Donepezil in vascular dementia

A randomized, placebo-controlled study

D. Wilkinson, MBChB, MRCGP, FRCPsych; R. Doody, MD, PhD;
R. Helme, MBBS, PhD, FRACP, FFPMANZCA; K. Taubman, MBBS, FRACP; J. Mintzer, MD;
A. Kertesz, MD; R.D. Pratt, MD; and the Donepezil 308 Study Group*

Abstract—Objective: To evaluate the efficacy and tolerability of donepezil in patients with vascular dementia (VaD).
Methods: Patients (n ⫽ 616; mean age, 75.0 years) with probable or possible VaD, according to National Institute of
Neurological Disorders and Stroke–Association Internationale pour la Recherche en l’Enseignement en Neurosciences
criteria, were randomized to receive donepezil 5 mg/day (n ⫽ 208), donepezil 10 mg/day (after 5 mg/day for the first 28
days) (n ⫽ 215), or placebo (n ⫽ 193) for 24 weeks. Results: Seventy-six percent of the patients enrolled had probable VaD.
A total of 75.3% of the 10 mg donepezil group and 80.8% of the 5 mg group completed the study compared with 83.4% of
the placebo group. Both donepezil-treated groups showed improvements in cognitive function on the Alzheimer’s Disease
Assessment Scale– cognitive subscale compared with placebo, with a mean endpoint treatment difference, as measured by
the change from baseline score, of approximately 2 points (donepezil 5 mg, ⫺1.65 [p ⫽ 0.003]; 10 mg, ⫺2.09 [p ⫽ 0.0002]).
Greater improvements on the Clinician’s Interview-Based Impression of Change–plus version were observed with both
donepezil groups than with the placebo group (overall donepezil treatment vs placebo p ⫽ 0.008); 25% of the placebo group
showed improvement compared with 39% (p ⫽ 0.004) of the 5 mg group and 32% (p ⫽ 0.047) of the 10 mg group.
Withdrawal rates due to adverse events were low (placebo, 8.8%; donepezil 5 mg, 10.1%; 10 mg, 16.3%). Conclusions:
Donepezil-treated patients demonstrated significant improvements in cognition and global function compared with
placebo-treated patients, and donepezil was well tolerated.
NEUROLOGY 2003;61:479 –486

Vascular dementia (VaD)—also known as dementia with National Institute of Neurologic Disorders and
associated with cerebrovascular disease—accounts Stroke–Association Internationale pour la Recherche
for 15 to 20% of dementia cases.1-3 Dementia is an et l’Enseignement en Neurosciences (NINDS-
independent risk factor for mortality in patients with AIREN) criteria– diagnosed17 VaD.
ischemic stroke4; however, there are currently no ap-
proved or recommended treatments for the cognitive Methods. Study design. This was a 24-week, double-blind,
randomized, placebo-controlled, parallel-group study conducted at
impairment of VaD.5 51 sites in the United States, Europe, Canada, and Australia.
Preclinical6,7 and clinical8,9 evidence indicates that Patient population. Men or nonpregnant women of any race
a cholinergic deficit, similar to that seen in Alzhei- and at least 40 years of age were considered for inclusion. A
diagnosis of possible or probable VaD of more than 3 months in
mer disease (AD),10 may be associated with VaD, duration, as defined by NINDS-AIREN criteria,17 was required for
suggesting that patients with VaD may benefit from enrollment. All eligible patients were also required to exhibit clin-
treatment with cholinesterase (ChE) inhibitors. A re- ical and radiologic evidence of cerebrovascular disease. It was
inherent in the diagnosis that patients would have risk factors for
cent study has shown some benefits of a ChE inhibi- hypertension, diabetes (type 1 and type 2), cardiac disease, or
tor in patients with AD plus cerebrovascular stroke; patients with these conditions were enrolled providing the
disease.11 Moreover, treatment with donepezil, an diseases had been stable or controlled by medication for at least 3
months.
acetylcholinesterase inhibitor that significantly ben- The diagnosis of VaD by NINDS-AIREN criteria precludes
efits cognition, global function, and activities of daily other causes of dementia, thus excluding patients with AD and
living (ADL) in patients with mild to moderately se- other neurodegenerative disorders that may be responsible for the
dementia. Patients were also excluded from study participation if
vere AD,12-15 improved patients with probable VaD they had psychiatric disorders such as schizophrenia, a Mini-
during a 6-month, open-label, pilot study.16 Mental State Examination (MMSE) score18 below 10 or above 26,
We report the results of one of the first two new stroke within the prior 28 days, myocardial infarction within
the last 3 months (although patients could be reconsidered for
placebo-controlled clinical trials of the efficacy and inclusion after 3 months had elapsed), any clinically relevant or
tolerability of a ChE inhibitor in a cohort of patients life-threatening disease, drug or alcohol abuse, or a known hyper-

*See the Appendix for a complete list of the members of the Donepezil 308 Study Group.
From Memory Assessment and Research Centre (Dr. Wilkinson), Moorgreen Hospital, Southampton, UK; Alzheimer’s Disease Center (Dr. Doody), Baylor
College of Medicine, Houston, TX; National Ageing Research Institute (Dr. Helme), Parkville, Victoria, Australia; Austin and Repatriation Medical Centre
(Dr. Taubman), Heidelberg, Victoria, Australia; Medical University of South Carolina (Dr. Mintzer), Alzheimer’s Research and Clinical Programs, North
Charleston, SC; University of Western Ontario (Dr. Kertesz), St. Joseph’s Hospital, London, Ontario, Canada; and Eisai Inc. (Dr. Pratt), Teaneck, NJ.
This report presents data generated from protocol E2020-A001-308, which was sponsored by Eisai Inc.
Received November 14, 2002. Accepted in final form May 7, 2003.
Address correspondence and reprint requests to Dr. David Wilkinson, Memory Assessment and Research Centre, Moorgreen Hospital, Southampton SO30
3JB, UK; e-mail: david.wilkinson@wht.nhs.uk

