxic, non-irritating and non-sensitizing; (5) compatible with a large variety of

drugs; (6) chemically and physiologically inert; (7) contract slightly on cooling; (8)

easy to manufacture by fusion, compression and extrusion.

1.1.5 Manufacturing of suppositories

Suppositories can be manufactured via a number of methods, namely hand rolling and

shaping, cold-compression and fusion moulding. The method of choice greatly

depends on nature of incorporated drug and the scale of manufacturing process as it

will not be practical to produce large quantities of suppositories via hand-rolling and

shaping method (Allen et al., 2008).

12
1.1.5.1 Hand rolling

Grated CB and all other required ingredients for the suppository are triturated

manually in a mortar to form a plastic-like mass. The mass is then quickly formed into

a ball using palms previously cooled in ice water and rolled into a cylinder using a

broad bladed spatula over a pill tile. The formed cylindrical mass can then be cut into

desired lengths and then shaped as desired by hand (Allen et al., 2008; Ansel, 1981).

1.1.5.2 Cold compression method

In this method, the active drug, suppository base and excipients are blended

thoroughly and pulverised to form a uniform blend of mixture which then softens into

a paste-like consistency due to the friction of the mixing process (Allen et al., 2008;

Ansel, 1981). The paste is then extruded into a mould and compressed for shape

setting, the resultant suppositories are then forced out of the mould orifice.

A similar method produces suppositories with uniform circumference by extruding the

paste through a perforated plate and cutting the extruded mass into the desired length

(Ansel, 1981). This method is suitable for incorporation of thermolabile drugs as the

process involves minimal heat exposure. It also enables the incorporation of large

amount of drugs that are insoluble in the base as it is unlikely for the insoluble

material to settle or separate from the suppository base. Such phenomenon is

commonly observed in suppositories produced via fusion moulding method.

1.1.5.3 Fusion moulding method

Fusion moulding method is by far the most commonly employed method to

manufacture suppositories as it is compatible with most conventional bases. The

13
process involves base melting and subsequent incorporation of drug and other

excipients into the molten base and pouring the melt into moulds for solidification.

The overfilled (excess) base is then scraped off using a warmed spatula to form a

smooth flat surface. The nominal capacities of the common moulds are 1, 2, 4 and 8 g

(Cooper and Gunn, 1987). This method would require prior calibration of moulds as

the densities of bases and drug are different (Allen et al., 2008; Ansel, 1981).

1.2 Human rectum

1.2.1 Anatomy

The human large intestine begins at the colon and extends to the rectum and anal canal

at the terminal end (Kokate et al., 2006). The rectum is preceded by sigmoidal colon

and ends at anal canal (Watanabe, 2007). The rectum is approximately 15-20 cm in

length, with a comparatively small surface area of approximately 200–400 cm2; while

the anal canal is the final 2.5-5 cm of the large intestines leading to the anal verge

(Barleben and Mills, 2010).

The rectal wall is made up of three layers; mucosa which composes of several layers

of cylindrical epithelial cells; submucosa; tunica muscularis and the visceral

peritoneum (Allen et al., 2008). There are three rectal valves in the rectal ampulla -

superior, middle and inferior rectal valves. The rectum is usually non-motile and has

no villi or microvilli (Allen et al., 2008). When a suppository is administered into the

rectum, it either melts or dissolves in the rectal ampulla to release incorporated drug to

allow diffusion across the rectal mucosa and subsequent absorption into systemic

circulation (Watanabe, 2007).

14
1.2.2 Rectal mucus

The human rectum contains only 2-3 mL of inert mucus when devoid of faecal matter

(Allen et al., 2008). Mucus is a layer of viscous, gel-like secretion by goblet cells

which lines all organs of the human body such as the oculo-rhino-otolaryngeal tracts,

airways, gastrointestinal tract (GIT), and urogenital tract (Andrews et al., 2009; Bansil

and Turner, 2006). It is made up of a mixture of mucin glycoproteins, water,

electrolytes, enzymes, bacteria and sloughed epithelial cells (Irons and Robinson,

2003). The bulk of mucus content is approximately 95 % water with 0.5-5 % mucin

glycoproteins and lipids, while 0.5-1 % of the contents consist of mineral salts with

another 1 % free proteins (Edsman and Hagerstrom, 2005).

This mucus layer functions as a physical barrier to protect the internal environment

from pathogens and noxious stimuli; ensures sufficient hydration of the epithelium

surface; provides a permeable gel layer for exchange of excretion products, nutrients

and gases, and lubricates the epithelium to allow passage of objects (Bansil and

Turner, 2006; Irons and Robinson, 2003). Rectal mucus is therefore, the first barrier

against diffusion of administered drugs before absorption across the mucosa

membranes.

1.2.3 Rectal absorption

The absorptive capacity of human rectum is significantly lesser than upper GIT due to

the limited surface area and absence of microvilli compared to small intestines. Drug

absorption via rectal mucous membrane is a passive process where only lipophilic,

unionised form of drug is absorbed across the membrane (Allen et al., 2008).

15
The upper rectum is drained by superior hemorrhoidal vein directly into hepatic portal

system while the lower rectum is drained by inferior and middle haemorrhoidal veins

into the systemic circulation; bypassing first pass metabolic pathways (Kokate et al.,

2006). However, the presence of extensive anastomoses may decrease the avoidance

of first pass metabolism, although it is generally accepted that at least 50-70 % of

active ingredients administered rectally circumvents the first pass effect (Allen et al.,

2008; Watanabe, 2007).

1.2.3.1 Factors affecting rectal absorption

Since rectum is not naturally an absorptive organ, the amount of drug absorbed is

greatly influenced by various physiologic and physicochemical factors (Table 1.4).

Table 1.4: Summary of factors affecting rectal absorption of drugs (Allen et al., 2008).

Physiological factors Physicochemical factors Formulation factors

nature and form of active microsphere encapsulation of
colonic contents
drug drug
presence of permeation
circulation route physical state of drug
enhancers
rectal fluid pH and
nature of suppository base bioadhesive properties
buffering capacity
volume of rectal fluid presence of excipients

motility of rectal wall

1.2.3.1.1 Physiological factors

There is greater contact between administered suppository and rectal wall for drug

absorption to occur when the rectum is empty and devoid of faecal matter, enabling

16
greater absorption of drug than when it is distended with colonic contents. Likewise,

absorption of rectally administered drugs can also be altered by medical conditions

such as diarrhoea, colonic obstruction and tissue dehydration (Allen et al., 2005).

Since both upper and lower rectum are drained by superior and inferior haemorrhoidal

veins respectively, the position at which the suppository is retained within the rectum

could affect systemic bioavailability as drugs absorbed via inferior haemorrhoidal

veins avoid first-pass metabolism of the liver (De Leede et al., 1983).

The pH of the rectal fluid is essentially 7.2–7.4, with negligible buffering capacity

(Allen et al., 2008; Jantzen et al., 1989; McNeil et al., 1987). Hence, the form of drug

incorporated into the suppository would greatly remain chemically unchanged once it

is released from the dosage form.

1.2.3.1.2 Physicochemical factors

As with drug absorption across the gastric mucosa, only unionised, undissociated form

of drug with sufficient lipophilicity would be able to travel across the bilayer lipid

membrane structure of rectal mucosa due to the bilayer lipid membrane structure.

However, the drug also has to be sufficiently soluble in rectal fluids to partition away

from the lipophilic bases or dissolve from the hydrophilic bases prior to absorption.

The size of drug particles suspended within the suppository base can influence its rate

of dissolution in rectal fluids which then affects the rate of absorption. The smaller the

size of drug particles, the greater the surface area available for dissolution; thus a

faster absorption of drug can be expected (Cooper and Gunn, 1987).

17
Drugs which are highly soluble in the formulated suppository base tend to exhibit

slower drug release than when they are formulated in bases in which they are less

soluble (Allen et al., 2008; Ermiş and Tarimci, 1995; Ibrahim et al., 1990; Nair and

Bhargava, 1999). This was clearly demonstrated by Nair and Bhargava (1999) where a

lipophilic drug fluconazole (log P = 0.44), had highest release rates from the PEG

compared to the more lipophilic bases like Suppocire® AP, Witepsol W45 and CB.

Therefore, a general rule to optimise drug release would be to formulate hydrophilic

drugs in fatty oleaginous suppository bases and lipophilic drugs in hydrophilic bases

(Allen et al., 2008, 2005). An alternative method employed to improve drug release is

to incorporate surfactants or absorption enhancers into the suppository formulation

(Shegokar and Singh, 2010).

Conversely, when sustained release of drug from the suppository is desired, various

excipients have been employed to retard drug release from the bases via formation of

drug containing micellars using lecithin (Nishihata et al., 1985); solid-reverse-

micellar-solutions also using lecithin (Schneeweis and Müller-Goymann, 2000);

incorporation of hydrophobic polymers such as hydroxypropyl methylcellulose

phthalate HP-55 (Ohnishi et al., 1986) or waxy hydrophobic materials such as

aluminium stearate and dioctyl sodium succinate (Ahmed et al., 2000).

1.3 Bioadhesion

Adhesion is the term used to describe the bond produced by interfacial forces when a

pressure-sensitive adhesive, either a natural or synthetic polymer, comes into contact

with a surface to allow prolonged attachment of the adhesive on the contact surface.

Bioadhesion is therefore the interaction which results in the adhesion of the polymer

18
to a biological surface (Ahuja et al., 1997; Roy and Prabhakar, 2010). Various regions

of the body, particularly the GIT is lined by mucosal epithelial which is covered by a

layer of continuous mucus (Ahuja et al., 1997; Roy and Prabhakar, 2010).

1.3.1 Theories and mechanisms of bioadhesion

Although exact mechanisms of bioadhesion are not known, it is generally accepted

that it involves initial wetting and swelling of the bioadhesive polymer. This is

followed by interpenetration between polymer chains and mucosal surface, with

subsequent formation of chemical bonds between entangled chains which constitutes

the bioadhesion phenomenon. Various theories have been hypothesised to explain the

phenomenon.

a) Diffusion theory

This theory postulated that polymer chains of bioadhesive material diffuse into the

glycoprotein network and vice versa in a time-dependent manner until there is

sufficient interpenetration to form mechanical interlocking and subsequent semi-

permanent adhesive bonds, producing a networked structure which results in adhesion

(Ahuja et al., 1997; Andrews et al., 2009; Roy and Prabhakar, 2010).

b) Adsorption theory

Adhesion is a result of intermolecular forces acting between atoms in bioadhesive

polymer and mucus (Ahuja et al., 1997). Both primary and secondary forces were

thought to be involved although adhesion is mainly due to secondary forces such as

electrostatic forces, van der Waals forces, hydrogen bonds and hydrophobic

interactions (Ahuja et al., 1997; Andrews et al., 2009; Roy and Prabhakar, 2010).

19
c) Electronic theory

This theory generally stems from the fact that adhesive polymers and mucus typically

have different electronic characteristics (Lee et al., 2000). Electron transfer happens

when adhesive polymer comes into close contact with the glycoprotein mucus

network, forming an electrical double layer. The attractive forces across this double

layer results in adhesion (Ahuja et al., 1997; Roy and Prabhakar, 2010).

d) Wetting theory

This theory is applicable for liquids or bioadhesive systems with low viscosity, where

the adhesive component would penetrate surface irregularities, harden and anchor

itself to the surface (Andrews et al., 2009). If the two adhering surfaces were brought

to close contact in the presence of fluid, the fluid could act as an adhesive to attach

both surfaces.

e) Fracture theory

The fracture theory evaluates bioadhesion based on difficulty (strain required) to

separate two adhering surfaces, which represents strength of the adhesive bond (Ahuja

et al., 1997; Andrews et al., 2009).

1.3.2 Factors affecting bioadhesion

The extent of bioadhesion between polymer and mucosa depends largely on polymer

properties, environment for bioadhesion and various physiological variables (Table

1.5).

20
Table 1.5: Factors affecting bioadhesion (Ahuja et al., 1997; Andrews et al., 2009).

Polymer properties Physiological factors

functional group mucin turnover

degree of hydration disease states

MW, chain length and degree of cross-linking pH

polymer concentration

pH and charge

swelling

Polymers containing carboxyl functional groups such as polycarbophils and

polyacrylic acid polymers are known to show better bioadhesion at acidic

environments whereby bioadhesion decreases with increasing pH. Park and Robinson

(1985) found that cross-linked polyacrylic acid has limited bioadhesive properties

above pH 6. This was attributed to the ionisation of the carboxyl groups which lead to

repulsion between negatively charged carboxylate anions and also a reduction in the

formation of hydrogen bonds (Andrews et al., 2009; Park and Robinson, 1985).

Degree of hydration and swelling characteristics of the bioadhesive polymer is also

important as swelling relaxes the polymer chains to facilitate interpenetration of the

chains. However, excessive swelling has been shown to significantly reduce

bioadhesion (Park and Robinson, 1985).

Studies have shown that there is an optimal MW and critical polymer chain length in

order to produce bioadhesive interactions (Tobyn et al., 1996, 1995). As polymer MW

increases, internal cohesion of the polymer molecule increases, resulting in higher

bioadhesion. However, the increase in MW also decreases aqueous dispersibility of

21
the polymer and hence fewer solubilised carboxylic groups are available for hydrogen

bonding (Tobyn et al., 1996). Extensive crosslinking of the polymer also limits

flexibility and mobility of the polymer chains, thus impeding penetration and

entanglement of the polymer–mucus matrix, resulting in lower bioadhesive forces

(Andrews et al., 2009).

Physiologically, mucus turnover is expected to limit residence of bioadhesive dosage

forms on the mucosal surface. The rectum has been known to possess relatively low

mucus turnover rates as compared to other regions of the GIT and this might be less

important in influencing rectal bioadhesion. However, diseased states like

inflammatory bowel disease could increase mucus production, or in diarrhoea where

there is significantly greater amount of water available (Allen et al., 2008).

1.4 Bioadhesive polymers

Although a large number of polymers exhibit bioadhesive properties, only a selected

number of polymers are suitable for pharmaceutical use due to safety considerations.

Some of the polymers adapted for pharmaceutical formulations are listed according to

their solubility in water and ionic charges in Table 1.6.

Anionic polymers are widely used as bioadhesive polymers due to their strong

bioadhesive properties and low toxicity. These polymers are characterised by the

presence of carboxyl, hydroxyl, amide or sulphate functional groups which dissociates

into negatively charged groups at pH larger than their respective pKa. Polyacrylic acid

polymer such as polycarbophil and carbopol are one of the most popular bioadhesive

polymers in pharmaceuticals (Yahagi et al., 2000, 1999).

22
The most widely used cationic polymer is chitosan; which is derived by hydrolysing

the aminoacetyl of chitin from crabs or shrimps. Various studies have reported on the

bioadhesivity of chitosan for development of bioadhesive dosage forms, including

rectal suppositories (Lehr and Bouwstra, 1992; Tarimci and Ermis, 1997) despite

reports that chitosan lacks bioadhesion (Wong et al., 1999a).

Table 1.6: Some of the bioadhesive polymers used in pharmaceutical dosage forms

(Irons and Robinson, 2003; Lehr and Bouwstra, 1992).

Water soluble Water insoluble

Anionic Alginic Acid Carbopol 934P
Carageenan Polycarbophil
Sodium carboxymethyl Cross-linked polymethacrylic acid
cellulose
Cationic Aminodextran Gelatin
Chitosan
Amphiprotic Carboxymethyl chitosan

Non-ionic Polyethylene Glycol Ethyl cellulose

Poly(vinylpyrrolidone)
Hydroxypropyl cellulose
Hydroxypropylmethyl cellulose

1.4.1 Carbomers

Carbomers are high MW carboxyvinyl polymers which are crosslinked with acrylic

acid using either allylsucrose or allyl pentaerythritol (Singla et al., 2000). A typical

example of carbomer is Carbopol® (CBP), its chemical structure shown in Figure 1.2a.

These polymers are commercially available in various grades depending on MW and

structure of the polymeric chain.

23
Figure 1.2 : The general chemical structure of bioadhesive polymers used, (a)

Carbopol®(CBP); (b) poly(vinylpyrrolidone) (PVP); (c) N,O-carboxymethyl chitosan

(CMCTS) and (d) hydroxypropyl methylcellulose (HPMC).

These polymers have been used in the development of various bioadhesive dosage

forms such as buccal tablets (Ikinci et al., 2004), transdermal gels (Shin et al., 2005),

ophthalmic gels (Edsman et al., 1996), pellets (Gomez-Carracedo et al., 2007),

suppositories (Ramadan, 2012; Yahagi et al., 2000, 1999) and more recently, as a

bioadhesive conjugate on liposome surfaces (Makhlof et al., 2011; Werle et al., 2010).

24
CBP is an anionic compound which appears as white, acidic, fluffy, hygroscopic

powder. These polymers are insoluble in water and have a pKa of approximately 6.0 ±

0.5. They are capable of swelling up to 1000 times their volume and 10 times their

diameter to produce a gel with pH 4.0–6.0 (Lubrizol Advanced Materials, 2009). At

pH above pKa, the carboxylate group on the polymer ionises causing repulsion

between the negatively charged polymer backbones, leading to swelling (Lubrizol

Advanced Materials, 2009; Singla et al., 2000).

1.4.2 Poly(vinylpyrrolidone)

Poly(vinylpyrrolidone) (PVP) is a synthetic, water soluble neutral polymer produced

by free-radical polymerisation of vinylpyrrolidone in water or isopropanolol (Guo et

al., 1998). Its general chemical structure is shown in Figure 1.2b. There is no general

consensus on the bioadhesive properties of PVP as various studies have reported the

absence or limited bioadhesion forces exerted by these polymers. Ivarsson and

Wahlgren (2012) and Jones et al. (2004) reported poor or negligible bioadhesion for

PVP while other researchers found that formulations containing PVP exhibited

considerable bioadhesive strength in the form of coated alginate beads (Suknuntha et

al., 2011), tablets (Hamzah et al., 2010), buccal patches (Wong et al., 1999b) and

bioadhesive thermogelling rectal gel (Barakat, 2009).

Apart from imparting bioadhesion, PVP was also reported to increase the rate of drug

release and sometimes result in a burst of drug release due to its high solubility which

promotes pore formation (Islam et al., 2012; Suknuntha et al., 2011).

25
1.4.3 Chitosan and carboxymethyl chitosan (CMCTS)

Chitosan is derived via partial deacetylation of the natural polysaccharide chitin.

Chitosan possesses both hydroxyl (-OH) and amine (-NH2) groups delivering many

useful properties such as gel and film forming capacity and bioadhesion (Tungtong et

al., 2012). However, the use of chitosan is hampered by its poor solubility at pH > 6.5.

This leads to the incorporation of carboxyl (-COOH) groups to form amphiprotic

carboxymethyl chitosan (CMCTS) which permits solubility even at neutral pH, yet

retaining good film forming (Mourya et al., 2010) and bioadhesive properties (Shinde

et al., 2013) of its parent chitosan. The structure of CMCTS is illustrated in Figure

1.2c. To date, CMCTS has been incorporated into pH sensitive hydrogels (Chen et al.,

2004; Vaghani et al., 2012), nanoparticles (Shinde et al., 2013) and transdermal

patches (Sarfaraz et al., 2012). However, CMCTS has yet to be formulated into

suppositories.

1.4.4 Hydroxypropyl methylcellulose (HPMC)

Hydroxypropyl methylcellulose (HPMC) is a non-ionic alkyl-hydroxyalkyl cellulose

ether derivative containing methoxyl and hydroxypropyl groups (Figure 1.2d). HPMC

dissolves in water to form a solution with pH 5.0–8.0 at a concentration of 2 %w/w

(Rowe et al., 2009).

HPMC has been incorporated into various drug delivery systems due to its ability to

gel on hydration and adhere to both mucin and mucosal surfaces. Due to its nonionic

and water soluble nature, possibilities of interaction with other components of the

formulation are greatly reduced. It has been incorporated into dosage forms to produce

26
buccal tablets (Akbari et al., 2010; Wong et al., 1999a), buccal patches (Vishnu et al.,

2007) oral bioadhesive tablets (Alladi et al., 2011), bioadhesive vaginal tablets (Bhat

and Shivakumar, 2010), bioadhesive transdermal gels (Cho and Choi, 2011) and nasal

insert (Bertram and Bodmeier, 2006).

1.5 Bioadhesive suppositories and its advantage

Retention of suppository within the lower rectum via incorporation of bioadhesive

polymers enables absorption of the released drug into the lower haemorrhoidal vein

which conveniently escapes first-pass degradation. This can be an excellent method to

deliver drugs which undergo extensive first-pass metabolism.

Furthermore, the rectum also has relatively low fluid content, thus dissolution of drugs

released from suppositories would theoretically be the rate limiting step for drug

absorption. Bioadhesive suppositories could potentially overcome this problem by

prolonging contact time between molten base and rectal mucosa where absorption

takes place. Securing dosage forms at the site of action also builds up a concentration

gradient for passive diffusion of drugs across the mucosa (Lehr et al., 1992; Roy and

Prabhakar, 2010).

Yahagi et al. (1999) developed double-phased bioadhesive suppositories containing

lidocaine using Carbopol 934P and beeswax in Witepsol H15. The group used 10 %

carbopol and 20 % beeswax in an attempt to anchor the suppository within the lower

rectum. As a result, increased systemic bioavailability of lidocaine from the double

phased bioadhesive suppository was observed.

27
Even when the drug is not extensively degraded during first pass metabolism,

suppositories containing bioadhesive polymers have been found to improve systemic

bioavailability of ramosetron (an antiemetic) by 2.5 times compared to when the

suppository was formulated without bioadhesive component (Yahagi et al., 2000).

1.6 Diclofenac sodium (DcNa)

As one of the more common NSAID, DcNa is prescribed for a myriad of conditions,

ranging from long term treatment in patients with rheumatoid arthritis to short term

treatment of muscular pains and aches. Recently, there has been great interest in using

rectal DcNa for postoperative pain management that could successfully improve pain

scores and reduce need for rescue analgesia (Dhawan et al., 2009).

DcNa is scientifically known as sodium (O-((2,6-dichlorophenyl)-amino)-phenyl)-

acetate and is a weak acid with pKa of 4 and a partition coefficient (n-octanol/ aqueous

buffer, pH 7.4) of 13.4. The ultraviolet (UV) absorbance of DcNa is detected at a

wavelength of 270-276 nm (Palomo et al., 1999).

It has a half-life of 2.3 hours after oral administration and is rapidly excreted from the

body. DcNa also precipitates under acidic conditions. Although the drug is fully

absorbed into the system, the absolute systemic bioavailability of active drug is only

approximately 50 % due to the extensive first-pass metabolism (Willis et al., 1979).

Therefore, dosage forms which could effectively avoid the hepatic first pass would

technically allow administration of DcNa at lower doses without compromising the

clinical responses (Palomo et al., 1999).

28
A study conducted in healthy volunteers revealed that peak serum levels of unchanged

DcNa upon administration of suppository containing 50 mg DcNa occurred within 1

hour, while peak serum levels were only attained after the 2 hours of administration in

the subjects given enteric-coated tablets containing equal amount of drug (Reiss et al.,

1978). The therapeutic plasma concentration of diclofenac in human was reported to

be 0.75–2.0 µg/mL (Winek et al., 2001) although other papers have reported that

concentrations as low as 100 ng/mL of plasma diclofenac was effective in initiating

analgesic effects (Radermacher et al., 1991). Meanwhile, trough concentrations of

Voltaren® SR were found to be 22-25 ng/mL, suggesting that the diclofenac minimum

therapeutic concentration might be within the range of 20-50 ng/mL.

To date, DcNa has been commercially marketed as enteric coated tablets, delayed

release tablets for oral administration, gel and ointments for topical applications,

suppositories for rectal administration and injectable solution for either IM or IV

administration. The conventional dose for DcNa orally is 50 mg twice or thrice a day

or 75 mg twice a day for adults. The latest addition to the market would be the

Voltaren®-XR, an extended release formulation of DcNa for once daily dosing.

1.7 Aims and objectives

Despite being a traditional suppository base, CB has been known to exist in various

polymorphic forms, therefore; an alternative base without the extensive polymorphism

exhibited by CB is highly desirable. Furthermore, suppositories seem to be an

excellent option for delivery of drugs which undergo extensive first pass metabolism

because absorption from the lower rectum enters systemic circulation directly.

29
The aim of this project was therefore to evaluate and characterise bioadhesive DcNa

suppositories produced using local HPKS.

Specific aims of the studies in this thesis were:

 to evaluate suitability of HPKS as an alternative to CB as suppository base.

HPKS are widely available in Malaysia as a by-product of palm oil at a low

cost

 to optimise manufacturing methods for suppositories made with both HPKS

and CB using differential scanning calorimetry(DSC)-simulated manufacturing

as well as extemporaneous methods

 to physically characterise DcNa suppositories containing bioadhesive

polymers made using HPKS in comparison to those made using CB

 to study release of DcNa from various suppository prototypes and the release

kinetics involved

 to fabricate and develop novel experimental prototypes for evaluation of

bioadhesive properties of suppository prototypes incorporated with

bioadhesive polymers (CBP, PVP, HPMC and CMCTS)

 to investigate the feasibility of synthetic membranes as substitute for biological

membranes in bioadhesion studies. Synthetic membranes are easily available,

standardised and doesn’t require prior processing.

 to evaluate physical stability of suppositories under different storage

conditions and duration in terms of thermal profile (DSC method), hardness,

softening time and drug release

30
CHAPTER 2

PREFORMULATION STUDIES

31
2.1 Introduction

The suitability of cocoa butter substitutes (CBS), namely ChocExa (CE) and

Supersocolate SpecialTM (SS) was assessed as part of the preformulation work for

development of a fast-acting bioadhesive suppository. Both CE and SS are HPKS

derived from kernel of oil palm fruits. CB, a traditional suppository base which was

indicated to possess qualities of an ideal suppository base is used for comparison

(Ansel, 1981; Kasture et al., 2007).

Although a suppository base is chemically inert and functions only as a carrier for the

active drug, properties of the base could affect physicochemical properties of resultant

suppositories. Therefore, it is pertinent to characterise the various base properties,

namely;

(a) Thermal profile

Two types of thermal analysis; thermogravimetric analysis (TGA) and DSC can be

used to study the thermal profile. TGA detects and quantify base decomposition (CB,

CE and SS) in terms of mass loss while the DSC shows phase changes encountered by

the base upon exposure to temperature range encountered during manufacturing

process (Giron, 1986; Schimdt, 2010).

(b) Solid fat content

Solid fat content (SFC) measures the solid-to-liquid ratio in fats over a temperature

range. It has traditionally been determined using dilatometry (Walker and Bosin,

1971) and more recently using pulsed nuclear magnetic resonance (p-NMR) (Leung et

al., 1985) or ultrasonic velocimetry (Singh et al., 2004). An alternative method of SFC

32
determination is by continuous integration of the DSC thermogram of a sample

subjected to heating across a desired temperature range (Menard and Sichina, 2000;

Nassu and Gonçalves, 1999).

(c) pH of molten base

Since the rectum is relatively devoid of buffering capacity, pH of the suppository base

would determine pH of rectal environment (Ahmad, 2001; Bottger et al., 1989).

Certain bases alter pH of rectal environment and this could be potentially problematic

if solubility of the drug is highly pH-dependent (Dash and Cudworth, 2001).

(d) Partition coefficient of DcNa in molten base / distilled water

Partition coefficient of DcNa between base and water determines how rapidly and

completely it partitions out of suppository in the rectum (Allen et al., 2008). A small

partition coefficient (favours aqueous phase) would permit rapid release of DcNa from

the bases, supporting the aim of this study to develop fast-release suppositories.

(e) Viscosity of molten base

Viscosity of the molten base during manufacturing affects uniformity of incorporated

drug within the dosage form (Allen et al., 2008; Coben and Lieberman, 1986). It is

practical that molten base be sufficiently viscous to prevent sedimentation of additives

but not too viscous that it makes manufacturing difficult or impedes drug release

(Azechi et al., 2000).

Both CE and SS have never been used in the pharmaceutical industry. Thus, proper

characterisation of these bases would allow a systematic approach in subsequent

33
formulation work. This chapter aims to characterise physical properties of HPKS in

comparison to CB as conventional suppository base and compares the robustness of

the bases towards various manufacturing parameters for subsequent optimisation of

CB and HPKS suppository manufacturing methods.

2.2 Materials

CB was purchased from JB Cocoa, Malaysia; while CE and SS were obtained as

samples from Lam Soon, Malaysia and Cargill, Malaysia respectively. DcNa was

purchased from Shreeji Pharma International, India. Product specifications are

included in Appendices 1-3.

2.3 Methods

2.3.1 Characterisation of suppository base

2.3.1.1 Thermal profile

2.3.1.1.1 Thermogravimetric analysis

TGA is conducted by placing 16.0 mg of unprocessed (raw) CB into a 40.0 μL

standard aluminium crucibles (Mettler Toledo, USA) sealed with a lid previously

pierced with a 50.0 μm hole to relieve vapour entrapment during the thermal cycle.

Analysis was carried out using a thermogravimetric analyser connected to a GC10 gas

controller system (Mettler Toledo, USA). Suppository bases were scanned at a speed

of 1 °C/min across a range of 25 to 70 °C. The flow of nitrogen purge was fixed at 5

mL/min. Experiment was repeated in triplicates.

34
2.3.1.1.2 Differential scanning calorimetry

DSC was conducted using TA Q2000 (TA Instruments, Delaware, USA) connected to

RCS 40 refrigeration system (TA Instruments, Delaware, USA) under nitrogen gas

flow of 50 mL/min. Unmanipulated (raw) base samples (CB, CE, SS) of 4-6 mg were

crimped into a hermetic Tzero aluminium pan and equilibrated to -10 °C before

subjected to heating to 60 °C at a rate of 5 °C/min (first heating). The samples were

then cooled to -10 °C at a rate of 5 °C/min before undergoing a second heating to

60 °C. Samples were held isothermally at 60 and -10 °C for 1 minute before cooling

and second heating phase respectively. Thermograms were analysed using TA

Universal Analysis 2000 software. The melting point is defined as the endothermic

peak. Samples were selectively replicated to ensure consistency.

2.3.1.2 Solid fat content (SFC)

Two methods were used to determine SFC in this study; namely the DSC method and

p-NMR method. The DSC method was modified from Leung et al. (1985) and was

conducted using instruments and methods mentioned in Section 2.3.1.1.2. SFC of the

raw bases was obtained by continuous integration of the thermogram generated.

Experiment was replicated using fresh samples. Determination of SFC using the p-

NMR was carried out by Malaysian Palm Oil Board (MPOB) using the MPOB

p4.9:2004 method.

2.3.1.3 pH of molten base

The pH of molten base (without additives) was determined using methods modified

from Dash (2001). 1.0 g of base was placed into a scintillation vial and added with

10.0 mL of distilled water. The vial was then left to shake for 6 hours at 100 rpm in an

35
isothermic shaker (WiseCube® WIS-20, Wisd Laboratory Instruments) kept at 37.0 ±

0.5 °C. The solution was filtered using a 0.45 µm filter and pH reading of the solution

was measured using a calibrated bench top pH meter (Eutech, USA) before and after

shaking. Procedures were conducted in triplicates for each base. ANOVA test

followed by post hoc Tukey’s HSD (Honestly Significant Difference) analysis (SPSS

Inc., version 20, USA) was used to analyse the results.

2.3.1.4 Partition coefficient of DcNa in molten base/ distilled water

The partition coefficient of DcNa in the bases were determined using methods

modified from studies by Ahmad (2001) and Nayak (2010). 3.0 g of base was weighed

into scintillation vials and added with 3.0 mL of 5 %w/v of aqueous DcNa. The vials

were sealed and left to shake for 24 hours at 100 rpm in an isothermic shaker

(WiseCube® WIS-20, Wisd Laboratory Instruments) kept at 37.0 ± 0.5 °C. The

aqueous phase was filtered with 0.45 µm nylon filters and absorbance measured at

276 nm. 3.0 g of base shaken with 3.0 mL of distilled water was used as control.

Procedures were conducted in triplicates for each base. ANOVA test followed by post

hoc Tukey’s HSD (Honestly Significant Difference) analysis (SPSS Inc., version 20,

USA) was used to analyse the results.

2.3.1.5 Viscosity of molten base

Rheological properties of only molten bases (without additives) were measured using

a LVDV-E rotational viscometer (Brookfield, USA) fitted with DAA cylindrical

spindle s87. The small volume sample chamber used was fitted with a water jacket

attached to a digital heating circulator unit (Protech, USA) maintained at 37.0 ± 0.5 °C.

Molten samples (2 mL) were added into the sample chamber and allowed to

36
equilibrate to 37.0 °C for 5 minutes. The viscosity measurement in units of centipoise

(cp) was obtained at various rotational speeds ranging from 20-100 rpm for torque

measurements above 10.0 %. Procedures were conducted in triplicates for each base

and expressed as mean ± SD.

2.3.2 Optimisation of manufacturing parameters

Due to similarities in thermal profiles between CE and SS (Section 2.4.1); SS was

used as an example of HPKS base in DSC-simulated suppository manufacturing

studies (Section 2.3.2.1) and extemporaneous manufacturing of suppositories at

various Tmax (Section 2.3.2.2). The robustness and ease of manufacturing of SS

suppositories were compared against conventional suppository base CB.

2.3.2.1 DSC–simulated suppository manufacturing

Simulation of the suppository manufacturing process was conducted using DSC

system mentioned in Section 2.3.1.2.2. Unmanipulated (raw) base (CB or SS) of 4-6

mg were subjected to a heat/cool/heat cycle across a predetermined temperature range.

The manufacturing parameters which potentially affect polymorphic behaviour of CB

and SS were investigated, namely; (a) maximum temperature the base was heated

before solidification (Tmax); (b) heating rate during melting of the base (Hrate) and

cooling rate for solidification of the molten base to suppositories (Crate). Samples were

selectively replicated to ensure consistency.

2.3.2.1.1 Effects of Tmax on phase behaviour

The first heating prior to a cooling cycle was used to simulate melting of base

followed by cooling and solidification in moulds. Independent, freshly prepared

37
samples were heated to various Tmax (CB = 34, 35, 36, 37, 38, 39 and 40 °C; SS = 36,

38, 40, 42 and 50 °C) during the first heat cycle. Molten CB was then cooled and

equilibrated at 4 °C before the second heating takes place. Second heating cycle was

conducted from -10 to 60 °C upon completion of the first heat/cool cycle to identify

the polymorphic behaviour of crystallised CB and SS after heating to various Tmax

(during first heat/cool cycle). Heating was conducted at 5 °C/min and cooling at

2 °C/min. Samples were held isothermally for 0.2 minutes in between each heat or

cool cycle.

2.3.2.1.2 Effects of Hrate and Crate on phase behaviour

To study the effects of Hrate, CB samples were heated from -10 to 37 °C while SS

samples were heated to 42 °C at various Hrate (1, 5 and 10 °C/min) during the first heat

cycle and cooled to 4 °C at 2 °C/min. In the Crate studies, CB and SS samples were at

5 °C/min to 37 and 42 ºC respectively followed by cooling down to 4 °C at various

rates (0.5, 2, 5 and 10 °C/min). The second heat cycle thermograms obtained by

heating the samples from -10 to 60 °C at 5 °C/min were used to identify the

polymorphic behaviour of crystallised CB and SS after being subjected to different

Hrate and Crate in the first heat/cool cycle.

2.3.2.2 Extemporaneous manufacturing of suppositories at various T max

Two types of suppositories were produced in this section for evaluation, base-only

suppositories (blank) and suppositories containing the model drug DcNa. The

suppositories were prepared by fusion method. Base (CB or SS) was heated over a

water bath (Julabo, Germany) to various Tmax temperatures (CB heated to 32, 34, 36,

37 and 42 ± 0.5 °C; SS to 36, 40, 42 and 50 ± 0.5 °C).

38
In DcNa-containing suppositories, DcNa was added into the molten base with gradual

stirring before pouring into the 1.0 g steel suppository mould cavities to cool at

controlled room temperature of 22 ± 1.5 °C; relative humidity (RH) 63 ± 3 % until

solidification of molten base. The produced suppositories each contained 50 mg DcNa.

Blanks were moulded without DcNa into the molten. Suppositories were then placed

into the refrigerator (3.5 ± 1.5 °C) for additional 10 minutes. The excess base

overfilling the mould cavities were scrapped off using a warm spatula to produce

suppositories with a smooth, flat end. The produced blank suppositories were scraped

and 4-6 mg of sample was crimped into a hermetic aluminium pan and heated in the

DSC from -10 to 60 °C at 5 °C/min. The ease of manufacturing suppositories at

various molten temperatures as well as the quality of the manufactured suppositories

was assessed. Quality of manufactured suppositories was evaluated based on the

presence or absence of cracks, bubbles or discoloration and well as smoothness to

touch.

2.3.2.3 Determination of displacement value (DV)

Blank suppositories (CB, CE and SS) were produced using calibrated six-cavity steel

suppository moulds (each cavity 1.0 g) and the suppositories produced were weighed.

The same moulds were used to manufacture suppositories containing 10 %w/w DcNa.

The medicated suppositories were weighed.

Displacement value (DV) determination was repeated in triplicates for each base and

calculated using Equation 2.1 as stated by Mollel (2006) and Vidras et al. (1982) :

39
 𝑿𝑩
Equation 2.1 𝑭=
𝟏𝟎𝟎 ( 𝑨 − 𝑩) − 𝑿𝑩
Where,

F = displacement value

X = percentage of additive used (DcNa)

B = weight of suppositories containing X % additive

A = weight of blank suppositories made without any additives

2.4 Results and discussion

2.4.1 Characterisation of suppository base

Based on the thermogravimetric data generated, the bases appear to be stable over the

temperature range of 25 to 70 °C. Weight loss of the base was minimal and

insignificant, thus it was concluded that decomposition did not occur across the range

of temperature that they are exposed to during manufacturing process.

Due to the large variability in naming of polymorphic forms in published literature,

the polymorphic forms for CB and both HPKS (CE and SS) encountered in this thesis

will be referred to as described in Tables 2.1-2.2 respectively. Designation of

polymorphic forms was based on their respective melting points in comparison to

previous published literature.

40
Table 2.1 : Reported polymorphic forms of CB and the variability in their nomenclature. Table includes nomenclature used to describe CB

polymorphs in this thesis.

