Systematic Review and Meta-Analysis Medicine ®

OPEN

Apnea–hypopnea index decreased significantly

after nasal surgery for obstructive sleep apnea
A meta-analysis
Jun Wu, MD, Guoqiang Zhao, MD, Yunchuan Li, MD, Hongrui Zang, MD, Tong Wang, MD,
∗
Dongbo Wang, MD, Demin Han, MD, PhD

Abstract
Background: Nasal surgeries have been applied to obstructive sleep apnea (OSA) patients with nasal obstruction for decades.
However, the efficiency of nasal surgery in improving OSA remains controversial.
The aim of this study was to identify whether isolated nasal surgery can improve apnea–hypopnea index (AHI).
Methods: Computerized searches were performed in MEDLINE, Web of Science, Cochrane Library, and Scopus from January 1,
2000 to April 30, 2016. A total of 18 articles and 587 participants were included. There were 1 randomized controlled trials, 2
nonrandomized trials, 11 prospective studies, and 4 retrospective studies. Data regarding study design (prospective/retrospective
clinical trial, randomized, and controlled), population size, participant characteristics (age, gender, and body mass index), surgical
intervention, and outcomes (AHI, Epworth sleep scale [ESS]) was collected.
Results: Statistically significant improvement in AHI (subgroup 1: weighted mean difference [WMD] [95%confidence interval (CI)],
4.17 [ 7.62, 0.73]; subgroup 2: WMD [95%CI], 4.19 [ 7.51, 0.88]; overall: WMD [95%CI], 4.15 [ 6.48, 1.82]) and ESS
(subgroup 1: WMD [95%CI], 2.14 [ 3.08, 1.19]; subgroup 2: WMD [95%CI], 4.70 [ 5.95, 3.44]; overall: WMD [95%CI],
4.08 [ 5.27, 2.88]) was revealed.
Conclusion: Both AHI and ESS improved significantly after isolated nasal surgery, but the improvement of AHI is slightly significant.
Future randomized controlled trials are needed to confirm the long-term benefits of nasal surgery on OSA.
Abbreviations: AHI = apnea–hypopnea index, BMI = body mass index, CI = confidence interval, ESS = Epworth sleep scale, NR
= nasal resistance, SD = standard deviation, WMD = weighted mean difference.
Keywords: apnea–hypopnea index, Epworth sleep scale, meta-analysis, nasal surgery, obstructive sleep apnea

1. Introduction The relationship between nasal resistance (NR) and sleep

disordered breathing has aroused the attention of researchers since
Obstructive sleep apnea (OSA) is characterized by recurrently
19 century.[4,5] Nasal structure malformation has been related to
complete or partial obstruction of the upper airway during sleep,
upper airway collapses in some patients with OSA as one of the
resulting in major cardiovascular and neurocognitive sequelae if
principal causes.[6–8] And epidemiological studies indicated that
not treated.[1,2] The obstruction could occur at multiple levels of
adults with nasal obstruction are more likely to have habitual
the upper airway, such as nasal cavity, pharyngeal cavity, and
snoring.[9,10] In addition, acute nasal obstruction in healthy adults
retroglottal region, among which nasal cavity accounts for 1/2 to
such as acute rhinitis can cause sleep disordered breathing.[11,12]
2/3 of the general airway resistance.[3]
Vice versa, the incidence of sleep disordered breathing in patients
with nasal septum deviation is far higher than that of normal
Editor: Giuseppe Insalaco. people.[13] Meanwhile, OSA patients have been proved to suffer
The authors have no funding and conflicts of interest to disclose. from a higher probability of nasal obstruction.[14,15]
Department of Otorhinolaryngology Head and Neck Surgery, Beijing Tongren
As one primary treatment for nasal obstruction, nasal
Hospital, Capital Medical University, and Key Laboratory of Otolaryngology Head surgeries, including septoplasty or/and functional sinus surgery
and Neck Surgery, Ministry of Education, Beijing, P.R. China. or/and turbinate displacement, as well as nasal cavity ventilation
∗
Correspondence: Demin Han, Department of Otorhinolaryngology Head and expansion techniques[16] have been applied to OSA patients for
Neck Surgery, Beijing Tongren Hospital, Capital Medical University, and Key decades. However, the efficiency of nasal surgery in improving
Laboratory of Otolaryngology Head and Neck Surgery, Ministry of Education, OSA remains controversial. Two meta-analyses on this topic
Dong Jiao Min Xiang, Eastern District, Beijing 100730, P.R. China
(e-mail: handemin_ent@163.com).
indicated that nasal surgery can improve Epworth sleepiness scale
(ESS) score, which is an indicator of daytime sleepiness.
Copyright © 2017 the Author(s). Published by Wolters Kluwer Health, Inc.
This is an open access article distributed under the Creative Commons However, apnea–hypopnea index (AHI), which is regarded as
Attribution License 4.0 (CCBY), which permits unrestricted use, distribution, and a key factor evaluating OSA severity and treatment effect, did not
reproduction in any medium, provided the original work is properly cited. reduce significantly.[17,18]
Medicine (2017) 96:5(e6008) Nevertheless, some newly published articles on this topic
Received: 9 December 2016 / Received in final form: 3 January 2017 / showed nasal surgery can decrease AHI in recent 3 years.[19–22]
Accepted: 5 January 2017 Here, we performed this meta-analysis of studies reporting data
http://dx.doi.org/10.1097/MD.0000000000006008 of treating OSA with nasal surgery.

