- Email: [email protected]
The Effects of Long-term CPAP on Weight Change in Patients With Comorbid OSA and Cardiovascular Disease
[
1
Original Research
]
56
2
57
3
58
4
59
5
60
The Effects of Long-term CPAP on Weight Change in Patients With Comorbid OSA and Cardiovascular Disease
6 7 8 9 10 11
61 62 63 64 65 66
Data From the SAVE Trial
12
67
13 14
68 Q20
15 16 17
Qiong Ou, MD; Baixin Chen, MMed; Kelly A. Loffler, PhD; Yuanming Luo, MD, PhD; Xilong Zhang, PhD; Rui Chen, PhD;
69
Qian Wang, MMed; Luciano F. Drager, MD, PhD; Geraldo Lorenzi-Filho, MD, PhD; Michael Hlavac, PhD;
70 71
Nigel McArdle, MD; Sutapa Mukherjee, MD, PhD; Olga Mediano, MD, PhD; Ferran Barbe, MD; Q1
18
72
Craig S. Anderson, MD, PhD; R. Doug McEvoy, MD; Richard J. Woodman, PhD; on behalf of the SAVE investigators
73
19
74
20
75
21
Although recent evidence suggests that OSA treatment may cause weight gain, the long-term effects of CPAP on weight are not well established.
BACKGROUND:
22 23
This study was a post hoc analysis of the Sleep Apnea Cardiovascular Endpoints (SAVE) study, a multicenter, randomized trial of CPAP plus standard care vs standard care alone in adults with a history of cardiac or cerebrovascular events and moderate to severe OSA. Participants with weight, BMI, and neck and waist circumferences measured at baseline and during follow-up were included. Linear mixed models were used to examine sex-specific temporal differences, and a sensitivity analysis compared high CPAP adherers ($ 4 h per night) with propensity-matched control participants.
METHODS:
24 25 26 27 28 29 30 31
A total of 2,483 adults (1,248 in the CPAP group and 1,235 in the control group) were included (mean 6.1 1.5 measures of weight available). After a mean follow-up of 3.78 years, there was no difference in weight change between the CPAP and control groups, for male subjects (mean [95% CI] between-group difference, 0.07 kg [–0.40 to 0.54]; P ¼ .773) or female subjects (mean [95% CI] between-group difference, –0.14 kg [–0.37 to 0.09]; P ¼ .233). Similarly, there were no significant differences in BMI or other anthropometric measures. Although male participants who used CPAP $ 4 h per night gained slightly more weight than matched male control subjects without CPAP (mean difference, 0.38 kg [95% CI, 0.04 to 0.73]; P ¼ .031), there were no between-group differences in other anthropometric variables, nor were there any differences between female high CPAP adherers and matched control subjects.
RESULTS:
32 33 34 35 36 37 38 39 40 41 42
Long-term CPAP use in patients with comorbid OSA and cardiovascular disease does not result in clinically significant weight change.
CONCLUSIONS:
43 44
TRIAL REGISTRY:
45
ClinicalTrials.gov; No.: NCT00738179; URL: www.clinicaltrials.gov.
46 47 48
CHEST 2018; Q7
KEY WORDS:
76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100
-(-):---
101 102
CPAP; long-term; OSA; weight
103
49
104
50
105
51
106
ABBREVIATIONS:
52 53 54 55
Q2
AHI = apnea-hypopnea index; CV = cardiovascular; ESS = Epworth Sleepiness Scale; ODI = oxygen desaturation index AFFILIATIONS: From the Sleep Center (Drs Ou, B. Chen, and Wang), Respiratory and Critical Medical Department, Guangdong Geriatric Institute, Guangdong General Hospital, Guangdong Academy of Medical Sciences, Guangzhou, Guangdong, China; Adelaide Institute
for Sleep Health (Drs Loffler and McEvoy), Flinders University, Adelaide, SA, Australia; The First Affiliated Hospital of Guangzhou Medical University (Dr Luo), State Key Laboratory of Respiratory Disease, Guangzhou, Guangdong, China; The First Affiliated Hospital of Nanjing Medical University (Dr Zhang), Nanjing, Jiangsu, China;
1
chestjournal.org
FLA 5.5.0 DTD CHEST1963_proof 23 October 2018 1:54 am EO: CHEST-18-1406
107 108 109 110
111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130
Obesity is one of the main causes of OSA. The resultant narrowing of the pharyngeal airway predisposes to airflow obstruction during sleep, when compensatory dilator neuromuscular reflexes are suppressed.1-3 It is likely that obesity in OSA is due to a complex interaction of environmental, socioeconomic, and hereditary factors. However, recently reviewed evidence suggests that the relationship may be reciprocal, whereby OSA leads to weight gain by inducing behavioral, metabolic, and/or hormonal changes that adversely affect the balance between energy intake and expenditure.4 Compared with BMI-matched control subjects, patients with OSA have increased circulating levels of leptin and ghrelin, which together may predispose to increased food (energy) intake.5 Studies that have used dietary questionnaires suggest that patients with severe OSA are predisposed toward eating high-fat foods,6
131 132 133
The Second Affiliated Hospital of Soochow University (Dr R. Chen); Instituto do Coracao and Hospital Universitario (Drs Drager and 134 Lorenzi-Filho), Faculty of Medicine, University of São Paulo, São 135 Paulo, Brazil; Christchurch Hospital (Dr Hlavac), Christchurch, New Zealand; West Australian Sleep Disorders Research Institute (Drs 136 McArdle and Mukherjee), Department of Pulmonary Physiology and 137 Sleep Medicine, Sir Charles Gairdner Hospital, Nedlands, WA, 138 Australia; University Hospital of Guadalajara (Dr Mediano), Guadalajara, Spain; Centro de Investigacion Biomedica en Red de Enferme139 dades Respiratorias (CIBERES) (Drs Mediano and Barbe), Madrid, 140 Spain; Respiratory Department (Dr Barbe), Institut de Recerca Bio141 medica de Lleida, Lleida, Spain; The George Institute for Global Health (Dr Anderson), Faculty of Medicine, University of New South Wales, 142 Sydney, NSW, Australia; Neurology Department (Dr Anderson), Royal 143 Prince Alfred Hospital, Sydney Health Partners, Sydney, NSW, 144 Australia; Sleep Health Service (Dr McEvoy), Respiratory and Sleep Services, Southern Adelaide Local Health Network, Bedford Park, SA, 145 Australia; and Flinders Centre for Epidemiology and Biostatistics (Dr 146 Woodman), College of Medicine and Public Health, Flinders Univer147 sity, Adelaide, SA, Australia. 148 FUNDING/SUPPORT: The Sleep Apnea Cardiovascular Endpoints (SAVE) trial was funded by project grants [1006501 and 1060078] 149 from the National Health and Medical Research Council of Australia 150 and by the Respironics Sleep and Respiratory Research Foundation and 151 Philips Respironics. Supplementary funding was provided by Fisher & Paykel Healthcare and the Australasian Sleep Trials Network [enabling 152 grant 343020 from the National Health and Medical Research Coun153 cil]. In-kind donations were provided by Respironics for the CPAP 154 equipment and by ResMed for the sleep apnea diagnostic devices. This 155 Q4 Q5 substudy was supported by a grant [NSFC 81870077] to Q. O. Dr Mukherjee is currently with the Adelaide Institute for Sleep Health, 156 Flinders University, Adelaide, SA, Australia; and the Sleep Health 157 Service, Respiratory and Sleep Services, Southern Adelaide Local 158 Health Network, Bedford Park, SA, Australia. CORRESPONDENCE TO: Qiong Ou, MD, Guangdong General Hospital, 159 Guangdong Academy of Medical Sciences, Guangzhou 510000, 160 Q6 Guangdong, China; e-mail: [email protected] OR R. Doug 161 McEvoy, MD, Adelaide Institute for Sleep Health, Flinders University (Repatriation General Hospital), GPO Box 2100, Adelaide SA 5001, 162 Australia; e-mail: doug.mcevoy@flinders.edu.au 163 Copyright Ó 2018 American College of Chest Physicians. Published by 164 Elsevier Inc. All rights reserved. 165 DOI: https://doi.org/10.1016/j.chest.2018.08.1082
2 Original Research
although no studies have objectively measured actual food intake in those with OSA. Physical activity seems to decline with increasing severity of OSA, associated with excessive sleepiness and obesity,7-9 which may not be reversed with treatment.10 Conversely, resting metabolic rate, a major determinant of energy expenditure, seems to be increased in OSA.11-13 Thus, uncertainty persists regarding the effects of OSA on energy balance (energy intake vs energy expenditure). Because CPAP, the first-line treatment for OSA, improves sleep disturbance, nocturnal hypoxia, sympathetic nervous system activity, and daytime sleepiness, measuring the course of body weight after this treatment may provide insight into the effects of OSA on overall energy balance. Early reports produced conflicting results, with studies indicating weight loss14 and weight gain.15,16 In an attempt to resolve this uncertainty, Drager et al17 undertook a meta-analysis of 25 randomized controlled trials with 3,181 patients. The median study duration in this meta-analysis was 3 months, with follow-up exceeding 6 months in only one of the 25 studies. CPAP treatment was found to promote a small increase in BMI and weight, raising the concern that CPAP might have an unintended detrimental effect on cardiometabolic health if such weight gain were to continue with long-term use. The Sleep Apnea Cardiovascular Endpoints (SAVE) study was a large, international, randomized controlled trial that compared the effects of CPAP treatment plus usual care vs usual care alone on the secondary prevention of serious cardiovascular (CV) events in patients with co-occurring moderate to severe OSA.18 Secondary analysis of the SAVE dataset provides an ideal opportunity to assess whether longterm CPAP treatment promotes weight gain in patients with OSA, as participants had regular anthropometric measurements for a mean follow-up period of 3.78 years. The present article reviews the changes in weight and other anthropometric measurements compared between the CPAP-treated and usual care groups.
