Sleep Breath (2013) 17:227234

DOI 10.1007/s11325-012-0677-3

ORIGINAL ARTICLE

Effects of dietary weight loss on obstructive sleep apnea:

a meta-analysis
Anil Anandam & Morohunfolu Akinnusi &
Thomas Kufel & Jahan Porhomayon & Ali A. El-Solh

Received: 24 November 2011 / Revised: 6 February 2012 / Accepted: 21 February 2012 / Published online: 29 February 2012
# Springer-Verlag 2012

Abstract
Purpose Clinical and epidemiologic investigations suggest
a strong association between obesity and obstructive sleep
apnea (OSA). The purpose of this study is to evaluate the
currently available literature reporting on the effectiveness
of dietary weight loss in treating OSA among obese patients.
Methods Relevant studies were identified by computerized
searches of PubMed, EMBASE, CINAHL, Web of Science,
and The Cochrane Central Register of Controlled Trials
through September 2011 as well as the reference lists of
all obtained articles. Information on study design, patient
characteristics, pre- and post-dietary weight loss measures
of OSA and body mass index (BMI), and study quality was

T. Kufel : J. Porhomayon : A. A. El-Solh

The Veterans Affairs Western New York Healthcare System,
Buffalo, NY, USA

obtained. Data were extracted by two independent analysts.

Weighted averages using a random-effects model are
reported with 95 % confidence intervals.
Results Nine articles representing 577 patients were selected.
Dietary weight loss program resulted in a pooled mean BMI
reduction of 4.8 kg/m2 (95 % confidence interval [CI] 3.85.9). The random-effects pooled apnea hypopnea (AHI) indices at pre- and post-dietary intervention were 52.5 (range
10.091.0) and 28.3 events/h (range 5.464.5), respectively
(p<0.001). Compared to control, the weighted mean difference of AHI was decreased by 14.3 events/h (95 % CI 23.5
to 5.1; p 00.002) in favor of the dietary weight loss
programs.
Conclusions Dietary weight loss programs are effective in
reducing the severity of OSA but not adequate in relieving
all respiratory events. Weight reduction programs should be
considered as adjunct rather than curative therapy.

A. Anandam : M. Akinnusi : T. Kufel : J. Porhomayon :

Keywords Sleep apnea . Obesity . Weight reduction .

Meta-analysis

A. A. El-Solh
Western New York Respiratory Research Center, Division
of Pulmonary, Critical Care, and Sleep Medicine, Department
of Medicine, State University of New York at Buffalo School
of Medicine and Biomedical Sciences and School of Public Health
and Health Professions,
Buffalo, NY, USA

Introduction

A. A. El-Solh
Department of Social and Preventive Medicine, State University
of New York at Buffalo School of Medicine and Biomedical
Sciences and School of Public Health and Health Professions,
Buffalo, NY, USA
A. A. El-Solh (*)
Medical Research, VA Western New York Healthcare System,
Bldg. 20 (151) VISN02, 3495 Bailey Avenue,
Buffalo, NY 14215-1199, USA
e-mail: solh@buffalo.edu

Obesity is an increasing health problem all over the world.

Based on the World Health Organization and the National
Institutes of Health criteria for defining overweight and
obesity [1], the prevalence of overweight and obese adults
reached 68 % in 20072008 [2]. The increase was detected
in all ages, socioeconomic strata, and regions [3, 4]. More
importantly, the prevalence of clinically severe obesity is
increasing much faster than that of moderate obesity, and the
number of morbidly obese patients quadrupled between
1986 and 2000, whereas the prevalence of obesity approximately doubled during the same period [5].

228

Obesity is considered one of the main risk factors for

obstructive sleep apnea (OSA) [68]. The prevalence and
severity of OSA rise in parallel with the body mass index
(BMI) [9]. OSA is nearly twice as common in obese and
severely obese patients as normal-weight adults. A onestandard-deviation difference BMI is associated with a fourfold increase in disease prevalence [9]. Furthermore,
patients with mild OSA who gain extra 10 % of body weight
are at sixfold increased risk of progressing to a higher OSA
severity [7]. An increase of body weight by 10 % over time
increases the apnea hypopnea index (AHI), on average, by
30 %. Alternatively, a 1015 % reduction in body weight
can reduce the AHI by 50 % [10].
Growing evidence indicates that OSA is an independent
risk factor for development of hypertension [11, 12], cardiovascular morbidity and mortality [1315], and sudden death
[16, 17]. Nasal continuous positive airway pressure (CPAP)
remains the gold standard treatment with robust evidence
from several randomized, placebo-controlled trials [18, 19]
demonstrating improvement of symptoms, cognitive function, and quality of life in OSA patients treated with nasal
CPAP. Weight loss has also been shown to improve apnea
frequency by relieving mechanical load on the upper airway,
reducing propensity for its collapse [20]. However, there are
few studies indicating whether the amount of weight loss
recommended by expert and governmental panels (10 %) is
sufficient to improve OSA. With the exception of four
randomized trials [10, 2123] that have been published on
the effect of dietary weight loss on OSA, series of trials
addressing non-surgical weight loss have been limited by
small sample sizes, selection bias, and follow-up bias. The
purpose of this investigation is to systematically review the
biomedical literature and to critically evaluate published
studies using meta-analysis techniques on the effectiveness
of dietary weight loss on sleep apnea indices.