1
Wu et al. Medicine (2017) 96:5 Medicine

2. Materials and methods We divided the studies into 2 subgroups (subgroup 1, subgroup
2) according to SD of change while conducting the meta-analysis.
2.1. Information source and search strategy
Forest plots were graphically inspected, and Cochran Q test (P <
Computerized and manual searches of 4 databases (MEDLINE, 0.1, a significant difference between studies) and I2 statistic (low:
Web of Science, Cochrane Library, and Scopus) were performed 25%, moderate: 50%, and high: 75%) were applied for
from January 1, 2000 to April 30, 2016 to identify all data of determining heterogeneity. We used fixed effects model for
relevance. The following keywords and MeSH terms were used: pooling effects if no or low heterogeneity of treatment effects was
“nasal surgery/sleep disorder,” “nasal surgery/sleep apnea,” found and a random effects model if moderate or high
“nasal surgery/snoring,” “nose/sleep disorder,” “nose/sleep heterogeneity was found.
apnea,” “nose/snoring,” “nasal obstruction/surgery,” “rhino-
plasty/sleep disorder,” “septorhinoplasty/sleep disorder” and 2.5. Ethical approval
“turbinectomy/sleep disorder,” “rhinoplasty/sleep apnea,” “sep-
torhinoplasty/sleep apnea” and “turbinectomy/sleep apnea,” This is a meta-analysis about literatures; therefore, ethical
“rhinoplasty/snoring,” “septorhinoplasty/snoring,” and “turbi- approval was not necessary.
nectomy/snoring.” The cited references in the relevant articles
were also reviewed to identify additional published work. Two
reviewers conducted the searches independently, and duplicates 3. Results
were excluded. A 3rd reviewer would resolve disagreements by
discussion. 3.1. Study characteristics
Eighteen studies dealing with nasal surgery for OSA met our
2.2. Eligibility criteria and study selection inclusion criteria, and they included 587 participants.[19–22,24–37]
The article selection flow chart is shown in Fig. 1. Overall, 1
Articles were screened by titles and abstracts then reviewed if full randomized controlled trial, 2 nonrandomized controlled trials,
texts were eligible. Inclusion criteria for the studies consisted of: 11 prospective studies, and 4 retrospective studies were included.
patients with OSA; isolated nasal surgery applied, such as The characteristics of the studies are shown in Table 1. The types
septorhinoplasty, rhinoplasty, turbinectomy, or sinus surgery; of nasal surgery performed in the studies are as the following:
both post- and preoperative quantitative outcomes data septorhinoplasty/septoplasty/submucosal septal resection/septal
evaluating AHI/ESS; and articles published only in English. surgery in 15 studies,[19,21,22,24,26–28,30–37] inferior turbinectomy/
Studies were excluded for the following criteria: age < 18 years partial inferior turbinectomy/submucosal turbinectomy/concha
old; case reports, letters to the editor, and review articles; and cauterization/turbinate surgery/radiofrequency ablation of infe-
additional level surgery described (tonsillectomy, uvulopalato- rior turbinate/turbinoplasty in 14 studies,[19,21,22,24,26–28,30–36]
pharyngoplasty, maxillomandibular advancement, etc.). and endoscopic sinus surgery in 7 studies.[19–21,26,27,29,32] The
treatment protocols are shown in Table 1.
2.3. Data extraction Regarding the sleep apnea severity assessment, polysomnog-
Data regarding study design (prospective/retrospective clinical raphy (PSG) was performed in 17 studies,[19–21,24–37] watch-PAT
trial, randomized, and controlled), population size, participant in 1 study.[22] The most commonly used subjective assessment is
characteristics (age, gender, and body mass index [BMI]), ESS, which is performed in 16 studies.[20–22,24–30,32–37]
surgical intervention, and outcomes (AHI, ESS) were collected.
Two authors independently checked the data to ensure accuracy. 3.2. Patients characteristics
Disagreements were resolved by discussion with a 3rd author.
The age of participants ranged from 20 to 70 years, with an
average age of 44 years. The overall proportion of male patients
2.4. Data analysis
was 90.5%, ranged 63.0% to 100%. Reported baseline BMI, but
The statistical analysis was performed with IBM SPSS Statistics only 8 studies (44.4%) reported postoperative BMI. The detailed
software version 18.0 (Chicago, IL) and the Cochrane Collab- data are shown in Table 2.
oration’s Review Manager (REVMAN) Software version 5.2. We
calculated the means, standard deviations (SDs), and 95%
3.3. Treatment outcomes
confidence intervals (CIs). The weighted mean differences
(WMDs) of AHI and ESS were obtained according to the The study by Victores and Takashima[25] was not included in the
differences of post- and preoperative values from the original analysis of AHI change because it did not provide the SD of
articles. A correlation coefficient between intervention effect and outcome parameter. Thus, 17 studies were analyzed. Heteroge-
baseline AHI/ESS of a study was calculated as described in neity test results of AHI (subgroup 1: I2 = 66.7%, P = 0.010;
Cochrane Handbook for Systematic Reviews of Interventions,[23] subgroup 2: I2 = 70.5%, P = 0.000; overall: I2 = 67.4%, P =
in which SD of change was provided. CorrE = (SD2_E baseline + 0.000) scores indicate heterogeneity in the studies. Random
SD2E final SD2_E change)/(2  SDE baseline  SDE final). And subse- effects model therefore was used to conduct the meta-analysis.
quently, the average correlation coefficient between intervention According to the meta-analysis results, statistically significant
effect and baseline AHI or ESS was applied to impute the SD of improvement in AHI (subgroup 1: WMD [95%CI], 4.17
change for AHI or ESS in studies of which the SDs of change were [ 7.62, 0.73]; subgroup 2: WMD [95%CI], 4.19 [ 7.51,
not provided. For the current meta-analysis, the average 0.88]; overall: WMD [95%CI], 4.15 [ 6.48, 1.82]), Fig. 2.
correlation coefficient between intervention effect and baseline Five studies were not included in the ESS score analysis since
AHI was 0.667 among the 6 studies with SD of change data, they did not obtain detailed post- or preoperative mean ± SD of
while the average correlation coefficient or ESS was 0.638 in the 3 ESS scores.[20,21,28,33,35] Thus, ESS scores were only analyzed in
studies with related information. the other 11 studies.[22,24–27,29,30,32,34,36,37]