Copyright © 2003 by AAN Enterprises, Inc. 479

sensitivity to donepezil. Most concomitant medications were per-
mitted during the study, except other cholinesterase inhibitors
and anticholinergics.
The caregiver and the patient (or a legal representative) pro-
vided written informed consent prior to screening for the study.
The study was conducted in accordance with the Declaration of
Helsinki and its subsequent amendments,19 and in compliance
with United States, European Union, Canadian, and Australian
clinical trials regulations.
Protocol. Patients were assigned to treatment groups by a
computer-generated randomization protocol. Patients were admin-
istered placebo, donepezil 5 mg/day, or donepezil 10 mg/day as
single daily doses each evening before bedtime. Blinding was en-
sured by the use of identical-appearing placebo and donepezil
tablets. Patients in the donepezil 10 mg/day group received done-
pezil 5 mg/day for 4 weeks and then 10 mg/day until week 24.
Reduction in treatment dosage was not permitted—if patients
could not tolerate their assigned dose, they were required to dis-
continue from the study.
Clinic visits occurred at screening, at baseline, and at weeks 6,
12, 18, and 24. During these visits, psychometric evaluations,
Figure 1. Subject disposition. *Twenty-one patients were
physical and neurologic examinations, laboratory determinations, excluded from the intent-to-treat (ITT) population, as they
vital signs measurements, medication compliance checks, and ad- did not take at least one dose of study medication, and did
verse event (AE) monitoring took place. If CT or MRI scans had not have a baseline, or at least one postbaseline, assess-
not been undertaken within the last 6 months, they were carried ment of at least one primary efficacy variable. †Six pa-
out at screening.
Outcome measures. Primary efficacy assessments. The pri-
tients died during the study (one in the placebo group,
mary efficacy measures were the AD Assessment Scale– cognitive three in the donepezil 5 mg/day group, and two in the
subscale (ADAS-cog)20 and the Clinician’s Interview-Based Im- donepezil 10 mg/day group). ‡Examples of other reasons
pression of Change–plus version (CIBIC-plus).21 The ADAS-cog is for discontinuation included medication noncompliance (1
a sensitive psychometric scale for measuring cognitive function, patient), protocol violation (5 patients), nursing home
routinely utilized in trials conducted in patients with AD.12,13
ADAS-cog assessment was performed by a trained clinician. An placement (4 patients), patient moved (6 patients), request
independent clinician, who was blinded to the patient’s psycho- of caregiver (2 patients). AE ⫽ adverse events; MMSE ⫽
metric test scores and AE information, performed the CIBIC-plus. Mini-Mental State Examination; AD ⫽ Alzheimer disease.
The severity of disease (Clinician’s Interview-Based Impression of
Severity [CIBIS]) was assessed at baseline, and then CIBIC-plus
change from baseline scores were rated at weeks 6, 12, 18, and 24. difference in CIBIC-plus between active treatment and placebo at
The CIBIC-plus used in this study was developed and validated the 0.05 significance level, allowing for a discontinuation rate of
for patients with AD by the AD Cooperative Study Group.22 It is 20%. Patients who withdrew from the study were not replaced.
derived from a semi-structured interview conducted by a clinician Efficacy analyses were performed on the intent-to-treat (ITT)
with the caregiver and the patient. The CIBIC-plus interview population, defined as all patients who were randomized to treat-
domain questions were modified to include the range of impair- ment, received at least one dose of study medication, and who
ments considered by the investigators to be important for patients provided data at baseline and at least one postbaseline efficacy
with VaD. A range of assessments validated for use in AD were assessment. Within the ITT population, analyses of both observed
utilized in this study, as assessments specific for VaD had not cases (OC) at each scheduled visit (weeks 6, 12, 18, and 24) and
been developed at the time of study design. last observation carried forward (LOCF) at week 24 were con-
Secondary efficacy assessments. The secondary efficacy mea- ducted. Week 24 LOCF using the ADAS-cog and the CIBIC-plus
sures utilized in this study included the MMSE18 (conducted by a was defined as the primary endpoint evaluation for each patient.
trained clinician) and the Sum of the Boxes of the Clinical Demen- Additional subgroup analyses were performed on patients with
tia Rating23 (CDR-SB); this score was agreed upon by consensus possible VaD and on patients with probable VaD, as classified by
among members of each patient’s assessment team. The AD Func- the investigator according to NINDS-AIREN criteria. Analysis of
tional Assessment and Change Scale (ADFACS) was used to mon- safety was performed on the population that included all patients
itor function. The ADFACS assesses instrumental and basic who received at least one dose of study medication and who pro-
ADL,15 and it should therefore be applicable to patients with VaD, vided any postbaseline follow-up data.
provided physical impairments due to stroke or other medical Demographic characteristics and efficacy measure outcomes
conditions are taken into account.24 This was done by categorizing were examined by an analysis of covariance (ANCOVA) (for con-
those patients with permanent motor or sensory deficits, in whom tinuous variables) or by Cochran-Mantel-Haenszel test (for cate-
change in certain ADL may not be observed, as “not assessable” in gorical variables). The least squares (LS) mean change from
affected domains, and assessing only physically unaffected baseline scores to weeks 6, 12, 18, and 24 and endpoint are pre-
domains. sented for all variables analyzed with an ANCOVA model.
Safety assessments. The safety and tolerability of the study All tests were two-tailed and conducted at the 0.05 significance
medication were assessed continually from baseline to endpoint by level.
monitoring discontinuations from the study and by comparing
treatment groups with respect to rates of AE, median changes in Results. Subject disposition. A total of 887 patients
laboratory test values, rates of clinically relevant laboratory ab- were screened and 616 entered the study and were ran-
normalities, changes in vital signs, electrocardiogram abnormali- domized to receive placebo (n ⫽ 193), donepezil 5 mg/day
ties, concomitant medication use, and treatment-emergent
physical examination findings.
(n ⫽ 208), or donepezil 10 mg/day (n ⫽ 215) (figure 1).
Statistical analysis. Sample size was determined following a Reasons for screening failure included not meeting entry
review of the results of phase III clinical trials of donepezil in criteria (e.g., 63 patients had an MMSE score ⬍10 or ⬎26),
AD.12,13,25 However, the sample size was increased from 450 pa- withdrawal of consent, a diagnosis of AD, and unstable
tients after a blinded analysis of the variance in CIBIC-plus scores medical conditions (see figure 1).
in the first 100 patients enrolled in the study indicated that the
SD was larger than observed in the AD studies. A revised sample Baseline characteristics: VaD population. Overall, 76%
size of 600 patients (200 placebo, 200 donepezil 5 mg/day, 200 (468/616) of enrolled patients had a diagnosis of probable
donepezil 10 mg/day) was considered sufficient to detect a 0.3 unit VaD, and 24% (148/616) had a diagnosis of possible VaD.
480 NEUROLOGY 61 August (2 of 2) 2003
Table 1 Baseline characteristics