Methods
Search strategy
We generally followed the Quality of Reporting of Metaanalyses statement guidelines [24]. We searched the electronic databases PubMed, EMBASE, CINAHL, Web of
Science, and The Cochrane Central Register of Controlled
Trials for relevant publications. We searched combinations
of the keywords obstructive sleep apnea, obstructive sleep
apnea, sleep-disordered breathing, weight loss, and obesity.
We included articles in English, French, or Spanish published from January 1975 to September 2011. We identified
unpublished studies in conference abstracts or in registers of
clinical trials (ClinicalTrials.gov and current controlled trials). We also consulted bibliographies of relevant articles,

Sleep Breath (2013) 17:227234

science citation index, and Google Scholar. We also

searched bibliographies of retrieved meta-analyses and systematic reviews and included studies and abstracts that were
presented at the most recent congresses and suggested by an
expert reader.
Study selection
We narrowed the list of publications obtained to studies that
meet our predetermined inclusion criteria. To be eligible for
inclusion, articles had to report data on patients (n5) who
underwent both pre- and post-dietary weight loss polysomnography to include at a minimum the measurement of the
AHI. We excluded studies where the reference standard at
baseline was not overnight monitored polysomnography.
Studies that included surgical interventions for the treatment
of obesity were also excluded. Two investigators (AA and
JP) independently screened the titles and abstracts of all
articles identified by the search strategy and individually
determined inclusion of studies for the analysis. In case of
disagreement, assessment of eligibility was done by manually rechecking the data extraction from each disputed article. In those situations where there was lack of agreement or
clarity, further advice was sought from independent sleep
specialists (MA and TK). Full copies of all selected articles
were retrieved. Where data had been published multiple
times, we abstracted data from the most recent publications
with the longest follow-up period.
Data extraction
We extracted data on patients clinical and demographic
characteristics, duration of weight loss program, and polysomnographic indices at baseline and following dietary
weight loss using a standardized form and cross-checked
for accuracy. Additional information that was abstracted
included year of publication, country of origin, time from
dietary intervention to polysomnography, and sample size.
The intraclass correlation coefficient for agreement between
the two raters on overall quality rating for all included
studies was 0.89. Disagreement was solved by discussion
and consensus finding. When important data were missing,
the author was contacted.
Validity assessment
To limit the heterogeneity secondary to differences among
study designs, the quality of each study was evaluated
according to the guidelines developed by the United States
Preventive Task Force [25] and the Evidence-Based
Medicine Working Group [26]. A point score system was
applied according to the quality of the study. The following
characteristics were assessed: (1) clear description of

228

Obesity is considered one of the main risk factors for

obstructive sleep apnea (OSA) [68]. The prevalence and
severity of OSA rise in parallel with the body mass index
(BMI) [9]. OSA is nearly twice as common in obese and
severely obese patients as normal-weight adults. A onestandard-deviation difference BMI is associated with a fourfold increase in disease prevalence [9]. Furthermore,
patients with mild OSA who gain extra 10 % of body weight
are at sixfold increased risk of progressing to a higher OSA
severity [7]. An increase of body weight by 10 % over time
increases the apnea hypopnea index (AHI), on average, by
30 %. Alternatively, a 1015 % reduction in body weight
can reduce the AHI by 50 % [10].
Growing evidence indicates that OSA is an independent
risk factor for development of hypertension [11, 12], cardiovascular morbidity and mortality [1315], and sudden death
[16, 17]. Nasal continuous positive airway pressure (CPAP)
remains the gold standard treatment with robust evidence
from several randomized, placebo-controlled trials [18, 19]
demonstrating improvement of symptoms, cognitive function, and quality of life in OSA patients treated with nasal
CPAP. Weight loss has also been shown to improve apnea
frequency by relieving mechanical load on the upper airway,
reducing propensity for its collapse [20]. However, there are
few studies indicating whether the amount of weight loss
recommended by expert and governmental panels (10 %) is
sufficient to improve OSA. With the exception of four
randomized trials [10, 2123] that have been published on
the effect of dietary weight loss on OSA, series of trials
addressing non-surgical weight loss have been limited by
small sample sizes, selection bias, and follow-up bias. The
purpose of this investigation is to systematically review the
biomedical literature and to critically evaluate published
studies using meta-analysis techniques on the effectiveness
of dietary weight loss on sleep apnea indices.