2
Wu et al. Medicine (2017) 96:5 www.md-journal.com

Figure 1. Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow chart.

Heterogeneity test of ESS showed heterogeneity in the studies included studies published from 1999 to 2009, and concluded
(subgroup 1: I2 = 60.6%, P = 0.079; subgroup 2: I2 = 86.1%, P = that nasal surgery could effectively improve daytime sleepiness
0.000; overall: I2 = 91.1%, P = 0.000). ESS decreased significant- (evaluated by ESS) while not the AHI.[18] A more recent
ly, indicating improved day time sleepiness in these patients systematic review and meta-analysis in 2015 by Ishii et al,[17] in
(subgroup 1: WMD [95%CI], 2.14 [ 3.08, 1.19]; subgroup defect of endoscopic sinus surgery, revealed a similar result that
2: WMD [95%CI], 4.70 [ 5.95, 3.44]; overall: WMD [95% ESS improved significantly, but AHI did not. However, Park et al
CI], 4.08 [ 5.27, 2.88]), Fig. 3. and some other researchers[19,20,22] reported AHI improvement
in selected patients recently.
We attributed the varying conclusions with previous studies to:
4. Discussion
we included more recent studies accessing sleep apnea severity
Although different interventions aimed at multiple airway levels with a similar criterion; we adopted a more reliable analyzing
have been applied for airway collapse, isolated nasal surgery is measurement, which subgrouped and analyzed the collected
rarely considered for the specific management of OSA. However, data. As a measure of surgical intervention, nasal surgery first
the reported effects were inconsistent. The present study showed opens middle nasal meatus and sinuses, therefore helping in
that AHI could be significantly improved postoperatively by draining of nasal and sinus cavity secretion and maintaining their
isolated nasal surgery. Our analysis used a subgroup analysis, the normal physiological function; second decreased upper airway
index before and after the operation according to the reliability of resistance reduces episodes of mouth breathing, negative pressure
the group, and once again combined to get the same positive of the nasopharynx and improves the collapsibility of the
results, which provided a more validated result. oropharyngeal cavity. It is reported that nasal surgeries improved
The present study included studies evaluating the efficiency of the compliances of continues positive pressure therapy.[38,39] In
isolated nasal surgery on OSA from 2002 to 2016 and draws a addition, improvement of psychological symptoms, such as
conclusion that both ESS and AHI improved significantly. Our depression as well as daytime sleepiness, were reported even
results differed from previous meta-analysis of the efficiency of without substantial AHI change.[30,32–34,36,37] Because of the
nasal surgery for OSA. In 2011, a systematic review by Li et al subjective clinical effect on patients, correcting nasal obstruction