Donepezil 5 mg, Donepezil 10 mg,

Characteristics Placebo, n ⫽ 193 n ⫽ 208 n ⫽ 215 Total, n ⫽ 616

Mean age, y, ⫾ SE (range) 74.4 ⫾ 0.6 (41–89) 74.7 ⫾ 0.6 (38–95) 75.7 ⫾ 0.6 (47–90) 75.0 ⫾ 0.3 (38–95)
Women, n (%) 88 (45.6) 78 (37.5) 81 (37.7) 247 (40.1)
Diagnosis, n (%)
Probable VaD 145 (75.1) 161 (77.4) 162 (75.3) 468 (76.0)
Possible VaD 48 (24.9) 47 (22.6) 53 (24.7) 148 (24.0)
Abnormal CT/MRI, n (%) 193 (100) 207 (99.5) 213 (99.1) 613 (99.5)
Medical history, n (%)
Cardiovascular disease 176 (91.2) 183 (88.0) 198 (92.1) 557 (90.4)
Stroke* 130 (67.4) 138 (66.3) 142 (66.0) 410 (66.6)
Cerebral infarction* 141 (73.1) 167 (80.3) 161 (74.9) 469 (76.1)
Hachinski score, mean ⫾ SE 9.6 ⫾ 0.2 (1–17) 9.4 ⫾ 0.2 (1–17) 9.5 ⫾ 0.2 (1–16) 9.5 ⫾ 0.1 (1–17)
(range)
Psychometric score, LS mean ⫾ SE
ADAS-cog 18.8 ⫾ 0.7 20.8 ⫾ 0.7† 20.6 ⫾ 0.7
MMSE 22.2 ⫾ 0.3 21.8 ⫾ 0.3 21.5 ⫾ 0.3
CDR-SB 5.6 ⫾ 0.2 6.0 ⫾ 0.2 6.1 ⫾ 0.2
ADFACS 15.1 ⫾ 0.7 15.7 ⫾ 0.7 16.1 ⫾ 0.7

* The presence of stroke was determined clinically, and the presence of cerebral infarction was determined radiologically.
† p ⫽ 0.048 vs placebo.

VaD ⫽ vascular dementia; LS ⫽ least squares; ADAS-cog ⫽ Alzheimer’s Disease Assessment Scale– cognitive; MMSE ⫽ Mini-Mental
State Examination; CDR-SB ⫽ subscale/sum of the boxes of the Clinical Dementia Rating; ADFACS ⫽ Alzheimer’s Disease Functional
Assessment and Change Scale.

Almost all patients (90%) had comorbid cardiovascular dis- and 10 mg, 75.3%) (see figure 1). Approximately half (73/
ease, and most had a history of cerebral infarction (deter- 125) of all discontinuations were due to treatment-
mined radiologically) (76%) and stroke (determined emergent AE. Additional reasons for discontinuation
clinically) (67%) (table 1). The majority of the population included the request of the patient or investigator, protocol
had hypertension (72%), most had a history of smoking violation, and nursing home placement (see figure 1).
(61%), and many had a history of hypercholesterolemia The overall mean compliance rate for all visits was
(37%). The prevalence of these conditions was similar 97.8% (placebo, 96.7%; donepezil 5 mg, 100.2%; and 10 mg,
among the treatment groups, with the exception of a his- 96.3%).
tory of hypertension, which was more common in the pla- At least one concomitant medication was used by 614/
cebo group than in the active treatment groups (placebo, 616 patients (99.7%) during the study (table 2). There were
79%, donepezil 5 mg, 70%, p ⫽ 0.04; and 10 mg, 67%, p ⫽ no differences in concomitant medication usage among
0.01). treatment groups (placebo, 99.0%; donepezil 5 mg, 100%;
Hachinski Ischemia Scale (HIS) scores were similar and 10 mg, 100%). The most common concomitant medica-
among treatment groups (see table 1), with an average tions were antithrombotics (81.5%; aspirin was taken by
total score of more than nine. The majority of patients had 65.6%) and agents acting on the renin-angiotensin system
an abrupt onset of dementia (placebo, 72%; donepezil 5 mg, (36.5%).
72%; and 10 mg, 68%), and many had shown stepwise Primary efficacy outcomes. ADAS-cog. Significant
deterioration in cognition prior to the study (placebo, 55%; differences in favor of donepezil 5 and 10 mg/day, com-
donepezil 5 mg, 51%; and 10 mg, 58%). pared with placebo, were observed on the ADAS-cog at
The three treatment groups were similar at baseline weeks 12, 18, and 24, and also at endpoint (figure 2). The
with respect to their demographic characteristics and psy- LS mean ⫾ SE change from baseline score in placebo-
chometric test scores (see table 1), with the exception of treated patients was ⫺0.58 ⫾ 0.42 at week 24 OC, and
the baseline ADAS-cog score, which was lower (better) in ⫺0.10 ⫾ 0.39 at endpoint. In comparison, the LS mean ⫾
the placebo group than in the donepezil 5 mg/day group SE change from baseline scores in donepezil-treated pa-
(p ⫽ 0.048) (see table 1). However, ADAS-cog scores were tients were ⫺2.02 ⫾ 0.36 at week 24 OC and ⫺1.75 ⫾ 0.33
similar among the treatment groups at screening (mean ⫾ at endpoint in the 5 mg group, and ⫺2.65 ⫾ 0.48 at week
SE: placebo, 21.8 ⫾ 0.8; donepezil 5 mg, 23.5 ⫾ 0.8; and 10 24 OC and ⫺2.19 ⫾ 0.44 at endpoint in the 10 mg group.
mg, 23.6 ⫾ 0.8). CIBIC-plus. More patients in the donepezil 5 mg and
The overall study completion rate was 79.7%, and the 10 mg treatment groups than in the placebo group were
three treatment groups showed only small differences in rated on the CIBIC-plus as showing improvement at end-
completion rates (placebo, 83.4%; donepezil 5 mg, 80.8%; point (% of each group showing minimal, moderate, or
August (2 of 2) 2003 NEUROLOGY 61 481
Table 2 Common concomitant medications taken during the study

Placebo, Donepezil 5 mg, Donepezil 10 mg,

Medications n ⫽ 193 n ⫽ 208 n ⫽ 215

ⱖ1 Concomitant medication 191 (99.0) 208 (100) 215 (100)

Antithrombotic agents 161 (83.4) 166 (79.8) 175 (81.4)
Aspirin 134 (69.4) 133 (63.9) 137 (63.7)
Agents acting on renin-angiotensin system 86 (44.6) 72 (34.6) 67 (31.2)
Vitamins 68 (35.2) 75 (36.1) 78 (36.3)
Calcium channel blockers 74 (38.3) 59 (28.4) 72 (33.5)
Analgesics 62 (32.1) 71 (34.1) 67 (31.2)
Diuretics 54 (28.0) 71 (34.1) 68 (31.6)
Antibacterials for systemic use 61 (31.6) 65 (31.3) 63 (29.3)
Serum lipid reducing agents 57 (29.5) 47 (22.6) 63 (29.3)
Cardiac therapy* 46 (23.8) 42 (20.2) 48 (22.3)

Values are n (%).