Methods
Search strategy
We generally followed the Quality of Reporting of Metaanalyses statement guidelines [24]. We searched the electronic databases PubMed, EMBASE, CINAHL, Web of
Science, and The Cochrane Central Register of Controlled
Trials for relevant publications. We searched combinations
of the keywords obstructive sleep apnea, obstructive sleep
apnea, sleep-disordered breathing, weight loss, and obesity.
We included articles in English, French, or Spanish published from January 1975 to September 2011. We identified
unpublished studies in conference abstracts or in registers of
clinical trials (ClinicalTrials.gov and current controlled trials). We also consulted bibliographies of relevant articles,

Sleep Breath (2013) 17:227234

science citation index, and Google Scholar. We also

searched bibliographies of retrieved meta-analyses and systematic reviews and included studies and abstracts that were
presented at the most recent congresses and suggested by an
expert reader.
Study selection
We narrowed the list of publications obtained to studies that
meet our predetermined inclusion criteria. To be eligible for
inclusion, articles had to report data on patients (n5) who
underwent both pre- and post-dietary weight loss polysomnography to include at a minimum the measurement of the
AHI. We excluded studies where the reference standard at
baseline was not overnight monitored polysomnography.
Studies that included surgical interventions for the treatment
of obesity were also excluded. Two investigators (AA and
JP) independently screened the titles and abstracts of all
articles identified by the search strategy and individually
determined inclusion of studies for the analysis. In case of
disagreement, assessment of eligibility was done by manually rechecking the data extraction from each disputed article. In those situations where there was lack of agreement or
clarity, further advice was sought from independent sleep
specialists (MA and TK). Full copies of all selected articles
were retrieved. Where data had been published multiple
times, we abstracted data from the most recent publications
with the longest follow-up period.
Data extraction
We extracted data on patients clinical and demographic
characteristics, duration of weight loss program, and polysomnographic indices at baseline and following dietary
weight loss using a standardized form and cross-checked
for accuracy. Additional information that was abstracted
included year of publication, country of origin, time from
dietary intervention to polysomnography, and sample size.
The intraclass correlation coefficient for agreement between
the two raters on overall quality rating for all included
studies was 0.89. Disagreement was solved by discussion
and consensus finding. When important data were missing,
the author was contacted.
Validity assessment
To limit the heterogeneity secondary to differences among
study designs, the quality of each study was evaluated
according to the guidelines developed by the United States
Preventive Task Force [25] and the Evidence-Based
Medicine Working Group [26]. A point score system was
applied according to the quality of the study. The following
characteristics were assessed: (1) clear description of

Sleep Breath (2013) 17:227234

inclusion and exclusion criteria, (2) study sample representative for mentioned population, (3) clear description of
sample selection, (4) full specification of clinical and demographic variables, (5) reporting loss of follow-up, (6) clear
definition of OSA, (7) clear definition of outcomes and
outcome assessment, and (8) adjustment of possible confounders on multivariate analysis. If a study did not clearly
mention one of these key points, we considered that it had
not been performed.
Statistical analysis
We did perform a random-effects meta-analysis using
Review Manager 5.0 (Cochrane Collaboration, 2008) following the MantelHaenszel model to assess changes in
BMI and AHI before and after dietary weight loss. A pooled
linear regression analysis was conducted across selected
studies to examine the coefficient of variability between
reduction of BMI and change in AHI. We determined clinical heterogeneity by comparing protocol, populations, and
methodology of the studies included. We measured statistical heterogeneity using the I2 statistic that assesses the
degree of inconsistency across studies; it results in a 0
100 % range quantifying the proportion of variation in the
effect, which is due to inter-study variation. We predefined
heterogeneity (I2 25 % for low, 25 %<I2 <50 % for moderate, and I2 50 % for high) [27]. Meta-regression was used
to examine potential sources of heterogeneity. If metaregression results indicated a variable to contribute significantly to heterogeneity between studies, subgroup analysis
by this variable was conducted, testing whether there was an
effect of treatment on outcomes within each subgroup. If
heterogeneity was reduced, a subgroup analysis provided a
more reliable estimate of pooled effect size between the treatment groups. Publication bias was evaluated visually using
the funnel plot and calculation of Eggers intercept to determine the degree of funnel plot asymmetry [28]. A p value less
than 0.05 was used to denote statistical significance.