3
 Table 1
Study characteristics.
Study design ∗
Author and year (sample size) Assessment Treatment protocol Outcome measures Time interval Major conclusion
[37]
Verse et al, 2002 Prospective (26) PSG, RM, ESS 7 SRP, 13 SP, 4 SP with bilateral PSS, 1 AHI, ODI, NR, ESS 12.7 (11.9) Isolated nasal surgery significantly decreased NR and ESS, thus
NTS, 1 NVS [3–50] (month) improving sleep quality, but not AHI.
[31]
Kim et al, 2004 Retrospective (21) PSG 10 SP, 11 SP with ITE AHI, RDI, AI, OSI,DS 1 (month) RDI, AI, and OSI improved significantly after surgery; nasal surgery
alone may benefit some SDB patients.
[36]
Nakata et al, 2005 Prospective (12) PSG, CPAP titration, 10 ITE with SMR, 2 ITE AHI, CPAP level, NR, ESS NA Nasal surgery decreased NR, LSaO2, and ESS significantly,
RM, ESS facilitating CPAP, but not AHI.
[35]
Virkkula et al, 2006 Prospective (40) PSG, RM, ESS, CM 8 SP, 2 SP with pITE, 2 SP with SRP AHI, ODI, TST, NR, DS, SII, ESS 113 [63–176] NR decreased after nasal surgery, but the snoring did not decrease,
Wu et al. Medicine (2017) 96:5

(day) and ESS did not change significantly.

[32]
Nakata, et al, 2008 Prospective (49) PSG, RM, ESS 38 SMR with ITE, 4 SMR with ITE and FESS, AHI, AI, ArI, LSaO2, NR, ESS, NA Nasal surgery decreased NR and ESS but did not change AHI and
6 ITE, 1 ITE with FESS ODT, TST may be useful to improve sleep.
Koutsourelakis Randomized PSG, RM, ESS 9 SMR, 18 SMR with SMT/Sham surgery AHI, NR, ESS, NOBE, MSaO2 [3–4] (month) Nasal surgery rarely treats OSA effectively. Baseline nasal breathing
et al,[34] 2008 controlled (27/22) epochs can predict surgery outcome.
Li et al,[33] 2008 Prospective (51) PSG, VAS, SOS, SMP: SMR, pITE AHI, VAS, SOS, SBPS, ESS, SF- 3 (month) Snoring and daytime sleepiness improve greater than other generic
SBPS, ESS 36 health status after nasal surgery.
Li et al,[30] 2009 Nonrandomized PSG, RM, ESS, SOS, SMP: SMR, pITE/Medication AHI, NR, ESS, SOS, SBPS 3 (month) Changes of AHI significantly differed between FTP I/II (LOW) and III/IV
controlled (44/22) SBPS, AA (HIGH) groups; surgical success was significantly higher in the
LOW group.
Tosun et al,[29] 2009 Prospective (27) PSG, RM, ESS, VAS, FESS with NPR AHI, NR, ESS, snoring score 3 (month) FESS with polypectomy significantly improved snoring and daytime
AR VAS, LSaO2, MSaO2, Ar sleepiness in nasal polyposis patients with chronic nasal
obstruction but not AHI.
Bican et al,[28] 2010 Prospective (20) PSG, ESS, MR 12 SP/NVR with spreader grafts and 1-sided AHI, LSaO2, ESS, CPAP level, 117 [80–189] AHI, SaO2, and ESS improved significantly after surgery; MR
CC; 8 SP/NVR with spreader grafts and IVVV, EVA (day) indicates increased nasal volume postoperatively.
double-sided CC
Choi et al,[27] 2011 Prospective (22) PSG, ESS 4 FESS with SS and TS; 1 FESS with TS; 17 AHI, SE, AI, ArI, LSaO2, TST, 3 (month) Nasal surgery alone is effective to improve snoring in OSAS with
SS with TS TIB, ESS nasal obstruction.