* Cardiac therapy includes glycosides (e.g., digoxin) and nitrates (e.g., glyceryl trinitrate).

marked improvement: placebo, 25%; donepezil 5 mg, 39%; CDR-SB. At week 12, patients receiving donepezil 5 or
and 10 mg, 32%) (figure 3). Greater improvements on the 10 mg/day demonstrated greater improvements on the
CIBIC-plus were observed with both donepezil treatment CDR-SB than placebo-treated patients. At weeks 18 and
groups than with the placebo group at endpoint (overall 24, and at endpoint, placebo-treated patients showed a
donepezil treatment, p ⫽ 0.008; donepezil 5 mg, p ⫽ 0.004; decline from baseline levels, whereas patients receiving
and 10 mg, p ⫽ 0.047). Benefits over placebo were also donepezil (5 or 10 mg/day) continued to show improvement
observed in favor of both doses of donepezil at week 12 above baseline, with greater benefits vs placebo in the 10
(overall donepezil treatment, p ⫽ 0.0004; 5 mg, p ⫽ 0.003; mg/day group (LS mean ⫾ SE change from baseline score
10 mg, p ⫽ 0.001), week 18 (overall donepezil treatment, p at week 24 OC: placebo, 0.08 ⫾ 0.14; donepezil 5 mg,
⫽ 0.002; 5 mg, p ⫽ 0.041; 10 mg, p ⫽ 0.001), and week 24 ⫺0.19 ⫾ 0.14, p ⫽ 0.18; and 10 mg, ⫺0.36 ⫾ 0.14, p ⫽ 0.03;
(overall donepezil treatment, p ⫽ 0.01; 5 mg, p ⫽ 0.006; 10 LS mean ⫾ SE change from baseline score at endpoint:
mg, p ⫽ 0.078). placebo, 0.16 ⫾ 0.13; donepezil 5 mg, ⫺0.15 ⫾ 0.12, p ⫽
Secondary efficacy outcomes. MMSE. Patients 0.07; and 10 mg, ⫺0.21 ⫾ 0.12, p ⫽ 0.03).
treated with donepezil 5 or 10 mg/day demonstrated signif- ADFACS. ADFACS scores were maintained above
icantly greater improvements on the MMSE than placebo- baseline in patients administered donepezil 10 mg/day (LS
treated patients at weeks 12, 18, and 24, and at endpoint
(figure 4). Placebo-treated patients showed small postbase-
line improvements in MMSE score (LS mean ⫾ SE change
from baseline score at week 24 OC, 0.41 ⫾ 0.25; at end-
point, 0.23 ⫾ 0.24).

Figure 3. Clinician’s Interview-Based Impression of

Change–plus version ratings at endpoint (week 24, last
observation carried forward) in donepezil- and placebo-
Figure 2. Alzheimer Disease Assessment Scale– cognitive treated patients. Overall treatment p ⬍ 0.01. p ⬍ 0.01
subscale least squares mean change from baseline score in donepezil 5 mg/day vs placebo, p ⬍ 0.05 donepezil 10 mg/
donepezil- and placebo-treated patients. *p ⬍ 0.05, **p ⬍ day vs placebo. Black bars ⫽ donepezil 10 mg/d (n ⫽
0.01, ***p ⬍ 0.001 vs placebo. ⫽ Donepezil 10 202); dark gray bars ⫽ donepezil 5 mg/d (n ⫽ 202); light
mg/d; ⫽ donepezil 5 mg/d; —»— ⫽ placebo. gray bars ⫽ placebo (n ⫽ 188).
482 NEUROLOGY 61 August (2 of 2) 2003
 Table 3 Adverse events (AE) occurring in ⬎5% of patients in any
group

Donepezil Donepezil
Placebo, 5 mg, 10 mg,
Adverse event n ⫽ 193 n ⫽ 208 n ⫽ 215

Any AE 167 (86.5) 188 (90.4) 197 (91.6)