229

studies that did not meet the inclusion criteria, we identified

nine studies representing 577 patients for inclusion. A flow
diagram of the included studies is shown in Fig. 1.
The characteristics of the included studies [10, 20, 22,
2934] are shown in Table 1. Most studies were performed
in the USA. Four studies were conducted in Sweden,
Finland, Italy, and Spain. Three were randomized control
trials and six were conducted in a prospective non-randomized
design. The mean sample size was 64.1 (range 8264), and the
pooled mean patient age was 53.1 years (range 46.4
61.2 years). The length of dietary intervention varied from
3 weeks to 1 year.
The mean pre- and post-dietary intervention BMI was
available for seven studies and was 39.6 (range 22.0
54.8 kg/m2) and 33.8 kg/m2 (range 27.246.6 kg/m2), respectively (Fig. 2). The weighted mean drop in BMI was 4.8
(95 % confidence interval [CI] 3.85.9; P <0.001). The
value of I2 was 54 %, indicating a high degree of heterogeneity between the studies.
Each selected article presented data on AHI at pre- and
post-dietary weight loss from overnight polysomnogram.
The pooled mean pre- and post-dietary intervention AHI

Role of the funding source

The source of funding had no input into the design or the
outcome of the study. The corresponding author had full
access to all the data in the study and had final responsibility
for the decision to submit for publication.

Results
A total of 684 publications were identified by the search.
Review of the titles led to the exclusion of 647 references.
Of the 37 publications examined, 21 were removed from
further analysis. After excluding duplicate studies and

Fig. 1 Flow diagram for the review and selection of included studies

2010

2009

2010

1998

1985

1992

1992

1991

1990

Tuomilehto

Foster

Nerfeldt

Sampol

Smith

Suratt

Nahmias

Schwartz

Pasquali

Italy

USA

USA

USA

USA

Spain

Sweden

USA

Finland

Country

Prospective

Prospective controlled

Prospective

Prospective

RCT

Prospective

Prospective

RCT

RCT

Study design

VLCD

Weight loss program

LCD

VLCD

LCD

LCD

LCD

LCD plus moderate

intensity exercise

VLCD

Intervention

VLCD very low calorie diet (600800 kcal/d), LCD low calorie diet (8001,800 kcal/d)

Year

Authors

Table 1 Summary of selected studies

314 months

18.49.5 months

2076 weeks

36 weeks

30 months

61132 months

2 years

1 year

2 years

Follow-up

23/

13/13

28/

8/

15/8

101/

33/

125/139

35/36

Intervention/control

Compliance with the 12-week intervention,

83 %. Intervention lowered risk of OSA
by adjusted OR 0.35 (95 % CI 0.120.97)
Compliance with the 12-month intervention,
82.4 %. Prevalence of OSA at 1 year in the
intervention group was half that of the control
Compliance at 2 years 70 % with the weight
reduction program. AHI showed non-significant
decrease after intervention (p00.054)
No correlation between changes in AHI and BMI.
58 % had OSA recurrence at 5 year follow-up
No correlation between changes in weight
and apnea frequency after weight loss
Despite initial improvement in sleep indices,
all subjects returned to baseline within 2 years
All but three had 50 % reduction in AHI with
68 % considered cure of OSA based on
AHI<15
Not a randomized but age-, weight-, and gendermatched study
Significant correlation between weight loss and
AHI. Failure of improvement associated with
otorhinolaryngoiatric pathology

Comments

230
Sleep Breath (2013) 17:227234

Sleep Breath (2013) 17:227234

231

Fig. 2 Forest plot for change in BMI following dietary weight loss program using the radom-effects model. The size of the box indicates the
studys relative weight based on standard error. The diamond reflects the 95 % confidence interval of the summary estimate

was 52.5 (range 10.091.0) and 28.3 events/h (range 5.4

64.5), respectively (Fig. 3). All studies showed a mean
improvement in the AHI value after diet therapy with a
range of 4.6 to 50.8 (Fig. 3). The weighted mean decrease
in AHI was 23.1 /h (95 % CI, 8.937.3; P00.001). The
value of I2 was 99 %, indicating a high degree of heterogeneity across the studies.
Two studies have examined the degree of excessive daytime sleepiness following weight reduction program. The
first study by Nerfeldt et al. [30] reported a significant drop
in the Epworth Sleepiness Scale (ESS) from 94 to 53
(p00.03) in the intervention group after a 2-year weight
reduction program, but the study lacked a control arm. The
second study by Tuomilehto et al. [34] found no significant
difference in the ESS change from baseline between the
intervention and the control group (ESS 2.3 3.6
versus 2.9 4.4; p 00.86) after 1 year of follow-up.
Further, there was no significant difference in the quality of life derived from the SF36 questionnaire between
baseline and after 24 months of treatment in the only
study that addressed this question [30]. Pooling studies
by country of origin had no impact on AHI following