166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215
Patients and Methods
216
Design
217
The SAVE study was an international, multicenter, randomized, open-label clinical trial, the details of which have been outlined elsewhere.18,19 In brief, patients aged 45 to 75 years, with a diagnosis of moderate to severe OSA (defined as an oxygen
218
[
-#- CHEST - 2018
FLA 5.5.0 DTD CHEST1963_proof 23 October 2018 1:54 am EO: CHEST-18-1406
]
219 220
221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242
Q8
desaturation index [ODI] of $ 12 of $ 4% oxygen desaturation events per hour during a home overnight screening sleep test [ApneaLink, ResMed]) and a previous diagnosis of coronary or cerebrovascular disease were included from 89 clinical centers in seven countries between December 2008 and November 2013. Patients were excluded if they had severe daytime sleepiness (Epworth Sleepiness Scale [ESS] score > 15), very severe nocturnal hypoxemia (oxygen saturation < 80% for > 10% of the monitoring time), or a predominantly Cheyne-Stokes respiration pattern. Potential participants were required to undergo a 1-week trial of sham CPAP delivered at subtherapeutic pressure to identify those most likely to tolerate mask CPAP treatment and adhere to procedures; those who recorded a daily average sham CPAP usage < 3 h on the machine recording card were excluded. The SAVE trial was conducted in accordance with the amended Declaration of Helsinki. Local institutional review boards or independent ethics committees at recruiting sites approved the protocol (e-Table 1), and all participants provided written informed consent. The trial is registered with ClinicalTrials.gov.20 Procedures Eligible participants were randomly assigned via a secure central server to receive either CPAP treatment plus usual CV care (CPAP group) or usual CV care alone (control group).18 All participants were advised on healthy sleep habits and lifestyle changes (including advice on healthy
276
eating and weight control) to minimize OSA symptoms. In the CPAP group, home pressure titration on an auto-PAP machine (REMstar Auto, M or PR series, Philips Respironics) was undertaken over the first week postrandomization. Data were extracted from recording cards to guide further therapy, with CPAP pressure fixed for longterm use at the 90th percentile of pressure achieved during the 1week auto-PAP trial period.
277 278 279 280 281
Clinic visits were scheduled for all participants at 1, 3, 6, and Q9 12 months and annually thereafter; participants were also contacted by telephone 6 monthly between the annual clinic visits. Additional end-of-study visits were performed between September 2015 and January 2016 on all participants who remained actively enrolled in the trial.
TABLE 1
287 289 290 291 292 293 294 295 296 297 299
] Baseline Characteristics
246 248
285 286
298
244
247
283 284
288
Following randomization, baseline anthropometric measurements of weight, height, and neck, waist, and hip circumferences were taken, and the BMI and waist-to-hip ratio (WHR) were calculated (e-Appendix 1). OSA characteristics included apneahypopnea index (AHI), ODI, mean nocturnal oxygen saturation, and lowest nocturnal oxygen saturation, computed from an overnight recording. Information was also obtained on ESS score, medical history, country of residence, and ethnicity. Anthropometric measurements were repeated at 6 and 12 months, and then annually for the duration of follow-up, with CPAP usage time downloaded from machine data cards at each clinic visit for patients allocated to CPAP.
243 245
282
Q16
CPAP Group (n ¼ 1,248) Characteristic
301
Control Group (n ¼ 1,235)
Male (n ¼ 1,010)
Female (n ¼ 238)
Male (n ¼ 996)
302
Female (n ¼ 239)
249
Age, y
60.8 7.7
63.0 7.1
60.7 8.1
63.4 7.0
250
Obese
307 (30.4)
92 (38.7)
286 (28.7)
82 (34.3)
251
OSA characteristics
300
303 304 305 306
252
AHI, events/h
29 16
28 16
30 16
27 15
307
253
ODI, events/h
27 14
29 15
27 14
308
254
28 14
Average SpO2, %
93.4 1.9
93.3 2.2
93.4 1.9
93.2 2.2
Lowest SpO2, %
77.1 6.0
76.9 5.7
77.3 6.0
76.3 6.7
7.4 3.6
7.1 3.8
7.5 3.5
7.5 3.9
255 256 257
ESS score
258
Medical history
309 310 311 312 313
259
Coronary artery disease
544 (53.9)
89 (37.4)
525 (52.7)
97 (40.6)
260
Cerebrovascular diseasea
424 (42.0)
144 (60.5)
425 (42.7)
132 (55.2)
315
261
42 (4.2)
5 (2.1)
46 (4.6)
10 (4.2)
316
262
Coronary and cerebrovascular disease
263
Diabetes mellitus
292 (28.9)
79 (33.2)
284 (28.5)
79 (33.1)
318 320
264 265
314
317 319
Country Australia
92 (9.1)
11 (4.6)
95 (9.5)
13 (5.4)
Brazil
88 (8.7)
37 (15.5)
90 (9.0)
35 (14.6)
268
China
643 (63.7)
155 (65.1)
615 (61.7)
156 (65.3)
269
India
45 (4.5)
9 (3.8)
44 (4.4)
14 (5.9)
270
New Zealand
40 (4.0)
8 (3.4)
41 (4.1)
2 (0.8)
325
271
Spain
101 (10.0)
18 (7.6)
109 (10.9)
19 (7.9)
326
272
United States
1 (0.1)
0
2 (0.2)
0
266 267
273 274 275
321 322 323 324
327
Data are presented as mean SD or No. (percentiles). Subjects indicated as obese had a BMI $ 30 kg/m2. AHI ¼ apnea-hypopnea index; ODI ¼ oxygen desaturation index; SpO2 ¼ blood oxygen saturation. a Stroke or transient ischemic attack.