Body as a whole 93 (48.2) 116 (55.8) 99 (46.0)
Accidental injury 19 (9.8) 35 (16.8)* 27 (12.6)
Asthenia 16 (8.3) 18 (8.7) 24 (11.2)
Back pain 6 (3.1) 12 (5.8) 3 (1.4)
Headache 16 (8.3) 15 (7.2) 14 (6.5)
Infection 26 (13.5) 36 (17.3) 30 (14.0)
Figure 4. Mini-Mental State Examination least squares Pain 19 (9.8) 16 (7.7) 13 (6.0)
mean change from baseline score in donepezil- and
Cardiovascular system 42 (21.8) 39 (18.8) 42 (19.5)
placebo-treated patients. *p ⬍ 0.05, **p ⬍ 0.01, ***p ⬍
0.001 vs placebo. ⫽ Donepezil 10 mg/d; ⫽ Hypertension 12 (6.2) 11 (5.3) 13 (6.0)
donepezil 5 mg/d; —»— ⫽ placebo. Syncope 7 (3.6) 3 (1.4) 11 (5.1)
Digestive system 65 (33.7) 83 (39.9) 96 (44.7)
Anorexia 11 (5.7) 17 (8.2) 19 (8.8)
mean ⫾ SE change from baseline score at week 24 OC,
Constipation 9 (4.7) 11 (5.3) 5 (2.3)
⫺0.38 ⫾ 0.45; at endpoint, ⫺0.23 ⫾ 0.40), reaching signif-
icance compared with placebo at week 12. Patients treated Diarrhea 20 (10.4) 24 (11.5) 40 (18.6)*
with donepezil 5 mg/day showed less deterioration than Dyspepsia 12 (6.2) 10 (4.8) 8 (3.7)
placebo, although this difference was not significant (LS Nausea 15 (7.8) 23 (11.1) 36 (16.7)†
mean ⫾ SE change from baseline score at week 24 OC,
⫺0.35 ⫾ 0.48; at endpoint, 0.11 ⫾ 0.45). In contrast, Vomiting 16 (8.3) 10 (4.8) 11 (5.1)
placebo-treated patients demonstrated a progressive de- Nervous system 86 (44.6) 96 (46.2) 107 (49.8)
cline in function, as assessed by the ADFACS (LS mean ⫾ Abnormal dreams 1 (0.5) 4 (1.9) 11 (5.1)†
SE change from baseline score at week 24 OC, 0.27 ⫾ 0.37;
Agitation 10 (5.2) 11 (5.3) 12 (5.6)
at endpoint, 0.76 ⫾ 0.39).
The results for the instrumental ADL subscale of the CVA (stroke) 11 (5.7) 7 (3.4) 8 (3.7)
ADFACS showed similar trends (LS mean ⫾ SE change Confusion 12 (6.2) 14 (6.7) 8 (3.7)
from baseline score at week 24 OC, placebo, 0.27 ⫾ 0.28; Depression 12 (6.2) 13 (6.3) 12 (5.6)
donepezil 5 mg, ⫺0.24 ⫾ 0.37; and 10 mg, ⫺0.48 ⫾ 0.34,
Dizziness 21 (10.9) 16 (7.7) 17 (7.9)
p ⫽ 0.109; LS mean ⫾ SE change from baseline score at
endpoint, placebo, 0.46 ⫾ 0.27; donepezil 5 mg, ⫺0.03 ⫾ Insomnia 10 (5.2) 22 (10.6)* 21 (9.8)
0.34; and 10 mg, ⫺0.34 ⫾ 0.29, p ⫽ 0.054). Other
Subgroup analyses. Probable VaD. Differences in fa- Peripheral edema 10 (5.2) 13 (6.3) 7 (3.3)
vor of both doses of donepezil, compared with placebo, were
observed on the ADAS-cog (LS mean ⫾ SE change from Leg cramps 1 (0.5) 11 (5.3)† 13 (6.0)†
baseline score at endpoint, placebo, ⫺0.30 ⫾ 0.46; donepe- Rhinitis 1 (0.5) 8 (3.8)* 11 (5.1)†
zil 5 mg, ⫺1.69 ⫾ 0.37, p ⫽ 0.028; and 10 mg, ⫺2.37 ⫾ Skin abrasion 7 (3.6) 11 (5.3) 12 (5.6)
0.50, p ⫽ 0.001), with additional significant differences for
Skin rash 8 (4.1) 11 (5.3) 6 (2.8)
5 and 10 mg/day on the CIBIC-plus and MMSE, and for 10
mg/day only on the CDR-SB at week 24. Results in the Urinary 8 (4.1) 6 (2.9) 17 (7.9)
probable VaD subgroup (placebo, n ⫽ 145; donepezil 5 mg, incontinence
n ⫽ 161; and 10 mg, n ⫽ 162) were therefore very similar Urinary tract 15 (7.8) 22 (10.6) 14 (6.5)
to those reported for the whole cohort. infection
Possible VaD. The possible VaD subgroup (placebo,
Values are n (%).
n ⫽ 48; donepezil 5 mg, n ⫽ 47; and 10 mg, n ⫽ 53) showed
a similar pattern of responses to those observed in the * p ⬍ 0.05 and †p ⬍ 0.01 vs placebo group.
overall cohort. Benefits in both active treatment groups
CVA ⫽ cerebrovascular accident.
compared with placebo were seen on the ADAS-cog (LS
mean ⫾ SE change from baseline score at endpoint, pla-
cebo, 0.37 ⫾ 0.74; donepezil 5 mg, ⫺1.13 ⫾ 0.72, p ⫽ 0.22; emergent AE experienced by 86.5% of the placebo group,
and 10 mg, ⫺1.03 ⫾ 0.88, p ⫽ 0.24), CIBIC-plus, MMSE, 90.4% of the donepezil 5 mg group, and 91.6% of the 10 mg
and CDR-SB. Most of these differences were not signifi- group. The AE occurring in more than 5% of either donepe-
cant; however, statistical conclusions in this subgroup are zil treatment group and at least twice the incidence of the
not robust owing to the small sample size. placebo group were nausea, abnormal dreams, insomnia,
Safety. The proportion of patients with AE was similar leg cramps, and rhinitis (table 3). There were no overall
among treatment groups, with at least one treatment- differences among the treatment groups in the incidences
August (2 of 2) 2003 NEUROLOGY 61 483
Table 4 Serious adverse events (SAE) occurring in ⬎0.5% of Discussion. In this study, patients with VaD
patients overall treated with donepezil 5 mg or 10 mg once daily
Donepezil Donepezil demonstrated significant benefits over placebo-
Serious adverse Placebo, 5 mg, 10 mg, treated patients on measures of cognition (ADAS-cog
event n ⫽ 193 n ⫽ 208 n ⫽ 215 and MMSE) and global function (CIBIC-plus). In ad-
dition, the donepezil 10 mg/day group showed signif-
Total SAE* 33 (17.1) 31 (14.9) 33 (15.3)
icant improvements compared with placebo on a
CVA (stroke) 9 (4.7) 7 (3.4) 5 (2.3) multidimensional rating of dementia severity, the
Skin carcinoma 8 (4.1) 1 (0.5) 4 (1.9) CDR-SB. Donepezil also slowed functional deteriora-
Pathologic fracture 2 (1.0) 3 (1.4) 3 (1.4) tion in these patients, as assessed on the ADFACS,
although this did not reach significance compared
Pneumonia 4 (2.1) 3 (1.4) 1 (0.5)
with placebo.
Confusion 6 (3.1) 0 1 (0.5)
In order to differentiate patients with VaD from
Transient ischemic 1 (0.5) 2 (1.0) 2 (0.9) those with AD or mixed dementia and thus ensure
attack the inclusion of only patients with VaD, enrolled pa-
Angina 2 (1.0) 1 (0.5) 2 (0.9) tients were required to satisfy NINDS-AIREN crite-
Syncope 1 (0.5) 2 (1.0) 2 (0.9) ria.17 Application of these criteria resulted in a study
Convulsion 3 (1.6) 1 (0.5) 1 (0.5)
population that was classified as 76% probable VaD
and 24% possible VaD.
Colitis 1 (0.5) 0 3 (1.4)
The population studied here included subjects
Values are n (%). with a strong history of cerebrovascular and cardiac
disease, and a high prevalence of hypertension and
* Including death.
hypercholesterolemia. Most of these patients had
CVA ⫽ cerebrovascular accident. had at least one TIA or clinical stroke prior to the
diagnosis of dementia. In addition, patients had a
mean HIS score of 9.5 (a score over 7 is indicative of
of agitation or other behavioral symptoms (see table 3).
VaD),26 the majority had an abrupt onset of demen-
The incidence of cardiovascular AE was similar among all
tia, and many had stepwise cognitive deterioration.
three treatment groups. Hypertension was the most com-
These characteristics are different from those of pa-
mon cardiovascular AE (placebo, 12 patients; donepezil 5
mg, 11 patients; and 10 mg, 13 patients), followed by syn-
tients who participated in previous trials of donepe-
cope (placebo, 7 patients; donepezil 5 mg, 3 patients; and zil in patients with AD.12-15,27
10 mg, 11 patients) (see table 3). Bradycardia was reported In this study, the placebo group did not show cog-
by less than 2% of patients overall (placebo, 4 patients; nitive deterioration over the 24 weeks of observation.
donepezil 5 mg, 2 patients; and 10 mg, 6 patients). Most Progression of VaD may be delayed by treatment of
AE were mild to moderate in severity, and considered by vascular disease risk factors,28 and in the current
the investigator to be unrelated or to be possibly related to study, patients received optimal therapy for their
the study drug. concomitant conditions. This treatment may there-
Six patients died during the study (one in the placebo fore have played some unknown role in delaying de-
group, three in the donepezil 5 mg group, and two in the 10 cline in cognitive function. The relative lack of
mg group), and each of these deaths was considered by the deterioration observed in placebo-treated patients
investigator to be unrelated to the study medication (three with VaD in this trial is supported by the results of
patients died of stroke, one of pneumonia, one of pneumo- previous studies that also utilized NINDS-AIREN
nia/stroke, and one death occurred following hospitaliza- criteria to enroll patients with VaD, in which
tion of the patient for evaluation of hypotension and atrial placebo-treated groups demonstrated minimal cogni-
fibrillation, although the exact cause of death was not tive decline over a period of up to 48 weeks.29,30
specified). Serious AE (SAE) were reported at a similar Placebo-treated patients enrolled in a recent