dietary intervention. Analysis by meta-regression did not

reveal an effect on the follow-up AHI by age, type of
weight loss program, or baseline AHI.
Sensitivity analysis
Because of significant heterogeneity, we assessed for differences in years of publication. Pooled analysis of studies
published prior to 1998 showed a significantly higher reduction in weighted mean AHI (34.3 events/h [95 % CI 24.9
43.7]) compared to those published after 1998 (8.8 events/
h [95 % CI 3.314.3]). We also examined differences in
measures of effect between prospective and randomized studies. There was a trend for non-randomized prospective studies
to report a higher drop in AHI from baseline following dietary
weight loss compared to randomized studies: 29.5
events/h (95 % CI 17.141.8) versus 11.5 events/h (95 % CI
0.222.8). Three studies included a control arm while using a
randomized design. Dietary weight loss resulted in a weighted
mean difference of 14.3 events/h (95 % CI 23.5 to 5.1; p0
0.002) in change of AHI favoring the intervention group
compared to the control group (Fig. 4). Stratified analyses of

Fig. 3 Forest plot for change in AHI following dietary weight loss program using the radom-effects model. The size of the box indicates the studys
relative weight based on standard error. The diamond reflects the 95 % confidence interval of the summary estimate

232

Sleep Breath (2013) 17:227234

Fig. 4 Forest plot for the difference in AHI between control and intervention program using the radom-effects model. The size of the box indicates
the studys relative weight based on standard error. The diamond reflects the 95 % confidence interval of the summary estimate

weight loss and AHI improvement found that the category

with highest weight loss was associated with the greatest
improvement in AHI. Pooled analysis of selected investigations revealed a modest correlation in coefficient of variation
between these two variables (R2 063 %; p00.03) (Fig. 5).
Publication bias
Publication bias was evaluated by a visual review of a
funnel plot in which standard error is plotted against the
difference in means. Publication bias was not observed for
both BMI and AHI from the graphical review. Eggers
intercept was 2.1 (p00.48) for BMI and 1.53 (p00.67) for
AHI, suggesting a low likelihood of publication bias.

Discussion
The results of this meta-analysis corroborate the general
consensus that dietary weight loss can result in significant
improvement in the severity of sleep apnea. The overall
effect size of the pooled weighted data shows an absolute
difference reduction in AHI of 23.1 events/h corresponding
to a combined reduction of 44 %.

Fig. 5 Regression analysis of pooled data on the effect of weight loss

on AHI (the outer dotted lines represent the 95 % confidence interval)

The clinical relevance of modest weight loss on OSA has

not been fully determined particularly because evidence that
dietary weight loss reduces the cognitive, metabolic, and
cardiovascular event rates or mortality rates from OSA is
limited. Two studies examined the long-term effect of very
low energy diet on the quality of life, dyslipidemia, and
insulin resistance in obese patients with OSA [30, 35].
Among patients with metabolic disease at baseline,
Johansson and colleagues [35] reported resolution of dyslipidemia and insulin resistance in 19 and 50 % of cases,
respectively, following a 1-year weight loss program. We
have excluded this study from our analysis because the
outcome was not based on polysomnographic data. Similar
improvements were denoted by Nerfeldt and coworkers [30]
after a 2-year weight reduction program. However, as many
of the included studies show, dietary and lifestyle changes
resulting in modest weight loss consistently reduced AHI.
This improvement in the severity of sleep apnea has been
attributed in part to the reduction in upper airway collapsibility from pharyngeal fat deposit in the oropharynx [20].
These anatomical changes are supported by the findings of
Suratt and coworkers [33] who demonstrated decreased
upper airway resistance in awake patients with OSA after
weight loss. More recent data suggest that abatement of
inflammatory cytokines and resolution of hormonal dysregulation may also contribute to the normalization of neuromechanical control of upper airway patency during sleep
[3638].
Changing behavior is a major goal for lifelong success in
obesity [39]. The type of intervention will impact the level
of adherence in the long term. Indeed the complexity of the
program and the frequency of ongoing interactions with the
patient influence the rate of adherence and, consequently,
the rate of success in the follow-up period. Given the dose
response relationship between weight loss and AHI, it is
important that weight loss programs achieve a mean weight
loss of at least 10 kg or more [22]. However, short-term
intense rehabilitation combined with a low-calorie diet and
exercise might not be sufficient to ensure a high rate of
adherence and sustained response [40] in these individuals.
The OSA cure rate reported from our analysis ranged from
61 % at 3 months to <10 % at 1 year [35]. Maintenance of