3
chestjournal.org
FLA 5.5.0 DTD CHEST1963_proof 23 October 2018 1:54 am EO: CHEST-18-1406
328 Q17
329 330
331
Statistical Analysis
332
Summary data for each group are presented as mean SD for normally distributed variables, and median (interquartile range) for non-normally distributed variables. We excluded participants with extreme outlying values for any of the five anthropometric variables (weight, BMI, neck or waist circumference, or WHR) at any time point according to a change across the study duration of > 4 SDs; the goal was to eliminate probable typographical errors or effects of unrelated health issues and to ensure a consistent case base for all analyses. A linear mixed effects model was used to determine the mean difference between treatments for each of the variables during follow-up, with fixed effect terms used for baseline values, treatment, sex, visit (categorical), and interactions for sex baseline value, treatment visit, treatment sex, visit sex, and treatment visit sex. Interaction effects between treatment and each of the main effects included in the final model were also tested. A random intercept term for each participant was also included, and the estimated mean between-group differences were estimated with the marginal treatment effect for each sex included in the same model.
333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349
The predictors of weight change were also determined for end-of-study minus baseline weight. A simple linear regression was first tested for crude associations, with inclusion of age, sex, follow-up duration,
country, baseline weight, AHI, ODI, ESS score, and presence of coronary or cerebrovascular CV disease, previous or current CV medications, hypertension, diabetes mellitus, myocardial infarction, angina, heart failure, valvular heart disease, percutaneous coronary revascularization with or without a stent, coronary artery bypass graft, stroke and transient ischemic attack, and treatment group allocation. Variables that were significant at P < .20 were then entered into a multivariate linear regression model that a priori included follow-up duration as a predictor to determine the independent predictors of weight change. All estimated differences were considered significant for P < .05 using two-tailed hypothesis tests. Finally, a sensitivity analysis was conducted to determine whether higher levels of CPAP adherence were associated with weight change. Anthropometric values were compared between the two previously reported groups of propensity score-matched patients19 after excluding those with unsatisfactory anthropometric measurements. Details of the method whereby “CPAP-adherent” patients (defined as a mean $ 4 h of treatment per day over the study period) were matched 1:1 with control patients who never used CPAP are provided in e-Table 2 of the main SAVE study report.19 All analyses were performed by using Stata version 14.2 (StataCorp).
386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404
350
405
351
406
352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381
Results There were 2,717 participants in the SAVE study, but 30 were excluded because of withdrawal of consent, nonadherence to the protocol, or data irregularities identified during monitoring at one site. Eight patients were removed from these analyses because their baseline anthropometric measurements were incomplete, and another 196 were excluded because of extreme values (> 4 SDs) recorded during follow-up. Thus, a total of 2,483 subjects (1,248 in the CPAP group and 1,235 in the control group) were included with a mean 6.1 1.5 (range, 1-8) measures of weight, including the baseline visit. Their mean length of follow-up was 3.78 1.46 years. Table 1 outlines the baseline characteristics of participants according to sex; male and female subjects in the CPAP group used their device for a daily mean of 3.33 2.29 h and 3.15 2.19 h, respectively, over the study duration. Table 2 describes the summary measures for each of the anthropometric variables and shows that there were no significant differences in the estimated mean differences between treatment groups for either male or female subjects. Figure 1 illustrates that median weights remained stable for both CPAP and control participants, in both male and female subjects. There were no significant treatment time interactions.
382 383 384
Table 3 presents the results of the multivariate linear regression analysis for the change in weight between
baseline and the end of follow-up. Older age (P < .001), higher baseline weight (P < .001), and recruitment in China (P < .001) or India (P ¼ .007) vs Australia were associated with significant weight loss. Being male (P ¼ .003), having previous percutaneous coronary revascularization with or without a stent (P ¼ .001), and recruitment in New Zealand (P ¼ .006 compared with Australia) were associated with significant weight gain. Interactions between treatment status and other variables were not significant, indicating that there was no subgroup in which CPAP had a significant effect. Despite different patterns of weight change indicated by the regression coefficients according to country, there is no evidence that CPAP treatment had a differential effect on weight between countries. A sensitivity analysis compared good CPAP adherers ($ 4 h per night; n ¼ 516) with propensity-matched control subjects (n ¼ 531) (e-Table 2). Each individual contributed 6.1 1.5 weight measures (range, 1-8), including the baseline visit. Table 4 describes the summary measures for each anthropometric outcome as well as the estimated mean difference between groups. There was a significantly higher weight gain after CPAP treatment compared with control for male subjects: the mean (95% CI) baseline-adjusted difference was 0.38 kg (0.04-0.73) (P ¼ .031). No significant differences were found for any of the other anthropometric measures, either in male or female subjects.
385
407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440
4 Original Research
[
-#- CHEST - 2018
FLA 5.5.0 DTD CHEST1963_proof 23 October 2018 1:54 am EO: CHEST-18-1406
]
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
chestjournal.org
FLA 5.5.0 DTD CHEST1963_proof 23 October 2018 1:54 am EO: CHEST-18-1406
TABLE 2
] Anthropometric Variables for Each Treatment Group, Separated According to Sex CPAP Group
Variable
Control Group
Baseline
Follow-up
Baseline
Follow-up
Median (IQR) (No.)
Median (IQR) (No.)
Median (IQR) (No.)
Median (IQR) (No.)
Adjusted D in Change From Baseline (CPAP vs Control), Mean (95% CI)
P
Weight, kg Female
70 (64-80) (238)
70 (63-79) (1,049)
70 (62-77) (239)
69.5 (61-76) (1,084)
Male
80 (73-90) (1,010)
80 (73-90) (4,737)
80 (73-90) (996)
80 (74-89) (4,684)
0.07 (–0.40 to 0.54)
.773
–0.14 (–0.37 to 0.09)
.233
Mean SD (No.)
Mean SD (No.)
Mean SD (No.)
Mean SD (No.)
Female
29.58 5.78 (238)
29.35 5.62 (1,049)
28.71 4.99 (239)
28.34 4.94 (1,084)
0.035 (–0.13 to 0.20)
.68
Male
28.43 3.96 (1,010)
28.38 4.06 (4,737)
28.39 3.93 (996)
28.45 4.07 (4,684)
–0.05 (-0.13 to 0.030)
.22
0.124 (–0.104 to 0.352)
.29
BMI, kg/m2
Neck circumference, cm Female
37.5 3.8 (235)
37.3 3.8 (679)
36.9 3.9 (239)
36.8 4.0 (695)
Male
41.5 3.6 (1,004)
41.2 3.5 (3,008)
41.5 3.7 (995)
41.4 3.8 (2,989)
–0.108 (–0.218 to 0.0017)
.054
Waist circumference, cm 98.5 13.3 (236)
98.5 13.8 (681)
101.5 11.1 (1,009)
101.4 11.2 (3,007)
Female
0.92 0.08 (235)
0.93 0.09 (679)
0.90 0.08 (239)
0.92 0.08 (696)
Male
0.97 0.06 (1,003)
0.97 0.07 (3,005)
0.96 0.07 (995)
0.97 0.07 (2,990)
Female Male
96.2 13.2 (239)
96.1 13.6 (696)
101.3 11.1 (996)
101.4 11.6 (2,995)
0.013 (–0.58 to 0.61)
.97
–0.015 (–0.30 to 0.27)
.92
–0.002 (–0.007 to 0.004)
.55
–0.0004 (–0.003 to 0.002)
.78
WHR
No. indicates the total number of measurements performed during either baseline or follow-up. A single measure per subject was performed at baseline, but multiple measures were performed during follow-up. IQR ¼ interquartile range; WHR ¼ waist-to-hip ratio.
5 497
496
499
498
501
500
502
503
505
504
507
506
509
508
511
510
512
513
515
514
517
516
519
518
521
520
522
523
525
524
527
526
529
528
531
530
532
533
535
534
537
536
539
538
541
540
542
543
545
544
547
546
548
550
549
551 552
140
553 555 556 557 558
*
*
120
*
*
*
*
Female
Weight (kg)
554
*
*
100 80 60
559 560
40
561 562
140
563 565 566 567 568 570
575 576 577 578 579 580
En
d
of
St
ud
m
y
o
o m
60
m
o
48
m
o 36
o
24
m 12
m
o
e 6
se
lin Q21 Q18
Ba
print & web 4C=FPO
574
80
40
571 573
100
60
569
572
120 Male
Weight (kg)
564
Control
CPAP
Figure 1 – Boxplots of observed weight at baseline and during follow-up according to treatment group and sex. The end-of-study visit indicates the last measurement for each individual. Each box shows median and 25th to 75th percentiles.