Melting range Von Vaeck Wille and Lovegren et Van Malssen Allen et al. Beckett Nomenclature in thesis

(°C) (1960) Lutton (1966) al. (1976) et al. (1999) (2008) (2008) (CB)

-5 to +5 γ 1 (Not observed)

12 to 15 VI I
2
16 to 20 γ I V α γ

21 to 24 α II IV α II
§ ‡
25 to 27 III III β’ III 3A

27 to 30 β’ IV II β’ IV 3B

29 to 34 β V I β V 4A

34 to 36 VI β VI 4B

‡ Van Malssen et al. (1999) suggested that β’ exist as a range rather than two distinct forms, recorded by Beckett (2008) and Wille and Lutton
(1966) as forms III and IV; and forms II, III and IV by Lovegren et al. (1976).
§ Form III was subsequently found to be a mixture of different proportions of forms II and IV, as confirmed in studies by Aronhime et al. (1988).
41
Table 2.2 : The reported polymorphic forms of lauric fats, PKO blends, HPKS and the variability in their nomenclature. Table includes

nomenclature used to describe CE and SS polymorphs in this thesis.

Lauric fats PKO blend** HPKS

Melting range Anihouvi et al. Noordin and Chung Peyronel and Rossell Nomenclature in
Siew (2001)
(°C) (2013) (2009) Marangoni (2014) (1975) thesis (CE and SS) ††

-30 to -10 α Not observed

20 to 28 Form I Not observed

28 to 32 Form II β’2
β’
33 to 34 Form III β’ β’ β’1
‡‡
35 to 37 β’ β

**Custom blend composed of HPKS: PKS: virgin PKO at ratio of 5:2:3.

†† Iodine value of 0.4 in CE and 0.29 in SS.
‡‡ Melting point quoted for β’ forms from HPKS are based on four different degrees of hydrogenation and iodine value (IV) of 8.3, 4.4, 2.5 and

0.4.
42
The first heating thermograms in Figure 2.1 (solid lines) depicted melting profiles of

the unmanipulated (raw) bases; while the second melt thermogram (dash lines)

obtained from reheating the bases after first heating and cooling showed tolerability of

the bases to subsequent solidification after heating to high temperatures.

At least three distinct polymorphs were observed in the unmanipulated CB based on

the first heating thermogram (Figure 2.1a, solid line). They were form 2 (11.9 °C),

form 3A (22.1 °C), form 3B (28.8 °C) and form 4B (34.7 °C). Form 1 was not

observed throughout this study due to its metastable nature which resulted in its

spontaneous crystallisation into form 2 polymorph (Dewettinck and Foubert, 2004).

The CB stock used in this study has been stored for about 3 years at 3.5 ± 1.5 °C,

stabilisation of the form 4A into form 4B polymorph over the storage duration

resulted in the existence of form 4B as the major component in unmanipulated CB.

Upon cooling after the first heating, CB existed mainly in form 2 which exhibits an

endotherm peak at 20.3 °C (melting of form 2) and to a much lesser extent, form 3B

which melted at 26.6 °C (Figure 2.1a, dashed line). The presence of these

polymorphic forms is unfavourable, as the suppositories would theoretically remain in

the liquid state at room temperatures, causing sedimentation and separation of the

active drug or additives from the base.

Previous studies by Siew (2001) suggested that HPKS predominantly crystallises into

β’ polymorph which has a melting endotherm at 32.5 °C followed by a shoulder at

35–38 °C. The author also observed an exothermic event at 22-25 °C suggestive of a

transition from lower melting polymorphs to the more stable β’.

43
Both the HPKS used in this study were found to have lesser polymorphic forms

compared to CB. In the first heating cycle, CE (Figure 2.1b, solid line) yielded three

distinct melting endotherms; β’2 presented as a small peak at 27.0 °C, a larger

endotherm at 34.1 °C by the β’1 form and a less visible shoulder around 37.1 °C

attributed to the presence of β form. This was consistent with the HPKS melting points

reported by Chapman et al. (1971). Upon cooling, molten CE crystallised into the β’2

form which exhibited a single melting endotherm at 31.7 °C, as observed in the

second heating thermogram (Figure 2.1b, dash line).

Similarly, SS displayed three different forms; the β’2 form (27.9 °C), β’1 form

(35.7 °C), and β form (present as a shoulder around 38.4 °C) (Figure 2.1c, solid line).

Upon first heating and cooling, SS crystallised into β’2 which melts at 31.9 °C (Figure

2.1, dash line). Both CE and SS existed in polymorphic forms with melting points at

approximately 32 °C after the first heat/cool cycle, despite a slight shift in the

endotherm peak to a lower temperature compared to the first heating. This reflected

the robustness of HPKS over CB towards temperature manipulation.

All three bases had melting points close to the human body temperature, which is an

essential characteristic of a suppository base.

44
Figure 2.1: DSC thermograms showing the first and second heating profiles of

unadulterated (raw) bases (a) CB; (b) CE and (c) SS. Solid lines and dash lines

indicate the first and second heating respectively. Thermograms were offset for

clarity. Enthalpy of fusion was included on far right of each thermogram.

45
Figure 2.2a shows the SFC of the three bases determined using the DSC method. The

melting integral of a thermogram corresponds to the proportion of melted fat at a

particular temperature (Marquez et al., 2013). The SFC curves (Figure 2.2a) of all

three bases (CB, CE, SS) were similar and have very steep slopes between 32-38 °C.

Comparatively, the SFC generated using the p-NMR (Figure 2.2b) showed that curves

for both CE and SS were almost identical but were skewed to the right of the CB

curve. This was due to the experimental methods where CB samples have been heated

up to 80 °C to clear thermal memory before analysis, thus the SFC curve is likely to

correspond to the less stable polymorphic form of CB. The DSC method allows

continuous measurement of a single sample across the entire range of temperature and

allows the measurement of unadulterated (raw) samples; hence preferred for studying

the processing behaviour of the base during suppository manufacturing. In general,

both the HPKS (CE and SS) remained solid until higher temperatures than CB before

undergoing a similar sharp melting over a narrow range of temperature.

The SFC of fats can be used as a prognostic indicator of essential properties; for

example, the SFC below 25 °C provides an indication of hardness of the particular fat,

while the SFC between 25-30 °C reflects its resistance to heating (Torbica et al.,

2005). Meanwhile, the steepness of the slope at 35 ± 2 °C is more relevant in

suppositories as it reflects how rapidly the suppository melts at human body

temperature.

46
Figure 2.2 : SFC (%) of unadulterated (raw) suppository bases (CB, CE and SS)

determined using (a) DSC method and (b) p-NMR method (MPOB p4.9 : 2004).

Meanwhile, none of the three bases significantly altered pH of distilled water upon

melting (p> 0.05). The model drug DcNa have been reported to have a solubility of

0.0036 mg/mL in pH 4.5 acetate buffer to 0.036 mg/mL at pH 5.3 where a modest

increment of pH by 0.8 lead to a 10-fold increase in solubility of DcNa (Chuasuwan et

al., 2009). Therefore, any changes in pH brought about by melting of bases could

47
potentially affect the solubility of DcNa and its release from the bases; considering the

lack of buffering capacity in the rectal cavity.

An important attribute of a suppository base is its ability to release the incorporated

drug completely upon melting. The melting profile, viscosity, partition ratio and

solubility of drug in surrounding aqueous environment impacts the ease of drug

release. If a drug has higher solubility in the fatty base over the aqueous phase (rectum

environment) then it may be incompletely released from the suppositories or released

slowly (Allen et al., 2008). DcNa was found to preferentially partition into the

aqueous phase over oily phase in all the three studied bases. This suggested that all the

three bases release DcNa rapidly, although the partition ratio (base/water) of DcNa in

CB (0.245 ± 0.007) was significantly higher (p< 0.05) than both CE (0.031 ± 0.003)

and SS (0.037 ± 0.008).

70

65
CB
Viscosity ( cP)

60

55 CE

50 SS

45

40
0 0.5 1 1.5 2
Shear rate (1/s)

Figure 2.3 : The viscosity of the bases (CB, CE and SS) under different shear rates

maintained at 37 ± 0.5 ºC. Mean ± SD, n=3.

48
Figure 2.3 showed that the viscosities of all three bases decrease with increasing shear

rate. CE generally had similar viscosity to CB over shear rate of 20–100 rpm at 37 ºC

while SS appeared less viscous than both CE and CB. CE and SS are mainly

composed of shorter-chained lauric acid (C12) while the longer stearic acid (C18)

predominates in CB (Allen et al., 1999; Gros and Feuge, 1952). A molten base with

lower viscosity may offer better spreadability in the rectum upon melting, thus

increasing surface area for drug release, but at the same time extremely low viscosities

risk sedimentation of drug particles during manufacturing. The lower viscosities in CE

and SS in this situation were compensated by a steep SFC profile (Figure 2.2), as

rapid solidification reduces risk of drug sedimentation.

2.4.2 Optimisation of suppository manufacturing methods

The first heating in the heat/cool/heat cycle in the DSC-simulated manufacturing of

suppositories was conducted to simulate heating up of the base; this is then followed

by cooling of the molten to produce solid suppositories. Finally, the second heating

was used to determine the melting profile of the suppositories produced.

The small endothermic peak observed in the 10-20 °C regions (Figures 2.4a–d, inset)

in CB samples was attributed to melting of metastable form 2. The entailing

exothermic peak observed at the range of 15–25 °C corresponded to direct solid-solid

transition of the metastable form 2 to form 3A (Dewettinck and Foubert, 2004), which

intensity increases proportionally to increasing amounts of form 2 present. CB molten

heated to a Tmax between 34-37 °C (Figures 2.4a–d) crystallised mainly into form 2 as

well as a mixture of form 3A and 3B. Similar to previous studies, the experiments

carried out in this study using DSC were unable to distinguish between the form 2 to

49
form 3 transitions and the direct crystallisation of form 3 from the melt (Toro-Vazquez

et al., 2004). The presence of a small peak at approximately 36 °C (Figures 2.4a–c,

asterisk) simply implied the incomplete melting of form 4B in the initial raw base

before cooling down rather than as a result of the thermal manipulation imposed. The

size of form 4B peak decreased as the Tmax was increased (Figures 2.4a–c) and

completely diminished when Tmax was above 37 °C (Figures 2.4d–g).

When CB was heated to Tmax > 37 °C, the main polymorphic form produced was form

2 with small amounts of form 3A which was present as a shoulder at 25–28 °C

(Figures 2.4e-g, solid arrow). Complete removal of crystal ‘memory’ followed by

continuous cooling to 4 °C at 2 °C/min resulted in formation of unstable form 2

polymorph. The small amounts of form 3A observed could be a result of the

spontaneous transition from any form 2 polymorph produced (Dewettinck and Foubert,

2004). This highlighted the narrow temperature derangement tolerated by CB during

manufacturing before the base preferentially crystallises into form 2 polymorph.

50
Figure 2.4 : Thermograms of the second heating cycle of CB samples. CB was

subjected to various Tmax (5 °C/min) before being cooled at 2 °C/min to 4 °C. Subtle

peaks are enlarged in the respective figure inset. Arrow denotes small amounts of

form 3A polymorph present as a shoulder. Asterisk denotes residual form 4B

polymorph from first heating.

Figures 2.5a–f showed that SS base heated to between 36 and 50 °C recrystallised into

a single polymorphic form, the stable β’2 form (31-32 °C). This was consistent with

previous studies which reported the preferential crystallisation of lauric fats into the

β’2 polymorph (Anihouvi et al., 2013; Ehlers, 2012; Rossell, 1975; Timms, 1984).

51
Figure 2.5 : Thermograms of the second heating cycle of SS samples. SS was

subjected to various Tmax heated at 5 °C/min before being cooled at 2 °C/min to 4

°C. Asterisk denotes residual β polymorph from first heating.

Presence of a small peak around 37-39 °C (Figures 2.5a–b, asterisk) was thought to be

due to incomplete melting of β form in the base before the cooling cycle. SS has a

wide range of Tmax tolerance of at least 10 °C (Figures 2.5c–f), a contrasting difference

to CB. This simplifies manufacturing of suppositories using SS as it required less

stringent temperature control. As long as SS is completely molten (>38 °C), it would

solidify into a stable β’2 polymorph. This trend was observed for Tmax up to 50 °C.

52
On the contrary, there was a narrow range of Tmax between 34-38 °C which the CB

base has to conform to during manufacturing. CB has to be heated up to the

temperature at which it completely melts, yet ensuring that it does not overheat to

produce form 1 (α form) or require excessively long time for solidification. Unlike CB,

crystallisation of SS was not affected by Tmax during manufacturing using DSC-

simulated methods across the Tmax range of 40-50 °C.

Figures 2.6a–c showed the crystallisation of molten CB previously heated to 37 °C

upon cooling at 2, 0.5 and 0.2 °C/min. CB molten cooled at 2 °C/min has three

distinct crystallisation exotherms at 21.9, 18.3 and 14.4 °C respectively (Figure 2.6a).

At cooling rate of 0.5 °C/min, there were two crystallisation exotherms at 21.5 and

10.5 °C (Figure 2.6b), while cooling rate of 0.2 °C/min resulted in a single

crystallisation exotherm at 23.3 °C (Figure 2.6c). This was translated into the different

polymorphic forms as characterised by distinct endothermic peaks observed upon

reheating (Figures 2.6d-f). At the cooling rates investigated, CB crystallised mainly

into form 4A (Figures 2.6d-f, open arrow). CB cooled at 2 °C/min contained

considerable amounts of form 2 (Figures 2.6d-e, solid arrow) which diminished as

Crate was lowered. Presence of form 2 is supported by the appearance of an exothermic

peak around 18–20 °C (Figures 2.6a-b, asterisk) which corresponded to its transition

to 3A polymorph. Melting of this 3A polymorph appeared as a faint shoulder around

22–26 °C (Figures 2.6a-b; hash). This exothermic peak at 18–20 °C was absent when

the Crate was reduced to 0.5 and 0.2 °C/min (Figure 2.6f).

53
Figure 2.6 : Thermograms of the cooling cycle of CB samples at (a) 2 °C/min, (b) 0.5 °C/min and (c) 0.2 °C/min after undergoing first heating to

Tmax of 37 °C. Thermograms of the second heating cycle are shown in the corresponding thermograms (d–f). First heating was conducted at

5 °C/min and subjected to solidification at various cooling rates. Subtle peaks are shown in figure insets. Solid arrows denote presence of form 2

while open arrows denote form 4A. Asterisks denote exothermic event while hashes indicate presence of form 3A.
54
Even when CB was heated to 38 °C, it may still be possible for the molten to

crystallise into form 3B polymorph (Figure 2.7d, asterisk) with a melting point at

30.2 °C via slow cooling at 0.5 °C/min, albeit in mixture with substantial amounts of

metastable form 2 (Figure 2.7d, solid arrow). However, the formation of 3B

polymorph decreased when CB was heated to 39 °C (Figure 2.7e, asterisk) and

completely diminished with Tmax of 40 °C (Figure 2.7f). In CB heated to 40 °C, slow

cooling at 0.5 °C/min produced only form 2 polymorphs (Figure 2.7f, solid arrow).

CB heated to 38-40 °C crystallised mainly into form 2 (Figures 2.7a-c, solid arrow)

with small amounts of 3A polymorphs (Figures 2.7a-c, open arrow) upon cooling at

2 °C/min. The difference in Crate greatly affects crystallisation and polymorphic form

of molten CB. Rapid cooling resulted in an unstable diffuse crystalline phase made up

of low energy polymorphs, while slow cooling allows more time for the TAG chains

to pack into a lamellae to form a stable, 3-D structure (Metin and Hartel, 2005).

In molten SS however, the Crate had less influence on the resultant polymorphic forms.

SS consistently underwent a single step crystallisation process (Figures 2.8(i)a-d) into

the stable β’ polymorph (mixture of β’1 and β’2) with a melting endotherm at 31-33 °C

(Figures 2.8(ii)a-d, solid arrow); regardless of the Crate between 0.2-5 °C/min. This

robustness observed in SS is favourable especially for the manufacturing of

suppositories as it eliminates requirement for rate controlled cooling and prevents

variability of polymorph formation in the final product.

The Hrate on the other hand, was found to not affect the polymorphic forms produced

in both CB and SS between 1–10 °C/min (results not shown).

55
Figure 2.7 : Thermograms of the second heating cycle of the CB. First heating was conducted to different Tmax 38, 39 and 40 °C (5 °C/min) and

subjected to solidification at either cooling rate of 2 °C/min (a–c) or 0.5 °C/min (d–f). Smaller peaks are shown in the figure insets. Solid arrows

denote presence of form 2; while open arrows denote form 3A. Asterisks denote form 3B.
56
Figure 2.8 : Thermograms of (i) cooling cycle and (ii) second heating thermogram after first ‘heat–cool’ cycle of SS samples. The SS samples

were subjected to cooling rate of (a) 5 °C/min, (b) 2 °C/min, (c) 0.5 °C/min and (d) 0.2 °C/min after first heating to Tmax of 42 °C. Both first and

second heating cycles were conducted at rate of 5 °C/min. Solid arrows denote presence of β’ form (mixture of β’1 and β’2).
57
The Crate of molten base in extemporaneously prepared suppositories was ~1 °C/min,

projected based on the time required for molten base temperature to decrease by 10 °C

during actual manufacturing.

All extemporaneously produced suppositories from CB molten at various Tmax

resulted in similar profiles, with the exception of molten CB at Tmax of 32 °C where an

additional small endothermic peak at 36.5 °C (Figure 2.9a, asterisk) was a result of

the incomplete melting of the unmanipulated (raw) CB base during manufacturing.

This was consistent with observations in DSC simulated studies (Figures 2.4a–c). The

main endothermic peak at 32.4 ± 0.3 °C indicated that CB suppositories existed

mainly in the stable form 4A (Figures 2.9a-e, open arrow), with a shoulder at 25–

29 °C attributed to the presence of one or more form 3 polymorphs (Figures 2.9a-e,

solid arrow). The proportion of form 3 polymorphs present increased with increasing

Tmax. A small peak at 10–20 °C was observed which corresponded to very small

amounts of form 2 polymorph (Figures 2.9a-e, inset marked with hash).

This is an interesting finding since even molten which has been heated to 40 °C

congealed to produce suppositories in the stable form 4A polymorphs (Figures 2.9a-e),

rather than into form 2 polymorphs as observed in the DSC-simulated studies and as

suggested by various literatures (Collett, 1990). It was deduced from these

observations, that even when molten CB has been heated above 36 °C, it was still

possible to manufacture suppositories in the stable form 4A as long as the molten CB

has been allowed to congeal over longer periods of time (ie: >55 mins) at

temperatures between 21.5–23.5 °C. If the same molten had been rapidly cooled in the

refrigerator (simulated by the DSC cooling), it would have congealed under

58
refrigeration into form 2 polymorph which reverts to a molten once removed from the

refrigerator, owing to its low melting point.

Figure 2.9 : Thermograms of extemporaneously produced CB suppositories

manufactured from molten heated to various Tmax followed by solidification at

regulated lab conditions of 22.5 ± 1 °C; RH 63 ± 3 %. Smaller peaks are shown in the

enlarged insets. Hashes indicate the form 2 polymorph, while solid and open arrows

denote the presence of form 3 and form 4A polymorphs respectively. Asterisk shows

the unmelted residual (raw) base.

On the other hand, molten SS appeared white and creamy when heated up to 39 °C.

At 40 °C, SS becomes a colourless liquid. The SS suppositories produced from

59
molten at Tmax of 40 to 50 °C consisted mainly of the stable β’1 polymorph (Figures

2.10a-d, open arrow) with a melting point of 34.3 ± 0.1 °C, together with a small

fraction of β’2 (Figure 2.10, solid arrow), evident as a small shoulder at 29.4 ± 0.5 °C.

The small peak present at 40.1 °C (Figure 2.10a, asterisk) was a result of residual

unmanipulated (raw) base due to incomplete melting.

Figure 2.10 : Thermograms of extemporaneously produced SS suppositories

manufactured from molten heated to various Tmax followed by solidification at

regulated lab conditions of 22.5 C ± 1 °C; RH 63 ± 3 %. Solid arrow denotes the

presence of small amounts of β’2 form, while open arrow denotes β’1 form. Asterisk

shows the unmelted residual base.

60
A summary of the qualitative assessment can be found in Table 2.3. It was found that

molten CB heated to 32 °C and below was too thick for DcNa incorporation and made

pouring into moulds difficult. On the other hand, molten CB heated to 38 °C and

42 °C resulted in solidification time more than 60 minutes and sedimentation of DcNa

particles to the bottom of the suppositories was observed. Incorporation of DcNa into

SS on the other hand, was easy in molten heated to temperatures above 36 °C.

The observed differences for CB and SS between both methods (DSC-simulated

method and extemporaneous manufacturing) were mainly due to differences in

experimental setup. Extemporaneously compounded suppositories were manufactured

in a lab with temperatures regulated at 22 ± 1.5 °C; RH 63 ± 3 %, while the samples

in DSC-simulated studies were crimped into a pan and cooled at a fixed rate.

Moreover, the molten bases in extemporaneous manufactured suppositories were

subjected to stirring (shear) which facilitated rearrangement of the fatty acid chains

into a more stable arrangement (Dhonsi and Stapley, 2006).

The effects of temperature and shear on crystallisation and polymorphic forms

produced has been well demonstrated by various groups (Dhonsi and Stapley, 2006;

Macmillan and Roberts, 2002; Sato, 2001). In the presence of shearing in CB, form

4A polymorphs were produced as a result of transformation from the precursor form

3A (Dhonsi and Stapley, 2006; Sato, 2001). While in the absence of shearing, only the

form 3 variants (both form 3A and 3B) were observed, with form 3A polymorph again

acting as a precursor for the formation of more stable form 3B (Sato, 2001).

61
Table 2.3 : Qualitative assessment of CB and SS molten properties, ease of manufacturing process and visual appearance of compounded

suppositories subjected to various Tmax, as determined by lab-scale extemporaneous manufacturing process. §§

CB SS
Tmax (°C)
30 32 34 37 40 36 40 42 50
Molten Colour Pale Milky Dark Dark
Yellow White Colourless Colourless Colourless
properties yellow yellow yellow yellow
Bubbles - - - + + - - - -
Turbidity ++ + - - - ++ - - -
Viscosity ++ + - - - ++ - - -
Ease of Drug Difficult Difficult
Difficult
manufacturing incorporation (thick Easy Easy (thick Easy Easy Easy
(sediment)
base) base)
Pouring into Difficult Difficult
mould Easy Easy Easy Easy Easy Easy
(viscous) (viscous)
Leaking from
- - + ++ - - - -
mould
Unable
Time to to mould
solidification 15 18 54 65 5 7 10 18
(min)
Produced Colour Yellow Yellow Yellow Yellow White White White White
suppositories Smoothness Smooth Smooth Smooth Smooth Smooth Smooth Smooth Smooth
Cracks - - - - - - - -
Airholes
+ - - - + + + +
(Bubbles)

§§
‘+’ implies presence of a particular observation; ‘++’ implies presence of the observation with greater intensity; while ‘-’ implies the absence of the observation.
62
Similarly, extemporaneously manufactured SS suppositories comprised of the higher

melting point β’1 polymorph (33-34 °C) in the presence of shear (from stirring of

molten) rather than the lower melting point β’2 polymorph observed in the DSC-

simulated studies.

Based on both DSC-simulated and extemporaneous manufacturing of suppositories,

Tmax of 34 and 42 °C were used in production of CB and SS suppositories respectively

for subsequent analysis considering their practical solidification time.

The DV of a compound is the weight of medicament or additive which displaces 1.0 g

of the base in which they are to be formulated in (Allen et al., 2008). These values are

crucial to determine the exact quantities of excipients, active drug and bases needed

to formulate each suppository as the moulds used during manufacturing are of a fixed

volume (Allen et al., 2008; Babar et al., 1999; Vidras et al., 1982).

DV of the same drug could be different for different bases due to difference in density,

hence complicating the manufacturing process and resulting in inaccurate dosage

strength. This study found that the DV of DcNa was similar among the three bases

(CB = 1.40 ± 0.06; CE = 1.40 ± 0.02 and; SS = 1.48 ± 0.07). CE and SS may be a

good substitute or equivalent for CB as a suppository base in existing formulations

using CB.

2.5 Conclusion

Both the HPKS used in this study, CE and SS were found to be comparable to CB in

terms of thermal profile, SFC, pH, viscosity and DV.

63
This chapter also found that CB has rigid processing parameters during manufacturing.

When manufactured extemporaneously, it is crucial that the CB molten base is (1)

maintained between 34 ± 0.5 °C during heating and (2) allowed to cool slowly to

between 20–24 °C to produce suppositories with the desirable form 4A polymorph (3)

molten CB should not be placed into a refrigerator before solidification is complete as

it will result in rapid crystallisation into the form 2 polymorph. These restrictions

however, were not relevant in HPKS where crystallisation was independent of the

heating and cooling process during manufacturing. Stirring of molten (shear)

promotes the formation of higher melting point polymorphic forms in both bases.

HPKS is a suitable alternative suppository base due to its superior manufacturing

flexibility compared to CB.

Based on the findings in this chapter, the two HPKS investigated (CE and SS) appear

to be suitable candidates as an alternative lipophilic base for the development of DcNa

suppositories.

64
CHAPTER 3

FORMULATION AND

ASSESSMENT OF SUPPOSITORY

DOSAGE FORM

65
3.1 Introduction

3.1.1 Routine analysis for manufactured suppositories

Analytical assessments of manufactured pharmaceutical products are routinely

conducted to ensure quality standards are met. Following manufacture, suppositories

are physically inspected for its appearance; examined for weight uniformity, melting

point, viscosity, active drug content, resistance to mechanical fracture (hardness) and

softening time (Coben and Lieberman, 1986; De Blaey and Tukker, 1996).

Visual appearance of suppositories can sometimes be a good indication of the

homogeneity of suspended ingredients as well as suitability of the manufacturing

parameters (Allen et al., 2008; De Blaey and Tukker, 1996). Meanwhile, significant

weight variation in a batch of suppositories could be an indication of either

inhomogeneity of additives or the presence of a cavity which ultimately affects

content uniformity of dosage forms. Proper standardisation of manufacturing

parameters would minimise these variations (Mollel, 2006).

Melting point of a lipophilic suppository is crucial for drug release upon insertion into

the rectum. A depressed melting point leads to melting and damage of the

suppositories during handling while those with melting point above body temperature

will not melt completely. Various methods have previously been used to study melting

point of suppositories, including the DSC (Kauss et al., 2013; Mollel, 2006; Yahagi et

al., 1999), capillary method (Ahmad, 2001; Hammouda et al., 1993; Pugunes and

Ugandar, 2013) and water bath method (Shegokar and Singh, 2010; Soremekun et al.,

2012). The DSC method offers a more precise measurement as it records actual phase

changes of the base across the temperature range tested.

66
Viscosity of melted suppositories would affect spreading and release of medication in

situ (Azechi et al., 2000). Thicker molten may hinder drug release and limit spreading

while excessively watery molten could leak from the rectum. Suppository viscosities

have been measured using viscometers of the Ubbelohde type (Yahagi et al., 2000),

cone and plate type (Takatori et al., 2004) and spindle viscometer (Reanmongkol et al.,

2011; Victoria and David, 2003).

Determination of the active drug (DcNa) content of suppositories is crucial to

demonstrate dose-to-dose consistency. Method of active drug quantification ranges

from the more conventional solvent or aqueous extraction of the drug followed by

subsequent spectrophotometric analysis (Shegokar and Singh, 2010; Swamy et al.,

2012; Zawar and Bhandari, 2012); by way of DSC micro quantification method

(Noordin and Chung, 2004) or by using the partial least squares treatment of the FT-

Raman spectra (Szostak and Mazurek, 2013).

Additives are known to affect hardness of suppositories (Coben and Lieberman, 1986;

Güneri et al., 2004; Kosior, 2001; Shegokar and Singh, 2010). Although there is no

standard method to evaluate hardness of suppositories, measurements have been done

using suppository hardness tester type SBT by Erweka (Babar et al., 1999; Ghorab et

al., 2011; Güneri et al., 2004; Hanaee et al., 2004; Shegokar and Singh, 2010), bench

top tablet hardness tester (El-Majri and Sharma, 2010; Kurosawa et al., 1985; Oribe et

al., 1995), hand-held Monsanto tablet hardness tester (Noman and Kadi, 2011; Saleem

et al., 2008; Varshney Himanshu and Tanwar, 2009), modified rheometer with tooth

press stick B (Ramadan, 2012; Yahagi et al., 1999), and texture analyser (Gugulothu

et al., 2010).

67
Softening time is especially important for lipophilic suppositories. A long softening

time may result in premature expulsion from the rectum before drug absorption can

occur, yet suppositories which soften too soon are easily damaged during handling

prior to administration. The European Pharmacopoeia (2010) specified an apparatus in

Section 2.9.22 which consisted of a glass tube immersed in a temperature controlled

water bath for the measurement of time elapsed for lipophilic suppositories to soften

permitting penetration of rod probe. Apart from that, softening time can also be

determined using the liquefaction and softening time apparatus (Moghimipour et al.,

2009) and U-tube submerged in a water-bath (Varshney Himanshu and Tanwar, 2009).

This chapter aims to manufacture bioadhesive suppositories containing DcNa (model

drug) and four different types of bioadhesive polymers using the fusion method. The

polymers used were the anionic polymer CBP grade 974P NF; amphiprotic polymer

CMCTS and non-ionic polymers PVP grade K30 and HPMC grade 2910. The

manufactured suppositories were then subjected to routine analysis in terms of

physical inspection, weight variation, melting point, DcNa content, viscosity, hardness

and softening time in order to ensure the quality standards are met.

3.2 Materials

The base and model drug used have been described in Section 2.2. CBP 974P NF was

obtained as a sample from Drex-chem (M) Sdn. Bhd., Malaysia; HPMC 2910 was

purchased from Newstar Chem Enterprise, China; PVP K30 was acquired from

Brightchem Sdn Bhd., Malaysia; while CMCTS was procured from China Eastar Co.

Ltd., China. Potassium dihydrogen phosphate (KH2PO4), dipotassium hydrogen

68
phosphate (K2HPO4) and potassium bromide were purchased from Nacalai Tesque

Inc., Kyoto, Japan. Product specifications are included in Appendices 4-7.

3.3 Methods

3.3.1 Manufacturing of DcNa suppositories containing bioadhesive polymers

Methods of suppository manufacturing have been optimised in Section 2.4. The DV

for bioadhesive polymers were determined using methods described in Section 2.3.3.3

for all three bases used.

The actual amount of suppository base required for a specific batch of suppositories

was calculated using Equation 3.1. The base required was accurately weighed into a

ceramic evaporating dish using a lab analytical balance (Sartorius, Germany) and

heated over a water bath (Julabo, Germany) to 34.0 °C for CB and 42.0 °C for both

the HPKS (CE and SS). Accurately weighed DcNa and bioadhesive polymers were

evenly mixed into the molten with gradual stirring before pouring into the 1.0 g steel

suppository mould cavities. Suppositories were allowed to congeal at room

temperature for 30 minutes before the overfilled excess was scraped-off using a heated

spatula. The suppositories produced were stored at 3.5 ± 1.5 °C, RH 29 ± 3 %; and

used within 2 weeks of manufacturing. Five suppository formulations were produced

for the subsequent analyses; base-only suppositories (blank), suppositories containing

only 50 mg DcNa (DcNa only) and suppositories containing 50 mg DcNa added with

1, 2 or 5 %w/w bioadhesive polymers (CBP, PVP, HPMC, CMCTS). The resultant

suppositories were 25 mm in length and 8 mm in diameter at the barrel end.

69
 𝐴1 𝐴
Equation 3.1 𝐵 = ( 𝑁 𝑥 𝐹 ) − [( ) + ( 2 )]
𝐷1 𝐷2

Where,

B = amount of base required

N = number of suppositories to prepare

F = fill weight of mould cavities obtained from mould calibration

A1 = required amount of additive-1

D1 = displacement values of additive-1 in the base used

A2 = required amount of additive-2

D2 = displacement values of additive-2 in the base used

3.3.2 Dosage form analysis

3.3.2.1 Visual inspection

Physical appearances of the external surface of 10 randomly selected suppositories

(n=10) from each formulation were examined for odour, shape integrity, colour

uniformity, and manufacturing defects.

3.3.2.2 Weight variation

The mass of 10 randomly selected suppositories (n=10) from each formulation were

individually weighed. Data obtained was presented as mean weight ± SD.

The target weight (g) of individual suppositories for each formulation was calculated

as a comparison to the actual weight of the suppositories produced, using the equation:

70
 𝑊𝐵𝑁 + 𝑊𝐷𝑁 + 𝑊𝑃𝑁
Equation 3.2 𝑇𝑎𝑟𝑔𝑒𝑡 𝑤𝑒𝑖𝑔ℎ𝑡 =
𝑁

Where,

WBN = weight of base used (CB, CE or SS) to produce one batch

WDN = weight of active drug used (DcNa) to produce one batch

WPN = weight of bioadhesive polymer (CBP, HPMC, PVP, CMCTS) for one batch

N = number of suppositories to produce in one batch

3.3.2.3 Melting point

The melting point of all the formulations was determined using DSC parameters

described for determination of melting point of the bases in Section 2.3.1.1. The

endothermic peak minimum on the DSC thermogram was identified as melting point

of the formulation tested.

3.3.2.4 Determination of DcNa content

The suppositories were stirred in 200 mL phosphate buffer (pH 7.4) solution on a

Variomag® Telesystem magnetic stirrer plate (Thermoelectron Corp, WI, USA)

immersed in a 37.0 °C water bath (Fisher Scientific, PA, USA) for 3 hours. The

content of DcNa released from each suppository was quantified

spectrophotometrically. Only formulations incorporated with 5 %w/w bioadhesive

polymers (CBP, PVP, HPMC and CMCTS) were examined for content uniformity.

Suppositories containing 50 mg DcNa without bioadhesive polymer was examined for

comparison purposes.

71
3.3.2.5 Viscosity

Viscosity measurements of the molten suppositories were carried out in the same

manner as Section 2.3.1.5. The measurement obtained at shear rate of 50 rpm was

used for comparison purposes between the base-only suppositories, suppositories

containing only DcNa and suppositories containing DcNa and 1–5 %w/w bioadhesive

polymers. The experiment was carried out in triplicates and the results were expressed

as mean ± SD.

3.3.2.6 Hardness

Hardness of the formulated suppositories was measured using Dr. Schleuniger 8M

Tablet Hardness Tester (Pharmatron, USA). The hardness tests were carried out at

room temperature. Hardness measurements in units of Newton (N) were taken both at

the cylindrical circumference (Figure 3.1a) as well as the pointed tip (Figure 3.1b) of

the suppositories. Measurements were repeated with six independent samples (n=6)

for each batches tested for both the forces required to break the barrel and tip of the

suppository.

Figure 3.1: Direction of force exerted at (a) cylindrical circumference (barrel) (b)

pointed tip of suppository.

72
3.3.2.7 Softening time

The softening time of suppositories were measured using a modified apparatus (Figure

3.2) adapted from the Erweka Suppository Penetration Tester PM 30 (Erweka,

Germany) and SPT-6 Penetration Tester (Pharmatest, USA).

Figure 3.2 : The experimental setup of the softening point apparatus.

The setup of the apparatus illustrated in Figure 3.2, comprised of a rounded end pyrex

test tube (inner diameter = 0.9 cm; length = 7.4 cm) (Iwaki, Japan) vertically

immersed in a water bath maintained at 37.0 ± 0.5 °C using a test tube rack secured

with a polystyrene holder. The polystyrene holder serves as an insulator to prevent

heat loss as well as to secure the test tube at a 90° angle.

A suppository was introduced into the test tube pointed end first and a timer was

started simultaneously as a graduated brass probe (diameter = 0.9 cm, length = 8.9 cm,

73
weight = 51.8 g) was placed over the suppository. The rod slowly penetrated the

suppository as it softens. The end point (softening time) was measured as the time

elapsed for the probe to penetrate 1 cm depth of the suppository. Experiment was

carried out in triplicates for each formulation at room temperature of 24.5 ± 2.5 °C,

and RH of 58 ± 5 %. Results were expressed as mean time ± SD.

3.3.2.8 Statistical analysis

The results obtained from Sections 3.3.2.5, 3.3.2.6 and 3.3.2.7 were analysed using

analysis of variance (ANOVA) (SPSS Inc., version 20, USA). Post hoc analysis using

Tukey’s HSD test was carried out where appropriate when a statistically significant

difference at p < 0.05 was obtained.

3.4 Results and discussion

3.4.1 Physical inspection

The visual appearances of the suppositories were acceptable for all the formulations.

Suppositories were intact; surface was smooth; colour was even and uniformed with

the absence of visible defects such as those illustrated in Figure 3.3 on the external

surface of suppositories. There was no abnormal odour from the suppositories. Only

suppositories which have passed the visual inspection were included in subsequent

analysis.

74
Figure 3.3 : The common defects found on suppositories: (a) fissures or cracks; (b)

pitting; (c) fat-blooming; (d) particle sedimentation; (e) chipping.

3.4.2 Weight variation

The target weight, actual weight, melting point and content uniformity of the test

formulations were tabulated in Tables 3.1-3.3 for CB, CE and SS respectively.

Suppositories produced weighed between 1.09 to 1.14 g (Tables 3.1-3.3). Mean actual

weight of manufactured suppositories was in good agreement with the target weight,

and deviations from respective target weights were ± 1 %. The SD value of the

samples for each formulation was small (± 0.011 g), this was suggestive of good

weight uniformity and that the additives were well dispersed within the suppositories.

The British Pharmacopoeia (2015) stated that weight deviation in a batch of

suppositories should be within the range of ± 5%. A large variation in weight could be

due to either non-standardised mould cavity fill or non-homogeneity of molten

mixture which would affect content uniformity of the suppositories (Mollel, 2006).

All batches of suppositories produced throughout this study were assessed for

compliance to the weight uniformity (± 5 %) before further analysis. This provides

assurance that suppositories used in subsequent analyses were homogenous and of

satisfactory quality.

75
3.4.3 Melting point

Although thermal profiles of the investigated bases have been adequately

characterised in Chapter 2, the melting point of suppositories produced may be altered

by incorporation of additives (DcNa and bioadhesive polymers) or by the

manufacturing process itself (El-Majri and Sharma, 2010; Mollel, 2006; Shegokar

and Singh, 2010; Takatori et al., 2004; Yahagi et al., 1999). This study found that

manufactured blanks for all three bases melted between 32.5 to 34.5 °C (Tables 3.1-

3.3). Suppositories made using HPKS (CE and SS) melted at 1-2 ºC higher than the

corresponding formulation produced using CB. The addition of DcNa did not appear

to affect the melting point for suppositories made with all three bases (CB, CE and

SS).

The addition of 1-5 %w/w CMCTS to CB suppositories increased melting point of the

suppositories by 1 ºC, while addition of CMCTS decreased the melting point of CE

suppositories slightly. SS suppositories containing PVP had consistently higher

melting points compared to SS suppositories containing other bioadhesive polymers.