frequency in all treatment groups; the number of patients 6-month trial of patients with AD plus cerebrovascu-
experiencing SAE (other than death) was 32 (16.6%) in the lar disease (mixed dementia) and patients with prob-
placebo group, 28 (13.5%) in the donepezil 5 mg group, and
able VaD (43% of the total cohort) showed no decline
31 (14.4%) in the 10 mg group. Eighty-seven percent of
in the probable VaD subgroup after 6 months,
SAE were considered to be unrelated to the study drug; the
whereas the subgroup with AD plus cerebrovascular
remainder were considered to be possibly related. The
most common SAE in each group was stroke (table 4); the
disease showed a decline comparable with AD.11
numbers of patients experiencing stroke were not signifi- These observations indicate that patients classified
cantly different among the three groups. Most SAE (78% of as having probable VaD not only differ from those
placebo, 71% of donepezil 5 mg, and 74% of 10 mg) did not with AD but also from those with mixed dementia, a
lead to discontinuation from the study. group showing a pattern of deterioration more typi-
There were no clinically relevant mean changes from cal of AD than VaD. The consistency of these find-
baseline in vital signs, or in any clinical chemistry, hema- ings confirms that the NINDS-AIREN criteria for
tology, or urinalysis tests, in either active treatment group. probable VaD can select a patient population distinct
There were no significant differences from placebo in any from those of probable AD and mixed dementia.
treatment-emergent abnormal laboratory values. In contrast to their relative cognitive stability, pa-
484 NEUROLOGY 61 August (2 of 2) 2003
tients with VaD may show functional decline due to ments in donepezil-treated patients with mixed
repeated cerebrovascular events, particularly stroke, dementia.
that affect somatic function. As with patients with The overall safety profile of donepezil in this study
AD, patients with VaD may be unlikely to regain an is consistent with the clinical experience in patients
instrumental ADL or ADL once they have lost that with AD,33 and demonstrates that donepezil is well
activity, even if their cognition improves. Beneficial tolerated, even in a relatively frail population of el-
effects on function in patients with VaD may there- derly patients with considerable comorbidities. The
fore be best measured as the time to additional func- proportion of patients with AE and SAE was similar
tional loss, as with patients with AD.15 In this study, among the three treatment groups. The AE reported
the donepezil 10 mg/day treatment group demon- more frequently by the donepezil- compared with the
strated improved ability to perform ADL compared placebo-treated patients were due to cholinergic
with the placebo group, as measured by the stimulation and have previously been reported in
ADFACS, although this effect was small. patients with AD treated with donepezil and other
The magnitude of treatment effects (effect sizes) ChE inhibitors.12,14 Most of the SAE were associated
observed in this study is smaller than those observed with underlying cardiac, cerebrovascular, and pe-
in similar AD studies,12,13 because of the nature of ripheral vascular diseases, and did not occur with
disease progression in VaD. The lack of deterioration increased frequency in donepezil-treated patients.
in the VaD placebo group may make it more difficult Furthermore, most SAE did not lead to discontinua-
to demonstrate drug–placebo differences, because tion from the study, possibly because the majority
significant treatment-related differences are driven were considered to be unrelated to treatment.
by actual improvement, and not simply by reduced Compared with patients with stroke who do not
rate of decline, in the drug-treated group. It is likely develop dementia, patients who develop dementia
that the positive effects of donepezil treatment in following stroke (i.e., patients with VaD or mixed
this study are demonstrated against stability or im- dementia) have a two to three times greater risk of
death, and furthermore, deaths due to cerebrovascu-
provement in the placebo group. This observation is
lar disease (e.g., recurrent stroke) and infection (e.g.,
supported by the distribution of CIBIC-plus scores in
pneumonia) are observed more frequently.4 There
this study, because the majority of placebo-treated
were six deaths in this study, five caused by stroke
patients with VaD were rated as “no change” at end-
and/or pneumonia and all considered unrelated to
point. The sensitivity of the CIBIC-plus scale may
study drug, emphasizing the general frailty of pa-
therefore be reduced in this population, because the
tients with VaD.
rating is based on a seven-point scale with three of
In our study, donepezil 5 and 10 mg/day improved
these points representing decline. cognition and global function compared with placebo,
One perceived weakness of the study might be and 10 mg/day was associated with a greater drug–
that the assessments used were designed for AD, as placebo difference than 5 mg/day on a measure of
no specific assessments for VaD had yet been devel- dementia severity. Thus, donepezil 5 mg/day was
oped. Thus, not all aspects of cognition relevant to clinically effective in this population, whereas 10 mg/
VaD, such as executive function,31 were directly as- day yielded additional treatment benefits. In clinical
sessed. However, the CIBIC-plus interview was mod- practice, the optimal dose should be established for
ified slightly to address impairments commonly each patient, to ensure maximal efficacy while main-
encountered with VaD. taining good tolerability.
The relatively short duration of this trial, along
with the likelihood of no change in placebo-treated Acknowledgment
patients, may have made it more difficult for clini- The authors thank PPS International Communications (Worthing,
cians to detect treatment benefits. However, despite UK) for assisting in the development of this article.
these various mitigating factors, which make it more
difficult to detect a treatment difference, these re- Appendix
sults show detectable treatment benefits. The Donepezil 308 Study Group principal investigators are as follows: Jef-
frey T. Apter, Princeton Biomedical Research, Princeton, NJ; Peter M.
In our study, donepezil-treated patients with Aupperle, COPSA Institute for AD and Related Disorders, UMDNJ/UBHC,
probable or possible VaD demonstrated significant Piscataway, NJ; Nancy R. Barbas, University of Michigan Medical Center,
improvement on the ADAS-cog and the MMSE, rela- Ann Arbor, MI; Barry Baumel, Baumel Eisner Neuromedical Institute, Ft.
Lauderdale, FL; Patricia Blanchette, Honolulu, HI; Sharon Cohen, North
tive to placebo, over 24 weeks. Subanalyses revealed York Senior’s Health Center, North York, Ontario, Canada; Helen Creasey,
similar responses in both VaD groups—those with Tony Broe, Center for Education and Research on Aging, Concord Repatri-
ation General Hospital, Concord, NSW, Australia; Denis Crimmins, Central
probable VaD and those with possible VaD. Indeed, Coast Neuroscience Research, East Gosford, NSW, Australia; James Dex-
significant improvements in cognition compared to ter, University of Missouri, Columbia, MO; Rachelle S. Doody, Baylor Col-
lege of Medicine, Houston, TX; Eugene DuBoff, Denver Center for Medical
placebo are observed in the probable VaD subgroup Research, Denver, CO; Keith R. Edwards, Geriatric Neurology and Alzhei-
and the possible VaD subgroup of the combined co- mer’s Center, Bennington, VT; Larry Einbinder, Neurology Center, Rock-
ville, MD; Karl Einhaeupl, Universitaetsklinikum Charite, Berlin,
hort of patients enrolled in this and a second, simi- Germany; Phillip M. Green, Borgess Research Institute, Kalamazoo, MI;
lar, donepezil trial.32 It would be consistent with the John Grubbs, Four Rivers Clinical Research, Paducah, KY; Danilo Antonio
Guzman, SCOHS, Elizabeth Bruyere Center, Ottawa, Ontario, Canada;
findings in probable and possible VaD, and also with Hisham M. Hafez, The Institute for Clinical Research, Nashua, NH; Robert
the known benefits in AD, to see similar improve- Hausner, The Institute on Aging Research Center, San Francisco, CA; Roy