4
Sufioglu et al,[26] 2011 Prospective (28) PSG, ESS, VAS 3 SP; 2 SRP; 18 SP with RFA of IT; 4 FESS AHI, ESS, VAS, CPAP pressure 3 (month) Nasal surgery significantly improved nasal passage, snoring
with SP and RFA of IT; 4 RFA of IT† frequency, apnea and daytime sleepiness, but not AHI. As for
CPAP titration pressures, it decreased after isolated nasal surgery,
but not significantly.
Victores et al,[25] 2012 Retrospective (24) PSG, DISE, ESS nasal surgery AHI, ESS 3 (month) AHI and the pattern of upper airway obstruction did not change
significantly after nasal surgery.
Moxness et al,[24] 2014 Prospective (59) PSG, ESS, VAS 33 SP; 26 SP with RFA/ITE AHI, ODI, ESS, VAS sleep 3 (month) Nasal surgery seemed more beneficial to patients receiving combined
surgery of the inferior turbinates and the nasal septum than
patients receiving septoplasty alone.
Yalamanchali Retrospective (56) PSG 37 SP with bilateral SMT and pan-FESS; 19 AHI, LSaO2, MSaO2 [40.3–32.5] (day) Combined nasal and sinus surgery may slightly improve AHI in
et al,[21] 2014 SP with bilateral SMT and limited-sinus patients with moderate or severe OSA, but it does not cure OSA
surgery; 7/56 with concurrent NVR, 30/56 or have a significant clinical impact.
with concurrent NPE
Park et al,[22] 2014 Prospective (25) watch-PAT, AR, DISE, 25 SP with TP AHI, RDI, VST, REMST, ESS, 2 (month) Nasal surgery improved nasal airway patency and reduced OSA
ESS, MCA, VAS, autoPAP level severity in 56% and led to statistically significant decreases in
VAS, autoPAP AHI and ESS.
Shuaib et al,[20] 2015 Retrospective (26) PSG, NOSE, ESS EFRP AHI, LSaO2, ESS, NOBE 132 (83) (day) Functional rhinoplasty may significantly improve the severity of OSA
for patients with nasal obstruction, especially for patients with a
body mass index less than 30.
Xiao et al,[19] 2016 Nonrandomized PSG, PSQI, SCL-90, FESS‡ AHI, LSaO2, VAS, PSQI, SCL- 3 (month) The correction of nasal patency through nasal surgery decreases the
controlled (30/30) VAS, GSI 90, GSI severity of OSA in patients with nasal obstruction.

Statistics: data are presented as mean and (standard deviation). AA = anatomic assessment stage, AHI = apnea–hypopnea index, AI = apnea index, AR = acoustic rhinometry, Ar = arousal, ArI = arousal index, autoPAP = auto positive airway pressure, CC = concha cauterization, CM =
cephalometry, CPAP = continuous positive airway pressure, DISE = drug-induced sleep endoscopy, DS = duration of snoring, EFRP = endonasal functional rhinoplasty, ESS = Epworth sleepiness scale, EVA = external valve angle, FESS = functional endoscopic sinus surgery, GSI = Global
Symptom Index, IT = inferior turbinate, ITE = inferior turbinectomy, IVVV = internal valve volume value, LSaO2 = lowest arterial saturated oxygenation, MCA = mean cross-secional area, MR = magnetic resonance, MSaO2,mean arterial saturated oxygenation, NOBE = nasal and oral/oronasal
breathing, NPE = nasal polypectomy, NR = nasal resistance, NTS = nasal tip surgery, NVR = nasal valve reconstruction, NVS = nasal valves surgery, ODI = oxygen desaturation index, ODT = oxygen desaturation time with SpO2 < 90%, OSA = obstructive sleep apnea, OSI = oxygen
saturation index, pITE = partial inferior turbinectomy, PSQI = Pittsburgh Sleep Quality Index, PSS = paranasal sinus surgery, RDI = respiratory disturbance index, REMST = rapid eye move sleep time, RFA = radiofrequency ablation, RM = rhinomanometry, SBPS = sleep/bed partner survey,
SCL-90 = Symptom Check List 90, SE = sleep efficiency, SF-36 = 36-Item Short-Form Health Survey, SII = snoring intensity index, SMP = septomeatoplasty, SMR = submucosal septal resection, SMT = submucosal turbinectomy, SOS = snore outcomes survey, SP = septoplasty, SS =
septal surgery, TIB = time in bed, TP = turbinoplasty, TS = turbinate surgery, TST = total sleep time, VAS = visual analogue scale, VST = valid sleep time.
∗
Time between operation and controlled PSG.
†
Including 3 cases without controlled PSG.
Medicine

‡
Include septoplasty with 3 high-tension line resections, medial displacement and fixation of the middle turbinate, symmetrical bilateral ethmoidectomy and maxillary sinus surgery, and lateral displacement and fixation of the inferior turbinate.
Wu et al. Medicine (2017) 96:5 www.md-journal.com