In general, the melting points of all suppositories containing 50 mg DcNa remained

approximating body temperature; between 32.5 to 35.0 ºC. Yarnykh et al. (2011)

reported that the melting point of commercial suppository bases found in the market

ranged between 29.0 to 40.0 ºC. Suppositories produced using natural fatty bases, for

example fat from borneo tallow seeds (Robertson, 1961) and palm oil blends (Pugunes

and Ugandar, 2013) were found to have melting points ranging between 34.0 – 37.0

ºC and 37.0-37.1 ºC respectively.

76
Table 3.1 : The target weight, actual weight, melting point and DcNa content of CB suppository formulations containing 50 mg DcNa and

bioadhesive polymers. Mean ± SD.

Polymers (%w/w) Target weight Actual weight (g) Melting point DcNa content (mg)

CBP HPMC PVP CMCTS (g) (n=10) (°C) (n=3)

CB without DcNa - 1.090 ± 0.005 32.8 -

- - - - 1.104 1.103 ± 0.005 32.7 48.51 ± 1.81
1 - - - 1.108 1.105 ± 0.005 32.9 -
2 - - - 1.112 1.109 ± 0.006 32.7 -
5 - - - 1.125 1.118 ± 0.006 32.5 48.52 ± 1.07
- 1 - - 1.108 1.112 ± 0.004 33.1 -
- 2 - - 1.111 1.111 ± 0.004 33.0 -
- 5 - - 1.121 1.123 ± 0.004 33.0 49.52 ± 1.08
- - 1 - 1.104 1.100 ± 0.006 32.8 -
- - 2 - 1.104 1.101 ± 0.005 32.6 -
- - 5 - 1.104 1.104 ± 0.004 32.8 49.25 ± 1.01
- - - 1 1.108 1.113 ± 0.006 33.7 -
- - - 2 1.112 1.117 ± 0.005 33.7 -
- - - 5 1.124 1.129 ± 0.006 33.8 49.28 ± 0.97

77
Table 3.2 : The target weight, actual weight, melting point and DcNa content of CE suppository formulations containing 50 mg DcNa and
bioadhesive polymers. Mean ± SD.

Polymers (%w/w) Target weight Actual weight (g) Melting Point DcNa content (mg)
CBP HPMC PVP CMCTS (g) (n=10) (°C) (n=3)
CE without DcNa - 1.101 ± 0.006 34.0 -
- - - - 1.114 1.125 ± 0.006 34.4 47.97 ± 0.49
1 - - - 1.120 1.128 ± 0.009 34.3 -
2 - - - 1.122 1.126 ± 0.011 34.6 -
5 - - - 1.133 1.143 ± 0.006 34.2 48.17 ± 0.32
- 1 - - 1.118 1.118 ± 0.005 34.7 -
- 2 - - 1.121 1.121 ± 0.005 34.9 -
- 5 - - 1.132 1.137 ± 0.009 34.7 49.07 ± 0.50
- - 1 - 1.114 1.119 ± 0.009 34.8 -
- - 2 - 1.114 1.120 ± 0.009 34.3 -
- - 5 - 1.114 1.123 ± 0.011 34.4 49.39 ± 1.14
- - - 1 1.119 1.120 ± 0.006 33.6 -
- - - 2 1.121 1.128 ± 0.007 33.8 -
- - - 5 1.137 1.143 ± 0.005 33.9 49.03 ±0.24

78
Table 3.3 : The target weight, actual weight, melting point and DcNa content of SS suppository formulations containing 50 mg DcNa and
bioadhesive polymers. Mean ± SD.

Polymers (%w/w) Target weight Actual weight (g) Melting point DcNa content (mg)

CBP HPMC PVP CMCTS (g) (n=10) (°C) (n=3)

SS without DcNa - 1.098 ± 0.004 34.0 -

- - - - 1.116 1.124 ± 0.006 34.3 48.55 ± 1.81
1 - - - 1.120 1.126 ± 0.005 34.8 -
2 - - - 1.124 1.131 ± 0.004 34.8 -
5 - - - 1.136 1.142 ± 0.007 34.6 47.53 ± 0.47
- 1 - - 1.120 1.125 ± 0.008 34.4 -
- 2 - - 1.123 1.129 ± 0.005 34.7 -
- 5 - - 1.134 1.136 ± 0.010 34.8 48.33 ± 1.27
- - 1 - 1.116 1.113 ± 0.007 34.9 -
- - 2 - 1.116 1.120 ± 0.004 34.9 -
- - 5 - 1.116 1.124 ± 0.004 35.1 48.84 ± 0.48
- - - 1 1.120 1.130 ± 0.005 34.7 -
- - - 2 1.124 1.129 ± 0.011 34.6 -
- - - 5 1.137 1.141 ± 0.005 34.5 48.68 ±0.85

79
3.4.4 Determination of DcNa content

This is especially important when suppositories are manufactured manually, as the

manual process is highly variable. Quantification of DcNa content in selected samples

obtained from a large batch allows verification of the labelled content as well as the

uniformity of DcNa content between individual suppositories produced in a particular

batch (Allen et al., 2008). Noordin and Chung (2004) investigated both the content

uniformity between suppositories as well as within a suppository by comparing

between different sections of a single dosage form to address uniformity of a

suppository in the event of dose splitting. The active drug content of suppositories

have to be within the range of 90-110% of the stated amount (British Pharmacopoeia,

2015). Content uniformity is achieved when not more than one suppository has drug

content out of the 85-115 % range, and none outside the limit of 75-125% of the

average content (British Pharmacopoeia, 2015).

The results of content uniformity studies for CB, CE and SS suppositories were

summarized in Tables 3.1-3.3 respectively. All the formulations contained more than

95 % of the stipulated DcNa content and the SD values were ± 2 % of the average

DcNa content. This confirmed the accuracy of DV determined in Section 2.3.3.3

which allowed proper adjustments of the amount of base required to account for

different densities of additives added.

3.4.5 Viscosity

The bioadhesive polymers used in this study such as CBP and HPMC are known to

have thickening properties, thus are likely to alter the viscosities of the bases

determined in Section 2.3.1.5. Viscosity was generally found to be highest in CB

80
suppositories followed by CE and SS (Figures 3.4a-c). This could be due to the longer

carbon chain of the palmitic (C16) and stearic acid (C18) in CB, compared to lauric acid

(C12) in CE and SS (Lonchampt and Hartel, 2004; Peyronel and Marangoni, 2014).

(a)
140 *
120 Blank
* * *
100 * * * *
Viscosity (cp)

* * * * * * DcNa only
80 * *
60 1%
40
20 2%
0
5%
CBP HPMC PVP CMCTS

(b)
140
120 Blank
*
Viscosity (cp)

100
80
* DcNa only
* *
60 * * * * * * * 1%
*
40
20 2%
0
5%
CBP HPMC PVP CMCTS

(c)
140
120 Blank
100
Viscosity (cp)

* DcNa only
80 *
* *
60 1%
* * * * * * * * *
40
2%
20
0 5%
CBP HPMC PVP CMCTS

Figure 3.4 : The viscosity (cp) of suppositories made using (a) CB; (b) CE; and (c) SS;

each suppository contained 50 mg DcNa and 1-5 %w/w of bioadhesive polymers

(CBP, HPMC, PVP and CMCTS). Asterisks indicate formulations which are

significantly different from blank suppositories.

81
The addition of DcNa significantly increased viscosity of CB suppositories but

conversely, reduced the viscosity of CE and SS suppositories. Suppositories

containing DcNa and bioadhesive polymer was generally more viscous than their

corresponding DcNa only suppositories. CB suppositories incorporated with HPMC

(Figures 3.4a) were most viscous while CE and SS suppositories incorporated with

CBP (Figures 3.4b-c) were most viscous compared to suppositories with other

bioadhesive polymers. The same polymer affected the viscosity of the suppositories

differently. While PVP and CMCTS did not alter the viscosity of CE and SS

suppositories drastically, it resulted in a significant concentration dependent increase

in viscosity of suppositories when formulated with CB (Figures 3.4a-c). Generally, an

increase in viscosity was observed when concentration of polymer is increased from 1

to 5 %w/w. Statistical comparison of the formulations using Tukey’s HSD analysis is

tabulated in Appendices 8-10.

More viscous suppositories may be beneficial to confine drug absorption within the

lower rectum by limiting spread of the molten. However, excessively viscous

suppositories may also impede DcNa release from the base and thus, drug release

studies are required to determine if this increase in viscosity is unfavourable for drug

release (Azechi et al., 2000).

3.4.6 Hardness

Hardness measurements obtained from the tip of suppositories were omitted from

subsequent analysis as the results were less distinguishable between subsets (in

Tukey’s HSD analysis) compared to values obtained at the suppository barrel (Figures

3.5a-c). The values obtained depended on the crushed surface area at which the force

82
exerts on and this is made difficult by the cone-shaped pointed tip as the surface area

varies along the length of the tip.

(a) 140 *
* * Blank
120
* * * * * *
100 * * DcNa only
Hardness (N)

* * * * *
80
60 1%

40
2%
20
0 5%
CBP HPMC PVP CMCTS

(b) 140
* Blank
120 * *
* * *
100
DcNa only
Hardness (N)

80
60 1%
40
2%
20
0 5%
CBP HPMC PVP CMCTS

(c) 140
120 Blank
* * * * *
100 * * *
Hardness (N)

* * DcNa only
80 * * * * * *
60 1%
40
2%
20
0 5%
CBP HPMC PVP CMCTS

Figure 3.5: The hardness of suppositories produced using (a) CB; (b) CE and (c) SS;

each suppository contained 50 mg DcNa and 1-5 %w/w of bioadhesive polymers

(CBP, HPMC, PVP and CMCTS). Mean ± SD, n=6. Asterisks indicate formulations

which are significantly different from blank suppositories.

83
Figures 3.5a-c showed that blank CE suppositories (82.33 ± 5.20 N) were significantly

harder than the CB and SS suppositories (59.00 ± 2.97 N and 55.83 ± 1.60 N

respectively). Addition of DcNa increased the hardness of all three (CB, CE, SS)

suppositories. Statistical comparison of the formulations using Tukey’s HSD analysis

is tabulated in Appendices 11-13.

Suppositories incorporated with 1-5 %w/w of CMCTS for all three bases (CB, CE and

SS) were hardest compared to those containing other polymers (Figures 3.5a-c).

Addition of CBP showed an irregular hardness trend in all three bases. The presence

of CBP significantly increased the hardness of CB suppositories; however the

hardness of CE suppositories were not affected by addition of 1-2 %w/w CBP while

SS suppositories were harder with 1-2 %w/w of CBP but not in the presence of

5 %w/w CBP (Figure 3.5c). On the other hand, formulations containing PVP were

generally harder than those containing CBP and HPMC. HPMC suppositories were

the least hard (Figures 3.5a-c). There was however, no observable trend between

increment in polymer content and hardness of suppositories produced.

Sirisa-ard et al. (2014) found that suppositories produced using Krabok wax resulted

in hardness of 6 N were too soft and easily broken, while Allen et al. (2008) suggested

that suppositories should have hardness of 17-20 N. Meanwhile, studies by El-Majri

and Sharma (2010); Oribe et al. (1995); Shegokar and Singh (2010) have reported a

wide range of suppository hardness ranging from 10 to 60 N. These large differences

of reported values were a result of differences in experimental methods as well as

variability in the type of base and additives used. Although the addition of bioadhesive

polymers have variable effects on the hardness of the bases, all the formulations in

84
this study had hardness values between 60 to 130 N, suggestive of good resistance

towards the hazards of manufacturing and packing (Sirisa-ard et al., 2014).

3.4.7 Softening time

Figure 3.6 showed that all the formulations tested had a softening time between 3-7

minutes. Softening time of blank suppositories was found to be shortest in CB,

followed by SS and CE; the post hoc Tukey’s HSD test indicated that these outcomes

were significantly different (Appendices 14-16). However, for suppositories

containing 50 mg DcNa, the softening point of CB suppositories were increased while

a decrease in softening time was observed for CE and SS suppositories.

In all the formulations containing polymers, softening time for CB suppositories were

always faster than the corresponding formulations for CE and SS; both of which were

comparable (CB < SS = CE). Generally, there was an increase in softening time with

increasing polymer concentration. CE and SS suppositories incorporated with 5 %w/w

HPMC showed the longest softening time among all the formulations tested.

The longer softening time in suppositories made using CE and SS (4-7 minutes) is an

indication that the HPKS suppositories were slightly better at withstanding short

exposures to heat compared to CB (3-5 minutes) suppositories. The British

Pharmacopoeia ( 2015) has recommended that the softening time of lipophilic

suppositories should be less than 30 minutes while studies by Janicki et al. (2001) and

Moghimipour et al. (2009) produced suppositories with softening time which ranged

between 4-13 minutes and 11-16 minutes respectively. Despite being harder than

85
formulations produced in similar studies (Section 4.3.6), all the formulations produced

in this study had softening times below 7 minutes.

(a)
7
Blank
Softening time (min)

6
5 * * * * * * * * * * * * * * * * DcNa only
4
1%
3
2 2%
1 5%
0
CBP HPMC PVP CMCTS

(b)
7 * Blank
6 * * * * *
Softening time (min)

5 ** * * * * * * * DcNa only
*
4 1%
3
2%
2
1 5%
0
CBP HPMC PVP CMCTS

(c)
7
6 * Blank
* * *
Softening time (min)

5 * * * * * * * * * * * DcNa only
*
4
1%
3
2 2%
1
5%
0
CBP HPMC PVP CMCTS

Figure 3.6 : The softening time of suppository formulations produced using (a) CB;

(b) CE and (c) SS. Each suppository contained 50mg DcNa and 1-5 %w/w of

bioadhesive polymers (CBP, HPMC, PVP and CMCTS). Mean ± SD, n=3. Asterisks

indicate formulations which are significantly different from blank suppositories.

86
3.5 Conclusion

Suppositories containing bioadhesive polymers (CBP, PVP, HPMC, CMCTS) were

successfully manufactured using CB and both the HPKS (CE and SS) as suppository

bases through the optimised manufacturing methods from Section 2.3.3.

Dosage form analysis is essential for the evaluation of product quality. Suppositories

manufactured in this study have good weight uniformity and there was minimal

variation (± 1 %) between actual weight and the respective target weight. The melting

point of suppositories manufactured using CE and SS were 1-2 ºC higher than CB

suppositories but all formulations had melting points within the range of 32.5 °C to

35.5 °C. The addition of bioadhesive polymers did not significantly alter melting point

(less than ± 1 °C) of suppositories, maintaining a melting point close to body

temperature which is suitable for rectal administration. All the formulations tested

contained > 95 % of the stipulated DcNa content.

In general, there was no consistent trend between hardness and increasing

concentrations of bioadhesive polymers in the suppositories (CB, CE and SS) based

on the post hoc Tukey’s HSD analysis (p < 0.05). Conversely, viscosity of the molten

suppositories increased with increasing amount of bioadhesive polymers (CBP, PVP,

HPMC, CMCTS) incorporated into the formulations. Softening times on the other

hand, were longer in suppositories produced using the HPKS (CE and SS) than CB.

However, the softening times for all formulations were between 3–7 minutes which

was acceptable for rectal drug administration.

87
Physical properties of the suppositories examined in this chapter were found to be

suitable for rectal administration. The longer softening time in HPKS suppositories

coupled with a melting point of 34.5 ± 1 °C is an added advantage compared to CB

suppositories. This offers more resistance to melting or damage during handling prior

to insertion into the rectum. Although the physical properties are well characterised in

this chapter, further studies are required to examine the impact of various additives on

the drug release (Chapter 4) and bioadhesive properties (Chapter 5).

88
CHAPTER 4

IN VITRO DRUG RELEASE AND

RELEASE KINETICS

89
4.1 Introduction

4.1.1 Drug release studies

After suppository administration into the rectum, the base has to first melt to allow

diffusion of drug from the base into rectal environment before absorption through

rectal mucosa into systemic circulation (Allen et al., 2008). Previous studies have

found that data obtained from in vitro drug release studies for suppositories were well

correlated to plasma concentration profiles (Aiache et al., 1987; Babar et al., 1999;

Chicco et al., 1999; Gjellan et al., 1994; Vidras et al., 1982). Therefore, these studies

are useful for prediction of dosage form performance.

Various methods exist for studying in vitro drug release from suppositories. Among

those commonly used are the USP paddle apparatus (Ahmad, 2001; Moghimipour et

al., 2009; Young et al., 1987), basket apparatus (Babar et al., 1999; Ermiş and Tarimci,

1995; Gjellan et al., 1994; Swamy et al., 2012), dialysis rotating cell method (Aoyagi

et al., 1988; Lootvoet et al., 1992; Oribe et al., 1995), Muranishi method (Ermiş and

Tarimci, 1995; Umeda et al., 1985), and flow through cell method (Aiache et al.,

1987; Mollel, 2006; Tukker et al., 1981; Yahagi et al., 1999).

Apart from type of apparatus; the type, pH and volume of dissolution medium used;

temperature of the in vitro run as well as method of quantification of active drug has

to be carefully selected based on the type of dosage form studied.

90
4.1.2 Statistical comparison and mathematical modelling

The three main methods used to analyse data obtained from drug release studies are:

(1) exploratory data analysis; (2) model–independent and; (3) model-dependent

methods.

4.1.2.1 Exploratory data analysis

As the term suggests, exploratory data analysis is used during initial interpretation of

raw data to provide basic understanding of how drug is released from a formulation.

This method compares drug release both graphically and numerically.

In graphical comparison, cumulative drug release against time plots were visually

inspected for overlapping of error bars (extending two standard errors (SE) either side

of the mean) between the drug release profiles at each time point. Non overlapping of

error bars indicates that drug release profiles at that particular time point are

significantly different from each other (O’Hara et al., 1998; Yuksel et al., 2000).

Exploratory data analysis can also be conducted numerically by summarising data in

the form of 95 % confidence interval for the difference in the mean drug release

profiles at each time point. Difference at a particular time point is considered

significant when the p> 0.05 (O’Hara et al., 1998).

4.1.2.2 Model-independent methods

This method involved computing of raw data into mathematical formulae to obtain

single-value measurements of differences between two multiple point drug release

91
profiles. This is further differentiated into amodelistic dissolution parameters (ratio

test) and pair-wise procedures (Costa and Sousa Lobo, 2001; O’Hara et al., 1998).

Amodelistic parameters compare profiles using dissolution efficiency (DE) and mean

dissolution time (MDT) (Costa and Sousa Lobo, 2001; Mollel, 2006; Vaghela et al.,

2011), both determined using the following equations:

𝑡2
∫𝑡 𝑦 𝑑𝑡
1
Equation 4.1 𝐷𝐸 = 𝑥 100%
𝑦100 ∗ (𝑡2− 𝑡1 )

Where,

y = percentage of drug released at time t2

y100 = maximum amount of drug available for release

𝑛
𝑀𝑡
Equation 4.2 𝑀𝐷𝑇 = ∑ 𝑡̂𝑖
𝑀∞
𝑖=1

Where,

n = number of dissolution samples

i = sample number

𝑡̂𝑖 = time at midpoint between ti and ti-1

Mt = fraction of dose released at time ti

𝑀∞ = dose of formulation

Pair-wise procedures compare profiles in terms of ‘fit factors’ and Rescigno’s indices

(ξ). The ‘fit factors’ or ‘similarity indices’ known as ‘difference factor’ (ƒ1) and the

92
‘similarity factor’ (ƒ2), both described by Moore and Flanner using Equation 4.3 and

Equation 4.4 (Anderson et al., 1998; Costa, 2001; Moore and Flanner, 1996).

∑𝑛𝑡=1 𝑤𝑡 |𝑅𝑡 − 𝑇𝑡 |
Equation 4.3 𝑓1 = { } 𝑥 100
∑𝑛𝑡=1 𝑤𝑡 𝑅𝑡

−0.5
1 2
Equation 4.4 𝑓2 = 50 𝑙𝑜𝑔10 {[1 + ∑ 𝑤𝑡 (𝑅𝑡 − 𝑇𝑡 ) ] 𝑥 100}
𝑛

Where,

Rt = percentage of drug release for reference sample

Tt = percentage of drug release for test sample

n = number of sampling points taken into account

wt = optional weight factor which is usually kept at 1

The ƒ1 represents the average percentage difference between test formulation (Tt) and

reference formulation (Rt) across all time points analysed while the ƒ2 is a log function

of differences between the compared profiles and can assume any value between 0

and 100. The Food and Drug Administration (FDA) has set a standard where ƒ1

between 0 to 15 and ƒ2 between 50 to 100 indicate similarity between two drug release

profiles (Anderson et al., 1998; Costa and Sousa Lobo, 2001; Ministry of Health

Malaysia, 2000).

93
4.1.2.3 Model–dependent methods

In the model-dependent approach, data obtained from drug release studies are fitted

into selected mathematical equations describing release kinetics. Various parameters

are generated to determine goodness of fit of data to the selected kinetic equation.

These parameters include release constant, linear coefficient of determination (r2) and

lag time which describes the rate of release as well as the proximity of data to the

release model tested. It was suggested that this approach requires a minimum of four

data points excluding time point ‘zero’ up till 80 % of drug release or when asymptote

is achieved (O’Hara et al., 1998; Yuksel et al., 2000).

4.1.2.3.1 First order kinetics

First order release kinetics describes drug release in a concentration-dependent

manner, this was first described by Gibaldi and Feldman (1967) based on the Noyes-

Whitney equation where drug release occurs at a constant proportion to the amount

remaining within the dosage form (Costa and Sousa Lobo, 2001). This kinetic model

has been used to describe release of water soluble drugs from porous matrices, and

also lipophilic suppositories (Tarimci and Ermis, 1997). First order release is

explained by the following equation (Basak et al., 2008; Dash et al., 2010);

𝐾1 . 𝑡
Equation 4.5 𝑙𝑜𝑔10 𝐶𝑡 = 𝑙𝑜𝑔10 𝐶0 −
2.303
Where,

Ct = amount of drug remaining to be released at time t

C0 = initial amount of drug in the dosage form (usually 100 %)

K1 = first order rate constant

94
4.1.2.3.2 Higuchi release

The Higuchi model was initially developed to quantify mass transport of

homogenously dispersed drug particles in a planar matrix into perfect sink conditions

based on Fick’s Law (Dokoumetzidis and Macheras, 2006; Kalam et al., 2007;

Siepmann and Peppas, 2011). This model can be described by the simplified equation

(Costa and Sousa Lobo, 2001);

Equation 4.6 𝑄𝑡 = 𝐾𝐻 . 𝑡1/2

Where,

Qt = amount of cumulative drug released at time t

KH = Higuchi release constant

This equation assumes that drug release occurs in a thin, planar geometry, where drug

movement through the matrix (base) is a rate limiting step, provided that (1) amount

of drug in the dosage forms greatly exceeds drug solubility to achieve pseudo-steady-

state and, (2) the base does not swell or dissolve during drug release so that distance

of the base-medium interface remains unaltered (Siepmann and Peppas, 2011). The

Higuchi equation is only valid for analysis of the first 60 % of drug release.

Previous studies on drug release from lipophilic suppositories have suggested that the

release were in agreement with Higuchi equation (Nakajima et al., 1990; Takatori et

al., 2004). This equation could be useful to describe drug release from suppositories as

the suppositories melt and form a thin layer of molten base matrix containing

suspended drug particles, similar to that observed in the ointment-skin interface from

which the Higuchi equation stems from.

95
4.1.2.3.3 Korsmeyer – Peppas model

This mathematical model was initially developed by (Korsmeyer et al., 1983) to

explain diffusion-controlled drug release mechanisms from polymeric systems, and

subsequent studies lead to characterisation of the type of diffusional release based on

the release exponent ‘n’, taking into account variable geometries and swelling

capacities of dosage forms. A number of lipophilic suppositories demonstrated release

patterns with good fit to the Korsmeyers-Peppas model (Güneri et al., 2004; Mollel,

2006; Uzunkaya and Bergişadi, 2003). The release kinetics can be described as,

𝑀𝑡
Equation 4.7 log = n. log 𝑡 + log 𝑘𝑚
𝑀∞

Where,

Mt / M ∞ = fraction of drug released at time t

Km = release constant

n = release exponent

In suppositories (non-swellable cylindrical dosage forms); the ‘n’ value limits of n =

0.45 for Fickian diffusion; 0.45 < n < 1.0 non- Fickian or anomalous diffusion; n ≥ 1

for case-2 transport (zero order release) were adopted.

4.1.2.3.4 Weibull model

The Weibull model has been used empirically to describe drug release from Euclidian

and fractal matrices; despite criticism of the lack of kinetic basis and non-dissolution

specific nature of its parameters (Costa et al., 2003). Use of Weibull function in drug

release stemmed from the concept where a concentration gradient is present at the

96
releasing boundaries of dosage forms, which are either described as a Euclidian matrix

or fractal geometry (Kosmidis et al., 2003). The Weibull model is described by the

following equation;

Equation 4.8 −(𝑡 − 𝑇𝑙𝑎𝑔 )𝛽

𝑚 = 1 − exp(
𝛼

Where,

m = amount of drug released

α = scale parameter (rate constant)

β = shape parameter

Tlag = lag time

Description of drug release using the Weibull model is through the scale parameter (α)

and shape parameter (β) which represents the apparent rate of release and shape of the

curve respectively. When β=1, the curve corresponds exactly to a homogeneous model

similar to that of first order kinetics; while β>1 indicates a sigmoidal curve with an

inflexion point; and β<1 indicates a steeper initial slope than exponential curves

(Cupera, 2009; Dash et al., 2010).

4.1.2.3.5 Bi-exponential first-order kinetic model

This model had been used to describe drug release from fenbufen suppositories made

using lipophilic, hydrophilic and ampiphilic bases as well as in sustained release

indomethacin tablets and capsules (Laakso et al., 1984; Young et al., 1987). The

equation defining bi-exponential first-order kinetics is as the following;

97
 Equation 4.9 𝑤 = 𝐴𝑒 −𝑘1 𝑡 + 𝐵𝑒 −𝑘2 𝑡

Where,

w = drug remaining to be released at time t

k1 = release rate constant for initial phase

k2 = release rate constant for terminal phase

A = amount of drug released in initial phase

B = amount of drug released in terminal phase

The plot of remaining drug against time would generate a biphasic curve. Parameters

generated from linear regression would yield release constants, k1 and k2 which

reflects rate of release at each phase and their respective coefficient of determination

(r2) which indicates closeness of fit to the equation while the y-axis intercept

corresponds to lag time. A good fit to the equation is represented by r2 approximating

to 1 and small, reasonable lag times (Laakso et al., 1984).

4.1.3 Selection of the best fit model

To determine the best fit model, previous studies employed coefficient of

determination (r2) or adjusted coefficient of determination (r2 adjusted) (Ladani et al.,

2011), sum of squares residues (SSR) (Thakkar et al., 2009), mean square error (MSE),

Akaike Information Criterion (AIC) (Ozkan et al., 2000) and more recently the ratio

of SSR/r2 (Costa and Sousa Lobo, 2001; Singh et al., 2011). The r2 measures

proportion of variation between observed data and the mean generated through linear

regression model. The greater approximation of r2 to 1, the closer the fit of data to a

particular model (Ladani et al., 2011). The AIC, on the other hand, shows the quality

98
of fit by comparing models using the same weighting scheme. The model which

produces the smallest AIC value is regarded to provide the best fit out of a set of

models tested (Costa, 2001).

Within the context of this research project, in vitro drug release studies of DcNa from

the HPKS bases (CE and SS) were essential as the suppository bases used were non-

conventional. There were no reported data on drug release characteristics from these

bases and how they fared compared to CB. Furthermore, the inclusion of a

bioadhesive component in the form of polymers could alter drug release profiles. Thus,

this chapter aims to quantify DcNa release from the formulation prototypes developed

in Chapter 3 using suitable dissolution study methods, followed by comparison of

DcNa release kinetics from the formulations using exploratory data analysis methods,

model independent methods and mathematical models. This chapter will also assess

the effects of bioadhesive polymers (CBP, HPMC, PVP and CMCTS) on drug release

kinetics.

4.2 Materials and methods

4.2.1 Dissolution setup and evaluation of formulations

4.2.1.1 Experimental setup

The in vitro release of DcNa from suppositories was studied using the 8-vessel Distek

dissolution system 2100c (Distek Inc., New Jersey, USA) fitted with thermocirculator

TCS 0200 (Distek Inc., New Jersey, USA). The system is connected to Distek

Evolution 4300 syringe pump dissolution sampler (Distek Inc., New Jersey, USA).

99
Suppositories were enclosed in helix-shaped sinkers made using steel wires with

dimensions (length 2.8 cm x diameter 1.0 cm; interloop distance of 0.5 cm) to ensure

retention of suppositories at the bottom of the vessel. Dissolution of a blank

suppository made with only the base (CB, CE and SS) was used as experimental blank.

The parameters employed during dissolution studies were summarized in Table 4.1.

Samples were then analysed using a single cell UV-Visible spectrophotometer (Libra,

S12) at a wavelength of 276 nm as determined in Chapter 2.

Table 4.1 : Summary of parameters used for in vitro drug release studies.

Experimental Parameters Settings

Dissolution tester USP Apparatus II – Paddle
Dissolution media Distilled water
Volume of media 900 mL
Temperature 37.0 ± 0.5 °C
Paddle rotation speed 50.0 ± 0.2 rpm
Filter pore size 0.45 μm
Replicates 6 suppositories per experiment
Sampling time (min) 2, 6, 10, 15, 30, 45, 60, 90, 120, 180, 240, 300, 360
Sampling volume (mL) 5 mL, reconstituted

4.2.1.2 Evaluation of formulations

4.2.1.2.1 The effects of drug loading on DcNa release

The effects of drug loading on DcNa release was evaluated using suppositories made

from all three bases (CB, CE, SS) containing 25, 50 and 75 mg DcNa using

procedures mentioned in Section 4.2.1.1. Data obtained was analysed using

exploratory data analysis method.

100
4.2.1.2.2 The effects of bioadhesive polymers on DcNa release

DcNa release from the formulations containing bioadhesive polymers (CBP, HPMC,

PVP, CMCTS) at concentrations of 1-5 %w/w were studied using methods described

in Section 4.2.1.1. A suppository containing the corresponding amount of bioadhesive

polymers without DcNa was used as experimental blank. Comparisons were carried

out via model-independent methods and model dependent methods.

4.2.2 Statistical and mathematical analysis of data

4.2.2.1 Exploratory data analysis

Graphical comparison was employed to analyse DcNa release from suppositories (CB,

CE and SS) incorporated with 25, 50 and 75 mg of DcNa. Percentage cumulative drug

release was plotted against time with error bars indicative of two SE at both sides of

the error bar (95 % confidence interval) at each time point (Mollel, 2006; O’Hara et

al., 1998). The graph is then visually inspected for overlapping of error bars.

4.2.2.2 Model-independent method

The fit-factors (ƒ1 and ƒ2) were used to identify dissimilar drug release profiles by

comparing the Tt formulations containing bioadhesive polymers (CBP, HPMC, PVP,

CMCTS) and their corresponding Rt (suppositories containing only base and DcNa).

Meanwhile, other model-independent parameters such as DE and MDT were used to

describe the rate and extent of DcNa release from both Rt and Tt suppositories.

4.2.2.3 Model dependent method

Drug release data from all formulations were fitted into first order, Higuchi’s release,

Korsmeyer–Peppas model, Weibull’s equation and biphasic first-order release

101
equations using KinetDS® software version 3.0 (Aleksander Mendyk, Kraków,

Poland.). Goodness of fit for the various models were investigated and compared in

terms of r2 and AIC values.

4.3 Results and discussion

4.3.1 Method development

While the United States Pharmacopeia and National Formulary, (2008) and European

Pharmacopoeia, (2010) specified the use of paddle dissolution apparatus and flow

through cell apparatus respectively to study drug release from suppositories; various

alternatives have been attempted (Gjellan and Graffner, 1994; Nair and Bhargava,

1999; Palmieri, 1981; Webster et al., 1998). Palmieri (1981) however, reported erratic

and irreproducible results using the basket apparatus due to clogging of basket mesh

by melted base; while Gjellan and Graffner (1994) found that the flow through cell

resulted in more rapid drug release with a larger variance in data compared to basket

and paddle apparatus.

The rectal environment is simulated by the receptor medium during in vitro drug

release studies. Although it should mimic physiological environment closely, it is

impractical to carry out drug release studies under stringent rectal physiological

parameters especially when only 2-3 mL of mucus is present in the rectum. The use of

2-3 mL receptor media will not be able to provide ‘sink’ conditions for drug release

which should be at least 3 times the solubility of the drug tested (Brown et al., 2011;

Lee et al., 2008; Vaghela et al., 2011).

102
Meanwhile the rectal pH also affects ionisation of drug and its partitioning out of the

base which could alter drug release. pH of the rectum varies with age and has been

reported to be within the range of 6.29–6.45 in neonates; 6.68–7.12 in infants older

than 28 days while rectal pH in children aged between 1 to 14 years ranged between

7.2-12.1 (Jantzen et al., 1989; Turner et al., 2012). Adult rectal pH on the other hand

is approximately 7.2, but varies according to colonic content (Desai, 2007). Since the

rectum is void of buffering capacity; use of a buffered medium would not reflect

actual drug release conditions within the rectum, especially when aqueous solubility

of DcNa is pH dependant (Chuasuwan et al., 2009).

After careful considerations of the factors mentioned above, this research employed

the USP II paddle dissolution apparatus with the use of sinkers due to considerations

that lipophilic bases and bioadhesive polymers used are likely to clog the basket

apparatus mesh. Distilled water maintained at 37.0 ± 0.5 °C was selected as the

receptor medium for drug release studies to reflect lack of buffering capacity and

temperature within the rectum (Allen et al., 2008; Bottger et al., 1989; Grayson, 1951).

4.3.2 Exploratory data analysis

4.3.2.1 The effects of drug loading on DcNa release

Drug release profiles of suppositories incorporated with 25, 50 and 75 mg of DcNa

were superimposable upon visual inspection for all the three bases (CB, CE and SS)

investigated (Figures 4.1a-c). However, the exploratory analysis method produced

inconclusive results as there was overlapping of error bars at certain time points (no

significant difference) while others were significantly different.

103
(a)
100

Cumulative DcNa Release (%)

80

60
25 mg
40
50 mg
20 75 mg

0
0 60 120 180

Time (min)

(b)
100
Cumulative DcNa Release (%)

80

60 25 mg

40 50 mg
75 mg
20

0
0 60 120 180
Time (min)

(c)
100
Cumulative DcNa Release (%)

80

60 25 mg
50 mg
40
75 mg
20

0
0 60 120 180
Time (min)

Figure 4.1 : Cumulative percentage release of DcNa in (a) CB; (b) CE; (c) SS

suppositories containing 25, 50 and 75 mg of DcNa. Mean ± 2 SE, n=6.

104
Further ANOVA analysis followed by post hoc Tukey’s test showed that extent of

DcNa release from bases was not affected by initial amount of DcNa incorporated.

Figure 4.1a showed that although CB suppositories with 25 mg DcNa had

significantly lower initial rate of release from 10 to 60 minutes compared to those with

50 and 75 mg DcNa (p< 0.05), the extent of DcNa release was indifferent at 180

minutes (p = 0.155).

CE and SS suppositories containing 25 mg DcNa on the other hand, had significantly

lower release between 6 to 15 minutes (p< 0.05) and 15 to 60 minutes (p< 0.05)

respectively compared to the corresponding formulations containing 50 and 75 mg of

DcNa (Figures 4.1b-c). The extent of DcNa released at 180 minutes was not

significantly different between the three CE formulations (p= 0.977) and the three SS

formulations (p= 0.06). This indicated that while drug load did not affect the extent of

DcNa release; the initial rate of release was increased when drug loading is doubled

from 25 mg to 50 mg, but this effect was not significant when drug loading was

further increased to 75 mg.

This was further supported by the plot of actual DcNa released (Appendix 17) as each

time point interval showed that there was rapid release of DcNa during the first 45

minutes for all the suppositories. The actual amount released at each time interval

increased with increasing amount of DcNa incorporated (p< 0.05). This suggested that

initial release of DcNa did not occur at a fixed rate (zero order). The anomalous

increase in actual amount of DcNa released at time point interval of 30 minutes could

be a result of change in surface area for DcNa release due to melting and molten

accumulation at the dissolution media-air interface in the dissolution vessel. Further

105
investigation on kinetics of DcNa release was conducted using mathematical models

(Section 4.3.4).

Comparing DcNa release from suppositories containing 50 mg DcNa (Figure 4.2),

release profiles between CB and SS were not significantly different up to 240 minutes

while the profiles for CE were significantly different from CB and SS up to 120

minutes (p< 0.05). The release plateaued after 120 minutes in both CB and SS while it

was only achieved after 180 minutes in CE suppositories.

100
Cumulative DcNa Release (%)

80

60
CB
40 CE
SS
20

0
0 60 120 180 240
Time (min)

Figure 4.2 : DcNa release profiles from suppositories made with different bases, each

containing 50 mg of DcNa. Mean ± 2 SE, n=6.

CB, CE and SS suppositories containing 50 mg DcNa released more than 50 % DcNa

within the first 10 minutes, and more than 80 % within the first 60 minutes of

dissolution. Although there were no pharmacopoeia requirements for qualification as

fast-acting suppositories, the United States Pharmacopeia and National Formulary

(2008) required fast-acting tablets to achieve at least 75 % of drug release within the

106
first 60 minutes of dissolution. CB, CE and SS suppositories were found to provide

fast release of DcNa upon dissolution.

The ANOVA analysis with post hoc Tukey’s test used in this section was used to

provide descriptions for the observations via exploratory method of data interpretation.

However, this method was found to be tedious and generated large quantities of data

based on point-to-point comparisons between two profiles. The method did not

consider drug release process as a whole but provided only comparison at a specific

time point which limits its practicality when more than two time points or release

profiles are being compared.

4.3.3 Model- independent method

Due to difficulties in simultaneous interpretation of multiple drug release profiles

using methods in Section 4.3.2, fit factors (f1 and f2) were used to compare

dissimilarities between dissolution profiles in the course of this study. Subsequently,

DE and MDT were calculated for evaluation of the extent and rate of DcNa release of

each formulation. The corresponding drug release profiles were attached in

Appendices 18-21 for reference.

4.3.3.1 Effects of bioadhesive polymers on DcNa release

4.3.3.1.1 Fit factors (ƒ1 and ƒ2)

This method had been endorsed by both FDA and Malaysian Ministry of Health for

model-independent comparison of drug release profiles (Ministry of Health, 2000; US

Department of Health and Human Services & CDER, 1997).