581
Discussion
582
The principal finding of this large clinical trial was that there was no clinically significant weight change
583 584 585
TABLE 3
586
Variable
587 588 589
b (95% CI)a
Sex (1 ¼ female, 2 ¼ male)
591
Treatment group (1 ¼ CPAP, 2 ¼ control)
592
Duration in study (y)
593
PCR (yes vs no)
594
Country Australia
609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638
641 642 643
.003
–0.049 (–0.065 to –0.032)
607 608
639
< .001
0.789 (0.268 to 1.310)
606
640 Pa
–0.058 (–0.084 to –0.031)
Age (y)
Baseline weight (kg)
596
Only one other randomized controlled trial of CPAP vs usual care in overweight or obese patients with moderate to severe OSA recorded weight and BMI over a similar period (median follow-up, 4 years)21 as our study. Patients allocated to the control group were found to have had a significant decrease in BMI, whereas those allocated to CPAP treatment experienced no significant change in BMI over time. Q10 The reported between-group mean difference in change in BMI for CPAP vs control in that trial was 0.25 kg/m2 (95% CI, 0.04-0.46), and the corresponding difference in change in weight (F. Barbe, personal communication) was 0.51 kg (95% CI, 0.13-1.25). Neither the study by Barbe et al21
] Regression Coefficients From a Multivariate Linear Regression for Change in Weight
590
595
in patients with established CV disease and comorbid OSA over several years, and there was no differential effect of CPAP on other anthropometric measures. In a subanalysis, male subjects, but not female subjects, who achieved high use of CPAP each night had a small (< 400 g) but statistically significant increase in weight compared with matched control patients but without changes in other anthropometric indices such as BMI or neck or waist circumference. Taken together, these results indicate that long-term CPAP treatment is unlikely to exacerbate the problems of overweight and obesity that are common among patients with OSA. Such a small change in weight, even with good adherence over several years, is highly unlikely to have any serious clinical ramifications.
644
< .001
645
0.105 (–0.281 to 0.491)
.594
–0.103 (–0.278 to 0.072)
.249
647
0.698 (0.270 to 1.126)
.001
648
646
649 650
1.00
651
Brazil
–0.82 (–1.82 to 0.18)
598
China
–1.60 (–2.38 to –0.808)
< .001
599
India
–1.81 (–3.12 to –0.502)
.007
600
New Zealand
1.75 (0.51 to 3.00)
.006
655
601
Spain
0.16 (–0.84 to 1.15)
.758
656
602
United States
3.73 (–2.74 to 10.21)
.258
597
603 604 605
.107
652 653 654
657
The b-coefficients are adjusted for all other variables in the table. Age, sex, treatment group, and duration in study were included in the model a priori. PCR ¼ percutaneous coronary revascularization with or without a stent. a Using a multivariate linear regression for change in weight between baseline and end of follow-up.
6 Original Research
[
-#- CHEST - 2018
FLA 5.5.0 DTD CHEST1963_proof 23 October 2018 1:54 am EO: CHEST-18-1406
]
658 659 660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
chestjournal.org
FLA 5.5.0 DTD CHEST1963_proof 23 October 2018 1:54 am EO: CHEST-18-1406
TABLE 4 ]
Change in Anthropometric Variables in Patients With Good CPAP Compliance and Propensity-Matched Control Subjects, Separated According to Sex CPAP Group
Variable
Adjusted D in Change From Baseline (CPAP vs Control), Mean (95% CI)
Control Group
Baseline
Follow-up
Baseline
Follow-up
Median (IQR) (No.)
Median (IQR) (No.)
Median (IQR) (No.)
Median (IQR) (No.)
P
Weight, kg –0.217 (–0.97 to 0.53)
Female
70.0 (65.0-79.50) (92)
70.0 (63.0-78.0) (445)
70.0 (61.0-76.0) (99)
70.0 (60.0-76.0) (415)
Male
82.0 (74.0-91.0) (439)
82.0 (74.0-92.0) (2,196)
82.0 (74.0-90.0) (417)
82.0 (74.0-90.0) (1,955)
Mean SD (No.)
Mean SD (No.)
Mean SD (No.)
Mean SD (No.)
BMI, kg/m
.569
0.383 (0.035 to 0.730)
.031a
–0.059 (–0.324 to 0.207)
.665
2
Female
29.5 5.0 (92)
29.3 5.1 (445)
29.0 4.9 (99)
28.7 4.9 (415)
Male
28.8 4.1 (439)
28.9 4.2 (2,196)
28.8 4.2 (417)
28.8 4.3 (1,955)
0.123 (–0.001 to 0.246)
.052
Female
37.4 3.8 (90)
37.1 3.6 (285)
36.3 3.6 (99)
36.2 3.7 (272)
0.122 (–0.239 to 0.483)
.507
Male
41.8 3.6 (438)
41.6 3.6 (1,387)
41.8 3.6 (416)
41.6 3.7 (1,267)
0.043 (–0.123 to 0.209)
.610
98.5 12.7 (91)
98.9 14.2 (285)
97.0 12.9 (99)
95.9 13.2 (272)
102.0 11.1 (439)
102.5 11.3 (1,386)
102.4 11.8 (417)
102.5 12.2 (1,269)
Female
0.92 0.08 (90)
0.93 0.09 (285)
0.91 0.09 (99)
0.91 0.09 (272)
Male
0.97 0.06 (438)
0.97 0.07 (1,386)
0.97 0.07 (416)
0.97 0.07 (1,268)
Neck circumference, cm
Waist circumference, cm Females Male
0.89 (–0.06 to 1.84)
.068
0.290 (–0.154 to 0.733)
.200
0.004 (–0.005 to 0.013)
.357
–0.000 (–0.004 to 0.004)
.862
WHR
No. indicates the total number of measurements performed during either baseline or follow-up. A single measure per subject was performed at baseline, but multiple measures were performed during follow-up. See Table 1 legend for expansion of abbreviations. a P < .05.
7 717
716
719
718
721
720
722
723
725
724
727
726
729
728
731
730
732
733
735
734
737
736
739
738
741
740
742
743
745
744
747
746
749
748
751
750
752
753
755
754
757
756
759
758
761
760
762
763
765
764
767
766
768
770
769
771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823
nor the present study found evidence of clinically significant weight gain among CPAP-treated patients with long-term follow-up. The small differences in weight change noted between the CPAP and control groups in the two studies may relate to differences in the patient characteristics and levels of CPAP adherence achieved; for example, mean age (52 vs 61 years), prevalence of preexisting CV disease (0% vs 100%), and median CPAP use (5.00 h [interquartile range, 2.18-6.25 h] vs 3.26 h [interquartile range, 1.26-5.08 h]) per night. The meta-analysis by Drager et al17 had raised concern that CPAP may cause an increase in weight gain, although the scale of the effect was small (0.17 kg; 95% CI, 0.10-0.24). It is possible that this difference between our observations and that of the previous shorter studies in the meta-analysis may be related to differences in study populations such as baseline obesity, ethnicity, age, and chronic comorbidities. Encouragingly, the relatively small short-term changes seen in the meta-analysis17 do not seem to have been compounded during long-term follow-up. It is reassuring that we found no clinically significant weight gain in CPAP-treated patients with OSA who, in the main, were overweight or obese and had proven (mainly atherosclerotic) CV disease. Conversely, it is noteworthy and arguably disappointing from a clinical perspective that weight did not decrease in the male or female subjects in either the CPAP or the control group. All SAVE participants had their CV risk factors managed according to national clinical guidelines, including advice on weight loss for secondary CV prevention in overweight/obese patients.22-25 They were also given verbal lifestyle advice regarding weight control to minimize the severity of OSA. The absence of weight loss likely reflects the widely acknowledged dual problems of suboptimal adherence to clinical guidelines by practitioners26 and the difficulties that patients experience in adopting the lifestyle changes that are necessary to achieve sustained weight loss, as discussed elsewhere.25,27 Most people with OSA are overweight or obese and have a high frequency of CV risk factors (eg, hypertension,28 diabetes mellitus29) and disease.30
Results of the Wisconsin Sleep Cohort Study highlight the potential benefit of weight loss in the management of OSA: a 10% reduction in weight was associated with a 26% drop in AHI,31 and weight loss is recommended in OSA management guidelines32,33 and for CV risk reduction.23,25,27 Our study findings indicate that CPAP, and the symptom benefits that flow from it, cannot be relied upon to lead to meaningful weight loss.17 Enhanced weight management is therefore needed for overweight patients with OSA receiving long-term CPAP treatment and is ideally delivered as part of a personalized chronic disease management program.34 Indeed, combining weight loss with CPAP therapy for OSA seems necessary to improve CV risk markers.35 The importance of encouraging weight loss in those with OSA and CV disease living in affluent societies is emphasized in our multivariable regression model, which showed that male subjects were less likely to lose weight than female subjects, as were those from Australia and New Zealand compared with India and China, independent of CPAP status. These differences observed irrespective of CPAP treatment may relate to national differences in dietary patterns, health-care systems, socioeconomic status, or other unknown lifestyle factors.36 We recognize that our study has several limitations. First, the study population was restricted to older patients with moderate to severe OSA and established CV disease but with very severe cases (nocturnal oxygen desaturation < 80% for > 10% of recording time) and marked self-reported daytime sleepiness (ESS score > 15) excluded. Thus, our results may not be generalizable to all patients with OSA. However, the large size of the study and prospective recording of weight, CPAP adherence, and other measures over several years allowed sensitivity analyses to be undertaken according to levels of adherence to CPAP.