Table 2
Patients characteristics.
Body mass Apnea–hypopnea Epworth
Sample Gender index, kg/cm2 index sleepiness scale
Author and year size (male %) Age, year BMI pre BMI post AHI pre AHI post ESS pre ESS post
Verse et al,[37] 2002 26 96.15 52.5 (8.4) 29.16 (4.18) 29.2 (4.28) 31.57 (25.6) 28.93 (24.73) 11.87 (4.7) 7.73 (4.96)
Kim et al,[31] 2004 21 71.43 39 28.33 (2.53) 39 (14.03) 29.14 (14.42)
Nakata et al,[36] 2005 12 100.00 54.2 (9.2) 27 (3.9) 26 (2.8) 55.9 (18.2) 47.8 (20.4) 11.7 (4.1) 3.3 (1.3)
Virkkula et al,[35] 2006 40 100.00 44.2 (9.5) 27.9 (3.4) 13.6 (15.8) 14.9 (19.3) 6.3 (3.7)
Nakata, et al,[32] 2008 49 100.00 46.1 (12.3) 26.4 (3.8) 26.2 (3.4) 44.6 (22.5) 42.5 (22) 10.6 (4.1) 4.5 (2.6)
Koutsourelakis et al,[34] 2008 27 62.96 39 (7.5) 30.4 (3.2) 31 (3.6) 31.5 (16.7) 31.5 (18.2) 13.4 (2.9) 11.7 (3.4)
Li et al,[33]a 2008 51 98.04 39 (10) 26 (3.5) 37.4 (28.9) 38.1 (32.7)
Li et al,[30]b 2009 44 95.45 38.3 (9.9) 26.2 (3.5) 36.4 (29.1) 37.5 (31.6) 10.6 (3.9) 7.6 (4.5)
Tosun et al,[29]c 2009 27 81.48 40.37 (14.48) 23.87 (1.31) 6.7 (11.8) 5.5 (10.8) 9.44 (4.08) 4.15 (3.16)
Bican et al,[28] 2010 20 100.00 47.5 31.2 31 43.1 (27.1) 24.6 (22.2)
Choi et al,[27] 2011 22 100.00 41.3 (10.9) 25.5 (2.9) 25.2 (2.8) 28.9 (20.4) 26.1 (21.9) 8.8 (3.3) 6.3 (3.3)
∗ ∗
Sufioglu et al,[26] 2011 28 83.87 53 (9.6) 30.3 (4.1) 32.5 (22.6) 32.4 (24.6) 9.3 (5.1) 5.9 (3.9)
Victores et al,[25] 2012 24 79.17 44.8 (13.9) 30.3 (5.9) 27.3 (18.1) 24.4 12.3 (6.2) 6.6 (4.2)
Moxness et al,[24]d 2014 59 91.53 46.53 28.1 (3.23) 28.28 (2.95) 18.15 (13.71) 16.6 (12.9) 10.74 (3.67) 8.94 (3.84)
Yalamanchali et al,[21] 2014 56 85.71 43.6 (11.3) 33.5 (22) 29.4 (20.8)
Park et al,[22] 2014 25 92.00 47.4 21.3 23.9 (14.9) 12.2 (6.4) 9.68 (4.7) 5.84 (2.1)
Shuaib et al,[20]e 2015 26 65.38 42.7 (13.6) 27.6 24.7 (18.8) 16 (16.1) 11.5 7.5
Xiao et al,[19]f 2016 30 100.00 45.5 (11.37) 27.62 (3.69) 26.63 (3.39) 49.67 (19.49) 43.07 (21.86)
Statistics: data are presented as mean and (standard deviation). age > 60, BMI > 40 excluded. age > 60, BMI > 33 excluded. cwomen BMI > 27.3 or men > 27.8 excluded. dBMI ≥ 35 excluded. eBMI ≥ 40
a b

excluded. fage > 70 excluded. AHI = apnea–hypopnea index, BMI = body mass index, ESS = Epworth sleepiness scale, LSaO2 = lowest oxygen saturation.
∗
Characteristics of 31 cases.

is still considered an important measure of treating OSA. But the perform the sensitivity analysis by removing each study to
long-term improvement of both objective and subjective indexes investigate its effect on the summarized effect and heterogeneity.
needs to be confirmed by a more long-term observational This is because while collecting and organizing data we
experiment. performed imputation of SD of change in some studies, and
The present study compared the difference between post- and we subgrouped and analyzed the included studies basing on
preoperative values to evaluate the treatment effect. We did not whether they provided original individual data or SD of change

Figure 2. Subgroup 1 includes studies providing AHI SD of outcome parameter; subgroup 2 includes studies not providing AHI SD of outcome parameter. AHI =
apnea–hypopnea index, SD = standard deviation.

5
Wu et al. Medicine (2017) 96:5 Medicine

Figure 3. Subgroup 1 includes studies providing ESS SD of outcome parameter; subgroup 2 includes studies not providing ESS SD of outcome parameter. ESS =
Epworth sleepiness scale, SD = standard deviation.