107
Comparisons using fit factors (Table 4.2) were made between the Rt which contains

only 50 mg DcNa, against Tt which are incorporated with 1-5 %w/w bioadhesive

polymer. Table 4.2 showed that drug release profiles between CB-SS suppositories

and CE-SS suppositories appear more similar than CB-CE. The addition of CBP

increasingly altered drug release profiles with increasing polymer concentrations,

resulting in ƒ1 > 15 and ƒ2 < 50 for suppositories made with all three bases.

Generally, incorporation of 1-5 %w/w HPMC, PVP and CMCTS to CE and SS did

not alter drug release profiles compared to their respective Rt. In CB suppositories, the

addition of 1-2 %w/w HPMC, PVP and CMCTS did not alter dissolution profiles; but

when concentration of polymer was increased to 5 %w/w, DcNa release for all three

formulations were significantly different from Rt (ƒ1 > 15 and ƒ2 < 50).

Although fit factors were effective in describing the similarities and dissimilarities

between profiles, it was unable to reflect the reason of dissimilarities in terms of rate

and extent (maximum amount) of drug release or kinetics of drug release. In order to

understand DcNa release from these suppositories, it is thus essential to assess the DE

(Section 4.3.3.1.2) and MDT (Section 4.3.3.1.3) for evaluation of the extent and rate

of DcNa release from each formulation.

108
Table 4.2 : Statistical comparison of formulations containing DcNa only (Rt) and

formulations with DcNa and bioadhesive polymers (Tt). Dissolution profiles are

considered similar when; 0 < ƒ1 < 15 and 50 < ƒ2 < 100. The highlighted ƒ1 and ƒ2

values indicate similar dissolution profiles between Rt and Tt.

Formulation
Rt Tt Parameters
Polymer amount (%w/w)
Base
CBP HPMC PVP CMCTS ƒ1 ƒ2
CB CE 0 0 0 0 17.389 47.450
CB SS 0 0 0 0 3.729 78.410
CE SS 0 0 0 0 14.652 52.490
CB CB 1 - - - 36.418 29.433
CB CB 2 - - - 49.455 23.911
CB CB 5 - - - 88.266 1.867
CB CB - 1 - - 9.861 55.968
CB CB - 2 - - 7.395 61.663
CB CB - 5 - - 28.995 34.874
CB CB - - 1 - 14.161 48.072
CB CB - - 2 - 19.774 41.019
CB CB - - 5 - 25.740 35.515
CB CB - - - 1 10.914 55.759
CB CB - - - 2 9.693 57.316
CB CB - - - 5 22.288 38.843
CE CE 1 - - - 32.530 35.763
CE CE 2 - - - 66.797 21.675
CE CE 5 - - - 83.739 16.724
CE CE - 1 - - 12.553 55.443
CE CE - 2 - - 8.311 64.993
CE CE - 5 - - 7.086 65.839

109
“Table 4.2 : Continued…”

Formulation

Rt Tt Parameters

Polymer amount (% w/w)

Base
CBP HPMC PVP CMCTS ƒ1 ƒ2

CE CE - - 1 - 4.438 75.840

CE CE - - 2 - 4.318 74.878

CE CE - - 5 - 3.883 80.308

CE CE - - - 1 5.691 71.193

CE CE - - - 2 13.039 54.323

CE CE - - - 5 2.362 83.401

SS SS 1 - - - 38.049 29.091

SS SS 2 - - - 71.912 15.819

SS SS 5 - - - 89.125 11.003

SS SS - 1 - - 3.873 76.739

SS SS - 2 - - 3.269 79.135

SS SS - 5 - - 5.233 70.184

SS SS - - 1 - 5.967 67.582

SS SS - - 2 - 5.956 66.663

SS SS - - 5 - 6.636 65.966

SS SS - - - 1 10.454 57.512

SS SS - - - 2 6.524 68.781

SS SS - - - 5 3.937 78.985

110
4.3.3.1.2 Dissolution efficiency (DE)

The DE (Figure 4.3) provided an insight on the amount of DcNa released over a time

period, reflecting the extent of drug release from the formulations examined. Among

DcNa only formulations, CE showed a significantly lower DE compared to both CB

and SS; this correlated well to the observations from Figure 4.2 where DcNa release

rates were lower in CE. Statistical comparison of the formulations using Tukey’s HSD

analysis is tabulated in Appendices 22-24.

In general, the addition of CBP reduced extent of DcNa release in a concentration

dependent manner in all three bases (Figures 4.3a–c). The DE of suppositories

containing 5 %w/w CBP was decreased by 60 % in CB and 80 % in both CE and SS

suppositories.

On the other hand, addition of 5 %w/w PVP resulted in a slight but significantly lower

DE in CB, CE and SS suppositories. While the incorporation of 1-5 %w/w HPMC did

not significantly alter the DE of CB suppositories; it resulted in a slight decrease in

DE for both CE and SS suppositories. CB suppositories incorporated with 1-5 %w/w

CMCTS had a significantly lower DE while CE suppositories containing 2-5 %w/w

CMCTS resulted in a higher DE. The DE of CB suppositories containing 1-5 % w/w

HPMC and SS suppositories containing 1-5 %w/w CMCTS remained unchanged

compared to DcNa only suppositories. There were however, no consistent trends on

the effects of HPMC, PVP and CMCTS on DE of all formulations.

111
 (a)
100 * * * * *
0%

Dissolution efficiency
80
* 1%
60 *

(%)
* 2%
40

20 5%
0
CBP HPMC PVP CMCTS

(b)
100 * * 0%
Dissolution efficiency

* * *
80
* 1%
60
(%)

40 * 2%
*
20 5%
0
CBP HPMC PVP CMCTS

(c)
100 * * * * 0%
Dissolution efficiency

80
* 1%
60
(%)

40 2%
*
20 * 5%
0
CBP HPMC PVP CMCTS

Figure 4.3 : The DE of suppository formulations containing 50 mg DcNa and 1-

5%w/w bioadhesive polymers (CBP, HPMC, PVP, CMCTS) in (a) CB; (b) CE; and

(c) SS. Asterisks indicate formulations which are significantly different from

formulations without polymer (DcNa only). Mean ± SD, n=6.

4.3.3.1.3 Mean dissolution time (MDT)

MDT reflects amount of time required for completion of drug release. The longer the

MDT, the slower the rate of DcNa release from suppositories. Figures 4.4a-c showed

112
the comparison of MDT generated for each formulation. Tukey’s HSD post hoc test at

95 % confidence interval is tabulated in Appendices 25-27.

(a) *
120
0%
Min dissolution time

100
80 * 1%
*
(min)

60
40 2%
* * * *
*
20
5%
0
CBP HPMC PVP CMCTS

(b)
120
Min dissolution time

0%
100
*
80 1%
(min)

60
* * 2%
40
* * * * *
20
5%
0
CBP HPMC PVP CMCTS

(c) 120
*
Min dissolution time

100 0%

80
1%
(min)

60 *
40 * 2%

20 * * *
5%
0
CBP HPMC PVP CMCTS

Figure 4.4 : The MDT of suppository formulations containing 50 mg DcNa and

1-5 %w/w bioadhesive polymers (CBP, HPMC, PVP, CMCTS) made from (a)

CB; (b) CE; and (c) SS. Asterisks indicate formulations which are significantly

different from formulations without polymer (DcNa only). Mean ± SD, n=6.

113
Figures 4.4a-c showed that formulations containing 1-5 %w/w CBP had significantly

longer MDT compared to DcNa only formulations. This increase in MDT caused by

incorporation of CBP was concentration dependent and was consistent in all three

bases. The higher the CBP content, the slower the rate of DcNa release (reflected by

longer MDT, Figures 4.4a-c) accompanied by a lowering in the maximum percentage

of DcNa released at the end of 6 hours (reflected by DE, Figures 4.3a-c). This

observation was similar to the report of impaired ramosetron release from

suppositories containing 2-10 %w/w of Carbopol 934P, where the authors

hypothesised the formation of a viscous CBP gel that suppressed drug release (Yahagi

et al., 2000).

Decrease in DcNa release with increasing CBP content could also be due to reduction

in DcNa solubility as a result of decreased receptor medium pH brought about by

dispersion of CBP. CBP is acidic and produces a solution with pH 2.7–3.5 at

0.5 %w/v (Lubrizol Advanced Materials, 2009). Kincl et al. (2004) reported that

solubility of DcNa at pH 5.8 can be reduced by up to 86 folds compared to at pH 8.0.

In general, the addition of HPMC, PVP and CMCTS at 1 %w/w did not significantly

alter MDT of formulations in comparison to DcNa only formulations, while 2-

5 %w/w of polymers resulted in small but significant changes in MDT (Figures 4.4a-

c). There was however, no clear trend of concentration-dependent alteration of MDT

by HPMC, PVP and CMCTS formulations.

114
4.3.4 Mathematical modelling

Although fit factors were effective in showing dissimilarities between profiles while

DE and MDT allowed direct comparison of the rate and extent of DcNa release

between profiles; release kinetics can only be explained by substituting drug release

data into mathematical equations.

Release parameters for the selected mathematical model (first order, Higuchi’s,

Korsmeyer-Peppas and Weibull’s mathematical models) after empirical fitting of each

formulation were tabulated in Tables 4.3-4.5. Higuchi’s equation and Korsmeyer-

Peppas model were fitted with up to 60 % of the drug dissolution data in conjunction

with the assumption that these two models are valid for description of cumulative drug

release of 60 % (Korsmeyer et al., 1983). Meanwhile, both the first order and

Weibull’s model were fitted with data up to the first time point at plateau (maximum

dissolution).

115
Table 4.3 : The goodness of fit parameters obtained from equation fitting of drug release data from all CB formulations.

Amount of Polymer (%w/w) First order Higuchi Korsmeyer- Peppas Weibull model
CBP HPMC PVP CMCTS r2 AIC r2 AIC r2 AIC r2 AIC
0 0 0 0 0.241 118.946 0.797 28.180 0.953 26.289 0.924 43.695
1 0 0 0 0.272 84.845 -1.739 100.043 0.647 79.800 0.988 50.021
2 0 0 0 0.309 126.062 -0.8502 105.469 0.720 83.61 0.938 81.581
5 0 0 0 0.3 150.781 0.728 109.206 0.892 119.661 0.872 109.000
0 1 0 0 0.49 71.925 0.876 33.536 0.943 35.357 0.870 22.522
0 2 0 0 0.532 74.000 0.779 35.813 0.956 34.117 0.862 10.463
0 5 0 0 0.621 72.642 0.842 32.951 0.985 26.452 0.792 24.267
0 0 1 0 0.51 71.557 0.875 33.067 0.946 34.616 0.870 22.428
0 0 2 0 0.611 68.686 0.908 28.720 0.952 30.523 0.841 34.969
0 0 5 0 0.558 73.484 0.799 34.870 0.967 31.623 0.851 12.567
0 0 0 1 0.393 82.570 0.674 29.759 0.977 24.096 0.927 41.995
0 0 0 2 0.418 82.956 0.745 28.414 0.978 22.881 0.901 33.132
0 0 0 5 0.523 76.730 0.681 38.436 0.960 35.066 0.923 48.734
116
Table 4.4 : The goodness of fit parameters obtained from equation fitting of drug release data from all CE formulations.

Amount of Polymer (%w/w) First order Higuchi Korsmeyer- Peppas Weibull model
CBP HPMC PVP CMCTS r2 AIC r2 AIC r2 AIC r2 AIC
0 0 0 0 0.306 118.806 0.765 27.229 0.966 23.620 0.946 54.683
1 0 0 0 0.212 144.910 -0.886 108.831 0.675 94.124 0.983 91.430
2 0 0 0 0.226 137.598 -1.003 151.483 0.692 133.769 0.937 104.763
5 0 0 0 0.319 142.782 0.827 112.236 0.824 134.667 0.859 111.005
0 1 0 0 0.438 76.189 0.522 39.593 0.915 40.066 0.948 39.330
0 2 0 0 0.411 74.169 0.731 37.803 0.903 40.125 0.947 34.930
0 5 0 0 0.494 72.340 0.775 35.636 0.943 35.743 0.908 33.777
0 0 1 0 0.476 80.340 0.883 32.101 0.950 33.517 0.890 35.355
0 0 2 0 0.369 83.874 0.731 37.803 0.957 40.125 0.950 42.744
0 0 5 0 0.382 82.377 0.757 36.793 0.903 39.416 0.953 41.415
0 0 0 1 0.377 84.159 0.687 38.283 0.957 40.693 0.954 37.455
0 0 0 2 0.430 74.676 0.590 30.996 0.970 26.301 0.899 35.078
0 0 0 5 0.417 81.403 0.777 35.864 0.910 38.400 0.935 39.156
117
Table 4.5 : The goodness of fit parameters obtained from equation fitting of drug release data from all SS formulations.

Amount of Polymer (%w/w) First order Higuchi Korsmeyer- Peppas Weibull model
CBP HPMC PVP CMCTS r2 AIC r2 AIC r2 AIC r2 AIC
0 0 0 0 0.229 119.01 0.729 29.433 0.952 27.754 0.938 61.568
1 0 0 0 0.213 143.137 -6.839 183.66 0.643 133.642 0.986 87.630
2 0 0 0 0.183 132.781 -1.748 150.488 0.616 131.717 0.932 110.567
5 0 0 0 0.291 125.585 0.554 110.627 0.798 117.730 0.926 91.910
0 1 0 0 0.439 67.757 0.834 26.59 0.934 26.488 0.922 36.212
0 2 0 0 0.401 68.72 0.786 28.116 0.926 27.885 0.948 35.007
0 5 0 0 0.459 69.178 0.810 26.985 0.958 24.801 0.919 29.679
0 0 1 0 0.421 75.566 0.857 25.833 0.923 26.627 0.895 47.908
0 0 2 0 0.408 75.584 0.851 25.974 0.916 26.955 0.925 41.593
0 0 5 0 0.409 75.025 0.868 25.225 0.924 26.167 0.945 33.686
0 0 0 1 0.345 71.832 0.763 29.905 0.966 26.243 0.953 44.333
0 0 0 2 0.377 70.348 0.815 28.102 0.962 25.600 0.952 38.441
0 0 0 5 0.374 70.717 0.771 28.900 0.949 27.092 0.962 33.666
118
The formulations containing only DcNa (no bioadhesive polymers) displayed the best

‘goodness of fit’ to the Korsmeyer-Peppas model (r2> 0.95), supported by the AIC

values. The order of increasing possibilities of fit to the kinetic equations based on

AIC values were first order (lowest), Weibull’s model, Higuchi’s equation,

Korsmeyer-Peppas (highest).

The n-values obtained through fitting of data into the Korsmeyer-Peppas equation

(Table 4.6) were within the range of 0.45–1.00, which was indicative of non-Fickian

diffusion of DcNa from the suppositories.

In general, the incorporation of HPMC, PVP and CMCTS did not affect the release

mechanism of DcNa from the suppositories. The DcNa release was predominantly via

non-Fickian diffusional methods as the Korsmeyer-Peppas release exponent (n),

ranged between 0.580–0.980 (Table 4.6). DcNa release from the suppositories were

therefore diffusion and erosion-controlled. These polymers (HPMC, PVP and

CMCTS) dissolve upon contact with dissolution media to form gaps which facilitates

erosion of the suppository matrix while DcNa simultaneously diffuse out of the DcNa-

drug matrix.

119
Table 4.6 : The release constant (km) and release exponent (n) of formulations fitted to
Korsmeyer- Peppas model.

Formulation Korsmeyer –Peppas

Base Amount of Polymer (%w/w) parameters
CB HPMC PVP CMCTS km n
0 0 0 7.945 0.878
1 0 0 7.772 0.756
2 0 0 4.382 0.896
5 0 0 5.157 0.795
0 1 0 7.285 0.759
0 2 0 8.096 0.584
0 5 0 6.569 0.662
0 0 1 4.489 0.949
0 0 2 5.649 0.953
0 0 5 3.228 0.980
CE 0 0 0 5.539 0.931
1 0 0 7.109 0.802
2 0 0 7.704 0.642
5 0 0 4.366 0.898
0 1 0 7.080 0.747
0 2 0 4.370 0.957
0 5 0 4.770 0.915
0 0 1 3.589 0.913
0 0 2 3.460 0.916
0 0 5 4.992 0.882
SS 0 0 0 6.221 0.972
1 0 0 9.442 0.792
2 0 0 8.287 0.860
5 0 0 7.390 0.862
0 1 0 7.668 0.895
0 2 0 8.073 0.878
0 5 0 8.179 0.862
0 0 1 7.644 0.934
0 0 2 8.714 0.856
0 0 5 7.405 0.914

120
Meanwhile, addition of 1-5 %w/w CBP resulted in a change of DcNa release

mechanism in CB, CE and SS where poor fit of the data to Korsmeyers–Peppas model

(r2 < 0.9) was observed (Tables 4.3-4.5). Instead, these data was better fitted to the

Weibull’s model (Table 4.7) which describes release of matrix-like drugs (higher R2

and smaller AIC). This could be due to the higher viscosity of molten suppository

mixture (in the presence of DcNa and CBP) coupled with gelling properties of CBP

upon contact with water which subsequently lead to trapping of DcNa within a base-

DcNa matrix.

Based on the Weibull’s model, Table 4.7 showed that as the concentration of CBP

increased, the scale factor (α) which corresponded to apparent rate constant decreased.

This was in good agreement with the findings in Sections 4.3.3.1.2 and 4.3.3.1.3 that

showed prolonged MDT and lowered DE as the concentration of CBP increased.

On the other hand, the shape dependence factor (β) were within the range of 0.3–0.7

(<1), which described the shape of drug release curves as having a steeper initial slope

than exponential release curves. The β value of the formulations containing 1-5 %w/w

CBP in this study reflected drug release in accordance to Fickian’s diffusion (β <

0.75), as quoted by Papadopoulou et al. (2006). As the suppository melts and CBP

starts to gel in contact with dissolution medium, it traps DcNa within the matrix. As

gelling continues, the gel layer forms a barrier which impedes DcNa release, similar to

that observed in drug release for devices with fractal geometry (Kosmidis et al., 2003).

DcNa release from the dosage form is therefore expected to be proportionate to the

fraction of particles that are sufficiently close to the barrier surface to diffuse from the

121
dosage form, down its concentration gradient (Kosmidis et al., 2003; Papadopoulou et

al., 2006).

Table 4.7 : The release parameters of formulations containing CBP fitted with Weibull

equation.

Formulation Weibull model parameters

CBP α β
Base
(%w/w) (time dependence factor) (shape dependence factor)

1 15.216 0.297

CB 2 10.702 0.334

5 0.417 0.754

1 10.321 0.370

CE 2 3.779 0.404

5 0.795 0.620

1 16.605 0.239

SS 2 3.636 0.391

5 0.916 0.534

Despite a generally better fit of formulations containing CBP to the Weibull model

(smaller AIC), the goodness of fit in terms (r2) of the formulations at higher

concentrations of CBP were still < 0.95 (Tables 4.3-4.5). Thus, further investigation

using a biphasic release model (bi-exponential first-order kinetic model) was

attempted. Parameters obtained from fitting into bi-exponential first-order kinetic

equation were described in Table 4.8.

122
Table 4.8 : The release parameters of formulations containing 1-5 %w/w CBP fitted

with bi-exponential first-order kinetic equation. All the r2 values for initial and

terminal phases were > 0.95 when goodness of fit of the data was reviewed by linear

regression.

Formulation Bi-exponential first-order kinetics parameters

CBP k1 k2 Lag time

Base A B
(%w/w) (min-1) (min-1) (min)

1 49.422 0.282 54.495 0.002 0.290

CB 2 42.423 0.200 61.271 0.001 0.451

5 28.968 0.026 74.344 0.001 4.185

1 52.305 0.217 56.821 0.002 0.870

CE 2 27.975 0.102 75.384 0.001 1.205

5 21.313 0.057 82.914 0.001 3.701

1 49.979 0.247 58.096 0.004 0.699

SS 2 28.011 0.108 74.911 0.001 1.000

5 11.642 0.062 89.407 0.001 1.427

The CBP formulations showed good linearity and reasonable lag times with bi-

exponential first-order kinetic equation. The results suggested that DcNa was released

from suppositories via a rapid initial phase (k1) followed by a slow terminal phase

drug release (k2). The addition of CBP increased lag times and decreased k1 in a

concentration dependent manner.

The bi-exponential first-order kinetic fitting was initially modelled to characterise

rapid intravenous injections. Over time, it has been used to describe the release of

123
drugs from solid dosage forms such as tablets and capsules (Laakso et al., 1984). In

tablets, the rapid initial release was described as a result of increasing surface area for

dissolution following tablet disintegration while the slower phase describes diffusion

of drug from the dosage form.

This concept can be adapted to explain DcNa release form CBP suppositories. During

dissolution, the suppository undergoes initial melting followed by formation of fatty

globules (containing molten base, DcNa and CBP). At this stage, DcNa release is via

both diffusion from the molten base as well as erosion or deformation of the dosage

form during the melting process (k1). But as CBP comes in contact with the

dissolution medium, it starts to gel and eventually form a barrier between the matrix

(base with DcNa) and dissolution medium, whereby DcNa can now only be released

via a slow diffusion process across the barrier (k2). Weibull’s equation provided

information on possible mechanism of DcNa release from CBP suppositories while

the bi-exponential first-order equation described the release rate and kinetics during

the biphasic release process; thus both models were used concomitantly to describe

the release of DcNa from CBP suppositories.

4.4 Conclusion

The HPKS bases were comparable to CB in terms of drug release capacity and could

be good lipophilic base candidates for fast-acting DcNa suppository formulations.

Although convenient, the exploratory data analysis methods were unable to compare

between large number of drug release profiles and offered no explanations to drug

release mechanisms. The fit factors allowed quick detection of dissimilar profiles yet

124
does not identify cause of the differences between profiles. DE and MDT were useful

for comparing differences between rate and extent of drug release but only

mathematical modelling enabled the prediction of DcNa release mechanism. However,

none of these methods were sufficient as a standalone analysis and thus, they should

be used concomitantly to provide a complete picture of drug release profiles.

Generally, the model independent methods (fit factors, DE, MDT) provided strong

indication that DcNa release from formulations containing 1-5 %w/w CBP was

markedly different (statistically significant at 95 % confidence interval) from their

respective reference formulations (DcNa only suppositories). Although the fit factors

found that 5 %w/w PVP, HPMC and CMCTS made in CB were significantly different

from reference formulations (containing only DcNa), this difference was mainly

reflected in terms of DE rather than due to a change in mechanism of drug release.

Mathematical modelling of data found that suppositories containing only DcNa

released the drug via non-Fickian diffusion kinetics. Addition of 1-5 %w/w HPMC,

PVP and CMCTS to the formulations did not alter mechanism of DcNa release. They

are therefore suitable candidates of bioadhesive polymers for development of DcNa

suppositories.

However, the addition of CBP lead to considerable change in morphology of molten

suppository during dissolution via gelling, which resulted in biphasic DcNa release

process involving a rapid initial diffusion and erosion process followed by slow

diffusion process across the CBP gel layer. Furthermore, as CBP gels in the

dissolution medium, it decreases the environment pH which leads to decreased DcNa

125
solubility, thus further retarding the release of DcNa from suppositories. The

concentration dependent impedance of DcNa release from CBP suppositories indicate

that CBP should only be used at the lowest possible concentration to confer

bioadhesivity as concentrations of 1-5 %w/w had evidently suppressed drug release.

The drug release studies using distilled water were only preliminary in nature. Drug

release studies using buffered solutions at rectal pH of 7.2 should be considered after

further investigation on bioadhesive properties and enhancement on formulation

prototypes.

126
CHAPTER 5

BIOADHESION STUDIES

127
5.1 Introduction

5.1.1 Methods to study bioadhesion

Bioadhesion studies can be performed via in vitro, in vivo and ex vivo experimental

setups. In vitro bioadhesion studies are by far most commonly adopted due to ease of

experimental setup. This method involves the use of a suitable excised mucosal

membrane or synthetic membrane surface as the site of attachment under simulated

conditions.

To date, various methods have been developed and applied to study bioadhesive

properties of pharmaceutical formulations. Among the techniques employed were (a)

florescence probe technique (Park and Robinson, 1984) which measures change in

fluorescence upon binding of polymer to epithelial cells labelled with pyrene and

fluorescein, (b) detachment stress methods including Wilhelmy plate method (Sam et

al., 1992; Santos et al., 1999), tensile stress method (Smart, 1991; Thirawong et al.,

2007; Tobyn et al., 1995; Wong et al., 1999a) and shear stress method (Jiménez-

Castellanos et al., 1993; Leung and Robinson, 1988; Mortazavi and Smart, 1995;

Wang and Tang, 2008); (c) the wash-off method (Lehr and Bouwstra, 1992) which

quantifies amount of particulate remaining on the test surface after bouts of agitation;

and (d) mucin-particulate method (Takeuchi et al., 2005) which measures change in

zeta potential of mucin brought about by mucin-polymer interaction. The Biacore®

method to study bioadhesion was subsequently developed as an extension of this

concept, where the change in surface plasmon resonance brought about by interaction

between polymer and mucin reacted onto chitosan-immobilised sensor chips were

measured arbitrarily (Thongborisute and Takeuchi, 2008). Of these, the detachment

methods were most commonly used on solid and semi-solid dosage forms.

128
5.1.2 Detachment methods

5.1.2.1 Tensile stress

(a)

(b)

Figure 5.1 : Experimental setup for testing tensile stress of bioadhesion using texture

analyser; (a) without temperature control (Thirawong et al., 2007; Tobyn et al., 1995;

Wong et al., 1999a) and; (b) with temperature control (Thirawong et al., 2007).

The tensile setup measures force required to fracture bioadhesion bond at right angles

to the plane of contact between test sample and mucosa surface. Wilhelmy plate

method (Smart et al., 1984) measures the tensile stress generated upon detaching a

glass plate coated with polymer vertically immersed in mucin using a microforce

129
balance. This concept was subsequently modified to measure tensile stress generated

upon the detachment between a tablet and mucosa surface using the tensiometer

(Leung and Robinson, 1988; Ponchel et al., 1987) and texture analyser (Thirawong et

al., 2007; Tobyn et al., 1995; Wong et al., 1999a). Figures 5.1a-b shows some of the

experimental setup developed by other researchers to measure tensile stress of force

required to detach dosage form from the mucosa.

5.1.2.2 Shear stress

Figure 5.2 shows some of the experimental setups developed to measure tangential

shear stress. Shear setup measures the sliding force parallel to the plane of contact

required to dislodge a sample disc from mucosa surface.

Most studies measure the degree of bioadhesion in terms of tensile stress or tensile

strength, although in actual fact, dosage forms administered into the GIT or vagina or

buccal regions are most likely to undergo frictional and shear stress which occurs

parallel to the adhesive joint. This might be due to the difficulties in measuring shear

stress as well as the inadequacy of current methods to distinguish between actual force

from the bond joint fracture and the force contributed by the friction of both surfaces

(Jiménez-Castellanos et al., 1993; Leung and Robinson, 1988; Mortazavi and Smart,

1995). While Leung and Robinson (1988) reported good results with their

experimental setup (Figure 5.2b), Mortazavi and Smart (1995) were unable to yield

comprehendible shear stress readings using the setup in Figure 5.2a because the

readings were affected by friction and the occurrence of re-adhesion after joint

fracture under the influence of gravity.

130
(a) (b)

(c)

Figure 5.2 : Experimental setups used to investigate the shear stress of the bioadhesion

joint between bioadhesive material and membrane; (a) modified Dia-Stron rheometer

equipped with pulley system (Mortazavi and Smart, 1995); (b) dual modified

tensiometer method (Leung and Robinson, 1988) and; (c) vertical rod coupled to

tensiometer (Jiménez-Castellanos et al., 1993).

5.1.3 Experimental design considerations

Methods to measure stress of detachment vary greatly in terms of experimental setup,

choice of mucosa membrane and test medium used, in order to simulate in vivo

conditions at which bioadhesion is expected to take place.

131
Experimental setup designed for testing bioadhesion in the GIT involved complete

immersion of tissue and dosage form samples in a suitable test medium (Figure 5.1b

and Figures 5.2a-b); usually simulated gastric which served as both a hydration

medium and water bath to maintain the tissue at physiological temperature

(Thirawong et al., 2007; Tobyn et al., 1995). This however, may not be applicable for

the testing of bioadhesive suppositories as the rectum contains only 2–3 mL of inert

mucus over a surface area of 200–400 cm2 (Allen et al., 2008). Conversely, some

studies excluded immersion of tissue in test medium, but at the expense of

physiological temperature control (Figure 5.1a and Figure 5.2c) (Jiménez-Castellanos

et al., 1993; Ponchel et al., 1987; Wong et al., 1999a). Temperature control was

particularly important for current work as suppositories would soften and melt at

human body temperature, yet immersion of samples in test medium does not reflect

physiological rectal conditions.

Bioadhesive polymers were incorporated into rectal suppositories in this research as

an attempt to circumvent the pre-systemic first-pass metabolism. It is hoped that the

bioadhesive polymers would enable adherence of suppositories to the rectal mucosa,

thus preventing its movement towards the upper rectum where capillaries drain into

the hepatic portal system which is responsible for a substantial degree of pre-systemic

drug inactivation. Previous studies by Yahagi et al. (2000) and Ramadan (2012) have

incorporated CBP into suppositories, however, bioadhesive properties of the

formulations were not tested.

Therefore, this chapter aims to develop and optimise methods for in vitro assessment

of bioadhesivity of suppositories using the texture analyser for measurement of tensile

132
and shear stresses required to disrupt bioadhesion. Measurements were made in terms

of peak force of detachment (Fmax) and work of adhesion (Wad). This is followed by

evaluation of bioadhesive properties of the suppository formulations developed in

Chapter 3. Finally, as efforts to diversify experimental setup, synthetic cellulose

membrane was investigated as a potential alternative to biological membranes in

bioadhesion studies.

5.2 Materials and methods

5.2.1 Materials

Type III mucin from porcine stomach was purchased from Sigma Aldrich, Missouri,

USA. The regenerated cellulose membrane (nominal MW 12,000–14,000; thickness

33 mm) was purchased from Fisher Scientific. Other materials used have been

described in Sections 2.2 and 3.2. All reagents were analytical grade. The materials

were used as received.

5.2.2 Methods

5.2.2.1 Preparation of sample discs

Cylindrical sample discs with a radius of 1.3 cm and thickness of 0.5 cm were

prepared via fusion moulding. Method of manufacturing was similar to that of

suppositories (Section 3.3.1) with the exception of acrylate disc moulds in place of

suppository moulds. Bioadhesive polymers (CBP, HPMC, PVP and CMCTS) were

added to the molten base at concentrations of 1, 2 and 5 %w/w alongside 50 mg of

DcNa per disc.

133
5.2.2.2 Preparation of large intestinal tissue

Freshly excised porcine large intestines were obtained from a local slaughterhouse and

processed within 24 hours. The large intestines were split lengthwise and luminal

contents were removed by careful rinsing with distilled water. The serosa, tunica

muscularis and the submucosa layer were removed and the large intestines were

separated into three sections - the crown, rectum and colon (Figures 5.3a-b).

(a)

(b)

Figure 5.3: Different segments of freshly excised porcine large intestinal tissues used

in this study, (a) intact large intestines with serosa, tunica muscularis and submucosa

layer; (b) split large intestines, with mucosa facing upwards.

134
The crown was discarded and rectum region was harvested 3 cm after the dentate

while 10-15 cm of the mucosa between rectum and colon were discarded to omit

‘transitional regions’. The rectum and colon were carefully examined to ensure the

membrane is intact before cutting into 6 cm x 6 cm membrane segments. The

membranes were immersed in 0.9 %w/v sodium chloride and kept frozen until use,

within 1 week from the day of processing. The yield of sample membranes from the

intestines was tabulated.

5.2.2.3 Preparation of simulated rectal mucus

Simulated rectal mucus (SRM) was prepared by stirring 5 %w/w type III mucin in pH

7.4 simulated colonic fluid (SCF) for 3 hours. The SCF was prepared based on the

formula for SCF by Marques et al. (2011). SRM was stored refrigerated at 4 °C and

used within 72 hours from time of preparation.

5.2.2.4 Experimental setup

Bioadhesion measurement was conducted using a Ta.XT plus Texture Analyser

(Stable Microsystems, Surrey, UK) equipped with a 5 kg load cell. All measurements

were conducted at 29 ± 1 °C with RH of 55–65 %. All studies were carried out in 5–6

replicates. Fresh tissue and sample disc was used for each replicate.

5.2.2.4.1 Tensile measurement

The method used in the current study was modified from Thirawong et al. (2007) and

Wong et al. (1999) to allow temperature control of the membrane without immersion

of the setup in a water bath. The setup comprised of a probe affixed to the texture

analyser arm with a flat round surface and a copper membrane stage heated using a

135
ceramic top stirring hotplate (Fisher Scientific, Pittsburg, USA). The entire assembly

is as depicted in Figure 5.4.

Figure 5.4 : The tensile bioadhesion study experimental setup using texture analyser

with heated copper membrane stage.

Sample discs were securely mounted onto the flat surface of the cylindrical probe

using double sided tape. The tissue prepared using methods describe in Section 5.2.2.2

were allowed to thaw to room temperature and immersed in SCF for 30 minutes

before clamping on the copper stage using binder clips, luminal surface facing

upwards. The copper stage was then positioned below the texture analyser arm and

aligned to ensure the sample disc comes into direct contact with membrane surface

when probe is lowered. A fixed volume of SRM was dispensed onto the mucosa

surface and spread out evenly before lowering the sample disc to 5 cm above the

membrane. The membrane temperature was measured using an infrared thermometer

and allowed to equilibrate to 37.0 ± 0.5 °C on the copper stage before commencement

of experiments. The probe was lowered at a speed of 1 mm/s until contact was made

136
between the sample disc and the membrane (Figure 5.4, Step I). This contact was

maintained for a specific time (contact time) under a fixed contact force (Figure 5.4,

Step II). At the end of contact time, the probe was withdrawn at a predetermined

speed (probe withdrawal speed) to a 10 mm distance (Figure 5.4, Step III).

5.2.2.4.2 Shear measurement

The method used in the current study was modified from Chary et al. (1999) and

Wang and Tang (2008) to allow direct measurement of shear force required to disrupt

bioadhesion between sample disc and the membrane under temperature control. The

entire assembly is as depicted in Figure 5.5.

Figure 5.5 : The shear bioadhesion study experimental setup using texture analyser

with heated copper membrane stage.

The setup comprised of three components: a sample holder; a copper stage connected

to an acrylate stage affixed with a pulley wheel and; a texture analyser probe fitted

with a hook. Sample discs were securely mounted onto the flat surface of the sample

holder using double sided tape while the mucosa was attached to the copper stage

137
using binder clips (luminal surface upwards). The entire copper stage was maintained

at 37 °C using a ceramic top stirring hotplate (Fisher Scientific, Pittsburg, USA). A

fixed length of nylon string was attached from the sample holder to the hook on the

probe through a pulley at right angles. The position of the probe was adjusted to

ensure there is no slack along the length of the string. The membrane temperature was

allowed to equilibrate to 37.0 ± 0.5 °C on the copper stage and dispensed with a fixed

volume of SRM before commencement of experiments. Weights (contact force) were

placed on the sample holder for a predetermined duration of time (contact time) to

facilitate bioadhesion between sample disc and membrane (Figure 5.5, Step I). At the

end of the contact time, the probe was withdrawn at a predetermined speed (probe

withdrawal speed) to dislodge sample disc from membrane surface (Figure 5.5, Step

II).

5.2.2.5 Effects of instrument and test variables on the bioadhesive test

Fmax of the sample discs from the porcine large intestinal mucosa under different test

conditions was measured. Three instrumental variables were studied; the contact time,

probe withdrawal speed and contact force; while the two test variables studied were

the volume of mucin used and the type of large intestinal mucosa used (rectum or

colon). Studies on the instrument variables were conducted using the porcine colon

mucosa with 150 μL of SRM evenly spread over the surface. All studies were carried

out in 5–6 replicates.

5.2.2.5.1 Tensile measurement

Sample discs containing 50 mg DcNa and 5 %w/w CBP were used to optimise the

instrument and test variables used in tensile measurements. Four different contact

138
times, five probe withdrawal speed, five contact forces, five volumes of SRM and two

types of mucosa. The parameters are tabulated in Table 5.1.

5.2.2.5.2 Shear measurement

Sample discs containing 50 mg DcNa and 5 %w/w PVP were used to optimise the

instrument and test variables used in shear measurements. Four different contact times,

four probe withdrawal speed, four contact forces, four volumes of SRM and two types

of mucosa. The parameters are tabulated in Table 5.2.

5.2.2.6 Evaluation of the bioadhesive strengths in suppository formulations

using biological membranes

The instrumental and experimental parameters used to evaluate bioadhesive strength

of suppository sample disc were obtained from studies in Section 5.2.2.5.1 and

5.2.2.5.2. All measurements were carried out in 5-6 replicates.

5.2.2.7 Evaluation of synthetic regenerated cellulose membrane as an

alternative to biological membrane

Evaluation was carried out using settings and parameters used in Section 5.2.2.6 with

the substitution of synthetic regenerated cellulose membrane for porcine colon mucosa.

The regenerated cellulose membrane was cut to 6 cm x 6 cm squares and immersed in

SCF for 1 hour prior to use. All measurements were carried out in 5-6 replicates.

139
Table 5.1 : The fixed and variable parameters used for tensile force optimisation using sample discs containing 50 mg DcNa and 5 %w/w CBP.

Variable parameter
Fixed parameter
Contact time Probe withdrawal speed Contact force Volume of SRM Type of mucosa

Contact times 5, 10, 20, 30 s 20 s 20 s 20 s 20 s

Probe withdrawal speed 10 mm/s 1, 2, 5, 10, 20 mm/s 10 mm/s 10 mm/s 10 mm/s

Contact force 2N 2N 0.5, 1, 1.5, 2, 3 N 2N 2N

Volumes of SRM 150 μL 150 μL 150 μL 0, 50, 100, 150, 300 μL 150 μL

Type of mucosa colon colon colon colon colon, rectum

140
Table 5.2 : The fixed and variable parameters used for shear force optimisation using sample discs containing 50 mg DcNa and 5 %w/w PVP.