826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872
Conclusions The present study showed that patients with moderate to severe OSA and co-existing CV or cerebrovascular disease had no significant changes in BMI or neck or waist circumference from the use of CPAP over several years. We found no evidence that patients
873 874 875 876 877 878
824
879
825
880
8 Original Research
[
-#- CHEST - 2018
FLA 5.5.0 DTD CHEST1963_proof 23 October 2018 1:54 am EO: CHEST-18-1406
]
881
with moderate to severe OSA and coexisting CV disease are susceptible to clinically concerning
882 883
936
long-term weight change from the use of CPAP therapy.
937 938
884
939
885
940
886
Acknowledgments
887
Author contributions: R. J. W. and K. A. L. had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Q. O., B. C., C. S. A., R. D. M., and R. J. W. planned the study. Q. O., Y. L., X. Z., R. C., Q. W., L. F. D., G. L.-F., M. H., N. M., S. M., O. M., F. B., C. S. A., and R. D. M. recruited participants and collected data and samples. R. J. W. and K. A. L. conducted statistical analyses. Q. O., B. C., R. D. M., R. J. W., K. A. L., and C. S. A. interpreted data and drafted the manuscript. All authors contributed to drafting the manuscript, and read and approved the manuscript for submission.
888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903
Q12
904 905 906 907 908 909 910 911
Q13
912 913 914 915 916
Q14
917
Q11
918 920 921 922 923 925 926 927 928 929 930 931 932 933 934 935
Role of sponsors: The sponsor had no role in the design of the study, the collection and analysis of the data, or the preparation of the manuscript. Collaborators: The SAVE trial was conducted by a large team of investigators. Details and affiliations are given in eAppendix 2.
919
924
Financial/nonfinancial disclosures: The authors have reported to CHEST the following: L. F. D. received research study equipment from Philips Respironics. M. H. is a shareholder of Cansleep Ltd. N. M. has received research funding support from Oventus Pty Ltd (Brisbane, Australia), Nyxoah Pty Ltd (Mont-Saint Guibert, Belgium), and Zelda Therapeutics Pty Ltd (Australia). F. B. received research grants from ResMed and Philips. R. D. M. reports grants and nonfinancial support from Philips Respironics, nonfinancial support from ResMed, grants from the National Health and Medical Research Council of Australia, and grants from Fisher & Paykel, during the conduct of the study; nonfinancial support from Air Liquide; and speaker fees from ResMed. None declared (Q. O., B. C., K. A. L., Y. L., X. Z., R. C., Q. W., G. L.-F., S. M., O. M., C. S. A., R. J. W.).
Q19
Additional information: The e-Appendixes and e-Tables can be found in the Supplemental Materials section of the online article.
References 1. Genta PR, Schorr F, Eckert DJ, et al. Upper airway collapsibility is associated with obesity and hyoid position. Sleep. 2014;37(10):1673-1678. 2. Ryan S, Crinion SJ, McNicholas WT. Obesity and sleep-disordered breathing— when two ’bad guys’ meet. QJM. 2014;107(12):949-954. 3. Wheatley JR, Mezzanotte WS, Tangel DJ, White DP. Influence of sleep on genioglossus muscle activation by negative
pressure in normal men. Am Rev Respir Dis. 1993;148(3):597-605. 4. Shechter A. Obstructive sleep apnea and energy balance regulation: a systematic review. Sleep Med Rev. 2017;34:59-69. 5. Harsch IA, Konturek PC, Koebnick C, et al. Leptin and ghrelin levels in patients with obstructive sleep apnoea: effect of CPAP treatment. Eur Respir J. 2003;22(2): 251-257. 6. Cao Y, Wittert G, Taylor AW, Adams R, Shi Z. Associations between macronutrient intake and obstructive sleep apnoea as well as self-reported sleep symptoms: results from a cohort of community dwelling Australian men. Nutrients. 2016;8(4):207. 7. Basta M, Lin HM, Pejovic S, Sarrigiannidis A, Bixler E, Vgontzas AN. Lack of regular exercise, depression, and degree of apnea are predictors of excessive daytime sleepiness in patients with sleep apnea: sex differences. J Clin Sleep Med. 2008;4(1):19-25. 8. Lopes C, Esteves AM, Bittencourt LR, Tufik S, Mello MT. Relationship between the quality of life and the severity of obstructive sleep apnea syndrome. Braz J Med Biol Res. 2008;41(10):908-913. 9. Quan SF, O’Connor GT, Quan JS, et al. Association of physical activity with sleepdisordered breathing. Sleep Breath. 2007;11(3):149-157. 10. West SD, Kohler M, Nicoll DJ, Stradling JR. The effect of continuous positive airway pressure treatment on physical activity in patients with obstructive sleep apnoea: a randomised controlled trial. Sleep Med. 2009;10(9): 1056-1058. 11. Ryan CF, Love LL, Buckley PA. Energy expenditure in obstructive sleep apnea. Sleep. 1995;18(3):180-187. 12. Stenlof K, Grunstein R, Hedner J, Sjostrom L. Energy expenditure in obstructive sleep apnea: effects of treatment with continuous positive airway pressure. Am J Physiol. 1996;271(6 pt 1): E1036-E1043. 13. Ucok K, Aycicek A, Sezer M, et al. Resting metabolic rate and anthropometric measurements in male sleep apnea patients. Intern Med. 2011;50(8):833-838. 14. Loube DI, Loube AA, Erman MK. Continuous positive airway pressure treatment results in weight loss in obese and overweight patients with obstructive sleep apnea. J Am Diet Assoc. 1997;97(8): 896-897. 15. Quan SF, Budhiraja R, Clarke DP, et al. Impact of treatment with continuous positive airway pressure (CPAP) on weight in obstructive sleep apnea. J Clin Sleep Med. 2013;9(10):989-993.