or we estimated SD of change. Then the subgroups were (2) The sample size is small, with the largest size for 59 cases.[24]
combined and analyzed. In the meta-analysis of AHI, the result of (3) Surgical methods are not unified, and the selection of
subgroup 1 is 4.17( 7.62, 0.73), subgroup 2: 4.19( 7.51, symptoms suitable for the surgery is not the same.
0.88), overall: 4.15( 6.48, 1.82), which indicated consis- (4) Criterion evaluating sleep breathing inconsistent.
tence of the 2 subgroups with the correlation coefficient 0.667. In (5) There is a certain heterogeneity.
the meta-analysis of ESS, similar results were obtained by In addition, most of the studies were followed up for a short
aforementioned 2 meta-analyses. time, which may have a particular effect on the outcome.
In previous studies, researchers regarded increased NR as a
contributor in inducing and aggravating OSA, and made
conclusions that nasal surgery could decrease NR significantly 5. Conclusion
and improve other sleep indices such as oxygen saturation nadir, Both AHI and ESS improved significantly after isolated nasal
arousal index, sleep efficiency, and sleep architecture.[40] surgery, but the improvement of AHI is slightly significant. The
However, AHI did not show significant change, with some results of this study provide some evidence supporting for
studies even reporting aggravated AHI postoperatively.[38] A isolated nasal surgery in OSA patients especially those with nasal
possible explanation is that AHI along, as an indicator of how obstruction. However, in the future, there still should be more
respiratory events occur, is not sufficient to describe the randomized clinical controlled trials with multicenter coopera-
immigration of sleep stage, or decreases in ventilation (ie, length tion and a long-term follow-up to evaluate the efficacy of nasal
of the events or the fraction of events that are hypopneas).[41,42] surgery on OSA.
Some other studies reported opposite conclusion that nasal
surgery can significantly improve AHI,[20,22] which may result
from the different phenotypes of the patients. Multiple risk References
factors are reported to contribute to sleep apnea, such as [1] Mullington JM, Haack M, Toth M, et al. Cardiovascular, inflammatory,
abnormal anatomy of upper airway, unstable breathing control, and metabolic consequences of sleep deprivation. Prog Cardiovasc Dis
compromised upper airway muscle activation, and low arousal 2009;51:294–302.
threshold. These individual key factors could affect nasal surgery [2] Lal C, Strange C, Bachman D. Neurocognitive impairment in obstructive
sleep apnea. Chest 2012;141:1601–10.
efficacy. Moreover, the efficiency of unilevel upper airway [3] Ferris BGJr, Mead J, Opie LH. Partitioning of respiratory flow resistance
surgery may eliminate overtime,[35] which might be a reason for in man. J Appl Physiol 1964;19:653–8.
the discrepancy of the surgical outcomes. [4] Carpenter JG. Mental aberration attending hypertrophic rhinitis, with
However, there are still some limitations of this study: sub-acute otitis media: read in the section of laryngology and otology, at
the forty-third annual meeting of the american medical association, held
(1) Evidence level of the included literature is different. Only 1 at Detroit, MI, June, 1892. J Am Med Assoc 1892;XIX:539–42.
article is level 1, 2 articles are level 2, and other articles are [5] Welis WAMD. Some nervous and mental manifestations occurring in
connection with nasal disease. Am J Med Sci 1898;116:677–91.
levels 3–4. Study design is relatively rough, not grouping [6] Kohler M, Bloch KE, Stradling JR. The role of the nose in the
patients according to the severity of disease or preoperative pathogenesis of obstructive sleep apnoea and snoring. Eur Respir J
obstructive plane location. 2007;30:1208–15.