Variable parameter
Fixed parameter
Contact time Probe withdrawal speed Contact force Volume of SRM Type of mucosa

Contact times 20, 40, 60, 90 s 60 s 60 s 60 s 60 s

Probe withdrawal speed 30 mm/s 5, 10, 20, 30 mm/s 30 mm/s 30 mm/s 30 mm/s

Contact force 2N 2N 1, 2, 3, 4 N 2N 2N

Volumes of SRM 150 μL 150 μL 150 μL 0, 100, 150, 300 μL 150 μL

Type of mucosa colon colon colon colon colon, rectum

141
5.2.2.8 Data analysis

Figure 5.6 showed the typical plot of force versus distance data obtained through

tensile measurement. The maximum force required for separation of sample disc from

the membrane or Fmax was obtained directly from the force–distance curve while the

work of adhesion (Wad) was calculated using area under the force–distance curve

using the Texture Exponent 32 software. Bioadhesive properties of the different

formulations were evaluated and compared based on these two parameters. ANOVA

followed by a post hoc Tukey’s HSD test was performed to examine both effects of

instrument and experimental variables on bioadhesion as well as bioadhesive strength

of various formulations. The statistical analyses were conducted using SPSS version

20 (SPSS Inc., USA). A statistically significant difference was observed when p <0.05.

Figure 5.6: A typical plot of force versus distance data for suppository sample disc

(CB + 50 mg DcNa + 1 %w/w PVP) tested with colon mucosa using the tensile setup.

142
5.3 Results and discussion

The design of both experimental setup for tensile and shear measurement were

targeted at mimicking the internal environment of the rectum. The disc surface area

was designed to reflect total surface area for bioadhesion in an actual torpedo shaped

suppository with height of 2.5 cm and radius of 0.4 cm (at its barrel end). The

intestinal mucosa and SRM (pH 7.4) was used to reproduce rectal environment.

5.3.1 Preparation of large intestinal tissue

The yield of both rectum and colon samples were tabulated in Table 5.3. The yield of

colon membrane samples were usually 2-3 times more than the amount of rectum

mucosa obtained per intestine.

Table 5.3 : The yield of biological membrane mucosa.

Rectum Colon
Intestine Length Yield Length Yield
(cm) (6 cm x 6 cm) (cm) (6 cm x 6 cm)
A 25 12 122 34
B 25 7 116 19
C 25 8 95 26
D 19 8 90 22
E 22 7 100 25
F 30 8 127 20

5.3.2 Effects of instrument and test variables on the bioadhesive test

5.3.2.1 Tensile measurement

Figure 5.7 showed that Fmax and Wad increased as contact time was increased until 20

s, where a further 10 s of contact time did not significantly increase Fmax.

143
(a) 6

Peak force of detachment (N)

4

0
0 5 10 15 20 25 30 35
Contact time (s)

(b) 6
Work of adhesion (N.mm)

0
0 5 10 15 20 25 30 35
Contact time (s)

Figure 5.7: Effect of contact time on (a) Fmax and (b) Wad of SS discs containing 50

mg DcNa and 5 %w/w CBP against porcine colon mucosa using the tensile setup.

Mean ± SD, n=5-6.

Increasing contact time beyond 20 s did not result in any significant increase in Fmax

and Wad; contrary to previous studies by Thirawong et al. (2007) and Wong et al.

(1999) which reported that increment in contact time resulted in linear increase in W ad.

Initial increase in contact time may have allowed interdiffusion and chain

entanglement between CBP and mucin in the SRM; however as the contact time was

increased, a higher fraction of the sample disc melted, resulting in an oily, slippery

layer between the sample disc and the tissue mucosa. Therefore, 20 s was selected as

the suitable contact time.

144
The effect of increasing probe withdrawal speed on both the Fmax and Wad (Figure 5.8)

were similar to findings by Wong et al. (1999), and Thirawong et al. (2007).

Increasing probe withdrawal speed from 1 mm/s to 10 and 20 mm/s resulted in

statistically significant increase in Fmax and Wad, but there were no differences

between 10 and 20 mm/s. Higher probe speeds produced larger Fmax and Wad which

afforded higher sensitivities in measuring bioadhesion while the lower speeds resulted

in bigger standard deviations. Therefore, an intermediate probe speed of 10 mm/s was

selected for subsequent bioadhesion studies.

(a)
5
Peak force of detachment (N)

3
2
1
0
0 5 10 15 20 25
Probe withdrawal speed (mm/s)

(b)
5
Work of adhesion (N.mm)

4
3
2
1
0
0 5 10 15 20 25

Probe withrawal speed (mm/s)

Figure 5.8: Effect of probe withdrawal speed on (a) Fmax and (b) Wad of SS discs

containing 50 mg DcNa and 5 %w/w CBP against porcine colon mucosa using tensile

setup. Mean ± SD, n=5-6.

145
A significant increase in Fmax and Wad was observed only after contact force was

increased from 0.5 to 2 N (Figure 5.9). The observed trend was similar to that by

Thirawong et al. (2007) although the latter study investigated contact force at the

range of 0.05 to 0.5 N. Various studies have shown that basal rectal pressure is at the

range of 5–25 cmH2O (Farouk, 2003) and 20–25 mmHg (Rao et al., 1988) while anal

pressure is approximately 26–75 cmH2O (Hancock, 1976). Therefore, 2 N which

translated to 0.049–0.74 N/cm2 (equivalent to 5-75.5 cmH2O) was selected as contact

force for future studies, derived from contact surface area of sample disc (5.1 cm2).

(a)
Peak force of detachment (N)

0
0 0.5 1 1.5 2 2.5 3 3.5
Contact force (N)

(b)
6
Work of adhesion (N.mm)

0
0 0.5 1 1.5 2 2.5 3 3.5
Contact force (N)

Figure 5.9 : Effect of contact force on (a) Fmax and (b) Wad of SS discs containing 50

mg DcNa and 5 %w/w CBP against porcine colon mucosa using tensile setup. Mean

± SD, n=5-6.

146
The effect of different SRM volumes ranging from 0–300 μL on the Fmax and Wad

were investigated to select the suitable volume which simulates rectal conditions yet

produced reasonable measurements of Fmax and Wad. Figure 5.10 showed that different

volumes of SRM did not significantly affect the Fmax and Wad generated, although it

was suggested that adhesive forces weakens as mucus content increases (Mortazavi

and Smart, 1995). However, 150 μL was selected for subsequent studies as it is the

average volume of rectal conditions based on the surface area of the tissue used.

(a)
6
Peak force of detachment (N)

no mucin
4 50 μL
100 μL
2 150 μL
300 μL
0

(b)
6
Work of adhesion (N.mm)

no mucin
4 50 μL
100 μL
2 150 μL
300 μL

Figure 5.10 : Effect of volume of SRM on (a) Fmax and (b) Wad of SS discs containing

50 mg DcNa and 5 %w/w CBP against the porcine colon mucosa using the tensile

setup. Mean ± SD, n=5-6.

147
This study also investigated suitability and consistency of various segments of the

large intestine (rectum and colon) as model mucosa for bioadhesion studies. Generally,

formulations containing either CMCTS or CBP exhibited higher Fmax and Wad values

than those without polymer (Figure 5.11). Bioadhesive properties of CBP were

greater than CMCTS and increased in a concentration dependant manner.

(a)
5
Peak force of detachment (N)

DcNa
4
2 % CMCTS
3
5 % CMCTS
2
2 % CBP
1 5 % CBP
0
Rectum Colon

(b)
5
Work of adhesion (N.mm)

DcNa
4
2 % CMCTS
3
5 % CMCTS
2
2 % CBP
1 5 % CBP
0
Rectum Colon

Figure 5.11 : Effect of different segments (rectum and colon) of the porcine large

intestines on (a) Fmax and (b) Wad of SS discs containing 50 mg DcNa and 2–5 %w/w

of CMCTS or CBP using the tensile setup. Values expressed as mean ± SD, n=5-6.

The correlation coefficient of Fmax between the rectum and colon was found to be

0.910. This indicated that the colon provided a reasonable representation of the

rectum in terms of evaluation of bioadhesion properties of suppositories using the

148
tensile setup. Due to the low yield of rectum membranes (Table 5.3), the colon

mucosa was used in subsequent evaluative studies in Section 5.3.3.

5.3.2.2 Shear measurement

Figure 5.12 showed that Fmax and Wad increased as contact time was increased until 90

s, followed by a decrease in bioadhesion at 120 s. Shojaei et al. (2000) reported that

bioadhesion strength plateaued when contact time was increased beyond 120 s, and

attributed it to excessive water sorption. In this study, it is likely that a slippery

surface was produced as the disc melts, resulting in a decrease in bioadhesion forces

with longer contact times; thus 60 s was selected for subsequent studies.

(a) 2
Peak force of detachment (N)

0
0 30 60 90 120 150
Contact time (s)

(b)
4
Work of adhesion (N.mm)

0
0 30 60 90 120 150
Contact time (s)

Figure 5.12 : Effect of contact time on (a) Fmax and (b) Wad of SS discs containing 50

mg DcNa and 5 %w/w PVP against the porcine colon mucosa using the shear setup.

149
Mean ± SD, n=5-6.

The effects of increasing probe withdrawal speed on both the Fmax and Wad (Figure

5.13) were similar to findings from the tensile setup. Increment in probe withdrawal

speed from 5 mm/s to 10, 20 and 30 mm/s resulted in statistically significant increase

in Fmax and Wad. Therefore, probe speed of 30 mm/s was selected for subsequent

bioadhesion studies using the shear setup.

(a) 2
Peak force of detachment (N)

0
0 10 20 30 40
Probe withdrawal speed (mm/s)

(b)
4
Work of adhesion (N.mm)

0
0 10 20 30 40
Probe withrawal speed (mm/s)

Figure 5.13 : Effect of probe withdrawal speed on (a) Fmax and (b) Wad of SS discs

containing 50 mg DcNa and 5 %w/w PVP against porcine colon mucosa using the

shear setup. Mean ± SD, n=5-6.

150
A significant increase in Fmax and Wad was observed when contact force was increase

from 1 to 2, 3 and 4 N (Figure 5.14). There was a ceiling effect for the increment in

Fmax and Wad brought about by increasing contact force, as no significant difference in

Fmax and Wad between contact force of 2, 3 and 4 N. This was in agreement with

Wong et al. (1999) where the authors suggested that excessive contact force may lead

to mucosal damage without any improvements on bioadhesion. Contact force of 2 N

was used in subsequent studies.

(a)
2
Peak force of detachment (N)

0
0 1 2 3 4 5
Contact force (N)

(b)
5
Work of adhesion (N.mm)

0
0 1 2 3 4 5
Contact force (N)

Figure 5.14 : Effect of contact force on (a) Fmax and (b) Wad of SS discs containing

50 mg DcNa and 5 %w/w PVP against the porcine colon mucosa using the shear

setup. Mean ± SD, n=5-6.

151
Figure 5.15 showed that the different volumes of SRM used in this study did not

significantly affect the Fmax and Wad generated. SRM volume of 150 μL was selected

for the subsequent studies.

(a)
2
Peak force of detachment (N)

0 μL

100 μL

1 150 μL

300 μL

(b)
5
Work of adhesion (N.mm)

4 0 μL

3 100 μL

2 150 μL

1 300 μL

Figure 5.15 : Effect of volume of SRM on (a) Fmax and (b) Wad of SS discs

containing 50 mg DcNa and 5 %w/w PVP against the porcine colon mucosa using

the shear setup. Mean ± SD, n=5-6.

When different segments of the large intestine (rectum and colon) were investigated,

the rank orders of Fmax and Wad for the tested formulations were similar (Figure 5.16).

The correlation coefficient of Fmax between the rectum and colon using the shear setup

was found to be 0.965. As with the observations using the tensile experimental setup

152
in section 5.3.2.1, the colon mucosa was a suitable replacement of the rectum.

Generally, formulations containing either CBP or PVP exhibited higher F max and Wad

values than those without polymer, and the bioadhesion conferred by PVP was greater

than CBP and increased in a concentration dependant manner.

(a)
1
Peak force of detachment (N)

DcNa
2 % CBP
0.5 5 % CBP
2 % PVP
5 % PVP

0
Rectum Colon

(b)
4.5
4
Work of adhesion (N.mm)

3.5 DcNa
3 2 % CBP
2.5
5 % CBP
2
1.5 2 % PVP
1
5 % PVP
0.5
0
Rectum Colon

Figure 5.16 : Effect of different segments (rectum and colon) of the porcine large

intestines on (a) Fmax and (b) Wad of SS discs containing 50 mg DcNa and 2–5 %w/w

of CBP or PVP using the shear setup. Mean ± SD, n=5-6.

Due to similarity in the data obtained through Fmax and Wad in both the tensile and

shear measurements, only Fmax was reported in Sections 5.3.3 and 5.3.4.

153
5.3.3 Evaluation of bioadhesive strength in suppository formulations using

biological and synthetic membranes

5.3.3.1 Tensile measurement

A contact time of 20 s under a contact force of 2 N, followed by a probe withdrawal

speed of 10 mm/s was used in this segment of studies. Porcine colon mucosa spread

with 150 μL of SRM was used as a model mucosa. The Fmax of various bioadhesive

suppository formulations are shown in Figure 5.17.

For CB and CE suppositories, only 5 %w/w CBP, 5 %w/w CMCTS and 5 %w/w PVP

showed significantly higher Fmax compared to blank formulations. Although there

were small increases in Fmax for formulations containing 1–2 %w/w of bioadhesive

polymers, none of these were statistically significant.

Meanwhile, SS suppositories containing 2–5 %w/w of CBP and PVP and 5 %w/w of

CMCTS resulted in a significantly higher Fmax compared to suppositories without

polymers. Formulations with higher amounts of polymer generally exhibited greater

bioadhesive properties. The strength of bioadhesion conferred by the polymers was in

the ascending rank order of: HPMC < CMCTS < CBP < PVP. Formulations

containing HPMC at concentrations of up to 5 %w/w exhibited the weakest

bioadhesion (lowest Fmax) compared to other polymers. Outcomes of the statistical

analysis are included in the Appendices 28-30.

154
(a) 6

Peak force of detachment (N)

*
5 *
* 0%
4
1%
3

2 2%

1 5%

0
CBP HPMC PVP CMCTS

(b) 6
* *
Peak force of detachment (N)

5 *
0%
4
1%
3
2%
2

1 5%

0
CBP HPMC PVP CMCTS

(c) 6 *
Peak force of detachment (N)

*
5
* * * 0%
4
1%
3
2%
2

1 5%

0
CBP HPMC PVP CMCTS

Figure 5.17 : The Fmax of (a) CB; (b) CE and (c) SS formulations containing 50 mg

DcNa and 1–5 %w/w of bioadhesive polymer (CBP, HPMC, PVP, CMCTS) using

tensile setup. Asterisks indicate Fmax values which are significantly different from

formulations without bioadhesive polymers. Mean ± SD, n=5–6.

155
5.3.3.2 Shear measurement

A contact time of 60 s under a contact force of 2 N, followed by a probe withdrawal

speed of 30 mm/s was used in this segment of studies. Porcine colon mucosa spread

with 150 μL of SRM was used as a model mucosa.

Figure 5.18 showed the Fmax and Wad measured using the shear setup. In general,

bioadhesion measured using the shear setup was observed to increase in the following

order: CBP = HPMC < CMCTS < PVP; with formulations containing PVP exhibiting

the highest bioadhesive properties. Although all formulations containing 5 %w/w of

polymer had significantly higher Fmax, HPMC exhibited limited bioadhesive

properties. This was observed in both the tensile and shear measurements and strongly

suggests the limited benefit of using HPMC in the formulation of bioadhesive

suppositories. Outcomes of the statistical analysis are included in the Appendices 31-

33.

Similar to the tensile setup, formulations containing 5 %w/w PVP was found to

generate the highest Fmax, which indicated the superior bioadhesivity conferred by this

particular polymer when incorporated in lipophilic suppositories. The shear forces

required to detach sample discs increased with increasing amounts of PVP

incorporated into the sample disc. A similar albeit less obvious trend was observed in

formulations containing other polymers.

Formulations containing CBP exhibited poor bioadhesive properties when tested

using the shear setup; contrary to results obtained using the tensile setup in Section

5.3.3.1.

156
(a)
1
*

Peak force of detachment (N)

0.8 * * 0%
* *
0.6 * * *
* 1%

0.4 2%

5%
0.2

0
CBP HPMC PVP CMCTS

(b) 1
*
Peak force of detachment (N)

0.8 *
* * 0%
* * * * *
0.6 1%

0.4 2%

5%
0.2

0
CBP HPMC PVP CMCTS

(c)
1
*
Peak force of detachment (N)

0.8 * 0%
* *
0.6 * *
1%

0.4 2%

5%
0.2

0
CBP HPMC PVP CMCTS

Figure 5.18 : The Fmax of (a) CB; (b) CE and (c) SS formulations containing 50 mg

DcNa and 1–5 %w/w of bioadhesive polymer (CBP, HPMC, PVP, CMCTS) using

shear setup. Asterisks indicate Fmax values which are significantly different from

formulations without bioadhesive polymers. Mean ± SD, n=5-6.

157
The bioadhesion properties of CBP, an anionic polymer is attributed to presence of

carboxylic acid (-COOH) groups which facilitates the formation of hydrogen bonds.

Lehr and Bouwstra (1992) found that bioadhesion of anionic polycarbophil decreased

as the pH of the test medium increased. Furthermore, CBP is known to gel at higher

pH. Upon mixing with molten suppository base and simulated rectal fluid (pH 7.4),

CBP could have resulted in slippery mucilage which facilitates sliding between the

sample disc and mucosa surface; resulting in poor shear Fmax. Apart from that, this

observation could also be a result of the smooth and fine texture of CBP compared to

the grainier PVP, CMCTS and HPMC.

Previous studies had inconsistent findings on bioadhesive properties of PVP; Wong et

al. (1999) found that PVP K30 produced Fmax comparable to that of CBP 974P.

Conversely, Ivarsson and Wahlgren (2012) reported that PVP had limited

bioadhesivity via ellipsometry, tensile strength and rheology methods while Smart et

al. (1984) reported poor bioadhesivity in PVP using the Wilhelmy plate method.

Current study found that PVP exhibited similar bioadhesive performance compared to

CBP in tensile stress measurements and displayed superior bioadhesivity compared to

CBP in the shear measurements. PVP, although lack hydrophilic groups, possess

cyclic amide groups which could serve as potential sites for hydrogen bonding.

The secondary amine of CMCTS (structure depicted in Figure 1.2c) is protonated to

form a positive charge at lower pH while the carboxylic acid groups ionise to form

carboxylate groups as the pH increased. At the pH of SRM (pH 7.4), both the amine

and carboxylic acid groups would be protonated and could result in formation of

temporary bonds between the polymer chains, reducing polymer–mucin interaction

158
(Hombach and Bernkop-Schnurch, 2010). However, this work found that CMCTS has

more promising bioadhesivity compared to CBP when both the tensile and shear

measurements were considered collectively.

The poorest bioadhesion was observed in the formulations with HPMC, a linear,

nonionic polymer derived from etherified anhydro-glucose rings substituted with a

28–30 % hydrophobic methyl groups. Most of the previous studies which reported of

good bioadhesion properties in HPMC formulations employed the lesser methyl

substituted HPMC grade 2208 rather than 2910 used in this study (Akbari et al., 2010;

Mortazavi and Smart, 1995; Wong et al., 1999). This further affirms the importance of

potential hydrogen bonding groups in conferring bioadhesive properties.

5.3.4 Evaluation of synthetic regenerated cellulose membrane as an

alternative to biological membrane

Due to variability in mucosa surface properties and difficulties in obtaining biological

samples, a synthetic regenerated cellulose membrane was investigated as potential

substitute for biological mucosa used during in vitro measurement of bioadhesion.

5.3.4.1 Tensile measurement

Figure 5.19 showed that Fmax generated for the same formulations using synthetic

(regenerated cellulose) membrane were much higher than those generated using

biological (colon) membranes. However, both were similar in terms of rank order of

bioadhesion. The bioadhesivity in ascending manner was found to be: HPMC <

CMCTS < CBP < PVP. Formulations tested using the synthetic membranes resulted

in smaller SD but were less sensitive compared to the biological membrane, as

159
observed by the higher Fmax produced by blank samples. An outcome of the statistical

analysis is included in the Appendix 34.

(a) 6 *
Peak force of detachment (N)

5
*
* * * 0%
4 *
*
3
2%
2
5%
1

0
CBP HPMC PVP CMCTS

(b) 16 * *
* * * *
Peak force of detachment (N)

14
12 0%
10
8 2%
6
4 5%
2
0
CBP HPMC PVP CMCTS

Figure 5.19 : The Fmax of the SS formulations containing 50 mg DcNa and 0-5 %w/w

of bioadhesive polymer (CBP, HPMC, PVP, CMCTS) tested with (a) colon mucosa

as biological membrane and (b) synthetic membrane using the tensile experimental

setup. Mean ± SD, n=5-6.

The correlation coefficient of Fmax between the colon and synthetic regenerated

cellulose membrane was found to be 0.737. Despite a lower correlation coefficient

value, it still indicated a strong relationship between the Fmax generated by the colon

and the Fmax generated by regenerated cellulose membrane. Despite this strong

160
correlation, the usage of synthetic regenerated cellulose as an alternative membrane to

colon samples may only be feasible for qualitative comparison (rank order of

bioadhesion), as the Fmax values obtained using synthetic membranes were

approximately 3-4 times higher than those obtained using the rectum and colon

membranes.

5.3.4.2 Shear measurement

Contrary to the findings from tensile measurements (Section 5.3.4.1); both Fmax

generated using biological and synthetic membranes were comparable and of the same

rank order (Figure 5.20). An outcome of the statistical analysis is included in

Appendix 35.

The correlation coefficient of Fmax between the colon and synthetic regenerated

cellulose was found to be 0.959. Synthetic regenerated cellulose appeared to be a

suitable alternative to biological membrane in in vitro bioadhesion studies using the

shear setup as there were no marked differences between the results obtained from

colonic mucosa and synthetic regenerated cellulose. However, caution has to be

exercised while interpreting results in both situations as synthetic membranes have a

flat and even surface which is a stark contrast to biological mucosal surfaces.

161
(a)
1
*

Peak force of detachment

0.8 * 0%
* *
0.6 * *
2%
(N) 0.4
5%
0.2

0
CBP HPMC PVP CMCTS

(b)
1
Peak force of detachment

0.8 *
* 0%
* *
0.6 * *
2%
(N)

0.4 5%

0.2

0
CBP HPMC PVP CMCTS

Figure 5.20 : The Fmax of the SS formulations containing 50 mg DcNa and 0-5 %w/w

of bioadhesive polymer (CBP, HPMC, PVP, CMCTS) tested with (a) colon mucosa

as biological membrane and (b) synthetic membrane using the shear experimental

setup. Mean ± SD, n=5-6.

5.4 Conclusion

Two distinct methods for in vitro evaluation of bioadhesive properties of

suppositories using the texture analyser were developed and optimised in this chapter.

The first method involved the measurement of tensile forces while the second

quantified shear forces required to disrupt the bioadhesive bond. Both methods

involved temperature control at 37 °C in the presence of SRM to simulate in vivo

conditions. Formulations containing PVP exhibited superior bioadhesion compared to

162
the other polymers when subjected to both tensile and shear forces of detachment.

This finding was promising for the development of bioadhesive suppositories.

Conversely, HPMC exhibited poor bioadhesive properties in both tests and has

limited role in the development of bioadhesive suppositories. In addition to evaluation

of the bioadhesivity of the formulations, this study also investigated synthetic

regenerated cellulose membrane as a practical alternative to biological membranes as

mucosa surface for in vitro bioadhesion studies. Synthetic regenerated cellulose

membranes were generally found to be a good substitute for colon mucosa for

qualitative assessment of bioadhesion strengths in both the tensile and shear

measurements.

163
CHAPTER 6

STABILITY STUDIES

164
6.1 Introduction

6.1.1 Stability studies in suppositories

Suppository formulations are susceptible to both chemical and physical instability

when inappropriately stored (Coben and Lordi, 1980; Tukker et al., 1984). Storage

temperature and storage duration are common factors causing ageing which leads to

altered stability in suppositories (Hosny et al., 1990; Sah and Saini, 2008; Yoshida et

al., 1991; Yoshino et al., 1981).

Stability studies fundamentally involve testing both physical and chemical aspects of

a particular formulation to determine its shelf life and preferential storage conditions.

Physical analysis comprises visual inspection of the physical appearance, mechanical

strength (hardness), melting point and softening time, while analysis of active drug

compound as well drug release studies make up the crucial aspects of chemical testing.

6.1.2 Chemical testing

Certain drugs are susceptible to chemical degradation or may interact with

suppository base after prolonged exposure to warm temperatures (Tukker et al., 1984;

Whitworth et al., 1973). When drug degrades, the formulation may no longer be

clinically effective and in some cases, the degraded product may even be toxic.

Yoshida et al. (1991) observed that indomethacin suppositories aged by storage at

room temperature for one month resulted in slower drug release compared to those

stored refrigerated for the same duration of time. A separate study also found that

aminophylline suppositories made with CB kept at 30 ºC for 8 weeks resulted in

retarded in vitro drug release compared to freshly prepared suppositories (Tukker et

165
al., 1984). Although the authors attributed the slow and incomplete drug release to

degradation of aminophylline to theophylline, they also noted that shape of the

suppositories remained intact (not melted) for 30 minutes during drug release studies,

which could also imply changes in physical properties of the dosage form leading to

difficulty in melting or softening.

6.1.3 Physical testing

Lipophilic suppository base in particular, are at risk of a multitude of physical

instabilities as they are made up of a mixture of TAG with various polymorphic forms.

Visible signs of physical instability include deformation, separation of incorporated

additives and active drug from the base as well as the presence of blooming (Khan

and Craig, 2004). Bloom occurs as a dull grey surface haze which may sometimes

cause the surface of the suppository to feel grainy or crumbly to touch (Allen et al.,

2008).

Oleaginous suppositories also undergo transitions into forms exhibiting higher

melting points during storage, and these effects appear to be less pronounced when

stored at lower temperatures (Hosny et al., 1990; Liversidge et al., 1982; Webster et

al., 1998; Yoshino et al., 1981). When its melting point is elevated beyond 37 ºC, a

suppository may not melt completely upon administration into the rectum or result in

molten with a higher viscosity at body temperature, both occurrences impede drug

release (Tukker et al., 1984). Hardening or prolonged softening time of suppositories

leads to incomplete melting upon administration into the rectum and may cause local

irritation or trigger the defecatory reflex, resulting in expulsion of suppository (Coben

and Lordi, 1980).

166
The effects of ageing in suppositories were highly variable depending on the type of

drug and excipients used to formulate the suppositories (Hosny et al., 1990; Yoshino

et al., 1982). Furthermore, effects of DcNa and bioadhesive polymers (CBP, HPMC,

PVP and CMCTS) towards ageing of suppositories were unknown.

Among formulations developed and tested in the previous chapters, suppositories (CB,

CE and SS) containing bioadhesive polymers PVP and CMCTS appear to be

promising candidates for fast release DcNa suppositories with bioadhesive properties.

Thus, this chapter aims to assess stability of these suppositories as a function of

storage duration and storage condition. The suppositories were evaluated and

compared in terms of visual appearance, thermal profile, hardness, softening time and

DcNa release to ascertain consequences of ageing and the preferred storage conditions

in these formulations.

6.2 Materials and methods

6.2.1 Materials

The materials used to manufacture suppository samples used in this chapter have been

previously described in Sections 2.1 and 3.2. All other chemicals used have been

described in Sections 3.2 and 4.2.

6.2.2 Methods

Suppositories were prepared using methods previously described in Section 3.3.1.

Prepared suppositories were either stored refrigerated (3.5 ± 1.5 °C; RH of 29 ± 3%)

or kept at room temperature (24.5 ± 2.5 °C; RH 58 ± 5%). The samples were analysed

167
at three time points; freshly prepared, 100 ± 10 days and 200 ± 10 days on storage

until analysis.

6.2.2.1 Physical appearance

The samples were inspected in terms of changes in colour, surface texture or presence

of bloom compared to freshly manufactured suppositories. Changes to the

suppositories in terms of colour, surface glossiness and smoothness (tactile) after

storage for 100 and 200 days were evaluated both visually and by touch.

6.2.2.2 Thermal profile

Thermal analyses of the suppositories were conducted using the DSC system

mentioned in Section 2.3.1.1.2. The samples were prepared according to methods

described and heated from -10 ºC to 60 ºC. Thermograms were analysed to: (a)

determine melting point of the formulations (endothermic peak minimum on

thermogram); (b) identify presence of new endothermic peaks; and (c) quantify SFC

of the formulations (continuous integration of the thermogram).

6.2.2.3 Hardness

Hardness of suppositories was examined using method described in Section 3.3.2.6.

Measurements were repeated with 6 independent samples (n=6) for each of the

formulation tested.

6.2.2.4 Softening time

The softening time of suppositories was determined using method described in

Section 3.3.2.7. Experiment was carried out in triplicates (n=3) for each formulation.

168
6.2.2.5 DcNa release

The release of DcNa from aged suppository samples was investigated using method

described in Section 4.2.1. Drug release studies (n=3) were carried out for 180

minutes.

6.2.2.6 Statistical analysis of data

The results from hardness (Section 6.2.2.3) and softening time (Section 6.2.2.4) were

subjected to analysis of variance (ANOVA) followed by post hoc Tukey’s HSD test

to detect presence of significant differences between the formulations (freshly

prepared samples; samples stored refrigerated for 100 and 200 days; and samples

stored at room temperature for 100 and 200 days). The results from DcNa release

(Section 6.2.2.5) on the other hand, were analysed via visual comparison of the DcNa

release curves alongside DE and MDT.

6.3 Results and discussion

6.3.1 Physical appearance

The physical appearance of PVP and CMCTS suppositories made using CB, CE and

SS bases were examined (Table 6.1). There was a general trend of decreasing surface

glossiness and increasing graininess of the suppositories stored at room temperature

over a period of 200 days. This effect was more predominant in suppositories

containing 5 %w/w PVP. Suppositories which were kept refrigerated have less

detectable physical changes compared to freshly made samples.

169
Table 6.1: The physical appearance of PVP and CMCTS suppositories containing 50 mg DcNa after storage at various conditions up to 200 days.

100 days 200 days

Freshly prepared
Formulations Refrigerated Room temp Refrigerated Room temp

Surface Texture Surface Texture Surface Texture Surface Texture Surface Texture

No polymer +++ *** +++ *** ++ ** +++ *** ++ **

CB 5%w/w PVP +++ *** +++ *** ++ * +++ *** + *

5%w/w CMCTS +++ *** +++ *** ++ ** +++ *** + **

No polymer +++ *** +++ *** ++ *** +++ *** + **

CE 5%w/w PVP +++ *** +++ *** ++ ** +++ *** + *

5%w/w CMCTS +++ *** +++ *** ++ *** +++ *** + **

No polymer +++ *** +++ *** ++ *** +++ *** + **

SS 5%w/w PVP +++ *** +++ *** ++ ** +++ *** + *

5%w/w CMCTS +++ *** +++ *** ++ *** +++ *** + **

‘+’ denotes glossiness of the suppository surface, with ‘+++’ glossy and ‘+’ dull; while ‘*’ denotes smoothness of the suppository to touch, with

‘***’ smooth and ‘*’ grainy.

170
6.3.2 Thermal profile

Due to the natural composition of fats, their polymorphic transitions often involve

multiple TAG which by themselves exists in an array of polymorphs. Changes in

properties of fats have been mainly attributed to polymorphic transitions and

segregation of components within complex lipids.

Figures 6.1(ii-iii) showed that the CB suppositories containing DcNa and 5 %w/w PVP

which were kept refrigerated (3.5 ± 1.5 ºC) for up to 200 days did not result in a change

in melting point (endothermic peak). However, the onset of melting for these

suppositories shifted to a lower temperature, as demonstrated by the widening of the

endothermic peak (Figure 6.1, arrow). A similar observation was demonstrated in CB

suppositories containing DcNa and 5 %w/w CMCTS in Figure 6.2. This a potential

cause of concern as excessively low onset of melting could result in suppositories

liquefying during handling prior to insertion into the rectum.

On the other hand, melting point of CB suppositories kept at room temperature

increased throughout storage (Figures 6.1(iv-v) and Figures 6.2(iv-v)). This increase in

melting point was observed as early as 100 days of storage at room temperature. The

melting point was 34.5 and 35.1 ºC for suppositories stored at room temperature for 100

and 200 days respectively; compared to freshly prepared suppositories which melted at

32.9 ºC. This was believed to be due to gradual transformation of polymorphic forms

3B and 4A to the stable 4B (nomenclature as described in Table 2.1), which was a

commonly observed polymorphic transition in poorly stored chocolates (Lonchampt

and Hartel, 2004). Increment in melting point of suppositories stored at room

temperature was reflected as a rightward shift in the SFC curve in Figure 6.3.

171
Figure 6.1 : The DSC thermogram of CB suppositories containing 50 mg DcNa and 5 %w/w PVP. Individual thermograms show the melting

endotherm of suppositories which were (i) freshly prepared; stored refrigerated at for (ii) 100 days and (iii) 200 days; stored at room

temperature for (iv) 100 days and (v) 200 days. Inset shows enlarged portions of the thermogram.
172
Figure 6.2 : The DSC thermogram of CB suppositories containing 50 mg DcNa and 5 %w/w CMCTS. Individual thermograms show the

melting endotherm of suppositories which were (i) freshly prepared; stored refrigerated at for (ii) 100 days and (iii) 200 days; stored at room

temperature for (iv) 100 days and (v) 200 days. Inset shows enlarged portions of the thermogram.
173
(a) 120

100
freshly prepared sample
Solid fat content (%) 80 refrigerated 100 days

60 refrigerated 200 days

room temperature 100 days

40
room temperature 200 days
20

0
0 10 20 30 40 50
Temperature (°C)

(b) 120

100
freshly prepared sample
Solid fat content (%)

80
refrigerated 100 days

60
refrigerated 200 days

40 room temperature 100 days

room temperature 200 days

20

0
0 10 20 30 40 50
Temperature (°C)

Figure 6.3 : The SFC of CB suppositories containing 50 mg DcNa and (a) 5 %w/w

PVP and (b) 5 %w/w CMCTS.

174
A faint shoulder at approximately 42 ºC (Figures 6.1-6.2, Section A inset) was

observed in all freshly prepared, refrigerated and room temperature samples. This is

likely to be due to the presence of small amounts of the β form of the stearic acid-oleic

acid-stearic acid (SOS) TAG which were already present in the CB stock used to

manufacture suppositories. The melting points of CB TAG and their polymorphic

forms are tabulated in Table 6.2.

Table 6.2 : The melting points of various polymorphic forms of CB TAG. (Fatty acid

nomenclature under TAG column as follows: P = palmitic acid; O = oleic acid; S =

stearic acid)

Melting Points of Polymorphic forms (°C)

Composition in
(Arishima et al., 1991; McClements, 1999; Sato, 2001;
CB (%)
TAG Smith, 2009; Susumu and Konishi, 2011)
(Lonchampt and
β’ β
Hartel, 2004) α γ δ
β2’ β1’ β2 β1

POS 46.9 19.5 28.3 31.6 35.5

SOS 29.8 23.5 35.4 36.5 41.0 43.0

POP 12.6 15.2 27.0 29.2 30.3 33.5 35.1 36.7

POO 11.0 -4.0 18.2 – 19.0

SOO 1.8 24

Furthermore, a small endotherm was observed at 50 ºC (Figures 6.1-6.2, Sections B

and C; insets iv-v). This endotherm is believed to be due to increased fraction of

saturated TAG as a result of subsequent storage of CB suppositories at room

temperature. Loisel et al. (1998) observed that lipid segregation occurred in CB during

175
storage at 30 ºC which resulted in crystallisation of a saturated TAG, in addition to the

usual form V polymorph (Forms 4A and 4B as per nomenclature in Table 2.1). The

melting point of polymorphic forms observed in various saturated TAG was tabulated

in Table 6.3.

Table 6.3: The melting points of various polymorphic forms of saturated TAG.

Melting point of polymorphic forms (ºC)

TAG (Belitz et al., 2009; Da Silva et al., 2009; Sato and Kuroda, 1987)

α β’ β

Trilaurin 15.2 34 46.5

Trimyristin 32.8 45 58.5

Tripalmitin 44.7 56.5 66.4

Tristearin 54 65 72.5

Figures 6.4-6.7 showed the thermal profile of 5 %w/w PVP and CMCTS suppositories

made using HPKS (CE and SS). The progression of thermal changes in both CE and

SS were similar and findings from Figures 6.4-6.7 will be discussed using CE

suppositories containing DcNa and 5 %w/w PVP as reference (Figure 6.4).

CE suppositories containing 5 %w/w of PVP (Figure 6.4) which were kept

refrigerated for 100 and 200 days (melting point= 34.0 ºC on both occasions) did not

show any significant changes in melting point compared to freshly prepared samples

(melting point=33.9 ºC). The melting points were similar and there were no additional

endothermic or exothermic events even up to 200 days of refrigeration (3.5 ± 1.5 °C;

RH 29 ± 3 %).

176
Figure 6.4 : The DSC thermogram of suppositories made using CE as suppository base containing 5 %w/w PVP. Individual thermograms

show the melting endotherm of suppositories which were (i) freshly prepared; stored refrigerated at for (ii) 100 days and (iii) 200 days; stored

at room temperature for (iv) 100 days and (v) 200 days. Inset shows enlarged portions of the thermogram.
177
Figure 6.5 : The DSC thermogram of CE suppositories containing 50 mg DcNa and 5 %w/w CMCTS. Individual thermograms show the

melting endotherm of suppositories which were (i) freshly prepared; stored refrigerated at for (ii) 100 days and (iii) 200 days; stored at room

temperature for (iv) 100 days and (v) 200 days. Inset shows enlarged portions of the thermogram.
178
Figure 6.6 : The DSC thermogram of SS suppositories containing 50 mg DcNa and 5 %w/w PVP. Individual thermograms show the melting

endotherm of suppositories which were (i) freshly prepared; stored refrigerated for (ii) 100 days and (iii) 200 days; stored at room temperature

for (iv) 100 days and (v) 200 days. Inset shows enlarged portions of the thermogram.
179
Figure 6.7 : The DSC thermogram of SS suppositories containing 50 mg DcNa and 5 %w/w CMCTS. Individual thermograms show the

melting endotherm of suppositories which were (i) freshly prepared; stored refrigerated for (ii) 100 days and (iii) 200 days; stored at room

temperature for (iv) 100 days and (v) 200 days. Inset shows enlarged portions of the thermogram.
180
Conversely, an additional peak was observed at 27.3 and 29.4 ºC in suppositories

stored at room temperature for 100 and 200 days respectively (Figures 6.4(iv-v),

arrow). These peaks were likely due to separation of lower melting point TAG

from the complex multicomponent mixture which makes up HPKS. Furthermore,

another shoulder peak was observed at approximately 38 ºC in suppositories kept

at room temperature (Figures 6.4(iv-v); Section A). This could be a result of

gradual β stabilisation of trilaurin (C36) which is present as 26–29 % of HPKS

content (Chin et al., 2003; Smith et al., 2004; Siew, 2001).