16. Redenius R, Murphy C, O’Neill E, AlHamwi M, Zallek SN. Does CPAP lead to change in BMI? J Clin Sleep Med. 2008;4(3):205-209. 17. Drager LF, Brunoni AR, Jenner R, Lorenzi-Filho G, Bensenor IM, Lotufo PA. Effects of CPAP on body weight in patients with obstructive sleep apnoea: a meta-analysis of randomised trials. Thorax. 2015;70(3):258-264. 18. Antic NA, Heeley E, Anderson CS, et al. The Sleep Apnea cardioVascular Endpoints (SAVE) trial: rationale, ethics, design, and progress. Sleep. 2015;38(8): 1247-1257. 19. McEvoy RD, Antic NA, Heeley E, et al. CPAP for prevention of cardiovascular events in obstructive sleep apnea. N Engl J Med. 2016;375(10):919-931. 20. ClinicalTrials.gov. NCT00738179; ACTRN 126080000409370. https:// clinicaltrials.gov/ct2/show/ NCT0073817 9. 21. Barbe F, Duran-Cantolla J, Sanchez-de-laTorre M, et al. Effect of continuous positive airway pressure on the incidence of hypertension and cardiovascular events in nonsleepy patients with obstructive sleep apnea: a randomized controlled trial. JAMA. 2012;307(20):2161-2168. 22. National Vascular Disease Prevention Alliance. Guidelines for the Management of Absolute Cardiovascular Disease Risk. National Stroke Foundation; 2012. 23. National Heart Foundation of Australia; Cardiac Society of Australia and New Zealand. Reducing Risk in Heart Disease: An Expert Guide to Clinical Practice for Secondary Prevention of Coronary Heart Disease. Melbourne, Australia: National Heart Foundation of Australia; 2012. 9781-74345-016-1. 24. Piepoli MF, Hoes AW, Agewall S, et al. 2016 European guidelines on cardiovascular disease prevention in clinical practice: the Sixth Joint Task Force of the European Society of Cardiology and Other Societies on Cardiovascular Disease Prevention in Clinical Practice (constituted by representatives of 10 societies and by invited experts) developed with the special contribution of the European Association for Cardiovascular Prevention & Rehabilitation (EACPR). Eur Heart J. 2016;37(29):2315-2381. 25. Smith SC Jr, Benjamin EJ, Bonow RO, et al. AHA/ACCF secondary prevention and risk reduction therapy for patients with coronary and other atherosclerotic vascular disease: 2011 update: a guideline from the American Heart Association and American College of Cardiology Foundation. Circulation. 2011;124(22): 2458-2473. 26. Barnes ER, Theeke LA, Mallow J. Impact of the Provider and Healthcare team
9
chestjournal.org
FLA 5.5.0 DTD CHEST1963_proof 23 October 2018 1:54 am EO: CHEST-18-1406
941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990
991
Adherence to Treatment Guidelines (PHAT-G) intervention on adherence to national obesity clinical practice guidelines in a primary care centre. J Eval Clin Pract. 2015;21(2): 300-306.
992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005
Q15
30. Molnar MZ, Mucsi I, Novak M, et al. Association of incident obstructive sleep apnoea with outcomes in a large cohort of US veterans. Thorax. 2015;70(9):888-895.
27. World Heart Federation. Secondary Cardiovascular Disease Prevention and Control. Switzerland: 2014.
31. Peppard PE, Young T, Palta M, Dempsey J, Skatrud J. Longitudinal study of moderate weight change and sleepdisordered breathing. JAMA. 2000;284(23):3015-3021.
28. Marin JM, Agusti A, Villar I, et al. Association between treated and untreated obstructive sleep apnea and risk of hypertension. JAMA. 2012;307(20):21692176.
32. Epstein LJ, Kristo D, Strollo PJ Jr, et al. Clinical guideline for the evaluation, management and long-term care of obstructive sleep apnea in adults. J Clin Sleep Med. 2009;5(3):263-276.
29. Botros N, Concato J, Mohsenin V, Selim B, Doctor K, Yaggi HK. Obstructive sleep apnea as a risk factor for type 2 diabetes. Am J Med. 2009;122(12):11221127.
33. Qaseem A, Holty JE, Owens DK, et al. Management of obstructive sleep apnea in adults: a clinical practice guideline from the American College of Physicians. Ann Intern Med. 2013;159(7):471-483.
34. Heatley EM, Harris M, Battersby M, McEvoy RD, Chai-Coetzer CL, Antic NA. Obstructive sleep apnoea in adults: a common chronic condition in need of a comprehensive chronic condition management approach. Sleep Med Rev. 2013;17(5):349-355. 35. Chirinos JA, Gurubhagavatula I, Teff K, et al. CPAP, weight loss, or both for obstructive sleep apnea. N Engl J Med. 2014;370(24):2265-2275.
1046 1047 1048 1049 1050 1051 1052 1053 1054
36. Kim R, Kawachi I, Coull BA, Subramaniam SV. Contribution of socioeconomic factors to the variation in body-mass index in 58 low-income and middle-income countries: an econometric analysis of multilevel data. Lancet Glob Health. 2018;6(7):e777e786.
1055 1056 1057 1058 1059 1060
1006
1061
1007
1062
1008
1063
1009
1064
1010
1065
1011
1066
1012
1067
1013
1068
1014
1069
1015
1070
1016
1071
1017
1072
1018
1073
1019
1074
1020
1075
1021
1076
1022
1077
1023
1078
1024
1079
1025
1080
1026
1081
1027
1082
1028
1083
1029
1084
1030
1085
1031
1086
1032
1087
1033
1088
1034
1089
1035
1090
1036
1091
1037
1092
1038
1093
1039
1094
1040
1095
1041
1096
1042
1097
1043
1098
1044
1099
1045
1100
10 Original Research
[
-#- CHEST - 2018
FLA 5.5.0 DTD CHEST1963_proof 23 October 2018 1:54 am EO: CHEST-18-1406
]
1
Original Research
]
56
2
57
3
58
4
59
5
60
The Effects of Long-term CPAP on Weight Change in Patients With Comorbid OSA and Cardiovascular Disease
6 7 8 9 10 11
61 62 63 64 65 66
Data From the SAVE Trial
12
67
13 14
68 Q20
15 16 17
Qiong Ou, MD; Baixin Chen, MMed; Kelly A. Loffler, PhD; Yuanming Luo, MD, PhD; Xilong Zhang, PhD; Rui Chen, PhD;
69
Qian Wang, MMed; Luciano F. Drager, MD, PhD; Geraldo Lorenzi-Filho, MD, PhD; Michael Hlavac, PhD;
70 71
Nigel McArdle, MD; Sutapa Mukherjee, MD, PhD; Olga Mediano, MD, PhD; Ferran Barbe, MD; Q1
18
72
Craig S. Anderson, MD, PhD; R. Doug McEvoy, MD; Richard J. Woodman, PhD; on behalf of the SAVE investigators
73
19
74
20
75
21
Although recent evidence suggests that OSA treatment may cause weight gain, the long-term effects of CPAP on weight are not well established.
BACKGROUND:
22 23
This study was a post hoc analysis of the Sleep Apnea Cardiovascular Endpoints (SAVE) study, a multicenter, randomized trial of CPAP plus standard care vs standard care alone in adults with a history of cardiac or cerebrovascular events and moderate to severe OSA. Participants with weight, BMI, and neck and waist circumferences measured at baseline and during follow-up were included. Linear mixed models were used to examine sex-specific temporal differences, and a sensitivity analysis compared high CPAP adherers ($ 4 h per night) with propensity-matched control participants.
METHODS:
24 25 26 27 28 29 30 31
A total of 2,483 adults (1,248 in the CPAP group and 1,235 in the control group) were included (mean 6.1 1.5 measures of weight available). After a mean follow-up of 3.78 years, there was no difference in weight change between the CPAP and control groups, for male subjects (mean [95% CI] between-group difference, 0.07 kg [–0.40 to 0.54]; P ¼ .773) or female subjects (mean [95% CI] between-group difference, –0.14 kg [–0.37 to 0.09]; P ¼ .233). Similarly, there were no significant differences in BMI or other anthropometric measures. Although male participants who used CPAP $ 4 h per night gained slightly more weight than matched male control subjects without CPAP (mean difference, 0.38 kg [95% CI, 0.04 to 0.73]; P ¼ .031), there were no between-group differences in other anthropometric variables, nor were there any differences between female high CPAP adherers and matched control subjects.