Sleep Breath (2013) 17:227234

large amounts of weight loss is arguably one of the greatest

existing clinical challenges. Case in point, a mere 30 % of
all participants were able to maintain>10 % reduction of
weight after a 2-year weight reduction program, and all
amelioration in AHI was lost by the end of the study [30].
After 5 years, 46 % were unable to sustain a one half BMI
drop achieved following dietary restriction [32]. This
emphasizes the importance of continuing a maintenance
behavioral program on a long-term basis [41].
The extent of AHI improvement following dietary
restriction is considerably less in magnitude compared to the
reduction in severity of sleep apnea observed following surgical weight loss. A recent meta-analysis [42] of 12 studies
encompassing 342 patients that examined the effect of
bariatric surgery on OSA revealed a reduction of 38.2
events/h corresponding to a reduction of 71 %. The large
difference between diet and surgical approaches is not surprising given the significant shedding of excess body weight
following surgical intervention. Yet, even dramatic weight
loss via bariatric surgery may not cure OSA. Only 25 % of
patients achieved resolution of OSA after surgery despite a
mean body mass index decrease of 16.9 kg/m2 [42]. Thus,
persistence of some degree of OSA in many patients who
undergo dietary weight loss or bariatric surgery can be
expected because most of these subjects still remain overweight, if not obese. With the projected rise in the prevalence
of OSA in the wake of obesity epidemic, a dietary intervention
program combined with nasal CPAP would likely remain the
primary approach for treatment of obesity-related OSA as
provision of bariatric surgery to all obese patients is
unrealistic.
In our study, we avoided selection bias by pre-specified
inclusion criteria, a systematic search, and independent
evaluation of studies inclusion by two reviewers. These
methodological strengths have therefore enhanced the
validity and applicability of our findings, but our study still
has limits. There are several sources responsible for the study
heterogeneity. The study population is likely diverse in ethnicity, body habitus, cultural differences, and other unmeasured
factors. The technology of PSG has changed dramatically with
years, most notably the introduction of software designed to
assist providers with detection of respiratory events. The scoring criteria for apneas and hypopneas have also evolved overtime with significant differences present between earlier and
later studies over a 26-year span. This may reflect bias [43],
time-dependent efficacy (e.g., when the treatment benefit
decreases with longer follow-up) [44], quality differences
[45], or chance. Furthermore, studies varied in the design of
the weight loss programs, their duration, and the time points at
which outcomes were reported. These inherent difficulties in
meta-analyzing such data, in which the effect (weight loss)
changes over time, led invariably to heterogeneity [Section
9.5.1 of Cochrane Handbook, Cochrane Handbook for

233

Systematic Reviews of Interventions Version 5.1.0 (updated

March 2011)].
In conclusion, dietary weight loss is effective in reducing
the severity of sleep apnea. This finding was consistent
across the selected studies despite significant heterogeneity
in patient populations and study design. The response in
AHI to dietary intervention, however, is less than optimal
and the presence of residual OSA in a large number of
patients is a significant probability that requires further
study and intervention.
Funding support This work was supported in part by VA Merit
Review Award HSR&D 10-087-1 (AES). The opinions of the authors
herein are their private views and are not to be construed as reflecting
the views of the Department of Veterans Affairs.

References
1. Kuczmarski RJ, Flegal KM (2000) Criteria for definition of overweight in transition: background and recommendations for the
United States. Am J Clin Nutr 72:10741081
2. Flegal KM, Carroll MD, Ogden CL, Curtin LR (2010) Prevalence
and trends in obesity among US adults, 19992008. JAMA
303:235241
3. Ogden CL, Carroll MD, Curtin LR, McDowell MA, Tabak CJ,
Flegal KM (2006) Prevalence of overweight and obesity in the
United States, 19992004. JAMA 295:15491555
4. Johnson-Taylor WL, Fisher RA, Hubbard VS, Starke-Reed P,
Eggers PS (2008) The change in weight perception of weight
status among the overweight: comparison of NHANES III
(19881994) and 19992004 NHANES. Int J Behav Nutr Phys
Act 5:9
5. Sturm R (2003) Increases in clinically severe obesity in the United
States, 19862000. Arch Intern Med 163:21462148
6. Resta O, Foschino-Barbaro MP, Legari G, Talamo S, Bonfitto P,
Palumbo A, Minenna A, Giorgino R, De Pergola G (2001) Sleeprelated breathing disorders, loud snoring and excessive daytime
sleepiness in obese subjects. Int J Obes Relat Metab Disord
25:669675
7. Peppard PE, Young T, Palta M, Dempsey J, Skatrud J (2000)
Longitudinal study of moderate weight change and sleepdisordered breathing. JAMA 284:30153021
8. Carneiro G, Florio RT, Zanella MT, Pradella-Hallinan M, RibeiroFilho FF, Tufik S, Togeiro SM (2012) Is mandatory screening for
obstructive sleep apnea with polysomnography in all severely
obese patients indicated? Sleep Breath 16:163168
9. Young T, Palta M, Dempsey J, Skatrud J, Weber S, Badr S (1993)
The occurrence of sleep-disordered breathing among middle-aged
adults. N Engl J Med 328:12301235
10. Smith PL, Gold AR, Meyers DA, Haponik EF, Bleecker ER (1985)
Weight loss in mildly to moderately obese patients with obstructive
sleep apnea. Ann Intern Med 103:850855
11. Peppard PE, Young T, Palta M, Skatrud J (2000) Prospective study
of the association between sleep-disordered breathing and hypertension. N Engl J Med 342:13781384
12. Bixler EO, Vgontzas AN, Lin HM, Ten Have T, Leiby BE, VelaBueno A, Kales A (2000) Association of hypertension and sleepdisordered breathing. Arch Intern Med 160:22892295
13. Shahar E, Whitney CW, Redline S, Lee ET, Newman AB, Javier
Nieto F, O'Connor GT, Boland LL, Schwartz JE, Samet JM (2001)