6
Wu et al. Medicine (2017) 96:5 www.md-journal.com

[7] Boudewyns AN, Van de Heyning PH, De Backer WA. Site of upper [25] Victores AJ, Takashima M. Effects of nasal surgery on the upper airway: a
airway obstruction in obstructive apnoea and influence of sleep stage. drug-induced sleep endoscopy study. Laryngoscope 2012;122:2606–10.
Eur Respir J 1997;10:2566–72. [26] Sufioglu M, Ozmen OA, Kasapoglu F, et al. The efficacy of nasal surgery
[8] Exar EN, Collop NA. The upper airway resistance syndrome. Chest in obstructive sleep apnea syndrome: a prospective clinical study. Eur
1999;115:1127–39. Arch Otorhinolaryngol 2012;269:487–94.
[9] Deegan PC, McNicholas WT. Predictive value of clinical features for the [27] Choi JH, Kim EJ, Kim YS, et al. Effectiveness of nasal surgery alone on
obstructive sleep apnoea syndrome. Eur Respir J 1996;9:117–24. sleep quality, architecture, position, and sleep-disordered breathing in
[10] Young T, Finn L, Kim H. Nasal obstruction as a risk factor for sleep- obstructive sleep apnea syndrome with nasal obstruction. Am J Rhinol
disordered breathing. The University of Wisconsin Sleep and Respiratory Allergy 2011;25:338–41.
Research Group. J Allergy Clin Immunol 1997;99:S757–762. [28] Bican A, Kahraman A, Bora I, et al. What is the efficacy of nasal surgery
[11] Friedman M, Maley A, Kelley K, et al. Impact of nasal obstruction on in patients with obstructive sleep apnea syndrome? J Craniofac Surg
obstructive sleep apnea. Otolaryngol Head Neck Surg 2011;144: 2010;21:1801–6.
1000–4. [29] Tosun F, Kemikli K, Yetkin S, et al. Impact of endoscopic sinus surgery
[12] Badr MS. Pathophysiology of upper airway obstruction during sleep. on sleep quality in patients with chronic nasal obstruction due to nasal
Clin Chest Med 1998;19:21–32. polyposis. J Craniofac Surg 2009;20:446–9.
[13] Silvoniemi P, Suonpaa J, Sipila J, et al. Sleep disorders in patients with [30] Li HY, Lee LA, Wang PC, et al. Can nasal surgery improve obstructive
severe nasal obstruction due to septal deviation. Acta Otolaryngol Suppl sleep apnea: subjective or objective? Am J Rhinol Allergy 2009;23:e51–5.
1997;529:199–201. [31] Kim ST, Choi JH, Jeon HG, et al. Polysomnographic effects of nasal
[14] Jenkinson C, Stradling J, Petersen S. Comparison of three measures of surgery for snoring and obstructive sleep apnea. Acta Otolaryngol
quality of life outcome in the evaluation of continuous positive airways 2009;124:297–300.
pressure therapy for sleep apnoea. J Sleep Res 1997;6:199–204. [32] Nakata S, Noda A, Yasuma F, et al. Effects of nasal surgery on sleep
[15] Strohl KP, Redline S. Recognition of obstructive sleep apnea. Am J Respir quality in obstructive sleep apnea syndrome with nasal obstruction. Am J
Crit Care Med 1996;154(2 Pt 1):279–89. Rhinol 2008;22:59–63.
[16] Han D, Zhang L. Nasal cavity ventilation expansion techniques. Acta [33] Li HY, Lin Y, Chen NH, et al. Improvement in quality of life after nasal
Otolaryngol 2011;131:1244–8. surgery alone for patients with obstructive sleep apnea and nasal
[17] Ishii L, Roxbury C, Godoy A, et al. Does nasal surgery improve OSA in obstruction. Arch Otolaryngol Head Neck Surg 2008;134:429–33.
patients with nasal obstruction and OSA? A meta-analysis. Otolaryngol [34] Koutsourelakis I, Georgoulopoulos G, Perraki E, et al. Randomised trial
Head Neck Surg 2015;153:326–33. of nasal surgery for fixed nasal obstruction in obstructive sleep apnoea.
[18] Li HY, Wang PC, Chen YP, et al. Critical appraisal and meta-analysis of Eur Respir J 2008;31:110–7.
nasal surgery for obstructive sleep apnea. Am J Rhinol Allergy [35] Virkkula P, Bachour A, Hytönen M, et al. Snoring is not relieved by nasal
2011;25:45–9. surgery despite improvement in nasal resistance. Chest 2006;129:81–7.
[19] Xiao Y, Han D, Zang H, et al. The effectiveness of nasal surgery on [36] Nakata S, Noda A, Yagi H, et al. Nasal resistance for determinant factor
psychological symptoms in patients with obstructive sleep apnea and of nasal surgery in CPAP failure patients with obstructive sleep apnea
nasal obstruction. Acta Otolaryngol 2016;136:626–32. syndrome. Rhinology 2005;43:296–9.
[20] Shuaib SW, Undavia S, Lin J, et al. Can functional septorhinoplasty [37] Verse T, Maurer JT, Pirsig W. Effect of nasal surgery on sleep-related
independently treat obstructive sleep apnea? Plastic Reconstr Surg breathing disorders. Laryngoscope 2002;112:64–8.
2015;135:1554–65. [38] Friedman M, Tanyeri H, Lim JW, et al. Effect of improved nasal
[21] Yalamanchali S, Cipta S, Waxman J, et al. Effects of endoscopic sinus breathing on obstructive sleep apnea. Otolaryngol Head Neck Surg
surgery and nasal surgery in patients with obstructive sleep apnea. 2000;122:71–4.
Otolaryngol Head Neck Surg 2014;151:171–5. [39] Camacho M, Riaz M, Capasso R, et al. The effect of nasal surgery on
[22] Park CY, Hong JH, Lee JH, et al. Clinical effect of surgical correction for continuous positive airway pressure device use and therapeutic treatment
nasal pathology on the treatment of obstructive sleep apnea syndrome. pressures: a systematic review and meta-analysis. Sleep 2015;38:279–86.
PloS One 2014;9:e98765. [40] Fairbanks DN. Effect of nasal surgery on snoring. Southern Med J
[23] Higgins JPT, Green S. Cochrane Handbook for Systematic Reviews of 1985;78:268–70.
Interventions Version 5.1.0 [updated March 2011]. The Cochrane [41] Olsen KD, Kern EB, Westbrook PR. Sleep and breathing disturbance
Collaboration, 2011. secondary to nasal obstruction. Otolaryngol Head Neck Surg 1981;89:
[24] Moxness MH, Nordgard S. An observational cohort study of the 804–10.
effects of septoplasty with or without inferior turbinate reduction in [42] Ohki M, Usui N, Kanazawa H, et al. Relationship between oral
patients with obstructive sleep apnea. BMC Ear Nose Throat Disord breathing and nasal obstruction in patients with obstructive sleep apnea.
2014;14:11. Acta Otolaryngol Suppl 1996;523:228–30.