RESULTS:
32 33 34 35 36 37 38 39 40 41 42
Long-term CPAP use in patients with comorbid OSA and cardiovascular disease does not result in clinically significant weight change.
CONCLUSIONS:
43 44
TRIAL REGISTRY:
45
ClinicalTrials.gov; No.: NCT00738179; URL: www.clinicaltrials.gov.
46 47 48
CHEST 2018; Q7
KEY WORDS:
76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100
-(-):---
101 102
CPAP; long-term; OSA; weight
103
49
104
50
105
51
106
ABBREVIATIONS:
52 53 54 55
Q2
AHI = apnea-hypopnea index; CV = cardiovascular; ESS = Epworth Sleepiness Scale; ODI = oxygen desaturation index AFFILIATIONS: From the Sleep Center (Drs Ou, B. Chen, and Wang), Respiratory and Critical Medical Department, Guangdong Geriatric Institute, Guangdong General Hospital, Guangdong Academy of Medical Sciences, Guangzhou, Guangdong, China; Adelaide Institute
for Sleep Health (Drs Loffler and McEvoy), Flinders University, Adelaide, SA, Australia; The First Affiliated Hospital of Guangzhou Medical University (Dr Luo), State Key Laboratory of Respiratory Disease, Guangzhou, Guangdong, China; The First Affiliated Hospital of Nanjing Medical University (Dr Zhang), Nanjing, Jiangsu, China;
1
chestjournal.org
FLA 5.5.0 DTD CHEST1963_proof 23 October 2018 1:54 am EO: CHEST-18-1406
107 108 109 110
111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130
Obesity is one of the main causes of OSA. The resultant narrowing of the pharyngeal airway predisposes to airflow obstruction during sleep, when compensatory dilator neuromuscular reflexes are suppressed.1-3 It is likely that obesity in OSA is due to a complex interaction of environmental, socioeconomic, and hereditary factors. However, recently reviewed evidence suggests that the relationship may be reciprocal, whereby OSA leads to weight gain by inducing behavioral, metabolic, and/or hormonal changes that adversely affect the balance between energy intake and expenditure.4 Compared with BMI-matched control subjects, patients with OSA have increased circulating levels of leptin and ghrelin, which together may predispose to increased food (energy) intake.5 Studies that have used dietary questionnaires suggest that patients with severe OSA are predisposed toward eating high-fat foods,6
131 132 133
The Second Affiliated Hospital of Soochow University (Dr R. Chen); Instituto do Coracao and Hospital Universitario (Drs Drager and 134 Lorenzi-Filho), Faculty of Medicine, University of São Paulo, São 135 Paulo, Brazil; Christchurch Hospital (Dr Hlavac), Christchurch, New Zealand; West Australian Sleep Disorders Research Institute (Drs 136 McArdle and Mukherjee), Department of Pulmonary Physiology and 137 Sleep Medicine, Sir Charles Gairdner Hospital, Nedlands, WA, 138 Australia; University Hospital of Guadalajara (Dr Mediano), Guadalajara, Spain; Centro de Investigacion Biomedica en Red de Enferme139 dades Respiratorias (CIBERES) (Drs Mediano and Barbe), Madrid, 140 Spain; Respiratory Department (Dr Barbe), Institut de Recerca Bio141 medica de Lleida, Lleida, Spain; The George Institute for Global Health (Dr Anderson), Faculty of Medicine, University of New South Wales, 142 Sydney, NSW, Australia; Neurology Department (Dr Anderson), Royal 143 Prince Alfred Hospital, Sydney Health Partners, Sydney, NSW, 144 Australia; Sleep Health Service (Dr McEvoy), Respiratory and Sleep Services, Southern Adelaide Local Health Network, Bedford Park, SA, 145 Australia; and Flinders Centre for Epidemiology and Biostatistics (Dr 146 Woodman), College of Medicine and Public Health, Flinders Univer147 sity, Adelaide, SA, Australia. 148 FUNDING/SUPPORT: The Sleep Apnea Cardiovascular Endpoints (SAVE) trial was funded by project grants [1006501 and 1060078] 149 from the National Health and Medical Research Council of Australia 150 and by the Respironics Sleep and Respiratory Research Foundation and 151 Philips Respironics. Supplementary funding was provided by Fisher & Paykel Healthcare and the Australasian Sleep Trials Network [enabling 152 grant 343020 from the National Health and Medical Research Coun153 cil]. In-kind donations were provided by Respironics for the CPAP 154 equipment and by ResMed for the sleep apnea diagnostic devices. This 155 Q4 Q5 substudy was supported by a grant [NSFC 81870077] to Q. O. Dr Mukherjee is currently with the Adelaide Institute for Sleep Health, 156 Flinders University, Adelaide, SA, Australia; and the Sleep Health 157 Service, Respiratory and Sleep Services, Southern Adelaide Local 158 Health Network, Bedford Park, SA, Australia. CORRESPONDENCE TO: Qiong Ou, MD, Guangdong General Hospital, 159 Guangdong Academy of Medical Sciences, Guangzhou 510000, 160 Q6 Guangdong, China; e-mail: [email protected] OR R. Doug 161 McEvoy, MD, Adelaide Institute for Sleep Health, Flinders University (Repatriation General Hospital), GPO Box 2100, Adelaide SA 5001, 162 Australia; e-mail: doug.mcevoy@flinders.edu.au 163 Copyright Ó 2018 American College of Chest Physicians. Published by 164 Elsevier Inc. All rights reserved. 165 DOI: https://doi.org/10.1016/j.chest.2018.08.1082
2 Original Research
although no studies have objectively measured actual food intake in those with OSA. Physical activity seems to decline with increasing severity of OSA, associated with excessive sleepiness and obesity,7-9 which may not be reversed with treatment.10 Conversely, resting metabolic rate, a major determinant of energy expenditure, seems to be increased in OSA.11-13 Thus, uncertainty persists regarding the effects of OSA on energy balance (energy intake vs energy expenditure). Because CPAP, the first-line treatment for OSA, improves sleep disturbance, nocturnal hypoxia, sympathetic nervous system activity, and daytime sleepiness, measuring the course of body weight after this treatment may provide insight into the effects of OSA on overall energy balance. Early reports produced conflicting results, with studies indicating weight loss14 and weight gain.15,16 In an attempt to resolve this uncertainty, Drager et al17 undertook a meta-analysis of 25 randomized controlled trials with 3,181 patients. The median study duration in this meta-analysis was 3 months, with follow-up exceeding 6 months in only one of the 25 studies. CPAP treatment was found to promote a small increase in BMI and weight, raising the concern that CPAP might have an unintended detrimental effect on cardiometabolic health if such weight gain were to continue with long-term use. The Sleep Apnea Cardiovascular Endpoints (SAVE) study was a large, international, randomized controlled trial that compared the effects of CPAP treatment plus usual care vs usual care alone on the secondary prevention of serious cardiovascular (CV) events in patients with co-occurring moderate to severe OSA.18 Secondary analysis of the SAVE dataset provides an ideal opportunity to assess whether longterm CPAP treatment promotes weight gain in patients with OSA, as participants had regular anthropometric measurements for a mean follow-up period of 3.78 years. The present article reviews the changes in weight and other anthropometric measurements compared between the CPAP-treated and usual care groups.