234

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.
28.

Sleep Breath (2013) 17:227234

Sleep-disordered breathing and cardiovascular disease: crosssectional results of the Sleep Heart Health Study. Am J Respir
Crit Care Med 163:1925
Mooe T, Franklin KA, Holmstrom K, Rabben T, Wiklund U (2001)
Sleep-disordered breathing and coronary artery disease: long-term
prognosis. Am J Respir Crit Care Med 164:19101913
Chung S, Yoon IY, Lee CH, Kim JW (2011) The effects of nasal
continuous positive airway pressure on vascular functions and
serum cardiovascular risk factors in obstructive sleep apnea syndrome. Sleep Breath 15:7176
He J, Kryger MH, Zorick FJ, Conway W, Roth T (1988) Mortality
and apnea index in obstructive sleep apnea. Experience in 385
male patients. Chest 94:914
Lavie P, Herer P, Peled R, Berger I, Yoffe N, Zomer J, Rubin AH
(1995) Mortality in sleep apnea patients: a multivariate analysis of
risk factors. Sleep 18:149157
Engleman HM, Martin SE, Kingshott RN, Mackay TW, Deary
IJ, Douglas NJ (1998) Randomised placebo controlled trial of
daytime function after continuous positive airway pressure
(CPAP) therapy for the sleep apnoea/hypopnoea syndrome.
Thorax 53:341345
Engleman HM, Kingshott RN, Wraith PK, Mackay TW, Deary IJ,
Douglas NJ (1999) Randomized placebo-controlled crossover trial
of continuous positive airway pressure for mild sleep apnea/hypopnea syndrome. Am J Respir Crit Care Med 159:461467
Schwartz AR, Gold AR, Schubert N, Stryzak A, Wise RA, Permutt
S, Smith PL (1991) Effect of weight loss on upper airway collapsibility in obstructive sleep apnea. Am Rev Respir Dis 144:494498
Tuomilehto HP, Seppa JM, Partinen MM, Peltonen M, Gylling H,
Tuomilehto JO, Vanninen EJ, Kokkarinen J, Sahlman JK,
Martikainen T, Soini EJ, Randell J, Tukiainen H, Uusitupa M
(2009) Lifestyle intervention with weight reduction: first-line treatment in mild obstructive sleep apnea. Am J Respir Crit Care Med
179:320327
Foster GD, Borradaile KE, Sanders MH, Millman R, Zammit G,
Newman AB, Wadden TA, Kelley D, Wing RR, Pi-Sunyer FX,
Reboussin D, Kuna ST (2009) A randomized study on the effect of
weight loss on obstructive sleep apnea among obese patients with
type 2 diabetes: the Sleep AHEAD study. Arch Intern Med
169:16191626
Johansson K, Neovius M, Lagerros YT, Harlid R, Rossner S,
Granath F, Hemmingsson E (2009) Effect of a very low energy
diet on moderate and severe obstructive sleep apnoea in obese
men: a randomised controlled trial. BMJ 339:b4609
Moher D, Cook DJ, Eastwood S, Olkin I, Rennie D, Stroup DF
(1999) Improving the quality of reports of meta-analyses of randomised controlled trials: the QUOROM statement. Quality of
reporting of meta-analyses. Lancet 354:18961900
Harris RP, Helfand M, Woolf SH, Lohr KN, Mulrow CD, Teutsch
SM, Atkins D (2001) Current methods of the US Preventive
Services Task Force: a review of the process. Am J Prev Med
20:2135
Levine M, Walter S, Lee H, Haines T, Holbrook A, Moyer V
(1994) Users' guides to the medical literature. IV. How to use an
article about harm. Evidence-Based Medicine Working Group.
JAMA 271:16151619
Higgins JP, Thompson SG, Deeks JJ, Altman DG (2003)
Measuring inconsistency in meta-analyses. BMJ 327:557560
Egger M, Smith GD, Phillips AN (1997) Meta-analysis: principles
and procedures. BMJ 315:15331537