An endothermic event was observed at 50 ºC (Figure 6.4, insets B and C) but at a

bigger magnitude compared to CB. This is probably due to the melting of

trisaturated TAG. CB contains approximately 3 % of trisaturated TAG while

HPKS contains approximately 7 % these high melting trisaturated TAG (Smith et

al., 2004). The amount of TAG component which melts at 50 ºC in samples stored

at room temperature increased from 100 days (Figure 6.4, inset B) to 200 days

(Figure 6.4, inset C) of storage. This was consistent in all the HPKS formulations,

suggesting that storage at room temperature resulted in an increased formation of

this TAG species.

The SFC curves of CE and SS suppositories (Figures 6.8-6.9) stored under

refrigeration and at room temperature were similar up to 35 °C; after which the

curve for suppositories stored at room temperature shift rightward, indicating the

presence of TAG with a higher melting point. This was different from CB

suppositories where the entire SFC curve shifted to a higher temperature (Figure

6.3).

181
(a) 120

100 freshly prepared sample

Solid fat content (%)

80 refrigerated 100 days

60 refrigerated 200 days

40 room temperature 100 days

20 room temperature 200 days

0
0 10 20 30 40 50
Temperature (°C)

(b) 120

100

freshly prepared sample

Solid fat content (%)

80
refrigerated 100 days

60
refrigerated 200 days

40 room temperature 100 days

20 room temperature 200 days

0
0 10 20 30 40 50
Temperature (°C)

Figure 6.8 : The SFC of CE suppositories containing 50 mg DcNa and (a) 5 %w/w

PVP and (b) 5 %w/w CMCTS.

182
(a) 120

100
Solid fat content (%) freshly prepared sample
80
refrigerated 100 days

60
refrigerated 200 days

40 room temperature 100 days

20 room temperature 200 days

0
0 10 20 30 40 50
Temperature (°C)

(b) 120

100

freshly prepared sample

Solid fat content (%)

80
refrigerated 100 days

60
refrigerated 200 days

40 room temperature 100 days

20 room temperature 200 days

0
0 10 20 30 40 50
Temperature (°C)

Figure 6.9 : The SFC of SS suppositories containing 50 mg DcNa and (a) 5 %w/w

PVP and (b) 5 %w/w CMCTS.

183
6.3.3 Hardness

Figures 6.10a-c showed changes in hardness of suppositories stored under different

storage conditions for different durations.

(a) freshly prepared

160 * *
140 * * *
120 refrigerated 100
days
Hardness (N)

100
80 refrigerated 200
60 days
40 room temperature
20 100 days
0 room temperature
PVP CMCTS 200 days

(b) 140 * freshly prepared

*
120 *
100 refrigerated 100
Hardness (N)

days
80
refrigerated 200
60 days
40 room temperature
20 100 days
0 room temperature
PVP CMCTS 200 days

(c) 140 freshly prepared

120 * * *
refrigerated 100
100 days
Hardness (N)

80 refrigerated 200
60 days
40 room temperature
20 100 days

0 room temperature
200 days
PVP CMCTS

Figure 6.10 : The hardness of suppositories made using (a) CB; (b) CE and (c) SS

containing 50 mg DcNa and 5 %w/w PVP and 5 %w/w CMCTS after various storage

conditions up to 200 days. Asterisks indicate a significant difference in hardness from

‘freshly prepared’ suppositories. Mean ± SD, n=6.

184
Refrigerated suppositories containing 5 %w/w PVP demonstrated statistically

significant increase in hardness at both 100 and 200 days (p <0.05). In general,

refrigerated 5 %w/w PVP suppositories were harder than suppositories kept at room

temperature for the same period.

Meanwhile, suppositories with 5 %w/w CMCTS showed very minimal differences in

hardness even when stored for 200 days both refrigerated and at room temperature. A

significant increase in hardness was only observed in CB and SS suppositories

containing 5 %w/w CMCTS after refrigeration for 200 days. Statistical comparison of

the formulations using Tukey’s HSD analysis is tabulated in Appendix 36.

Sah and Saini (2008) found that lipophilic indomethacin suppositories made using

Mayol W45 and Hydrokote AP5 became harder after being subjected to freeze-thaw

cycles or accelerated stability test at 30 ºC. The current work however, did not find a

clear trend of changes in hardness of the suppositories containing 5 %w/w of PVP and

CMCTS against storage time and storage conditions.

6.3.4 Softening time

Figure 6.11 showed that suppositories stored at room temperature have longer

softening times. This was consistent for suppositories containing both PVP and

CMCTS. All the formulations stored at room temperature for 100 and 200 days

(except SS suppositories containing 5 %w/w PVP at room temperature for 100 days)

had significantly prolonged softening times compared to freshly prepared

suppositories.

185
(a) 8 freshly prepared
* *
7 * *

Softening Time (min)

6 refrigerated 100
5
* * days
4 refrigerated 200
3 days
2 room temperature
1 100 days
0 room temperature
PVP CMCTS 200 days

(b) 8 *
* freshly prepared
7 *
*
Softening Time (min)

6
* refrigerated 100
5 days
4
refrigerated 200
3 days
2
room temperature
1 100 days
0
room temperature
PVP CMCTS
200 days

(c) 8
* freshly prepared
7 * *
Softening Time (min)

6
refrigerated 100
5 *
days
4 refrigerated 200
3 days
2 room temperature
100 days
1
room temperature
0
200 days
PVP CMCTS

Figure 6.11 : The softening time of suppositories made using (a) CB; (b) CE and (c)

SS containing 50 mg DcNa and 5 %w/w PVP and 5 %w/w CMCTS under different

storage conditions. Asterisks indicate a significant difference in softening time

compared to ‘freshly prepared’ suppositories. Mean ± SD, n=6.

186
With the exception of CB suppositories containing 5 %w/w PVP (Figure 6.11a), all

other refrigerated formulations have shorter softening times compared to freshly

prepared samples. This observation however, was only statistically significant in CE

suppositories containing 5 %w/w PVP and SS suppositories containing 5 %w/w

CMCTS after refrigeration for 100 days. Statistical comparison of the formulations

using Tukey’s HSD analysis is tabulated in Appendix 37.

Moes and Jaminet (1976) observed marked increase in liquefaction time of

suppositories measured at 37 ºC after prolonged storage at 30 ºC. However, the same

authors also found that increment in liquefaction time may or may not affect rectal

absorption of drugs. Other factors such as change in viscosity and melting point as a

result of ageing could have brought about a cumulative synergistic or contradictory

effect on rectal absorption.

6.3.5 DcNa release

Both Figures 6.12-6.13 showed that refrigeration of suppositories up to 200 days did

not alter DcNa release from all the formulations tested, with the exception of CB

suppositories containing 5 %w/w CMCTS (Figure 6.13a). Refrigerated CB

suppositories containing 5 %w/w CMCTS showed a slight and gradual decrease in

rate of DcNa released over time (from freshly prepared to 100 and 200 days). Another

study reported reduction in ampicillin release from suppositories stored at 4 °C over a

period of 240 days and the degree of reduction was dependent on type of base used as

well as the medicament used (Hosny et al., 1990).

187
(a) 120 freshly prepared

Cummulative drug release (%)

100
refrigerated 100
days
80
refrigerated 200
60 days

40 room temperature
100 days
20
room temperature
200 days
0
0 30 60 90 120 150 180
Time (min)

(b) 120 freshly prepared

Cummulative drug release (%)

100
refrigerated 100
80 days

refrigerated 200
60
days

40 room temperature
100 days
20
room temperature
0 200 days
0 30 60 90 120 150 180
Time (min)

(c) 120 freshly prepared

Cummulative drug release (%)

100
refrigerated 100
80 days

60 refrigerated 200
days
40
room temperature
100 days
20
room temperature
0 200 days
0 30 60 90 120 150 180
Time (min)

Figure 6.12: The cumulative DcNa release from (a) CB; (b) CE and (c) SS

suppositories containing 50 mg DcNa and 5 %w/w PVP under different storage

conditions and duration. Mean ± 2 SE, n=3.

188
(a) 120

Cummulative drug release (%)

freshly prepared
100

80 refrigerated 100 days

60 refrigerated 200 days

40
room temperature
100 days
20
room temperature
0 200 days
0 30 60 90 120 150 180
Time (min)

(b) 120
Cummulative drug release (%)

freshly prepared
100
refrigerated 100 days
80

60 refrigerated 200 days

40 room temperature
100 days
20
room temperature
0 200 days
0 30 60 90 120 150 180
Time (min)

(c) 120
Cummulative drug release (%)

freshly prepared
100

80 refrigerated 100 days

60 refrigerated 200 days

40
room temperature
20 100 days

room temperature
0 200 days
0 30 60 90 120 150 180
Time (min)

Figure 6.13: The cumulative DcNa release from (a) CB; (b) CE and (c) SS

suppositories containing 50 mg DcNa and 5 %w/w CMCTS under different storage

conditions and duration. Mean ± 2 SE, n=3.

189
Conversely, suppositories stored at room temperature for 100 and 200 days generally

had lower rates as well as lesser extent of DcNa release at 180 minutes. This

observation was independent of type of suppository base (CB, CE and SS) and type of

bioadhesive polymer incorporated (PVP, CMCTS). A study by De Blaey and Rutten-

Kingma (1977) reported a drastic decrease (50-60 %) in aminophylline release from

CB suppositories as early as 4 weeks storage at 22 °C and 6 days at 30 °C.

The parameters of drug release were numerically presented as the DE and MDT

values in Tables 6.4-6.5 respectively for easier comparison. Suppositories stored at

room temperature had significantly lower DE and higher MDT compared to freshly

prepared samples and the refrigerated samples. Among the suppositories stored at

room temperature, formulations containing 5 %w/w CMCTS have lower DE and

longer MDT compared to those containing 5 %w/w PVP. This suggests that

suppositories containing PVP have better resistance towards accelerated ageing

conditions although a definite reason for this observation is not known.

The decreased DcNa release could be explained via findings of the DSC thermogram

and SFC curves in Section 6.3.2. Storage of CB suppositories at room temperature

resulted in a 2–3 ºC rightward shift of the endothermic peak (higher melting points);

while a new endothermic shoulder peak was observed at approximately 38 ºC in

HPKS suppositories. Both these observations lead to an increase in SFC at 37 ºC; thus

preventing complete melting of the base which in turn hinders DcNa release from the

base. Suppositories stored at room temperature also had lower initial rates of DcNa

release, consistent with the longer softening times observed in these samples in

Section 6.3.4.

190
Table 6.4 : The DE (%) of suppositories made using CB, CE and SS containing 50 mg DcNa and 5 %w/w bioadhesive polymer (PVP, CMCTS).

Asterisks indicate a significant difference in DE compared to ‘freshly prepared’ suppositories. Mean ± SD, n=3.

DE (%)
Formulation
Refrigerated Room temperature
Freshly prepared
Base Polymer 100 days 200 days 100 days 200 days

CB PVP 90.951 ± 0.641 92.337 ± 0.957 89.299 ± 1.410 88.925 ± 0.642 * 80.546 ± 0.630 *

CE PVP 85.829 ± 0.665 85.244 ± 1.406 87.656 ± 4.106 51.840 ± 0.651 * 56.099 ± 2.383 *

SS PVP 86.270 ± 1.382 87.557 ± 0.862 89.823 ± 2.104 * 75.081 ± 1.344 * 79.269 ± 0.421 *

CB CMCTS 91.673 ± 1.099 91.984 ± 1.861 85.497 ± 1.132 * 82.913 ± 0.774 * 51.139 ± 3.250 *

CE CMCTS 92.039 ± 1.021 90.238 ± 2.207 90.379 ± 0.134 59.904 ± 4.085 * 36.582 ± 0.697 *

SS CMCTS 91.638 ± 1.551 93.089 ± 0.236 93.620 ± 0.663 74.913 ± 1.386 * 67.064 ± 1.118 *
191
Table 6.5 : The MDT (minutes) of suppositories made using CB, CE and SS containing 50 mg DcNa and 5 %w/w bioadhesive polymer (PVP,

CMCTS). Asterisks indicate a significant difference in MDT compared to ‘freshly prepared’ suppositories. Mean ± SD, n=3.

MDT (min)
Formulation
Refrigerated Room temperature
Freshly prepared
Base Polymer 100 days 200 days 100 days 200 days

CB PVP 24.740 ± 2.285 14.199 ± 0.835 * 28.009 ± 3.507 27.639 ± 1.839 39.763 ± 5.428 *

CE PVP 21.478 ±1.932 18.757 ± 1.437 24.159 ± 0.955 106.426 ± 16.754 * 93.973 ± 7.685 *

SS PVP 17.531 ± 0.511 17.369 ± 2.012 15.405 ± 2.577 42.026 ± 5.608 * 32.946 ± 3.310 *

CB CMCTS 20.001 ± 2.660 26.794 ± 3.816 39.124 ± 2.632 * 52.453 ± 2.684 * 112.75 ± 6.897 *

CE CMCTS 24.616 ± 3.082 27.098 ± 2.320 25.317 ± 0.424 86.192 ± 4.466 * 92.735 ± 8.642 *

SS CMCTS 18.082 ± 3.263 16.896 ± 1.398 19.175 ± 1.055 52.315 ± 6.651 * 75.695 ± 15.255 *
192
6.4 Conclusion

In general, suppositories which were stored at room temperature for up to 200 days

(accelerated ageing conditions) resulted in: (1) compromised aesthetic values in terms

of loss of glossiness and increasing graininess; (2) increased melting point; (3)

possible TAG separation suggested by the presence of new endothermic peaks; (4)

higher solid fat content at 37 °C; (5) prolonged softening times; and (6) decreased rate

and amount of DcNa release. Such changes to the tested formulations were

unfavourable and could potentially lead to treatment failure in patients. Based on

stability studies conducted in this chapter, the suppositories tested were better suited

for storage in the refrigerator rather than at room temperature. There were no

conclusive finding of superior stability at room temperature among HPKS and CB

formulations tested as both exhibited notable changes in terms of thermal profiles and

drug release after subjection to the accelerated ageing process. However, SS

suppositories appear to release DcNa more efficiently than CE suppositories after

being exposed to the same accelerated ageing process.

193
CHAPTER 7

GENERAL CONCLUSION AND

FUTURE WORK

194
7.1 General conclusion

There has been substantial interest in utilising alternative fat sources as suppository

bases due to the occurrence of up to six polymorphic forms in CB, one of the oldest

lipophilic bases used in manufacturing of suppositories (Loisel et al., 1998;

Marangoni and McGauley, 2003). The presence of various polymorphic forms and

rigid processing requirements were a hindrance to stability and storage as well as

industrial scale up. The HPKS, which have long been used in the chocolate and

confectionary industry as CB substitutes in the coating of candies, caramel

centrefilling and manufacturing of chocolate bars were evaluated as a potential

alternative to CB for the manufacturing of suppositories.

The model drug DcNa, a nonsteroidal anti-inflammatory drug (NSAID) has been

marketed for well over 30 years as oral tablets, suppositories, injectables and topical

creams. Although not as popular as oral administration, suppositories may prove

valuable in conditions where patients are unable to swallow their medication or are

inaccessible to a qualified caregiver for parenteral administration. In fact, a recent

study by van der Marel et al. (2004) found that DcNa suppositories administered in

children undergoing tonsilectomy had higher relative bioavailability and needed a

shorter time to achieve maximum plasma concentration compared to oral enteric

coated tablets of equivalent doses.

This drug however, has been reported to undergo substantial first pass metabolism,

resulting in oral bioavailability of approximately 55 % (Willis et al., 1979). Drugs

administered rectally may avoid the presystemic circulation if absorbed in the lower

rectum which drains into the inferior and middle hemorrhoidal veins flowing directly

195
into the systemic circulation, bypassing presystemic metabolism pathways (Allen et

al., 2008; Kokate et al., 2006).

Therefore, this research sought to (1) evaluate HPKS as an alternative lipophilic base

and (2) characterise DcNa bioadhesive suppositories produced by incorporation of

bioadhesive polymers (CBP, PVP, HPMC, CMCTS) to produce suppositories which

preferentially be retained at the lower rectum, thus bypassing presystemic metabolism

pathways.

The two HPKS used in this study (CE and SS) were found to be suitable lipophilic

suppository bases. They were comparable to CB in terms of thermal profile, SFC, pH,

viscosity and DV but with added benefits of lesser polymorphic forms and less rigid

manufacturing requirements. This study concluded that to manufacture CB

suppositories extemporaneously, the molten base should be maintained between 34 ±

0.5 °C during heating and allowed to cool slowly to between 20–24 °C to produce

suppositories with the desirable form 4A polymorph. Molten CB should not be placed

into a refrigerator before solidification is complete as it will result in rapid

crystallisation into the form 2 polymorph. These rigid processing restrictions however,

were not relevant in HPKS. The HPKS allowed more manufacturing flexibility

compared to CB.

Furthermore, the finished products (HPKS bioadhesive suppositories) were deemed

suitable for rectal administration as melting points were within the range of 32.5-

35.5 °C. The addition of bioadhesive polymers did not significantly alter melting

point (less than ±1 °C) of the suppositories and all formulations tested contained

196
>95 % of the stipulated DcNa content. Softening times recorded for all the

formulations were between 3–7 minutes which were acceptable for rectal drug release.

Viscosity of the molten suppositories were enhanced with increasing amount of

bioadhesive polymers (CBP, PVP, HPMC, CMCTS) incorporated into the

formulation and this may be beneficial for retention of suppositories within the lower

rectum by limiting its spread.

Apart from physical evaluation, the HPKS bases appeared to be comparable to CB in

terms of drug release capacity and could be good lipophilic base candidates for fast-

acting DcNa suppository formulations. Mathematical modelling of the data found that

DcNa release in suppositories without polymer was via non-Fickian diffusion kinetics

(Korsmeyer-Peppas model). Addition of 1-5 %w/w HPMC, PVP and CMCTS to the

suppositories did not alter the mechanism of DcNa release, and these were decent

candidates of polymers for development of bioadhesive suppositories. However,

formulations added with 1-5 %w/w CBP significantly suppressed DcNa release

compared to their respective blanks (DcNa only suppositories). The addition of CBP

lead to considerable change in morphology of molten suppository during dissolution

via gelling, resulting in a biphasic DcNa release process involving a rapid initial

diffusion and erosion process followed by a slow diffusion process across the CBP gel

layer. Furthermore, as CBP gels in the dissolution medium, it decreases the

environment pH which lowers DcNa solubility, thus further retarding the release of

DcNa from the suppositories.

Evaluation of bioadhesive properties of suppository formulations using the two

fabricated in vitro methods in current research found that formulations containing

197
PVP exhibited superior bioadhesion compared to the other polymers when subjected

to both the tensile and shear forces of detachment. This finding was promising for the

development of bioadhesive suppositories. CBP displayed good bioadhesive

properties only via tensile stress measurement while CMCTS showed appreciable

shear stress of bioadhesion. Conversely, HPMC exhibited poor bioadhesivity in both

tests and has limited role in the development of bioadhesive suppositories. Although

both methods used were temperature controlled to simulate in vivo conditions,

evaluation of shear stress of bioadhesion is more likely to reflect actual attachment

and detachment of a suppository in the human rectum. Meanwhile, colon mucosa and

regenerated cellulose membranes were also found to be good substitutes for rectal

membranes in qualitative evaluation of bioadhesion in suppositories.

Only suppositories (CB, CE and SS) containing 5 %w/w PVP and 5 %w/w CMCTS

were subjected to stability assessment, while suppositories containing 1-5 %w/w of

CBP and 1-5 %w/w HPMC were omitted from further development due to their

limited benefit in the formulation of bioadhesive suppositories. The concentration

dependent suppression of DcNa release from CBP suppositories indicate that CBP

should only be used at the lowest possible concentration but lower concentrations of

CBP (1 %w/w) exhibited poor bioadhesive properties. HPMC suppositories on the

other hand demonstrated poor bioadhesive properties at all concentrations tested (1-

5 %w/w).

In general, suppositories stored at room temperature for up to 200 days (accelerated

ageing conditions) resulted in: (1) compromised aesthetic values in terms of loss of

glossiness and increasing graininess; (2) increase in melting point; (3) possible TAG

198
separation suggested by the presence of new endothermic peaks; (4) higher SFC at

37 °C; (5) prolonged softening times; and (6) decreased rate and amount of DcNa

release. Such changes to the tested formulations were unfavourable and could

potential lead to treatment failure in patients. Based on the stability studies conducted

in this study, the suppositories tested were better suited for storage in refrigerator

rather than at room temperature. There were no conclusive findings of formulations

with superior stability at room temperature among the HPKS and CB formulations

tested as both exhibited notable changes in terms of thermal profiles and drug release

after subjection to the accelerated aging process. However, SS suppositories appeared

to release DcNa more efficiently than CE suppositories after being exposed to the

same accelerated ageing process.

Among all the formulations developed and tested, SS suppositories containing

5 %w/w PVP appeared to be the most suitable candidate for future development of

bioadhesive DcNa suppositories because (1) it has less rigid manufacturing

requirements compared to CB; (2) melts rapidly at human body temperature; (3) did

not cause retardation of DcNa release and (4) exhibits good bioadhesive properties.

However, additional work is required for in vivo studies to evaluate the retention

capacity of the suppositories within the lower rectum as well as the actual

bioavailability of these bioadhesive suppositories.

7.2 Future work

The work described in this thesis has been concerned with characterisation of two

HPKS as an alternative suppository base to CB, as well as to develop bioadhesive

suppositories by incorporating bioadhesive polymers into the two HPKS bases. A

199
number of problems and challenges were encountered during the course of this work

which suggests research directions to be pursued for further development of a

bioadhesive suppository system.

7.2.1 Evaluation of bioadhesive suppository migration in the rectum

The bulk of this work has been focused on the development of methods to study

bioadhesion as well as evaluation of bioadhesive properties of suppository

formulations. However, further studies should be designed to investigate the

migration and disposition of these bioadhesive suppositories in the rectum compared

to conventional suppositories. One of the methods would be to administer

suppositories stained with brilliant blue to fasted Wistar rats. The rat rectums were

then excised and length of the coloured regions reflected distance travelled by

suppositories (Yahagi et al., 1999). An alternative method which avoids sacrificial of

test subjects is by using gamma-scintigraphy. Administration of suppositories

radiolabelled with technetium hydroxymethyldiphosphonate (99m Tc) into volunteers

would allow continuous observation of suppository migration along the human rectum

via external scintigraphy (Jay et al., 1985; Sugito et al., 1988).

7.2.2 Incorporation of DcNa-polymer granules

Incorporation of bioadhesive polymers in this study has been via a direct solid

dispersion in the semi-solid base. This was done to increase viscosity as well as to

adhere the base matrix to the lower rectum. An alternative for further investigation

would be to prepare bioadhesive granules containing drug and bioadhesive polymer

(PVP and CMCTS) for incorporation into the base. These bioadhesive granules can be

prepared using wet granulation methods similar to that in manufacturing of tablets

200
(Attama et al., 2000). This approach could improve adhesion of DcNa to the rectal

mucosa by concentrating the polymers around the DcNa granules

7.2.3 Introduction of synthetic cellulose membrane as alternative to

biological membranes

Current work has found that synthetic cellulose membranes were well correlated to

porcine colon and rectum membranes and could be a possible alternative to biological

membranes using the newly devised experimental setup for measurement of the shear

forces of bioadhesion. Further investigations should be conducted using other

biological and synthetic membranes to evaluate and validate this proposed correlation

between biological and synthetic membranes at different experimental parameters.

7.2.4 In vivo bioavailability studies

This work has evaluated in vitro release of DcNa from the suppositories

comprehensively; however, further in vivo evaluation using animal models or human

test subjects would provide a better depiction of the systemic bioavailability of DcNa

from bioadhesive suppositories. The pharmacokinetics and metabolism of DcNa in

Yucatan miniature pigs (Oberle et al., 1994) and rats (Reiss et al., 1978) were similar

to man. While avoidance of first pass metabolism via the rectal route was

demonstrated in rabbits (Kurosawa et al., 1998) and rats (De Leede et al., 1983),

where bioavailability of drugs increased as the site of absorption was closer to the

anus. These observations were similar to that observed in man. Therefore, rats would

be a reasonable model for in vivo studies. Aoyagi et al. (1988) studied bioavailability

of indomethacin suppository in rabbits and pigs by administering the suppository to

the fasted test subjects and plasma samples were withdrawn for analysis at specific

201
intervals. Comparison of DcNa bioavailability between bioadhesive and conventional

suppositories would provide an indication of effectiveness in avoidance of first pass

metabolism provided by the addition of bioadhesive polymers.

There is clearly much work to be done in the development of a bioadhesive

suppository system. Perhaps, the most direct extension to this work is the evaluation

of bioadhesive suppository migration in the rectum post administration. Comparison

of shortlisted formulations from this work against conventional suppositories would

elucidate feasibility and practicality of a bioadhesive suppository.

202
BIBLIOGRAPHY

203
Ahmad, W., 2001. Formulation, release and bioavailability study of some important
drugs from various suppository bases. University of the Punjab, Lahore.

Ahmed, M., Jalil, R., Islam, M.A., Shaheen, S.M., 2000. Preparation and stability of
diclofenac suppositories. Pak. J. Pharm. Sci. 3, 1755–1757.

Ahuja, A., Khar, R.K., Ali, J., 1997. Mucoadhesive drug delivery systems. Drug Dev.
Ind. Pharm. 23, 489–515.

Aiache, J.M., Islasse, M., Beyssac, E., Aiache, S., Renoux, R., Kantelip, J.P., 1987.
Kinetics of indomethacin release from suppositories. In vitro-in vivo correlation. Int.
J. Pharm. 39, 235–242.

Akbari, J., Saeedi, M., Enayatifard, R., Doost, M.S., 2010. Development and
evaluation of mucoadhesive chlorhexidine tablet formulations. Trop. J. Pharm. Res. 9,
321–327.

Akinoso, R., Raji, A.O., 2011. Physical properties of fruit, nut and kernel of oil palm.
Int. Agrophysics. 25, 85–88.

Alladi, K.K., Suram, R., Bela, M., Kiran, S., Ramesh, V., Narendra, Y., 2011.
Formulation and characterisation of clarithromycin controlled released bioadhesive
tablets. J. Chem. Pharm. Res. 3, 684–690.

Allen, C.A.W., Watts, K.C., Ackman, R.G., Pegg, M.J., 1999. Predicting the viscosity
of biodiesel fuels from their fatty acid ester composition. Fuel. 78, 1319–1326.

Allen, L.V., Worthen, D.B., Mink, B., 2008. Suppositories, First. ed. Pharmaceutical
Press, Wiltshire.

Allen, L.V.J., Popovich, N.G., Ansel, H.C., 2005. Suppositories and inserts, in: Allen,
L.V.J. (Ed.), Ansel’s Pharmaceutical Dosage Forms and Drug Delivery Systems.
Lippincott Williams & Wilkins, Philadelphia, pp. 316–335.

Anderson, N.H., Bauer, M., Boussac, N., Khan-Malek, R., Munden, P., Sardaro, M.,
1998. An evaluation of fit factors and dissolution efficiency for the comparison of in
vitro dissolution profiles. J. Pharm. Biomed. Anal. 17, 811–822.

Andrews, G.P., Laverty, T.P., Jones, D.S., 2009. Mucoadhesive polymeric platforms
for controlled drug delivery. Eur. J. Pharm. Biopharm. 71, 505–518.

Anihouvi, P.P., Blecker, C., Dombree, A., Danthine, S., 2013. Comparative study of
thermal and structural behavior of four industrial lauric fats. Food Bioprocess
Technol. 6, 3381–3391.

Ansel, H.C., 1981. Suppositories and Other Rectal, Vaginal, and Urethral
Preparations, in: Ansel, H.C. (Ed.), Introduction to Pharmaceutical Dosage Forms.
Lea & Febiger, Philadelphia, pp. 328–344.

204
Aoyagi, N., Kaniwa, N., Takeda, Y., Uchiyama, M., Takamura, F., Kido, Y., 1988.
Release rates of indomethacin from commercial witepsol suppositories and the
bioavailabilities in rabbits and pigs. Chem. Pharm. Bull. 36, 4933–4940.

Arishima, T., Sagi, N., Mori, H., Sato, K., 1991. Polymorphism of POS. I. occurrence
and polymorphic transformation. J. Am. Oil Chem. 68, 710–715.

Aronhime, J.S., Sarig, S., Garti, N., 1988. Reconsideration of polymorphic

transformations in cocoa butter using the DSC. J. Am. Oil Chem. Soc. 65, 1140–1143.

Attama, A.A., Adikwu, M.U., Okoli, N.D., 2000. Studies on bioadhesive granules I:
granules formulated with Prosopis africana (prosopis) gum. Chem. Pharm. Bull.
(Tokyo). 48, 734–737. doi:10.1248/cpb.48.734

Azechi, Y., Ishikawa, K., Mizuno, N., Takahashi, K., 2000. Sustained release of
diclofenac from polymer-containing suppository and the mechanism involved. Drug
Dev. Ind. Pharm. 26, 1177–1183.

Babar, A., Bellete, T., Plakogiannis, F.M., 1999. Ketoprofen suppository dosage
forms : In vitro release and in vivo absorption studies in rabbits. Drug Dev. Ind.
Pharm. 25, 241–245.

Babul, N., Darke, A.C., Anslow, J.A., Krishnamurthy, T.N., 1992. Pharmacokinetics
of two novel rectal controlled-release morphine formulations. J. Pain Symptom
Manage. 7, 400–405.

Bahar, M.M., Jangjoo, A., Soltani, E., Armand, M., Mozaffari, S., 2010. Effect of
preoperative rectal indomethacin on postoperative pain reduction after open
cholecystectomy. J. Perianesthesia Nurs. 25, 7–10.

Bansil, R., Turner, B.S., 2006. Mucin structure, aggregation, physiological functions
and biomedical applications. Curr. Opin. Colloid Interface Sci. 11, 164–170.

Barakat, N.S., 2009. In vitro and in vivo characteristics of a thermogelling rectal

delivery system of etodolac. AAPS PharmSciTech. 10, 724–731.

Barleben, A., Mills, S., 2010. Anorectal Anatomy and Physiology. Surg Clin N Am
90, 1–15.

Basak, S.C., Kumar, K.S., Ramalingam, M., 2008. Design and release characteristics
of sustained release tablet containing metformin HCl. Rev. Bras. Ciências Farm. 44,
477–483.

Beckett, S.T., 2008. The Science of Chocolate, Second. ed. RSC Publishing,
Cambridge, United Kingdoms.

Belitz, H.-D., Grosch, W., Schieberie, P., 2009. Lipids, in: Belitz, H.D., Grosch, W.,
Schieberie, P. (Eds.), Food Chemistry. Springer, Liepzig, pp. 158–247.

205
Bertram, U., Bodmeier, R., 2006. In situ gelling, bioadhesive nasal inserts for
extended drug delivery: In vitro characterization of a new nasal dosage form. Eur. J.
Pharm. Sci. 27, 62–71.

Bhat, S.R., Shivakumar, H.G., 2010. Bioadhesive controlled release clotrimazole

vaginal tablets. Trop. J. Pharm. Res. 9, 339–346.

Bonner, S., Mallinder, P., Dodds, C., 1996. Patient consent and rectal drug
administration. Anaesthesia.

Bottger, W.M., Schoonen, B.J.., Moolenaar, F., Visser, J., Meijer, K.F., 1989. A study
on the buffering activity of the human rectum. Pharm. Weekbl. Sci. Ed. 11, 9 – 12.

British Pharmacopoeia, 2015. . The stationary office, London.

Brown, C.K., Friedel, H.D., Barker, A.R., Buhse, L.F., Keitel, S., Cecil, T.L.,
Kraemer, J., Morris, J.M., Reppas, C., Stickelmeyer, M.P., Yomota, C., Shah, V.P.,
2011. Meeting report : FIP/AAPS joint workshop report : dissolution/in vitro release
testing of novel/special dosage forms. AAPS PharmSciTech. 12, 51–64.

Bryan, C.P., Smith, G.E., 1930. The Papyrus Ebers. Translated from the German
Version.

Buchan, B.E., 2011. Formulation studies on cysteamine for the treatment of

nephropathic cyctinosis. Robert Gordon University.

Carroll, M., Day, F., Hennessy, A., Buggy, D., Cooney, C., 1996. Patient attitudes to
perioperative suppository administration for postoperative analgesia. Irish J. Med. Sci.
165, 286–288.

Chaiseri, S., Dimick, P.S., 1989. Lipid and hardness characteristics of cocoa butters
from different geographic regions. J. Am. Oil Chem. 66, 1771–1776.

Chapman, G.M., Akehurst, E.E., Wright, W.B., 1971. Cocoa butter and confectionery
fats. Studies using programmed temperature X-ray diffraction and differential
scanning calorimetry. J. Am. Oil Chem. 48, 824–830.

Chary, R.B.R.., Vani, G., Rao, Y.M., 1999. In vitro and in vivo adhesion testing of
mucoadhesive drug delivery systems. Drug Dev. Ind. Pharm. 25, 685–690.

Chen, S.C., Wu, Y.C., Mi, F.L., Lin, Y.H., Yu, L.C., Sung, H.W., 2004. A novel pH-
sensitive hydrogel composed of N,O-carboxymethyl chitosan and alginate cross-
linked by genipin for protein drug delivery. J. Control. Release 96, 285–300.

Chicco, D., Grabnar, I., Škerjanec, A., 1999. Correlation of in vitro and in vivo
paracetamol availability from layered excipient suppositories. Int. J. Pharm. 189, 147–
160.

206
Chin, H.H., Jinap, S., Salmah, Y., Suria, A.M., 2003. Palm Mid-Fraction ( PMF ) and
Palm Kernel Stearin ( RBD PKS ) as Filling Fats in Dark Chocolate and Desiccated
Coconut Fillings. ASEAN Food J. 12, 183–190.

Cho, C.W., Choi, J.S., 2011. Enhanced local anesthetic action of mepivacaine from
the bioadhesive gels. Pakistan J. Pharm. 24, 87–93.

Choi, H.G., Oh, Y.K., Kim, C.K., 1998. In situ gelling and mucoadhesive liquid
suppository containing acetaminophen: enhanced bioavailability. Int. J. Pharm. 165,
23–32.

Chuasuwan, B., Binjesoh, V., Polli, J.E., Zhang, H., Amidon, G.L., Junginger, H.E.,
Midha, K.K., 2009. Biowaiver monographs for immediate release solid oral dosage
forms : diclofenac sodium and diclofenac potassium. J. Pharm. Sci. 98, 1206–1219.

Coben, L.J., Lieberman, H.A., 1986. Suppositories, in: Lachman, L., Lieberman,
H.A., Kanig, J.L. (Eds.), The Theory and Practice of Industrial Pharmacy. Lea &
Febiger, Philadelphia, pp. 564–588.

Coben, L.J., Lordi, N.G., 1980. Physical stability of semisynthetic suppository bases.
J. Pharm. Sci. 69, 955–960.

Colbert, S.A., O’ Hanlon, D., McAnena, O., Flynn, N., 1998. The attitudes of patients
and health care personnel to rectal drug administration following day case surgery.
Eur. J. Anaesthesiol. 15, 422–426.

Collett, D.M., 1990. Suppositories and pessaries, in: Collett, D.M., Aulton, M.E.
(Eds.), Pharmaceutical Practice. Churchill Livingstone, Edinburgh, pp. 135–144.

Cooper, J.W., Gunn, C., 1987. Suppositories and pessaries, in: Carter, S.. (Ed.),
Cooper and Gunn’s Dispensing for Pharmaceutical Students. CBS Publishers, New
Delhi, pp. 232–264.

Costa, F.O., Sousa, J.J.S., Pais, A.A.C.C., Formosinho, S.J., 2003. Comparison of
dissolution profiles of Ibuprofen pellets. J. Control. Release. 89, 199–212.

Costa, P., 2001. An alternative method to the evaluation of similarity factor in

dissolution testing. Int. J. Pharm. 220, 77–83.

Costa, P., Sousa Lobo, J.M., 2001. Modeling and comparison of dissolution profiles.
Eur. J. Pharm. Sci. 13, 123–133.

Cremer Oleo GmBH & Co. KG, 2015. Witepsol : fatty bases for suppositories.

Cupera, J., 2009. Mathematical models of dissolution. Masarykova Univerzita.

Da Silva, E., Bresson, S., Rousseau, D., 2009. Characterization of the three major
polymorphic forms and liquid state of tristearin by Raman spectroscopy. Chem. Phys.
Lipids 157, 113–119.

207
Dash, A.K., Cudworth, G.C., 2001. Evaluation of an acetic acid ester of
monoglyceride as a suppository base with unique properties. AAPS PharmSciTech. 2,
32–38.

Dash, S., Murthy, P.N., Nath, L., Chowdhury, P., 2010. Kinetic modelling on drug
releasd from controlled drug delivery systems. Acta Pol. Pharm. 67, 217–223.

Davis, M.P., Walsh, D., LeGrand, S.B., Naughton, M., 2002. Symptom control in
cancer patients: the clinical pharmacology and therapeutic role of suppositories and
rectal suspensions. Support. Care Cancer. 10, 117–138.

De Blaey, C.J., Rutten-Kingma, J.J., 1977. Biopharmaceutics of aminophylline

suppositories II. In vitro release rate during storage. Pharm. Acta Helv. 52, 11–14.

De Blaey, C.J., Tukker, J.J., 1996. Suppositories and pessaries, in: Aulton, M. (Ed.),
Pharmaceutics: The Science of Dosage Form Design. Churchill Livingstone,
Leicaster, UK, pp. 412–422.

De Leede, L.G., De Boer, A.G., Roozen, C.P., Breimer, D.D., 1983. Avoidance of
“first-pass” elimination of rectally administered lidocaine in relation to the site of
absorption in rats. J. Pharmacol. Exp. Ther. 225, 181–185.

Desai, A., 2007. Rectal, Vaginal and Urethral Delivery, in: Lee, M., Desai, A. (Eds.),
Gibaldi’s Drug Delivery System in Pharmaceutical Care. American Society of Health-
System Pharmacist, Maryland, pp. 95–102.

Dewettinck, K., Foubert, I., 2004. Phase behavior of cocoa butter in a two-step
isothermal crystallization. Cryst. Growth Des. 4, 1295–1302.

Dhawan, N., Das, S., Kiran, U., Chauhan, S., Bisoi, A.K., Makhija, N., 2009. Effect
of rectal diclofenac in reducing postoperative pain and rescue analgesia requirement
after cardiac surgery. Pain Pract. 9, 385–393.