August (2 of 2) 2003 NEUROLOGY 61 485

Jones, Research Institute for the Care of the Elderly, St. Martin’s Hospital, 10. Perry EK, Tomlinson BE, Blessed G, Bergmann K, Gibson PH, Perry
Bath, UK; Jeffrey Karrasch, Peninsula Specialist Center, Kippa Ring, RH. Correlation of cholinergic abnormalities with senile plaques and
Queensland, Australia; Andrew Kertesz, St. Joseph’s Health Center, Lon- mental test scores in senile dementia. BMJ 1978;2:1457–1459.
don, Ontario, Canada; John Knesevich, Neuroscience Options, Inc., Am- 11. Erkinjuntti T, Kurz A, Gauthier S, Bullock R, Lilienfeld S, Damaraju
herst, NH; Guy Lacombe, Institut Universitaire de Geriatrie de Sherbrooke, CV. Efficacy of galantamine in probable vascular dementia and Alzhei-
Sherbrooke, Quebec, Canada; Peter Londborg, Seattle Clinical Research mer’s disease combined with cerebrovascular disease: a randomised
Center, Inc., Seattle, WA; Helmi Lutsep, OHSU-Oregon Stroke Center, trial. Lancet 2002;359:1283–1290.
Portland, OR; Abe Marcadis, Maria Jurado, Clinical Studies, Boynton 12. Rogers SL, Farlow MR, Doody RS, Mohs R, Friedhoff LT. A 24-week,
Beach, FL; David Ira Margolin, Margolin Brain Institute, Fresno, CA; Fred double-blind, placebo-controlled trial of donepezil in patients with Alz-
Marshall, University of Rochester, Rochester, NY; Peter N. McCracken, heimer’s disease. Donepezil Study Group. Neurology 1998;50:136 –145.
Glenrose Rehabilitation Hospital, Edmonton, Canada; John Merory, Austin 13. Burns A, Rossor M, Hecker J, et al. The effects of donepezil in Alzhei-
Repatriation Hospital, Heidelberg West Vic, Australia; Barnett Meyers, NY mer’s disease—results from a multinational trial. Dement Geriatr Cogn
Hospital–Westchester, White Plains, NY; Jacobo Mintzer, Medical Univer- Disord 1999;10:237–244.
sity of South Carolina, Charleston, SC; David Nyenhuis, Stroke Research 14. Winblad B, Engedal K, Soininen H, et al. A 1-year, randomized,
Center, Rush Presbyterian St. Luke’s Medical Center, Chicago, IL; Barry placebo-controlled study of donepezil in patients with mild to moderate
Passini, Carolina Neurologic Clinic, PA, Charlotte, NC; Andrew P. Pass- AD. Neurology 2001;57:489 – 495.
more, Queen’s University, Belfast, Northern Ireland; Richard Pellegrino, 15. Mohs RC, Doody RS, Morris JC, et al. A 1-year, placebo-controlled,
Central Arkansas Research, Hot Springs, AR; Merrell Reiss, Beta Research, preservation of function survival study of donepezil in AD patients.
Inc., Westmont, IL; Peter Ripley, Clinical Studies, Cape Cod, South Yarm- Neurology 2001;57:481– 488.
outh, MA; Alain Robillard, Recherche Clinique de Neurologie, Hôpital 16. Mendez MF, Younesi FL, Perryman KM. Use of donepezil for vascular
Maisonneuve-Rosemont, Montreal, Quebec, Canada; Martin Rossor, The dementia: preliminary clinical experience. J Neuropsychiatry Clin Neu-
National Hospital for Neurology and Neurosurgery, London, UK; Carl H. rosci 1999;11:268 –270.
Sadowsky, Premiere Research Institute, West Palm Beach, FL; Douglas W. 17. Roman GC, Tatemichi TK, Erkinjuntti T, et al. Vascular dementia:
Scharre, OH State University, Columbus, OH; Alan Schneider, Pharmacol- diagnostic criteria for research studies. Report of the NINDS-AIREN
ogy Research Institute, Northridge, CA; Stuart R. Stark, The Neurology International Workshop. Neurology 1993;43:250 –260.
Center, Alexandria, VA; Elisabeth Steinhagen-Thiessen, Evangelisches 18. Folstein MF, Folstein FE, McHugh PR. “Mini-mental state”. A practical
Geriatrie-Zentrum, Berlin, Germany; Steven Targum, Clinical Studies, method for grading the cognitive state of patients for the clinician.
Philadelphia, PA; Kim Taubman, Robert Helme, National Aging Research J Psychiatr Res 1975;12:189 –198.
Institute, Parkville, Victoria, Australia; Larry Tune, Wesley Woods Geriat- 19. World Medical Association Declaration of Helsinki. Recommendations
ric Hospital, Emory University, Atlanta, GA; Mahmood A. Usman, The guiding physicians in biomedical research involving human subjects.
Alzheimer Center of Pittsburgh, Pittsburgh, PA; David G. Wilkinson, Thorn JAMA 1997;277:925–926.
Hill Research, Hants, UK. 20. Rosen WG, Mohs RC, Davis KL. A new rating scale for Alzheimer’s
disease. Am J Psychiatry 1984;141:1356 –1364.
21. Knopman DS, Knapp MJ, Gracon SI, Davis CS. The Clinician
Disclosure Interview-Based Impression (CIBI). A clinician’s global rating scale in
Alzheimer’s disease. Neurology 1994;44:2315–2321.
Drs. Wilkinson, Doody, Mintzer, and Kertesz have received consult- 22. Schneider LS, Olin JT, Doody RS, et al. Validity and reliability of the
ing fees and honoraria from Eisai Inc. for educational presentations. Alzheimer’s Disease Cooperative Study—Clinical Global Impression of
Dr. Doody has additionally received grants from Eisai Inc., with Change. The Alzheimer’s Disease Cooperative Study. Alzheimer Dis
personal compensation and grants together totaling more than Assoc Disord 1997;11:S22–S32.
$10,000/year. Dr. Pratt is a full-time employee of Eisai Inc. 23. Berg L. Clinical Dementia Rating (CDR). Psychopharmacol Bull 1988;
24:637– 639.
24. Gauthier S, Ferris S. Outcome measures for probable vascular demen-
References tia and Alzheimer’s disease with cerebrovascular disease. Int J Clin
Pract suppl 2001;120:29 –39.
1. Lobo A, Launer LJ, Fratiglioni L, et al. Prevalence of dementia and 25. Rogers SL, Doody RS, Mohs RC, Friedhoff LT. Donepezil improves
major subtypes in Europe: a collaborative study of population-based cognition and global function in Alzheimer disease: a 15-week, double-
cohorts. Neurologic Diseases in the Elderly Research Group. Neurology blind, placebo-controlled study. Donepezil Study Group. Arch Intern
2000;54:S4 –S9. Med 1998;158:1021–1031.
2. Ebly EM, Parhad IM, Hogan DB, Fung TS. Prevalence and types of 26. Hachinski VC, Iliff LD, Zilhka E, et al. Cerebral blood flow in dementia.
dementia in the very old: results from the Canadian Study of Health Arch Neurol 1975;32:632– 637.
and Aging. Neurology 1994;44:1593–1600. 27. Pratt RD. Patient populations in clinical trials of the efficacy and toler-
3. Ott A, Breteler MM, van Harskamp F, et al. Prevalence of Alzheimer’s ability of donepezil in patients with vascular dementia. J Neurol Sci
disease and vascular dementia: association with education. The Rotter- 2002;203–204:57– 65.
dam Study. BMJ 1995;310:970 –973. 28. Sachdev PS, Brodaty H, Looi JC. Vascular dementia: diagnosis, man-
4. Desmond DW, Moroney JT, Sano M, Stern Y. Mortality in patients with agement and possible prevention. Med J Aust 1999;170:81– 85.
dementia after ischemic stroke. Neurology 2002;59:537–543. 29. Kittner B, De Deyn PP, Erkinjuntti T. Investigating the natural course
5. Doody RS, Stevens JC, Beck C, et al. Practice parameter: management and treatment of vascular dementia and Alzheimer’s disease. Parallel
of dementia (an evidence-based review). Report of the Quality Stan- study populations in two randomized, placebo-controlled trials. Ann NY
dards Subcommittee of the American Academy of Neurology. Neurology Acad Sci 2000;903:535–541.
2001;56:1154 –1166. 30. Orgogozo JM, Rigaud AS, Stoffler A, Mobius HJ, Forette F. Efficacy and
6. Togashi H, Kimura S, Matsumoto M, Yoshioka M, Minami M, Saito H. safety of memantine in patients with mild to moderate vascular demen-
Cholinergic changes in the hippocampus of stroke-prone spontaneously tia: a randomized, placebo-controlled trial (MMM 300). Stroke 2002;33:
hypertensive rats. Stroke 1996;27:520 –525. 1834 –1839.
7. Kimura S, Saito H, Minami M, et al. Pathogenesis of vascular dementia 31. Padovani A, Di Piero V, Bragoni M, Iacoboni M, Gualdi GF, Lenzi GL.
in stroke-prone spontaneously hypertensive rats. Toxicology 2000;153: Patterns of neuropsychological impairment in mild dementia: a compar-
167–178. ison between Alzheimer’s disease and multi-infarct dementia. Acta
8. Tohgi H, Abe T, Kimura M, Saheki M, Takahashi S. Cerebrospinal fluid Neurol Scand 1995;92:433– 442.
acetylcholine and choline in vascular dementia of Binswanger and mul- 32. Perdomo C, Pratt R. A comparison of the effects of donepezil in patients
tiple small infarct types as compared with Alzheimer-type dementia. with probable and possible vascular dementia: a combined analysis of
J Neural Transm 1996;103:1211–1220. two 24-week, randomized, double-blind, placebo-controlled trials. Neu-
9. Swartz RH, Black SE. How common is vascular compromise of cholin- rology 2002;58(suppl 3):PO5.115.
ergic white matter pathways in a memory clinic sample? J Neurol Sci 33. Pratt RD, Perdomo CA, Surick IW, Ieni JR. Donepezil: tolerability and
2002;203–204:281. safety in Alzheimer’s disease. Int J Clin Pract 2002;56:710 –717.