29. Nahmias J, Kirschner M, Karetzky MS (1993) Weight loss and

OSA and pulmonary function in obesity. N J Med: J Med Soc N J
90:4853
30. Nerfeldt P, Nilsson BY, Mayor L, Udden J, Friberg D (2010) A
two-year weight reduction program in obese sleep apnea patients. J
Clin Sleep Med 6:479486
31. Pasquali R, Colella P, Cirignotta F, Mondini S, Gerardi R, Buratti
P, Rinaldi Ceroni A, Tartari F, Schiavina M, Melchionda N et al
(1990) Treatment of obese patients with obstructive sleep apnea
syndrome (OSAS): effect of weight loss and interference of otorhinolaryngoiatric pathology. Int J Obes 14:207217
32. Sampol G, Munoz X, Sagales MT, Marti S, Roca A, Dolors de la
Calzada M, Lloberes P, Morell F (1998) Long-term efficacy of
dietary weight loss in sleep apnoea/hypopnoea syndrome. Eur
Respir J 12:11561159
33. Suratt PM, McTier RF, Findley LJ, Pohl SL, Wilhoit SC (1992)
Effect of very-low-calorie diets with weight loss on obstructive
sleep apnea. Am J Clin Nutr 56:182S184S
34. Tuomilehto H, Gylling H, Peltonen M, Martikainen T, Sahlman J,
Kokkarinen J, Randell J, Tukiainen H, Vanninen E, Partinen M,
Tuomilehto J, Uusitupa M, Seppa J (2010) Sustained improvement
in mild obstructive sleep apnea after a diet- and physical activitybased lifestyle intervention: postinterventional follow-up. Am J
Clin Nutr 92:688696
35. Johansson K, Hemmingsson E, Harlid R, Trolle Lagerros Y,
Granath F, Rossner S, Neovius M (2011) Longer term effects of
very low energy diet on obstructive sleep apnoea in cohort derived
from randomised controlled trial: prospective observational
follow-up study. BMJ 342:d3017
36. Sahlman J, Seppa J, Herder C, Peltonen M, Peuhkurinen K,
Gylling H, Vanninen E, Tukiainen H, Punnonen K, Partinen M,
Uusitupa M, Tuomilehto H (2010) Effect of weight loss on inflammation in patients with mild obstructive sleep apnea.
Nutrition, Metabolism, and Cardiovascular Diseases, in press
37. Romero-Corral A, Caples SM, Lopez-Jimenez F, Somers VK
(2010) Interactions between obesity and obstructive sleep apnea:
implications for treatment. Chest 137:711719
38. Schwartz AR, Patil SP, Squier S, Schneider H, Kirkness JP, Smith
PL (2010) Obesity and upper airway control during sleep. J Appl
Physiol 108:430435
39. Foreyt JP, Goodrick GK (1993) Evidence for success of behavior
modification in weight loss and control. Ann Intern Med 119:698
701
40. Tumiati R, Mazzoni G, Crisafulli E, Serri B, Beneventi C, Lorenzi
CM, Grazzi G, Prato F, Conconi F, Fabbri LM, Clini EM (2008)
Home-centred physical fitness programme in morbidly obese individuals: a randomized controlled trial. Clin Rehabil 22:940950
41. Wadden TA, Butryn ML (2003) Behavioral treatment of obesity.
Endocrinol Metab Clin N Am 32:9811003
42. Greenburg DL, Lettieri CJ, Eliasson AH (2009) Effects of surgical
weight loss on measures of obstructive sleep apnea: a metaanalysis. Am J Med 122:535542
43. Ioannidis JP (1998) Effect of the statistical significance of results
on the time to completion and publication of randomized efficacy
trials. JAMA 279:281286
44. Ioannidis JP, Cappelleri JC, Sacks HS, Lau J (1997) The relationship between study design, results, and reporting of randomized
clinical trials of HIV infection. Control Clin Trials 18:431444
45. Ioannidis JP, Cappelleri JC, Lau J (1998) Issues in comparisons
between meta-analyses and large trials. JAMA 279:10891093