Dhonsi, D., Stapley, A.G.F., 2006. The effect of shear rate, temperature, sugar and
emulsifier on the tempering of cocoa butter. J. Food Eng. 77, 936–942.

Dodd, J.M., Hedayati, H., Pearce, E., Hotham, N., Crowther, C.A., 2004. Rectal
analgesia for the relief of perineal pain after childbirth: a randomised controlled trial
of diclofenac suppositories. BJOG An Int. J. Obstet. Gynaecol. 111, 1059–1064.

Dokoumetzidis, A., Macheras, P., 2006. A century of dissolution research: From

Noyes and Whitney to the biopharmaceutics classification system. Int. J. Pharm. 321,
1–11.

Edsman, K., Carlfors, J., Harju, K., 1996. Rheological evaluation and ocular contact
time of some carbomer gels for ophthalmic use. Int. J. Pharm. 137, 233–241.

Edsman, K., Hagerstrom, H., 2005. Pharmaceutical applications of mucoadhesion for

the non-oral routes. J. Pharm. Pharmacol. 57, 3–22.

208
Ehlers, D., 2012. Controlled crystallization of complex confectionery fats. University
of Karlsruhe.

El-Majri, M.A., Sharma, R.K., 2010. Formulation and evaluation of piroxicam

suppositories. Int. J. Drug Deliv. 2, 108–112.

Ermiş, D., Tarimci, N., 1995. Ketoprofen sustained-release suppositories containing

hydroxypropylmethylcellulose phthalate in polyethylene glycol bases. Int. J. Pharm.
113, 65–71.

European Pharmacopoeia, 7th ed, 2010. . European Directorate for the Quality of
Medicines and Health Care (EDQM), Nordlingen.

Farouk, R., 2003. Anorectal Physiology. Surgery. 21, 184a – 184e.

Fayaz, M.K., Abel, R.J., Pugh, S.C., Hall, J.E., Djaiani, G., Mecklenburgh, J.S., 2004.
Opioid-sparing effects of diclofenac and paracetamol lead to improved outcomes after
cardiac surgery. J. Cardiothorac. Vasc. Anesth. 18, 742–747.

Ghorab, D., Refai, H., Tag, R., 2011. Preparation and evaluation of fenoterol
hydrobromide suppositories. Drug Discov. Ther. 5, 311–318.

Gibaldi, M., Feldman, S., 1967. Establishment of sink conditions in dissolution rate
determinations. Theoretical considerations and application to nondisintegrating
dosage forms. J. Pharm. Sci. 56, 1238–1242.

Giron, D., 1986. Applications of thermal analysis in the pharmaceutical industry. J.

Pharm. Bomedical Anal. 4, 755–770.

Gjellan, K., Graffner, C., 1994. Comparative dissolution studies of rectal formulations
using the basket , the paddle and the flow-through methods : II. Ibuprofen in
suppositories of both hydrophilic and lipophilic types. Int. J. Pharm. 112, 233–240.

Gjellan, K., Graffner, C., Quiding, H., 1994. Influence of amount of hard fat in
suppositories on the in vitro release rate and bioavailability of paracetamol and
codeine . I. A comparison of three suppository compositions in vivo. Int. J. Pharm.
102, 71–80.

Goh, E.M., 1994. Formulation of lauric-oil containing food products and their
performance, in: Applewhite, T.H. (Ed.), The World Conference on Lauric Oils:
Sources, Processing and Applications. AOCS Press, Illinois, pp. 98–103.

Goldstein, D.A., Tan, Y.K., Soldin, S.J., 1987. Pharmacokinetics and absolute
bioavailability of salbutamol in healthy adult volunteers. Eur. J. Clin. Pharmacol. 32,
631–634.

Gomez-Carracedo, A., Souto, C., Martinez-Pacheco, R., Concheiro, A., Gomez-

Amoza, J.L., 2007. Microstructural and drug release properties of oven-dried and of
slowly or fast frozen freeze-dried MCC-Carbopol pellets. Eur. J. Pharm. Biopharm.
67, 236–245.

209
Grayson, J., 1951. Observations on the temperature of the human rectum. Br. Med. J.
2, 1379–1382.

Gros, A.T., Feuge, R.O., 1952. Surface and interfacial tensions, viscosities, and other
physical properties of some n-aliphatic acids and their methyl and ethyl esters. J. Am.
Oil Chem. Soc. 29, 313–317.

Gugulothu, D., Pathak, S., Suryavanshi, S., Sharma, S., Patravale, V., 2010. Self-
microemulsifiyng suppository formulation of β-artemether. AAPS PharmSciTech. 11,
1179–1184.

Güneri, T., Arici, M., Ertan, G., 2004. Preparation and diffusional evaluation of
sustained- release suppositories containing ibuprofen microspheres. FABAD J.
Pharm. Sci. 29, 177–184.

Guo, J.H., Skinner, G.W., Harcum, W.W., Barnum, P.E., 1998. Pharmaceutical
applications of naturally occuring water-soluble polymers. Pharm. Sci. Technol.
Today. 1, 1–13.

Hammouda, Y.E., Kasim, N.A., Nada, A.H., 1993. Formulation and in vitro
evaluation of verapamil HCl suppositories. Int. J. Pharm. 89, 111–118.

Hamzah, M.M., Reem, K.A., Ahmad, A.N., Othman, A.A., 2010. Effect of different
biological membranes on in vitro bioadhesion property. Drug Invent. Today. 2, 155–
159.

Hanaee, J., Javadzadeh, Y., Taftachi, S., Farid, D., Nokhodchi, A., 2004. The role of
various surfactants on the release of salbutamol from suppositories. Farm. 59, 903–
906.

Hancock, B.D., 1976. Measurement of anal pressure and motility. Gut. 17, 645–651.

Hinton, R.L., Auwun, A., Pongua, G., Oa, O., Davis, T.M.E., Karunajeewa, H.A.,
Reeder, J.C., 2007. Caregivers’ acceptance of using artesunate suppositories for
treating childhood malaria in Papua New Guinea. Am. J. Trop. Med. Hyg. 76, 634–
640.

Hombach, J., Bernkop-Schnurch, A., 2010. Mucoadhesive Drug Delivery, in:

Schafer-Korting, M. (Ed.), Drug Delivery. Berlin, Heidelberg, pp. 251–267.

Hosny, E.A., Kassem, A.A., Ei-Shattawy, H.H., 1990. Influence of aging on the
physical characteristics of ampicillin suppositories. Drug Dev. Ind. Pharm. 16, 1585–
1589.

Ibrahim, S.A., El-Faham, T.H., Tous, S.S., Mostafa, E.M., 1990. Formulation, release
characteristics and evaluation of ibuprofen suppositories. Int. J. Pharm. 61, 1–7.

Ikinci, G., Senel, S., Wilson, C.G., Sumnu, M., 2004. Development of a buccal
bioadhesive nicotine tablet formulation for smoking cessation. Int. J. Pharm. 277,
173–178.

210
Irons, B.K., Robinson, J.R., 2003. Bioadhesives in Drug Delivery, in: Pizzi, A.,
Mittal, K.. (Eds.), Handbook of Adhesive Technology. Marcel Dekker Inc., New
York, pp. 615–627.

Islam, M.S., Rahman Khan, M.Z., Khatun, M., Chowdhury, J.A., Reza, M.S., 2012.
Studies on bio-adhesion of matrix tablets: II.Comparison on bio-adhesion strength and
release profiles of theophylline anhydrous and its sodium glycinate salt. Dhaka Univ.
J. Pharm. Sci. 10, 1–7.

Ivarsson, D., Wahlgren, M., 2012. Comparison of in vitro methods of measuring

mucoadhesion: ellipsometry, tensile strength and rheological measurements. Colloids
surfaces B Biointerfaces. 92, 353–359.

Janicki, S., Sznitowska, M., Zebrowska, W., Gabiga, H., Kupiec, M., 2001.
Evaluation of paracetamol suppositories by a pharmacopoeial dissolution test-
comments on methodology. Eur. J. Pharm. Biopharm. 52, 249–254.

Jantzen, J.P.A.H., Tzanova, I., Witton, P.K., Klein, A.M., 1989. Rectal pH in
children. Can. J. Anaesth. 36, 665–667.

Jay, M., Beihn, R.M., Digenis, G.A., Deland, F.H., Caldwell, L., Mlodoseniec, A.R.,
1985. Disposition of radiolabelled suppositories in humans. J. Pharm. Pharmacol. 37,
266–268.

Jiménez-Castellanos, M.R., Zia, H., Rhodes, C.T., 1993. Assessment of an in vitro

method for measuring the bioadhesiveness of tablets. Int. J. Pharm. 89, 223–228.

Jones, D.S., Lawlor, M.S., Woolfson, A.D., 2004. Formulation and characterisation of
tetracycline-containing bioadhesive polymer networks designed for the treatment of
periodontal disease. Curr. Drug Deliv. 1, 17–25.

Kalam, M.A., Humayun, M., Parvez, N., Yadav, S., 2007. Release kinetics of
modified pharmaceutical dosage form: A review. Cont. J. Pharm. Sci. 1, 30–35.

Kasture, P.V., Parakh, S.R., Gokhale, S.B., Paradkaar, A.R., 2007. Suppositories and
Pessaries, in: Kasture, P.V. (Ed.), Pharmaceutics II Second Year Diploma in
Pharmacy. Nirali Prakashan, Pune, pp. 14.1 – 14.11.

Kauss, T., Gaubert, A., Boyer, C., Boubakar, B.B., Manse, M., Massip, S., Léger,
J.M., Fawaz, F., Lembege, M., Boiron, J.M., Lafarge, X., Lindegardh, N., White, N.J.,
Olliaro, P., Millet, P., Gaudin, K., 2013. Pharmaceutical development and
optimization of azithromycin suppository for paediatric use. Int. J. Pharm. 444, 218–
226.

Keny, R.V., Lourenco, C.F., 2010. Formulation and evaluation of thermoreversible in

situ gelling and mucoadhesive diltiazem hydrochloride liquid suppository. Int. J.
Pharma Bio Sci. 1, 1–17.

Khan, N., Craig, D.Q.M., 2004. Role of blooming in determining the storage stability
of lipid-based dosage forms. J. Pharm. Sci. 93, 2962–2971.

211
Kincl, M., Melej, M., Veber, M., Vrecer, F., 2004. Study of physicochemical
parameters affecting the release of diclofenac sodium from lipophilic matrix tablets.
Acta Chim. Slov. 51, 409–425.

Kokate, A., Marasanapalle, V.P., Jasti, B.R., Li, X., 2006. Physiological and
biochemical barriers to drug delivery, in: Li, X., Jasti, B.R. (Eds.), Design of
Controlled Release Drug Delivery System. The McGraw-Hill Companies, Inc, New
York, pp. 41–73.

Korsmeyer, R.W., Gurny, R., Doelker, E., Buri, P., Peppas, N.A., 1983. Mechanisms
of solute release from porous hydrophilic polymers. Int. J. Pharm. 15, 25–35.

Kosior, A., 2001. Investigation of physical and hypoglycaemic properties of rectal

suppositories with chosen Insulin. Acta Pol. Pharm. 59, 353–358.

Kosmidis, K., Argyrakis, P., Macheras, P., 2003. Fractal kinetics in drug release from
finite fractal matrices. J. Chem. Phys. 119, 6373–6377.

Kurosawa, N., Owada, E., Ito, K., 1998. Avoidance of hepatic first-pass effect in the
rabbit via rectal route of administration. Biopharm. Drug Dispos. 19, 589–594.
doi:10.1002/(SICI)1099-081X(199812)19:9<589::AID-BDD141>3.0.CO;2-2

Kurosawa, N., Owada, E., Ito, K., Ueda, K., Takahashi, A., Kikuiri, T., 1985.
Bioavailability of nifedipine suppository in healthy subjects. Int. J. Pharm. 27, 81–88.

Laakso, R., Kristoffersson, E., Marvola, M., 1984. Bi-exponential first-order release
kinetics of indomethacin from tablets containing polysorbate 80. Int. J. Pharm. 19,
35–42.

Ladani, A., Patel, M., Patel, V., 2011. Fabrication of multi-layer matrix tablets of
venlafaxine hydrochloride using hypromellose and xanthan gum. Int. J. Drug Dev.
Res. 3, 291–299.

Lee, J.W., Park, J.H., Robinson, J.R., 2000. Bioadhesive-based dosage forms: The
next generation. J. Pharm. Sci. 89, 850–866.

Lee, S.L., Raw, A.S., Yu, L., 2008. Dissolution testing, in: Krishna, R., Yu, L. (Eds.),
Biopharmaceutics Applications in Drug Development. Springer Science + Bussiness
Media, New York, pp. 47–74.

Lehr, C.M., Bouwstra, J.A., 1992. In vitro evaluation of mucoadhesive properties of

chitosan and some other natural polymers. Int. J. Pharm. 78, 43–48.

Lehr, C.M., Bouwstra, J.A., Kok, W., De Boer, A.G., Tukker, J.J., Verhoef, J.C.,
Breimer, D.D., Junginger, H.E., 1992. Effects of the mucoadhesive polymer
polycarbophil on the intestinal absorption of a peptide drug in the rat. J. Pharm.
Pharmacol. 44, 402–407.

212
Leung, H.K., Anderson, G.R., Norr, P.J., 1985. Rapid determination of total and solid
fat contents in chocolate products by pulsed nuclear magnetic resonance. J. Food Sci.
50, 942–945.

Leung, S.H., Robinson, J.R., 1988. The contribution of anionic polymer structural
features to mucoadhesion. J. Control. Release. 5, 223–231.

Liversidge, G.G., Grant, D.J.W., Padfield, J.M., 1982. Drug-excipient interactions and
polymorphism in triglyceride suppository. Anal. Proc. 19, 549 – 553.

Loisel, C., Keller, G., Lecq, G., Bourgaux, C., Ollivon, M., 1998. Phase transitions
and polymorphism of cocoa butter. J. Am. Oil Chem. Soc. 75, 425–439.

Lonchampt, P., Hartel, R.W., 2004. Fat bloom in chocolate and compound coatings.
Eur. J. Lipid Sci. Technol. 106, 241–274.

Lootvoet, G., Beyssac, E., Shiu, G.K., Aiache, J.-M., Ritschel, W.A., 1992. Study on
the release of indomethacin from suppositories: In vitro-in vivo correlation. Int. J.
Pharm. 85, 113–120.

Lovegren, N.V., Gray, M.S., Feuge, R.O., 1976. Polymorphic changes in

confectionery fats. J. Am. Oil Chem. 53, 83–88.

Lubrizol Advanced Materials, I., 2009. Product information leaflet of Carbopol 974P
NF polymer.

Macmillan, S.D., Roberts, K.J., 2002. In situ small angle X-ray scattering ( SAXS )
studies of polymorphism with the associated crystallization of cocoa butter fat using
shearing conditions. Cryst. Growth Des. 2, 221–226.

Makhlof, A., Fujimoto, S., Tozuka, Y., Takeuchi, H., 2011. In vitro and in vivo
evaluation of WGA-carbopol modified liposomes as carriers for oral peptide delivery.
Eur. J. Pharm. Biopharm. 77, 216–224.

Marangoni, A.G., McGauley, S.E., 2003. Relationship between crystallization

behavior and structure in cocoa butter. Cryst. Growth Des. 3, 95–108.

Marques, M.R.C., Loebenberg, R., Almukainzi, M., 2011. Simulated biological fluids
with possible application in dissolution testing. Dissolution Technol. 18, 15–28.

Marquez, A.L., Perez, M.P., Wagner, J.R., 2013. Solid fat content estimation by
differential scanning calorimetry: prior treatment and proposed correction. J. Am. Oil
Chem. Soc. 4, 467–473.

McClements, D.J., 1999. Emulsion ingredients, in: Food Emulsions: Principles,

Practice, and Techniques. CRC press, Florida, pp. 95–173.

McNeil, N.I., Ling, K.L., Wager, J., 1987. Mucosal surface pH of the large intestine
of the rat and of normal and inflamed large intestine in man. Gut. 28, 707–713.

213
Menard, K.F., Sichina, W.J., 2000. Prediction of solid fat index ( SFI ) values of food
fats using DSC. Perkin Elmer Therm. Anal. Appl. note.

Menasse, R., Hedwall, P.R., Kraetz, J., Pericin, C., Reisterer, L., Sallmann, A., Zeil,
R., Jaques, R., 1978. Pharmacological properties of diclofenac sodium and its
metabolites. Scand J Rheumatol. 7, 5–16.

Metin, S., Hartel, R.W., 2005. Crystallization of fats and oils, in: Shahidi, F. (Ed.),
Bailey’s Industrial Oil and Fat Products. John Wiley & Sons, Inc, pp. 45–76.

Ministry of Health Malaysia, 2000. Malaysian guidelines for the conduct of

bioavailability and bioequivalance studies.

Moes, A., Jaminet, F., 1976. Influence of ageing of suppositories on rectal absorption
of paracetamol. Pharm. Acta Helv. 51, 119–125.

Moghimipour, E., Dabbagh, M., Zarif, F., 2009. Characterization and in vitro
evaluation of piroxicam suppositories. Asian J. Pharm. Clin. Res. 2, 92–98.

Mollel, H., 2006. Development and assessment of azithromycin paediatric

suppository formulations. Rhodes University, Grahamstown, South Africa.

Moore, J.W., Flanner, H.H., 1996. Mathematical comparison of dissolution profiles.

Pharm. Technol. 20, 64–74.

Morgan, D.J., Paull, J.D., Richmond, B.H., Wilson-Evered, E., Ziccone, S.P., 1986.
Pharmacokinetics of intravenous and oral salbutamol and its sulphate conjugate. Br. J.
Clin. Pharmacol. 22, 587–593.

Mortazavi, S.A., Smart, J.D., 1995. An investigation of some factors influencing the
in vitro assessment of mucoadhesion. Int. J. Pharm. 116, 223–230.

Mourya, V.K., Inamdar, N.N., Tiwari, A., 2010. Carboxymethyl chitosan and its
applications. Adv. Mater. Lett. 1, 11–33.

Nair, L., Bhargava, H.N., 1999. Comparison of in vitro dissolution and permeation of
fluconazole from different suppository bases. Drug Dev. Ind. Pharm. 25, 691–694.

Nakajima, T., Takashima, Y., Furuya, A., Ozawa, Y., Kawashima, Y., 1990.
Indomethacin sustained-release suppositories containing sugar ester. Chem. Pharm.
Bull. 38, 1027 – 1031.

Nassu, R.T., Gonçalves, L.A.G., 1999. Determination of melting point of vegetable

oils and fats by differential scanning calorimetry (DSC) technique. Grasas y Aceites.
50, 16–21.

Nayak, A.K., 2010. Thermodynamic study of the diclofenac sodium solubility.

Chemistry. 19, 121–128.

214
Nishihata, T., Wada, H., Kamada, A., 1985. Sustained-release of sodium diclofenac
from suppository. Int. J. Pharm. 27, 245–253.

Noman, M.O., Kadi, H.O., 2011. Formulation and evaluation of meloxicam

suppositories. Cont. J. Pharm. Sci. 5, 20–24.

Noordin, M.I., Chung, L.Y., 2009. Thermostability and polymorphism of the

theobroma oil and palm kernel oil as suppository bases. J. Therm. Anal. Calorim. 95,
891–894.

Noordin, M.I., Chung, L.Y., 2007. Palm kernel oil blends as suppository bases in the
delivery of aspirin. J. Univ. Malaya Med. Cent. 10, 43–50.

Noordin, M.I., Chung, L.Y., 2004. A rapid micro quantification method of

paracetamol in suppositories using differential scanning calorimetry. Drug Dev. Ind.
Pharm. 30, 925–930.

O’Hara, T., Dunne, A., Butler, J., Devane, J., 1998. A review of methods used to
compare dissolution profile data. Pharm. Sci. Technol. Today. 1, 214–223.

Oberle, R.L., Das, H., Wong, S.L., Chan, K.H., Sawchuk, R.J., 1994.
Pharmacokinetics and metabolism of Diclofenac sodium in Yucatan miniature pigs.
Pharm. Res. 11, 698–703.

Ohnishi, N., Yokoyama, T., Umeda, T., Kiyohara, Y., Kuroda, T., Kita, Y., Kuroda,
K., 1986. Preparation of sustained-release suppositories of indomethacin using a solid
dispersion system and evaluation of bioavailability in rabbits. Chem. Pharm. Bull. 34,
2999–3004.

Oribe, T., Yamada, M., Takeuchi, K., Tsunemi, S., Imasaka, K., Shirakura, O.,
Ishimaru, S., Takayama, K., Nagai, T., 1995. Formulation and in vivo-in vitro
correlation of the dissolution property of lemildipine solid dispersions-incorporated
suppositories. Int. J. Pharm. 124, 27–35.

Ozkan, Y., Atay, T., Dikmen, N., Isimer, A., Aboul-enein, H., 2000. Improvement of
water solubility and in vitro dissolution rate of gliclazide by complexation with B-
cyclodextrin. Pharm. Acta Helv. 74, 365–370.

Palmieri, A., 1981. Suppository dissolution testing: apparatus design and release of
aspirin. Drug Dev. Ind. Pharm. 7, 247–259.

Palomo, M.E., Ballesteros, M.P., Frutos, P., 1999. Analysis of diclofenac sodium and
derivatives. J. Pharm. Biomed. Anal. 21, 83–94.

Pantzaris, T.P., Ahmad, M.J., 2002. Properties and utilization of palm kernel oil.
Palmas. 23, 46–58.

Papadopoulou, V., Kosmidis, K., Vlachou, M., Macheras, P., 2006. On the use of the
Weibull function for the discernment of drug release mechanisms. Int. J. Pharm. 309,
44–50.

215
Park, H., Robinson, J.R., 1985. Physicochemical properties of water insoluble
polymers important to mucin/epithelial adhesion. J. Control. Release. 2, 47–57.

Park, K., Robinson, J.R., 1984. Bioadhesive polymers as platforms for oral-controlled
drug delivery : Method to study bioadhesion. Int. J. Pharm. 19, 107–207.

Peyronel, F., Marangoni, A.G., 2014. In search of confectionary fat blends stable to
heat: Hydrogenated palm kernel oil stearin with sorbitan monostearate. Food Res. Int.
55, 93–102.

Ponchel, G., Touchard, F., Duchene, D., Peppas, N.A., 1987. Bioadhesive analysis of
controlled-release systems. I. Fracture and interpenetration analysis in poly(acrylic
acid)- containing systems. J. Control. Release. 5, 129–141.

Pugunes, S., Ugandar, R.E., 2013. Formulation and evaluation of natural palm oil
based diclofenac sodium suppositories. Int. J. Pharm. Sci. Res. 4, 617–621.

Radermacher, J., Jentsch, D., Scholl, M.A., Lustinetz, T., Frölich, J.C., 1991.
Diclofenac concentrations in synovial fluid and plasma after cutaneous application in
inflammatory and degenerative joint disease. Br. J. Clin. Pharmacol. 31, 537–541.

Ramadan, A.A., 2012. Preparation, characterization and in-vivo evaluation of double-

phased mucoadhesive suppositories containing diclofenac in rats. J. Appl. Sci. Res. 8,
746–752.

Rao, S., Read, N., Stobart, J., Haynes, W., Benjamin, S., Holdsworth, C., 1988.
Anorectal contractility under basal conditions and during rectal infusion of saline in
ulcerative colitis. Gut. 29, 769–777.

Reanmongkol, W., Kaewnopparat, N., Ratanajamit, C., 2011. Physicochemical

properties, in vitro release and in vivo evaluation of tramadol hydrochloride rectal
suppository and rectal gel. Asian Biomed. 5, 269–275.

Reiss, W., Stierlin, H., Degen, P., Faigle, J.W., Gerardin, A., Moppert, J., Sallmann,
A., Schmid, K., Schweizer, A., Sulc, M., Theobald, W., Wagner, J., 1978.
Pharmacokinetics and metabolism of the anti-inflammatory agent voltaren. Scand. J.
Rheumatol. 7, 17–29.

Robertson, J.S., 1961. Substitute for cocoa butter as a suppository base in tropical
countries. J. Pharm. Sci. 50, 21–23.

Rossell, J.B., 1975. Differential scanning calorimetry of palm kernel oil products. J.
Am. Oil Chem. Soc. 52, 505–511.

Rowe, R.C., Sheskey, P.J., Quinn, M.E., 2009. Handbook of Pharmaceutical

Excipients, Sixth. ed. Pharmaceutical Press, London, UK.

Roy, S.K., Prabhakar, B., 2010. Bioadhesive polymeric platforms for transmucosal
drug delivery systems – a review. Trop. J. Pharm. Res. 9, 91–104.

216
Sah, M.L., Saini, T.R., 2008. Formulation development and release studies of
indomethacin suppositories. Indian J. Pharm. Sci. 70, 498–501.

Saleem, M.A., Taher, M., Samaullah, S., Najmuddin, M., Javed, A., Humaira, S.,
Roshan, S., 2008. Formulation and evaluation of tramadol hydrochloride rectal
suppositories. Indian J. Pharm. Sci. 70, 640–644.

Sam, A.P., Van Den Heuij, J.T.M., Tukker, J.J., 1992. Mucoadhesion of both film-
forming and non-film-forming polymeric materials as evaluated with the Wilhelmy
plate method. Int. J. Pharm. 79, 97–105.

Santos, C.A., Jacob, J.S., Hertzog, B.A., Freedman, B.D., Press, D.L.,
Harnpicharnchai, P., Mathiowitz, E., 1999. Correlation of two bioadhesion assays: the
everted sac technique and the CAHN microbalance. J. Control. Release. 61, 113–122.

Sarfaraz, M.D., Jayendra Reddy, A., Hiremath, D., Udupi, R.H., 2012. Trimetazidine
hydrochloride transdermal patch: Formulation and in-vitro evaluation. Int. Res. J.
Pharm. 3, 178–182.

Sato, K., 2001. Crystallization behaviour of fats and lipids - a review. Chem. Eng. Sci.
56, 2255–2265.

Sato, K., Kuroda, T., 1987. Kinetics of melt crystallization and transformation of
tripalmitin polymorphs. J. Am. Oil Chem. 64, 124–127.

Schimdt, J., 2010. Thermal Analysis, in: Neilson, S.S. (Ed.), Food Analysis. Springer
Science, New York, pp. 555–573.

Schneeweis, A., Müller-Goymann, C.C., 2000. Controlled release of solid-reversed-

micellar-solution (SRMS) suppositories containing metoclopramide-HCl. Int. J.
Pharm. 196, 193–196.

Shah, V.P., Tsong, Y., Sathe, P., Liu, J.P., 1998. In vitro dissolution profile
comparison—statistics and analysis of the similarity factor, f2. Pharm. Res. 15, 889–
896.

Shegokar, R., Singh, K., 2010. In-vitro release of paracetamol from suppocire
suppositories: Role of additives. Malaysian J. Pharm. Sci. 8, 57–71.

Shin, S.C., Kim, H.J., Oh, I.J., Cho, C.W., Yang, K.H., 2005. Development of
tretinoin gels for enhanced transdermal delivery. Eur. J. Pharm. Biopharm. 60, 67–71.

Shinde, U., Ahmed, M.H., Singh, K., 2013. Development of dorzolamide loaded 6-O-
carboxymethyl chitosan nanoparticles for open angle glaucoma. J. Drug Deliv. 2013,
562727.

Shojaei, A.H., Paulson, J., Honary, S., 2000. Evaluation of poly(acrylic acid-co-
ethylhexyl acrylate) films for mucoadhesive transbuccal drug delivery: factors
affecting the force of mucoadhesion. J. Control. Release. 67, 223–232.

217
Siepmann, J., Peppas, N.A., 2011. Higuchi equation : Derivation , applications , use
and misuse. Int. J. Pharm. 418, 6–12. doi:10.1016/j.ijpharm.2011.03.051

Siew, W.L., 2001. Crystallisation and melting behaviour of palm kernel oil and
related products by differential scanning calorimetry. Eur. J. Lipid Sci. Technol. 103,
729–734.

Siew, W.L., Ng, W.L., 2000. Differential scanning thermograms of palm oil
triglycerides in the presence of diacylglycerides. J. Oil Palm Res. 12, 1–7.

Singh, A.P., McClements, D.J., Marangoni, A.G., 2004. Solid fat content
determination by ultrasonic velocimetry. Food Res. Int. 37, 545–555.

Singh, J., Gupta, S., Kaur, H., 2011. Prediction of in vitro drug release mechanisms
from extended matrix tablets using SSR/R2 technique. Trends Appl. Sci. Res. 6, 400–
409.

Singla, A.K., Chawla, M., Singh, A., 2000. Potential applications of carbomer in oral
mucoadhesive controlled drug delivery system : A review. Drug Dev. Ind. Pharm. 26,
913–924.

Sirisa-ard, P., Piyamongkol, S., Charumanee, S., Yotsawimonwat, S., Pholsongkram,

K., 2014. Development of Krabok (Irvingia Malayana) wax as a suppository base, in:
ISHS Acta Horticulturae. pp. 71–76.

Smart, J.D., 1991. An in vitro assessment of some mucosa-adhesive dosage forms.

Int. J. Pharm. 73, 69–74.

Smart, J.D., Kellaway, I.W., Worthington, H.E.C., 1984. An In-vitro Investigation of

mucosa-adhesive materials for use in controlled drug delivery. J. Pharm. Pharmacol.
36, 295–299.

Smith, K.W., 2009. Ingredient preparation: The science of tempering, in: Talbot, G.
(Ed.), Science and Technology of Enrobed and Filled Chocolate, Confectionery and
Bakery Products. CRS Press, Padstow, pp. 313–343.

Smith, K.W., Cain, F.W., Talbot, G., 2004. Nature and composition of fat bloom from
palm kernel stearin and hydrogenated palm kernel stearin compound chocolates. J.
Agric. Food Chem. 52, 5539–5544.

Soremekun, R.O., Silva, B.O., Tayo, F., Igwilo, C.I., 2012. Formulation of quinine
suppository for initiation of early treatment of malaria– a preliminary study.
MalariaWorld J. 3, 1–7.

Spangenberg, J.E., Dionisi, F., 2001. Characterization of cocoa butter and cocoa
butter equivalents by bulk and molecular carbon isotope analyses: implications for
vegetable fat quantification in chocolate. J. Agric. Food Chem. 49, 4271–4277.

Stepan Specialty Product, 2014. Wecobee FS.

218
Sugito, K., Ogata, H., Noguchi, M., Kogure, T., Takano, M., Maruyama, Y., Sasaki,
Y., 1988. The spreading of radiolabelled fatty suppository bases in the human rectum.
Int. J. Pharm. 47, 157–162. doi:10.1016/0378-5173(88)90226-8

Suknuntha, K., Tantishaiyakul, V., Worakul, N., Taweepreda, W., 2011.

Characterization of muco- and bioadhesive properties of chitosan, PVP, and
chitosan/PVP blends and release of amoxicillin from alginate beads coated with
chitosan/PVP. Drug Dev. Ind. Pharm. 37, 408–418.

Susumu, M., Konishi, H., 2011. Crystallization behaviour of 1,3-dipalmitoyl-2-

oleoyl-glycerol and 1-palmitoyl-2,3-dioleoyl-glycerol. Eur. J. Lipid Sci. Technol. 103,
1–7.

Swamy, P.V., Ali, M.Y., Anand Kumar, Y., Prasad, K., Srinivaslu, N., 2012. Design
and evaluation of rectal drug delivery systems of non-steroidal anti-inflammatory
drug. Int. Curr. Pharm. J. 1, 165–170.

Szostak, R., Mazurek, S., 2013. Quantification of active ingredients in suppositories

by FT-Raman spectroscopy. Drug Test. Anal. 5, 126–129.

Takatori, T., Shimono, N., Higaki, K., Kimura, T., 2004. Evaluation of sustained
release suppositories prepared with fatty base including solid fats with high melting
points. Int. J. Pharm. 278, 275–282.

Takeuchi, H., Thongborisute, J., Matsui, Y., Sugihara, H., Yamamoto, H.,
Kawashima, Y., 2005. Novel mucoadhesion tests for polymers and polymer-coated
particles to design optimal mucoadhesive drug delivery systems. Adv. Drug Deliv.
Rev. 57, 1583–1594.

Tarimci, N., Ermis, D., 1997. Sustained release characteristics and pharmacokinetic
parameters of ketoprofen suppositories using chitosan. Int. J. Pharm. 147, 71–77.

Thakkar, V.T., Shah, P.A., Soni, T.G., Parmar, M.Y., Gohel, M.C., Gandhi, T.R.,
2009. Goodness-of-fit model-dependent approach for release kinetics of levofloxacin
hemihydrates floating tablet. Dissolution Technol. 16, 35–39.

Thirawong, N., Nunthanid, J., Puttipipatkhachorn, S., Sriamornsak, P., 2007.

Mucoadhesive properties of various pectins on gastrointestinal mucosa: An in vitro
evaluation using texture analyzer. Eur. J. Pharm. Biopharm. 67, 132–140.

Thongborisute, J., Takeuchi, H., 2008. Evaluation of mucoadhesiveness of polymers

by BIACORE method and mucin-particle method. Int. J. Pharm. 354, 204–209.

Timms, R., 1984. Phase behavior of fats and their mixtures. Prog. Lipid Res. 23, 1–
38.

Tobyn, M.J., Johnson, J.R., Dettmar, P.W., 1996. Factors affecting in vitro gastric
mucoadhesion. II: Physical properties of polymers. Eur. J. Pharm. Biopharm. 42, 56–
61.

219
Tobyn, M.J., Johnson, J.R., Dettmar, P.W., 1995. Factors affecting in vitro gastric
mucoadhesion I. Test conditions and instrument variables. Eur. J. Pharm. Biopharm.
41, 235–241.

Torbica, A., Jovanovic, O., Pajin, B., 2005. The advantages of solid fat content
determination in cocoa butter and cocoa butter equivalents by the Karlshamns
method. Eur. Food Res. Technol. 222, 385–391.

Toro-Vazquez, J.F., Pérez-Martínez, D., Dibildox-Alvarado, E., Charó-Alonso, M.,

Reyes-Hernández, J., 2004. Rheometry and polymorphism of cocoa butter during
crystallization under static and stirring conditions. J. Am. Oil Chem. Soc. 81, 195–
202.

Tukker, J., Blaey, C.J., Charbon, G.A., 1984. Rectal motility and bioavailability.
Pharm. Res. 1, 173 – 178.

Tukker, J.J., Crombeen, J.P., Breimer, D.D., De Blaey, C.J., 1981. The bioavailability
of seven commercially available rectal aminophylline suppositories. Pharm. Weekbl.
Sci. Ed. 3, 1325–1335.

Tungtong, S., Okonogi, S., Chowwanapoonpohn, S., Phutdhawong, W.,

Yotsawimonwat, S., 2012. Solubility, viscosity and rheological properties of water-
soluble chitosan derivatives. Maejo Int. J. Sci. Technol. 6, 315–322.

Turner, C., Aye Mya Thein, N., Turner, P., Francois, N., White, N.J., 2012. Rectal pH
in well and unwell infants. J. Trop. Pediatr. 58, 311–313.

Umeda, T., Yokoyama, T., Ohnishi, N., Kuroda, T., Kita, Y., Kuroda, K., Asada, S.,
1985. Studies on sustained-releae dosage forms. III. Preparation of nifedipine
suppositories and bioavailability in rabbits. Chem. Pharm. Bull. 33, 3953–3959.

United States Pharmacopeia and National Formulary, 31 st. ed, 2008. . Port City
Press, Baltimore.

University of North Carolina Eshelman School of Pharmacy, 2015. The

pharmaceutics and compounding laboratory: Preparation of suppositories [WWW
Document]. URL http://pharmlabs.unc.edu/labs/suppository/intro.htm (accessed
5.11.15).

US Food and Drug Administration, 1997. Guidance for industry dissolution testing of
immediate release solid oral dosage forms. Dissolution Technol.

Uzunkaya, G., Bergişadi, N., 2003. In vitro drug liberation and kinetics of sustained
release indomethacin suppository. Farm. 58, 509–512.

Vaghani, S.S., Patel, M.M., Satish, C.S., 2012. Synthesis and characterization of pH-
sensitive hydrogel composed of carboxymethyl chitosan for colon targeted delivery of
ornidazole. Carbohydr. Res. 347, 76–82.

220
Vaghela, B., Kayastha, R., Bhatt, N., Pathak, N., Rathod, D., 2011. Development and
validation of dissolution procedures. J. Appl. Pharm. Sci. 1, 50–56.

Van Langevelde, A., Van Malssen, K., Peschar, R., Schenk, H., 2001. Effect of
temperature on recrystallization behavior of cocoa butter. J. Am. Oil Chem. Soc. 78,
919–925.

Van Malssen, K., Langevelde, A., Peschar, R., Schenk, H., 1999. Phase behavior and
extended phase scheme of static cocoa butter investigated with real-time X-ray
powder diffraction. J. Am. Oil Chem. Soc. 76, 669–676.

Varshney Himanshu, M., Tanwar, Y.S., 2009. Designing , release characteristics and
in vitro evaluation of flurbiprofen sodium suppositories. Int. J. Pharm. Clin. Res. 1,
31–34.

Victoria, M.M., David, C.J., 2003. Thermal and rheological study of lipophilic
ethosuximide suppositories. Eur. J. Pharm. Sci. 19, 123–128.

Vidras, N.J., Reid, V.E., Bohidar, N.R., Plakogiannis, F.M., 1982. Medicament
release from suppository bases I: Physicochemical characteristics and bioavailability
of indomethacin in rabbits. J. Pharm. Sci. 71, 945–949.

Vishnu, Y.V., Chandrasekhar, K., Ramesh, G., Rao Y.M., 2007. Development of
mucoadhesive patches for buccal administration of carvedilol. Curr. Drug Deliv. 4,
27–39.

Von Vaeck, S.V., 1960. Uber Kakaobutter und den Fettreif von Kakao-
Erzeugnissen*. Fette Seifen Anstrichm. 8, 709 – 722.

Vyvyan, H.A.L., Hanafiah, Z., 1995. Patients’ attitudes to rectal drug administration.
Anaesthesia. 50, 983–984.

Walker, R.A., Bosin, W.A., 1971. Comparison of SFI, DSC and NMR methods for
determining solid-liquid ratios in fats. J. Am. Oil Chem. Soc. 48, 50–53.

Wang, L., Tang, X., 2008. A novel ketoconazole bioadhesive effervescent tablet for
vaginal delivery: design, in vitro and “in vivo” evaluation. Int. J. Pharm. 350, 181–
187.

Warren, D.E., 1996. Practical use of rectal medications in palliative care. J. Pain
Symptom Manage. 11, 378–386.