486 NEUROLOGY 61 August (2 of 2) 2003

 Donepezil in vascular dementia: A randomized, placebo-controlled study
D. Wilkinson, R. Doody, R. Helme, et al.
Neurology 2003;61;479-486
DOI 10.1212/01.WNL.0000078943.50032.FC

This information is current as of August 25, 2003

Updated Information & including high resolution figures, can be found at:
Services http://www.neurology.org/content/61/4/479.full.html

References This article cites 26 articles, 11 of which you can access for free at:
http://www.neurology.org/content/61/4/479.full.html##ref-list-1
Citations This article has been cited by 17 HighWire-hosted articles:
http://www.neurology.org/content/61/4/479.full.html##otherarticles
Subspecialty Collections This article, along with others on similar topics, appears in the
following collection(s):
All Clinical trials
http://www.neurology.org//cgi/collection/all_clinical_trials
All Cognitive Disorders/Dementia
http://www.neurology.org//cgi/collection/all_cognitive_disorders_deme
ntia
Clinical trials Randomized controlled (CONSORT agreement)
http://www.neurology.org//cgi/collection/clinical_trials_randomized_c
ontrolled_consort_agreement
Vascular dementia
http://www.neurology.org//cgi/collection/vascular_dementia
Permissions & Licensing Information about reproducing this article in parts (figures,tables) or in
its entirety can be found online at:
http://www.neurology.org/misc/about.xhtml#permissions
Reprints Information about ordering reprints can be found online:
http://www.neurology.org/misc/addir.xhtml#reprintsus

Neurology ® is the official journal of the American Academy of Neurology. Published continuously since
1951, it is now a weekly with 48 issues per year. Copyright . All rights reserved. Print ISSN: 0028-3878.
Online ISSN: 1526-632X.