Watanabe, Y., 2007. Permeation pathways in rectal absorption, in: Touitou, E., Barry,
B.W. (Eds.), Enhancement in Drug Delivery. CRC Press, Massachusetts, pp. 135–
146.

Webster, J.A., Dowse, R., Walker, R.B., 1998. In-vitro release of amoxycillin from
lipophilic suppositories. Drug Dev. Ind. Pharm. 24, 395–399.

221
Werle, M., Makhlof, A., Takeuchi, H., 2010. Carbopol-lectin conjugate coated
liposomes for oral peptide delivery. Chem. Pharm. Bull. 58, 432–434.

Whitworth, C.W., Luzzi, L.A., Thompson, B.B., Jun, H.W., 1973. Stability of aspirin
in liquid and semisolid bases II: Effect of fatty additives on stability in a polyethylene
glycol base. J. Pharm. Sci. 62, 1372–1374.

Wille, R.L., Lutton, E.S., 1966. Polymorphism of cocoa butter. J. Am. Oil Chem. Soc.
43, 491–496.

Willis, J.V., Kendall, M.J., Flinn, R.M., Thornbill, D.P., Welling, P.G., 1979. The
pharmacokinetics of diclofenc sodium following intravenous and oral administration.
Eur. J. Clin. Pharmacol. 16, 405–410.

Winek, C.L., Wahba, W.W., Balzer, T.W., 2001. Drug and chemical blood-level data
2001. Forensic Sci. Int. 122, 107–123.

Wong, C.F., Yuen, K.H., Peh, K.K., 1999a. An in-vitro method for buccal adhesion
studies: Importance of instrument variables. Int. J. Pharm. 180, 47–57.

Wong, C.F., Yuen, K.H., Peh, K.K., 1999b. Formulation and evaluation of controlled
release Eudragit buccal patches. Int. J. Pharm. 178, 11–22.

Yahagi, R., Machida, Y., Onishi, H., 2000. Mucoadhesive suppositories of ramosetron
hydrochloride utilizing Carbopol. Int. J. Pharm. 193, 205–212.

Yahagi, R., Onishi, H., Machida, Y., 1999. Preparation and evaluation of double-
phased mucoadhesive suppositories of lidocaine utilizing Carbopol and white
beeswax. J. Control. Release. 61, 1–8.

Yarnykh, T.G., Tolochko, E.V., Chusenko, V.N., 2011. Drug synthesis methods and
manufacturing technology: studying an assortment of suppository bases (review).
Pharm. Chem. J. 44, 551–556.

Yoshida, T., Itoh, Y., Gomita, Y., Oishi, R., 1991. Influence of storage temperature on
indomethacin release from fatty-base suppositories in vitro and in vivo. Acta Medica
Okayama. 45, 37–42.

Yoshino, H., Kobayashi, M., Samejima, M., 1982. The influence of drugs on the
physical stability of fatty suppositories. Chem. Pharm. Bull. 30, 992–1001.

Yoshino, H., Kobayashi, M., Samejima, M., 1981. Polymorphic transition rate of
semisynthetic fatty suppository bases. Chem. Pharm. Bull. 29, 2661–2669.

Young, C., Palin, K.J., Reid, A.S., Thomas, N.W., Gould, P.L., 1987. Formulation of
fenbufen suppositories. II. Selection of a suppository base using dissolution studies
and histological studies in rats. Int. J. Pharm. 40, 187–191. doi:10.1016/0378-
5173(87)90167-0

222
Yuksel, N., Kanik, A.E., Baykara, T., 2000. Comparison of in vitro dissolution
profiles by ANOVA-based, model-dependent and -independent methods. Int. J.
Pharm. 209, 57–67.

Zawar, L.R., Bhandari, G.S., 2012. Formulation and evaluation of sustained release
ondansetron poloxamer based solid suppositories. J. Appl. Pharm. Sci. 2, 186–190.

Zhou, S.L., Zhang, F.Q., Jin, Q.Z., Liu, Y.F., Liang, S., Zhang, T., Zou, X.Q., Wang,
X.G., 2010. Characterization of palm kernel oil, palm stearin, and palm olein in
isosolid diagrams. Eur. J. Lipid Sci. Technol. 112, 1041–1047.

223
 APPENDIX

Appendix 1: Certificate of analysis of cocoa butter (CB).

224
Appendix 2 : Certificate of analysis for Chocexa (CE).

225
226
Appendix 3 : Certificate of analysis for Supersocolate SpecialTM (SS).

227
Appendix 4 : Certificate of analysis for diclofenac sodium (DcNa).

228
Appendix 5 : Certificate of analysis for Carbopol 974P NF (CBP).

229
Appendix 6 : Certificate of analysis of hydroxypropyl methylcellulose 2910 (HPMC).

230
Appendix 7 : Certificate of analysis for carboxymethyl chitosan (CMCTS).

231
Appendix 8 : The viscosity (cp) of CB suppositories measured at 50 rpm shear rate. Outcomes were a result of ANOVA and post hoc Tukey’s
HSD analysis. Mean ± SD, n=3.

FORMULATION

CBP HPMC PVP CMCTS

Blank (A) 54.6 ± 4.14 54.6 ± 4.14 54.6 ± 4.14 54.6 ± 4.14

DcNa only (B) 82.8 ± 1.29 82.8 ± 1.29 82.8 ± 1.29 82.8 ± 1.29

1 %w/w polymer (C) 78.43 ± 0.95 90.03 ± 2.95 84.83 ± 2.53 94.23 ± 2.58

2 %w/w polymer (D) 76.03 ± 0.76 104.8 ± 3.33 92.30 ± 0.40 98.73 ± 4.65

5 %w/w polymer (E) 87.87 ± 6.27 122.3 ± 1.06 97.67 ± 2.27 106.6 ± 0.92

ANOVA P < 0.05 P < 0.05 P < 0.05 P < 0.05

A&B D&E A&B B&E A&B B&E A&B B&D

A&C A&C C&D A&C C&E A&C B&E
TUKEY’S HSD SIGNIFICANT
A&D A&D C&E A&D A&D C&E
DIFFERENCE
A&E A&E D&E A&E A&E
C&E B&D B&D B&C
232
Appendix 9 : The viscosity (cp) of CE suppositories measured at 50 rpm shear rate. Outcomes were a result of ANOVA and post hoc Tukey’s
HSD analysis. Mean ± SD, n=3.

FORMULATION

CBP HPMC PVP CMCTS

Blank (A) 53.43 ± 4.74 53.43 ± 4.74 53.43 ± 4.74 53.43 ± 4.74

DcNa only (B) 42.00 ± 0.96 42.00 ± 0.96 42.00 ± 0.96 42.00 ± 0.96

1 %w/w polymer (C) 68.10 ± 2.08 46.20 ± 1.79 43.70 ± 2.02 41.70 ± 1.87

2 %w/w polymer (D) 79.33 ± 1.15 56.10 ± 1.06 44.17 ± 1.27 49.93 ± 2.12

5 %w/w polymer (E) 102.60 ± 6.52 67.77 ± 3.10 56.40 ± 4.06 52.47 ± 5.24

ANOVA P < 0.05 P < 0.05 P < 0.05 P < 0.05

A&B B&D A&B C&E A&B A&B

A&C B&E A&E D&E A&C A&C
TUKEY’S HSD SIGNIFICANT
A&D C&D B&D B&E B&E
DIFFERENCE
A&E C&E B&E C&E C&E
B&C D&E C&D D &E
233
Appendix 10 : The viscosity (cp) of SS suppositories measured at 50 rpm shear rate. Outcomes were a result of ANOVA and post hoc Tukey’s
HSD analysis. Mean ± SD, n=3.

FORMULATION

CBP HPMC PVP CMCTS

Blank (A) 49.03 ± 1.46 49.03 ± 1.46 49.03 ± 1.46 49.03 ± 1.46

DcNa only (B) 37.73 ± 1.06 37.73 ± 1.06 37.73 ± 1.06 37.73 ± 1.06

1 %w/w polymer (C) 44.33 ± 1.89 41.40 ± 0.53 39.00 ± 1.52 40.73 ± 0.45

2 %w/w polymer (D) 55.87 ± 1.70 48.83 ± 0.80 44.93 ± 1.00 41.30 ± 0.91

5 %w/w polymer (E) 81.8 ± 4.78 67.77 ± 3.10 56.50 ± 2.23 46.15 ± 1.91

ANOVA P < 0.05 P < 0.05 P < 0.05 P < 0.05

A&B B&E A&B C&D A&B C&D A&B

A&D C&D A&C C&E A&C C&E A&C
TUKEY’S HSD SIGNIFICANT
A&E C&E A&E D&E A&E D&E A&D
DIFFERENCE
B&C D&E B&D B&D B&E
B&D B&E B&E
234
Appendix 11 : The hardness (N) of CB suppositories. Outcomes were a result of ANOVA and post hoc Tukey’s HSD analysis. Mean ± SD, n=6.

FORMULATION

CBP HPMC PVP CMCTS

Blank (A) 59.00 ± 2.97 59.00 ± 2.97 59.00 ± 2.97 59.00 ± 2.97

DcNa only (B) 73.33 ± 2.07 73.33 ± 2.07 73.33 ± 2.07 73.33 ± 2.07

1 %w/w polymer (C) 100.67 ± 4.27 101.00 ± 4.56 89.00 ± 1.90 121.50 ± 2.43

2 %w/w polymer (D) 91.67 ± 5.28 92.50 ± 1.76 103.00 ± 1.41 116.67 ± 2.16

5 %w/w polymer (E) 100.67 ± 3.01 80.83 ± 4.36 99.83 ± 1.47 131.17 ± 2.32

ANOVA P < 0.05 P < 0.05 P < 0.05 P < 0.05

A&B A&B B&D A&B B&D A&B B&D

B&D
A&C A&C B&E A&C B&E A&C B&E
TUKEY’S HSD SIGNIFICANT B&E
A&D A&D C&D A&D C&D A&D C&E
DIFFERENCE C&D
A&E A&E C&E A&E C&E A&E D&E
D&E
B&C B&C D&E B&C B&C
235
Appendix 12 : The hardness (N) of CE suppositories. Outcomes were a result of ANOVA and post hoc Tukey’s HSD analysis. Mean ± SD, n=6.

FORMULATION

CBP HPMC PVP CMCTS

Blank (A) 82.33 ± 5.20 82.33 ± 5.20 82.33 ± 5.20 82.33 ± 5.20

DcNa only (B) 88.33 ± 2.34 88.33 ± 2.34 88.33 ± 2.34 88.33 ± 2.34

1 %w/w polymer (C) 89.00 ± 3.41 2.64 ± 2.64 99.33 ± 5.09 104.83 ± 7.57

2 %w/w polymer (D) 84.17 ± 5.85 83.83 ± 4.17 94.17± 2.79 122.00 ± 6.99

5 %w/w polymer (E) 103.50 ± 2.35 87.83 ± 4.22 91.33 ± 3.93 106.67 ± 8.09

ANOVA P < 0.05 P < 0.05 P < 0.05 P < 0.05

A&B B&C A&C A&C B&E

B&E A&D A&D C&D
TUKEY’S HSD SIGNIFICANT
C&E B&C A&E D&E
DIFFERENCE
D&E B&C
B&D
236
Appendix 13 : The hardness (N) of SS suppositories. Outcomes were a result of ANOVA and post hoc Tukey’s HSD analysis. Mean ± SD, n=6.

FORMULATION

CBP HPMC PVP CMCTS

Blank (A) 55.83 ± 1.60 55.83 ± 1.60 55.83 ± 1.60 55.83 ± 1.60

DcNa only (B) 70.17 ± 1.72 70.17 ± 1.72 70.17 ± 1.72 70.17 ± 1.72

1 %w/w polymer (C) 103.50 ± 2.88 83.00 ± 4.20 91.17 ± 2.48 100.00 ± 5.18

2 %w/w polymer (D) 98.33 ± 4.50 70.33 ± 5.85 96.00 ± 1.55 102.50 ± 5.32

5 %w/w polymer (E) 74.67 ± 4.97 69.83 ± 2.76 90.50 ± 1.87 101.50 ± 5.65

ANOVA P < 0.05 P < 0.05 P < 0.05 P < 0.05

A&B B&D A&B C&D A&B B&D A&B B&D

A&C C&E A&C C&E A&C B&E A&C B&E
TUKEY’S HSD SIGNIFICANT
A&D D&E A&D A&D A&D
DIFFERENCE
A&E A&E A&E A&E
B&C B&C B&C B&C
237
Appendix 14 : The softening time (min) of CB suppositories. Outcomes were a result of ANOVA and post hoc Tukey’s HSD analysis. Mean ±
SD, n=3.

FORMULATION

CBP HPMC PVP CMCTS

Blank (A) 3.03 ± 0.01 3.03 ± 0.01 3.03 ± 0.01 3.03 ± 0.01

DcNa only (B) 3.79 ± 0.11 3.79 ± 0.11 3.79 ± 0.11 3.79 ± 0.11

1 %w/w polymer (C) 4.06 ± 0.03 3.82 ± 0.126 3.83 ± 0.106 3.68 ± 0.202

2 %w/w polymer (D) 3.68 ± 0.09 4.08 ± 0.051 4.26 ± 0.025 3.77 ± 0.035

5 %w/w polymer (E) 4.16 ± 0.04 4.31 ± 0.054 4.03 ± 0.017 4.20 ± 0.033

ANOVA P < 0.05 P < 0.05 P < 0.05 P < 0.05

A&B B&E A&B B&E A&B C&D A&B C&E

A&C C&D A&C C&D A&C A&C D&E
TUKEY’S HSD SIGNIFICANT
A&D D&E A&D C&E A&D A&D
DIFFERENCE
A&E A&E A&E A&E
B&C B&D B&D B&E
238
Appendix 15 : The softening time (min) of CE suppositories. Outcomes were a result of ANOVA and post hoc Tukey’s HSD analysis. Mean ±
SD, n=3.

FORMULATION

CBP HPMC PVP CMCTS

Blank (A) 6.16 ± 0.17 6.16 ± 0.17 6.16 ± 0.17 6.16 ± 0.17

DcNa only (B) 4.36 ± 0.07 4.36 ± 0.07 4.36 ± 0.07 4.36 ± 0.07

1 %w/w polymer (C) 4.60 ± 0.09 4.79 ± 0.03 4.66 ± 0.06 4.92 ± 0.09

2 %w/w polymer (D) 5.68 ± 0.09 5.07 ± 0.04 5.07 ± 0.05 4.99 ± 0.13

5 %w/w polymer (E) 5.34 ± 0.08 6.84 ± 0.08 5.58 ± 0.08 4.92 ± 0.08

ANOVA P < 0.05 P < 0.05 P < 0.05 P < 0.05

A&B B&E A&B B&D A&B B&D A&B B&D

A&C C&D A&C B&E A&C B&E A&C B&E
TUKEY’S HSD SIGNIFICANT
A&D C&E A&D C&D A&D C&D A&D
DIFFERENCE
A&E D&E A&E C&E A&E C&E A&E
B&D B&C D&E B&C D&E B&C
239
Appendix 16 : The softening time (min) of SS suppositories. Outcomes were a result of ANOVA and post hoc Tukey’s HSD analysis. Mean ±
SD, n=3.

FORMULATION

CBP HPMC PVP CMCTS

Blank (A) 5.66 ± 0.10 5.66 ± 0.10 5.66 ± 0.10 5.66 ± 0.10

DcNa only (B) 4.36 ± 0.13 4.36 ± 0.13 4.36 ± 0.13 4.36 ± 0.13

1 %w/w polymer (C) 4.48 ± 0.13 4.81 ± 0.08 4.56 ± 0.06 4.93 ± 0.03

2 %w/w polymer (D) 4.72 ± 0.05 5.16 ± 0.05 4.76 ± 0.04 4.76 ± 0.09

5 %w/w polymer (E) 5.43 ± 0.04 6.14 ± 0.04 5.13 ± 0.04 5.29 ± 0.04

ANOVA P < 0.05 P < 0.05 P < 0.05 P < 0.05

A&B B&E A&B B&D A&B B&D A&B B&D

A&C C&D A&C B&E A&C B&E A&C B&E
TUKEY’S HSD SIGNIFICANT
A&D C&E A&D C&D A&D C&D A&D C&E
DIFFERENCE
A&E D&E A&E C&E A&E C&E A&E D&E
B&D B &C D&E B &C D&E B &C
240
(a)
30

Amount of DcNa release (mg)

25
25 mg
20
50 mg
15
10 75 mg

5
0
0 60 120 180 240
Time (min)

(b)
30
Amount of DcNa release (mg)

25 25 mg
20 50 mg
15 75 mg
10
5
0
0 60 120 180 240
Time (min)

(c)
30
Amount of DcNa release (mg)

25 25 mg

20 50 mg

15 75 mg
10
5
0
0 60 120 180 240
Time (min)

Appendix 17: Amount of DcNa (mg) released at each time interval in (a) CB; (b) CE;

and (c) SS suppositories containing 25, 50, 75 mg of DcNa. Mean ± 2 SE, n=6.
241
a) 100

Cummulative DcNa release (%)

80

60 0%
1%
40
2%
20 5%
0
0 60 120 180 240
Time (min)

b) 100
Cummulative DcNa release (%)

80

60 0%
1%
40
2%
20
5%
0
0 60 120 180 240
Time (min)

c) 100
Cummulative DcNa release (%)

80

60 0%
1%
40
2%
20 5%

0
0 60 120 180 240
Time (min)

Appendix 18: Cumulative percentage DcNa release in a) CB; b) CE; c) SS

suppositories incorporated with 1-5 %w/w of CBP. Mean ± 2 SE, n=6.

242
a) 100

Cummulative DcNa release (%)

80

60 0%
1%
40
2%
20 5%

0
0 60 120 180 240
Time (min)

b) 100
Cummulative DcNa release (%)

80

60 0%
1%
40
2%
20 5%

0
0 60 120 180 240
Time (min)

c) 100
Cummulative DcNa release (%)

80

60 0%
1%
40
2%
20 5%

0
0 60 120 180 240
Time (min)

Appendix 19 : Cumulative percentage of DcNa release in a) CB; b) CE; c) SS

suppositories incorporated with 1-5 %w/w of HPMC. Mean ± 2 SE, n= 6.

243
a)
100

Cummulative DcNa release (%)

80
0%
60
1%
40
2%
20 5%

0
0 60 120 180 240
Time (min)

b)
100
Cummulative DcNa release (%)

80

60 0%
1%
40
2%
20 5%
0
0 60 120 180 240
Time (min)

c)
100
Cummulative DcNa release (%)

80
0%
60
1%
40 2%
20 5%

0
0 60 120 180 240
Time (min)

Appendix 20 : Cumulative percentage of DcNa release in a) CB; b) CE; c) SS

suppositories incorporated with 1-5%w/w PVP. Mean ± 2 SE, n=6.

244
a)
100

Cummulative DcNa release (%)

80
0%
60
1%
40
2%
20 5%

0
0 60 120 180 240
Time (min)

b)
100
Cummulative DcNa release (%)

80

60 0%
1%
40
2%
20 5%
0
0 60 120 180 240
Time (min)

c)
100
Cummulative DcNa release (%)

80
0%
60
1%
40
2%
20 5%

0
0 60 120 180 240
Time (min)

Appendix 21 : Cumulative percentage of DcNa release in a) CB; b) CE; c) SS

suppositories incorporated with 1-5%w/w CMCTS. Mean ± 2 SE, n=6.

245
Appendix 22 : The dissolution efficiency (DE) of CB suppositories. Outcomes were a result of ANOVA and post hoc Tukey’s HSD analysis.

Mean ± SD, n=6.

FORMULATION

CBP HPMC PVP CMCTS

0 %w/w polymer (A) 93.79 ± 1.52 93.79 ± 1.52 93.79 ± 1.52 93.79 ± 1.52

1 %w/w polymer (B) 59.65 ± 1.74 94.27 ± 1.18 92.56 ± 0.76 89.52 ± 0.95

2 %w/w polymer (C) 51.78 ± 1.34 92.87 ± 0.88 88.83 ± 1.22 90.97 ± 2.19

5 %w/w polymer (D) 36.80 ± 1.14 92.68 ± 0.83 90.95 ± 0.64 91.67 ± 1.10

ANOVA P < 0.05 P > 0.05 (0.078) P < 0.05 P < 0.05

A&B C&D A&C A&B

A&C A&D A&C
TUKEY’S HSD SIGNIFICANT
A&D B&C A&D
DIFFERENCE
B&C
B&D
246
Appendix 23 : The dissolution efficiency (DE) of CE suppositories. Outcomes were a result of ANOVA and post hoc Tukey’s HSD analysis.

Mean ± SD, n=6.

FORMULATION

CBP HPMC PVP CMCTS

0 %w/w polymer (A) 88.73 ± 2.12 88.73 ± 2.12 88.73 ± 2.12 88.73 ± 2.12

1 %w/w polymer (B) 58.47 ± 1.56 83.55 ± 2.32 86.55 ± 0.60 87.62 ± 0.65

2 %w/w polymer (C) 33.20 ± 1.81 87.58 ± 1.03 87.91 ± 0.51 91.74 ± 1.14

5 %w/w polymer (D) 22.59 ± 1.37 85.52 ± 0.64 85.83 ± 0.67 92.04 ± 1.02

ANOVA P < 0.05 P < 0.05 P < 0.05 P < 0.05

A&B C&D A&B A&D A&C

A&C A&D A&D
TUKEY’S HSD SIGNIFICANT
A&D B&C B&C
DIFFERENCE
B&C B&D B&D
B&D
247
Appendix 24 : The dissolution efficiency (DE) of SS suppositories. Outcomes were a result of ANOVA and post hoc Tukey’s HSD analysis.

Mean ± SD, n=6.

FORMULATION

CBP HPMC PVP CMCTS

0 %w/w polymer (A) 91.93 ± 1.04 91.93 ± 1.04 91.93 ± 1.04 91.93 ± 1.04

1 %w/w polymer (B) 59.29 ± 1.71 90.76 ± 0.37 88.13 ± 0.37 92.04 ± 1.02

2 %w/w polymer (C) 29.91 ± 1.91 90.78 ± 1.17 87.62 ± 0.67 91.60 ± 0.95

5 %w/w polymer (D) 16.49 ± 2.56 89.70 ± 0.53 86.27 ± 1.38 91.64 ± 1.55

ANOVA P < 0.05 P < 0.05 P < 0.05 P > 0.05 (0.892)

A&B C&D A&D A&B

A&C A&C
TUKEY’S HSD SIGNIFICANT
A&D A&D
DIFFERENCE
B&C
B&D
248
Appendix 25 : The min dissolution time (MDT) of CB suppositories. Outcomes were a result of ANOVA and post hoc Tukey’s HSD analysis.

Mean ± SD, n=6.

FORMULATION

CBP HPMC PVP CMCTS

0 %w/w polymer (A) 14.73 ± 2.47 14.73 ± 2.47 14.73 ± 2.47 14.73 ± 2.47

1 %w/w polymer (B) 60.72 ± 2.09 19.04 ± 2.28 18.21 ± 1.60 18.89 ± 0.98

2 %w/w polymer (C) 69.24 ± 2.09 24.08 ± 1.78 27.98 ± 3.85 18.47 ± 2.14

5 %w/w polymer (D) 114.03 ± 4.42 26.25 ± 2.80 24.74 ± 2.28 20.03 ± 2.66

ANOVA P < 0.05 P < 0.05 P < 0.05 P < 0.05

A&B B&C A&C A&C A&D

A&C A&D A&D
TUKEY’S HSD SIGNIFICANT
A&D B&C B&C
DIFFERENCE
B&D B&D B&D
C&D
249
Appendix 26 : The min dissolution time (MDT) of CE suppositories. Outcomes were a result of ANOVA and post hoc Tukey’s HSD analysis.

Mean ± SD, n=6.

FORMULATION

CBP HPMC PVP CMCTS

0 %w/w polymer (A) 26.44 ± 3.32 26.44 ± 3.32 26.44 ± 3.32 26.44 ± 3.32

1 %w/w polymer (B) 36.36 ± 1.73 26.07 ± 2.12 23.17 ± 1.28 20.81 ± 1.91

2 %w/w polymer (C) 43.04 ± 3.13 19.85 ± 1.48 19.01 ± 1.08 17.34 ± 1.91

5 %w/w polymer (D) 74.15 ± 6.14 25.04 ± 0.95 21.48 ± 1.93 24.62 ± 3.08

ANOVA P < 0.05 P < 0.05 P < 0.05 P < 0.05

A&B C&D A&C A&C A&B

A&C B&C A&D A&C
TUKEY’S HSD SIGNIFICANT
A&D C&D C&D
DIFFERENCE
B&C
B&D
250
Appendix 27 : The min dissolution time (MDT) of SS suppositories. Outcomes were a result of ANOVA and post hoc Tukey’s HSD analysis.

Mean ± SD, n=6.

FORMULATION

CBP HPMC PVP CMCTS

0 %w/w polymer (A) 15.60 ± 1.07 15.60 ± 1.07 15.60 ± 1.07 15.60 ± 1.07

1 %w/w polymer (B) 30.63 ± 2.89 16.77 ± 0.50 18.49 ± 0.19 14.82 ± 3.00

2 %w/w polymer (C) 42.18 ± 7.35 17.15 ± 0.81 17.03 ± 1.28 17.43 ± 2.44

5 %w/w polymer (D) 83.01 ± 10.39 17.92 ± 0.52 17.53 ± 0.51 18.08 ± 3.26

ANOVA P < 0.05 P < 0.05 P < 0.05 P > 0.05 (0.134)

A&B C&D A&D A&B

A&C A&C
TUKEY’S HSD SIGNIFICANT
A&D A&D
DIFFERENCE
B&C
B&D
251
Appendix 28 : The peak force of detachment (Fmax) of CB suppositories measured using tensile setup against colon mucosa. Outcomes were a

result of ANOVA and post hoc Tukey’s HSD analysis. Mean ± SD, n=5-6.

FORMULATION

CBP HPMC PVP CMCTS

DcNa only (A) 2.17 ± 0.43 2.17 ± 0.43 2.17 ± 0.43 2.17 ± 0.43

1 %w/w polymer (B) 2.48 ± 0.43 2.19 ± 0.47 2.74 ± 0.55 2.51 ± 0.32

2 %w/w polymer (C) 3.04 ± 0.46 2.65 ± 0.40 3.07 ± 0.48 2.93 ± 0.44

5 %w/w polymer (D) 4.17 ± 0.74 2.74 ± 0.70 3.89 ± 0.38 3.52 ± 0.47

ANOVA P < 0.05 P > 0.05 (0.40) P < 0.05 P < 0.05

A&D A&D A&D

TUKEY’S HSD SIGNIFICANT
B&D B&D
DIFFERENCE
C&D
252
Appendix 29 : The peak force of detachment (Fmax) of CE suppositories measured using tensile setup against colon mucosa. Outcomes were a

result of ANOVA and post hoc Tukey’s HSD analysis. Mean ± SD, n=5-6.

FORMULATION

CBP HPMC PVP CMCTS

DcNa only (A) 2.12 ± 0.30 2.12 ± 0.30 2.12 ± 0.30 2.12 ± 0.30

1 %w/w polymer (B) 3.28 ± 0.41 2.18 ± 0.46 2.81 ± 0.54 3.00 ± 0.40

2 %w/w polymer (C) 3.08 ± 0.40 2.39 ± 0.30 3.22 ± 0.91 2.70 ± 0.49

5 %w/w polymer (D) 4.20 ± 0.73 2.68 ± 0.62 4.27 ± 0.83 3.78 ± 0.70

ANOVA P < 0.05 P > 0.05 (0.18) P < 0.05 P < 0.05

TUKEY’S HSD SIGNIFICANT A&D A&D A&D

DIFFERENCE B&D
253
Appendix 30 : The peak force of detachment (Fmax) of SS suppositories measured using tensile setup against colon mucosa. Outcomes were a

result of ANOVA and post hoc Tukey’s HSD analysis. Mean ± SD, n=5-6.

FORMULATION

CBP HPMC PVP CMCTS

DcNa only (A) 1.82 ± 0.13 1.82 ± 0.13 1.82 ± 0.13 1.82 ± 0.13

1 %w/w polymer (B) 2.68 ± 0.51 1.90 ± 0.49 2.91 ± 0.64 2.24 ± 0.25

2 %w/w polymer (C) 3.80 ± 0.17 2.27 ± 0.50 3.53 ± 0.45 2.79 ± 0.33

5 %w/w polymer (D) 4.57 ± 0.29 2.86 ± 0.50 4.94 ± 0.65 3.56 ± 0.22

ANOVA P < 0.05 P > 0.05 (0.07) P < 0.05 P < 0.05

A&C A&C A&D

TUKEY’S HSD SIGNIFICANT A&D A&D
DIFFERENCE B&C B&D
B&D C&D
254
Appendix 31 : The peak force of detachment (Fmax) of CB suppositories measured using shear setup against colon mucosa. Outcomes were a

result of ANOVA and post hoc Tukey’s HSD analysis. Mean ± SD, n=5-6.

FORMULATION

CBP HPMC PVP CMCTS

DcNa only (A) 0.33 ± 0.04 0.33 ± 0.04 0.33 ± 0.04 0.33 ± 0.04

1 %w/w polymer (B) 0.37 ± 0.02 0.43 ± 0.07 0.51 ± 0.03 0.57 ± 0.02

2 %w/w polymer (C) 0.41 ± 0.07 0.55 ± 0.03 0.55 ± 0.06 0.58 ± 0.03

5 %w/w polymer (D) 0.57 ± 0.05 0.63 ± 0.05 0.85 ± 0.01 0.71 ± 0.02

ANOVA P < 0.05 P < 0.05 P < 0.05 P < 0.05

A&D A&B A&B A&B

B&D A&C A&C A&C
TUKEY’S HSD SIGNIFICANT
C&D A&D A&D A&D
DIFFERENCE
B&C B&D B&D
B&D C&D C&D
255
Appendix 32 : The peak force of detachment (Fmax) of CE suppositories measured using shear setup against colon mucosa. Outcomes were a

result of ANOVA and post hoc Tukey’s HSD analysis. Mean ± SD, n=5-6.

FORMULATION

CBP HPMC PVP CMCTS

DcNa only (A) 0.46 ± 0.02 0.46 ± 0.02 0.46 ± 0.02 0.46 ± 0.02

1 %w/w polymer (B) 0.49 ± 0.04 0.56 ± 0.05 0.52 ± 0.05 0.56 ± 0.04

2 %w/w polymer (C) 0.49 ± 0.01 0.56 ± 0.01 0.69 ± 0.06 0.63 ± 0.02

5 %w/w polymer (D) 0.53 ± 0.04 0.56 ± 0.03 0.81 ± 0.07 0.65 ± 0.02

ANOVA P < 0.05 P < 0.05 P < 0.05 P < 0.05

A&D A&B A&C A&B

A&C A&D A&C
TUKEY’S HSD SIGNIFICANT
A&D B&C A&D
DIFFERENCE
B&D B&D
C&D
256
Appendix 33 : The peak force of detachment (Fmax) of SS suppositories measured using shear setup against colon mucosa. Outcomes were a

result of ANOVA and post hoc Tukey’s HSD analysis. Mean ± SD, n=5-6.

FORMULATION

CBP HPMC PVP CMCTS

DcNa only (A) 0.41 ± 0.03 0.41 ± 0.03 0.41 ± 0.03 0.41 ± 0.03

1 %w/w polymer (B) 0.40 ± 0.05 0.46 ± 0.04 0.45 ± 0.07 0.48 ± 0.05

2 %w/w polymer (C) 0.49 ± 0.05 0.49 ± 0.04 0.65 ± 0.04 0.53 ± 0.04

5 %w/w polymer (D) 0.59 ± 0.03 0.55 ± 0.04 0.78 ± 0.05 0.61 ± 0.04

ANOVA P < 0.05 P < 0.05 P < 0.05 P < 0.05

A&D A&D A&C A&C

B&D B&D A&D A&D
TUKEY’S HSD SIGNIFICANT
C&D C&D B&C B&D
DIFFERENCE
B&D C&D
C&D
257
Appendix 34 : The peak force of detachment (Fmax) of SS suppositories measured using tensile setup against colon mucosa and synthetic

regenerated cellulose. Outcomes were a result of ANOVA and post hoc Tukey’s HSD analysis. Mean ± SD, n=5-6.

FORMULATION

Colon mucosa Synthetic membrane

CBP HPMC PVP CMCTS CBP HPMC PVP CMCTS

DcNa only (A) 1.82 ± 0.13 1.82 ± 0.13 1.82 ± 0.13 1.82 ± 0.13 9.89 ± 0.86 9.89 ± 0.86 9.89 ± 0.86 9.89 ± 0.86

2 %w/w polymer (B) 3.80 ± 0.17 2.27 ± 0.50 3.53 ± 0.45 2.79 ± 0.33 12.00 ± 1.19 11.92 ± 0.85 13.13 ± 1.41 12.98 ± 0.73

5 %w/w polymer (C) 4.57 ± 0.29 2.86 ± 0.50 4.94 ± 0.65 3.56 ± 0.22 13.71 ± 0.48 12.60 ± 1.27 14.60 ± 0.48 13.69 ± 0.42

ANOVA P < 0.05 P < 0.05 P < 0.05 P < 0.05 P < 0.05 P < 0.05 P < 0.05 P < 0.05

TUKEY’S HSD A&C A&C A&B A&C A&C A&C A&B A&B
SIGNIFICANT A&B A&C A&B A&C A&C
DIFFERENCE B&C B&C B&C
258
Appendix 35 : The peak force of detachment (Fmax) of SS suppositories measured using shear setup against colon mucosa and synthetic

regenerated cellulose. Outcomes were a result of ANOVA and post hoc Tukey’s HSD analysis. Mean ± SD, n=5-6.

FORMULATION

Colon mucosa Synthetic membrane

CBP HPMC PVP CMCTS CBP HPMC PVP CMCTS

DcNa only (A) 0.41 ± 0.03 0.41 ± 0.03 0.41 ± 0.03 0.41 ± 0.03 0.40 ± 0.07 0.40 ± 0.07 0.40 ± 0.07 0.40 ± 0.07

2 %w/w polymer (B) 0.49 ± 0.05 0.49 ± 0.04 0.65 ± 0.04 0.53 ± 0.04 0.47 ± 0.06 0.47 ± 0.04 0.54 ± 0.05 0.53 ± 0.02

5 %w/w polymer (C) 0.59 ± 0.03 0.55 ± 0.04 0.78 ± 0.05 0.61 ± 0.04 0.53 ± 0.04 0.53 ± 0.05 0.70 ± 0.07 0.62 ± 0.07

ANOVA P < 0.05 P < 0.05 P < 0.05 P < 0.05 P < 0.05 P < 0.05 P < 0.05 P < 0.05

TUKEY’S HSD A&C A&C A&B A&B A&C A&C A&B A&B
SIGNIFICANT B&C B&C A&C A&C A&C A&C
DIFFERENCE B&C B&C
259
Appendix 36 : Hardness values (N) of suppositories subjected to various storage condition and duration. Outcomes were a result of ANOVA and

post hoc Tukey’s HSD analysis. Mean ± SD, n=6.

FORMULATION
CB + DcNa + CB + DcNa + CE + DcNa + CE + DcNa + SS + DcNa + SS + DcNa +
5 % PVP 5 % CMCTS 5 % PVP 5 % CMCTS 5 % PVP 5 % CMCTS

Fresh samples (A) 99.83 ± 2.32 131.17 ± 2.32 91.33 ± 3.94 106.67 ± 8.09 91.33 ± 3.93 106.67 ± 8.09
Refrigerated for 100
133.83 ± 5.64 114.50 ± 2.80 118.67 ± 5.05 106.50 ± 4.09 108.67 ± 5.05 106.50 ± 4.09
days (B)
Refrigerated for 200
128.33 ± 3.51 138.33 ± 5.32 127.67 ± 4.27 116.67 ± 8.71 100.83 ± 8.35 114.17 ± 2.93
days (C)
Room temperature
116.17 ± 1.60 127.67 ± 4.27 98.17 ± 6.55 111.17 ± 7.83 82.83 ± 6.05 111.17 ± 7.83
for 100 days (D)
Room temperature
123.50 ± 5.05 131.17 ± 3.87 112.67 ± 6.44 102.83 ± 7.78 86.83 ± 5.23 106.17 ± 2.79
for 200 days (E)

ANOVA P < 0.05 P < 0.05 P < 0.05 P < 0.05 P < 0.05 P < 0.05

A&B B&D A&C A&B C&D C&E A&B C&E A&C

TUKEY’S HSD A&C B&E B&C A&C C&E A&C B&C
SIGNIFICANT A&D C&E C&D A&E D&E B&C C&D
DIFFERENCE A&E D&E C&E B&D B&D
B&C B&E
260
Appendix 37 : Softening time (min) of suppositories subjected to various storage condition and duration. Outcomes were a result of ANOVA

and post hoc Tukey’s HSD analysis. Mean ± SD, n=3.

FORMULATION
CB + DcNa + CB + DcNa + CE + DcNa + CE + DcNa + SS + DcNa + SS + DcNa +
5 % PVP 5 % CMCTS 5 % PVP 5 % CMCTS 5 % PVP 5 % CMCTS
Fresh samples (A) 4.03 ± 0.02 4.20 ± 0.03 5.58 ± 0.08 4.92 ± 0.09 5.13 ± 0.05 5.29 ± 0.04
Refrigerated for 100
4.63 ± 0.08 3.99 ± 0.14 4.54 ± 0.04 4.45 ± 0.55 4.77 ± 0.06 4.20 ± 0.64
days (B)
Refrigerated for 200
4.27 ± 0.09 4.03 ± 0.06 5.44 ± 0.08 5.04 ± 0.04 5.14 ± 0.60 4.99 ± 0.06
days (C)
Room temperature
6.38 ± 0.04 6.18 ± 0.13 6.06 ± 0.09 6.24 ± 0.26 5.57 ± 0.05 6.06 ± 0.23
for 100 days (D)
Room temperature
6.87 ± 0.09 5.99 ± 0.07 7.08 ± 0.41 7.54 ± 0.06 6.5 ± 0.08 6.33 ± 0.08
for 200 days (E)
ANOVA P < 0.05 P < 0.05 P < 0.05 P < 0.05 P < 0.05 P < 0.05
A&B B&D A&D C&E A&B B&E A&D C&E A&E A&B C&D
TUKEY’S HSD A&C B&E A&E A&E C&D A&E D&E B&D A&C C&E
SIGNIFICANT A&D C&D B&D A&D C&E B&D B&E A&E
DIFFERENCE A&E C&E B&E B&C D&E B&E C&E B&D
B&C D&E C&D B&D C&D D&E B&E
261