166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215
Patients and Methods
216
Design
217
The SAVE study was an international, multicenter, randomized, open-label clinical trial, the details of which have been outlined elsewhere.18,19 In brief, patients aged 45 to 75 years, with a diagnosis of moderate to severe OSA (defined as an oxygen
218
[
-#- CHEST - 2018
FLA 5.5.0 DTD CHEST1963_proof 23 October 2018 1:54 am EO: CHEST-18-1406
]
219 220
221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242
Q8
desaturation index [ODI] of $ 12 of $ 4% oxygen desaturation events per hour during a home overnight screening sleep test [ApneaLink, ResMed]) and a previous diagnosis of coronary or cerebrovascular disease were included from 89 clinical centers in seven countries between December 2008 and November 2013. Patients were excluded if they had severe daytime sleepiness (Epworth Sleepiness Scale [ESS] score > 15), very severe nocturnal hypoxemia (oxygen saturation < 80% for > 10% of the monitoring time), or a predominantly Cheyne-Stokes respiration pattern. Potential participants were required to undergo a 1-week trial of sham CPAP delivered at subtherapeutic pressure to identify those most likely to tolerate mask CPAP treatment and adhere to procedures; those who recorded a daily average sham CPAP usage < 3 h on the machine recording card were excluded. The SAVE trial was conducted in accordance with the amended Declaration of Helsinki. Local institutional review boards or independent ethics committees at recruiting sites approved the protocol (e-Table 1), and all participants provided written informed consent. The trial is registered with ClinicalTrials.gov.20 Procedures Eligible participants were randomly assigned via a secure central server to receive either CPAP treatment plus usual CV care (CPAP group) or usual CV care alone (control group).18 All participants were advised on healthy sleep habits and lifestyle changes (including advice on healthy
276
eating and weight control) to minimize OSA symptoms. In the CPAP group, home pressure titration on an auto-PAP machine (REMstar Auto, M or PR series, Philips Respironics) was undertaken over the first week postrandomization. Data were extracted from recording cards to guide further therapy, with CPAP pressure fixed for longterm use at the 90th percentile of pressure achieved during the 1week auto-PAP trial period.
277 278 279 280 281
Clinic visits were scheduled for all participants at 1, 3, 6, and Q9 12 months and annually thereafter; participants were also contacted by telephone 6 monthly between the annual clinic visits. Additional end-of-study visits were performed between September 2015 and January 2016 on all participants who remained actively enrolled in the trial.
TABLE 1
287 289 290 291 292 293 294 295 296 297 299
] Baseline Characteristics
246 248
285 286
298
244
247
283 284
288
Following randomization, baseline anthropometric measurements of weight, height, and neck, waist, and hip circumferences were taken, and the BMI and waist-to-hip ratio (WHR) were calculated (e-Appendix 1). OSA characteristics included apneahypopnea index (AHI), ODI, mean nocturnal oxygen saturation, and lowest nocturnal oxygen saturation, computed from an overnight recording. Information was also obtained on ESS score, medical history, country of residence, and ethnicity. Anthropometric measurements were repeated at 6 and 12 months, and then annually for the duration of follow-up, with CPAP usage time downloaded from machine data cards at each clinic visit for patients allocated to CPAP.
243 245
282
Q16
CPAP Group (n ¼ 1,248) Characteristic
301
Control Group (n ¼ 1,235)
Male (n ¼ 1,010)
Female (n ¼ 238)
Male (n ¼ 996)
302
Female (n ¼ 239)
249
Age, y
60.8 7.7
63.0 7.1
60.7 8.1
63.4 7.0
250
Obese
307 (30.4)
92 (38.7)
286 (28.7)
82 (34.3)
251
OSA characteristics
300
303 304 305 306
252
AHI, events/h
29 16
28 16
30 16
27 15
307
253
ODI, events/h
27 14
29 15
27 14
308
254
28 14
Average SpO2, %
93.4 1.9
93.3 2.2
93.4 1.9
93.2 2.2
Lowest SpO2, %
77.1 6.0
76.9 5.7
77.3 6.0
76.3 6.7
7.4 3.6
7.1 3.8
7.5 3.5
7.5 3.9
255 256 257
ESS score
258
Medical history
309 310 311 312 313
259
Coronary artery disease
544 (53.9)
89 (37.4)
525 (52.7)
97 (40.6)
260
Cerebrovascular diseasea
424 (42.0)
144 (60.5)
425 (42.7)
132 (55.2)
315
261
42 (4.2)
5 (2.1)
46 (4.6)
10 (4.2)
316
262
Coronary and cerebrovascular disease
263
Diabetes mellitus
292 (28.9)
79 (33.2)
284 (28.5)
79 (33.1)
318 320
264 265
314
317 319
Country Australia
92 (9.1)
11 (4.6)
95 (9.5)
13 (5.4)
Brazil
88 (8.7)
37 (15.5)
90 (9.0)
35 (14.6)
268
China
643 (63.7)
155 (65.1)
615 (61.7)
156 (65.3)
269
India
45 (4.5)
9 (3.8)
44 (4.4)
14 (5.9)
270
New Zealand
40 (4.0)
8 (3.4)
41 (4.1)
2 (0.8)
325
271
Spain
101 (10.0)
18 (7.6)
109 (10.9)
19 (7.9)
326
272
United States
1 (0.1)
0
2 (0.2)
0
266 267
273 274 275
321 322 323 324
327
Data are presented as mean SD or No. (percentiles). Subjects indicated as obese had a BMI $ 30 kg/m2. AHI ¼ apnea-hypopnea index; ODI ¼ oxygen desaturation index; SpO2 ¼ blood oxygen saturation. a Stroke or transient ischemic attack.
3
chestjournal.org
FLA 5.5.0 DTD CHEST1963_proof 23 October 2018 1:54 am EO: CHEST-18-1406
328 Q17
329 330
331
Statistical Analysis
332
Summary data for each group are presented as mean SD for normally distributed variables, and median (interquartile range) for non-normally distributed variables. We excluded participants with extreme outlying values for any of the five anthropometric variables (weight, BMI, neck or waist circumference, or WHR) at any time point according to a change across the study duration of > 4 SDs; the goal was to eliminate probable typographical errors or effects of unrelated health issues and to ensure a consistent case base for all analyses. A linear mixed effects model was used to determine the mean difference between treatments for each of the variables during follow-up, with fixed effect terms used for baseline values, treatment, sex, visit (categorical), and interactions for sex baseline value, treatment visit, treatment sex, visit sex, and treatment visit sex. Interaction effects between treatment and each of the main effects included in the final model were also tested. A random intercept term for each participant was also included, and the estimated mean between-group differences were estimated with the marginal treatment effect for each sex included in the same model.
333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349
The predictors of weight change were also determined for end-of-study minus baseline weight. A simple linear regression was first tested for crude associations, with inclusion of age, sex, follow-up duration,
country, baseline weight, AHI, ODI, ESS score, and presence of coronary or cerebrovascular CV disease, previous or current CV medications, hypertension, diabetes mellitus, myocardial infarction, angina, heart failure, valvular heart disease, percutaneous coronary revascularization with or without a stent, coronary artery bypass graft, stroke and transient ischemic attack, and treatment group allocation. Variables that were significant at P < .20 were then entered into a multivariate linear regression model that a priori included follow-up duration as a predictor to determine the independent predictors of weight change. All estimated differences were considered significant for P < .05 using two-tailed hypothesis tests. Finally, a sensitivity analysis was conducted to determine whether higher levels of CPAP adherence were associated with weight change. Anthropometric values were compared between the two previously reported groups of propensity score-matched patients19 after excluding those with unsatisfactory anthropometric measurements. Details of the method whereby “CPAP-adherent” patients (defined as a mean $ 4 h of treatment per day over the study period) were matched 1:1 with control patients who never used CPAP are provided in e-Table 2 of the main SAVE study report.19 All analyses were performed by using Stata version 14.2 (StataCorp).
386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404
350
405
351
406
352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381
Results There were 2,717 participants in the SAVE study, but 30 were excluded because of withdrawal of consent, nonadherence to the protocol, or data irregularities identified during monitoring at one site. Eight patients were removed from these analyses because their baseline anthropometric measurements were incomplete, and another 196 were excluded because of extreme values (> 4 SDs) recorded during follow-up. Thus, a total of 2,483 subjects (1,248 in the CPAP group and 1,235 in the control group) were included with a mean 6.1 1.5 (range, 1-8) measures of weight, including the baseline visit. Their mean length of follow-up was 3.78 1.46 years. Table 1 outlines the baseline characteristics of participants according to sex; male and female subjects in the CPAP group used their device for a daily mean of 3.33 2.29 h and 3.15 2.19 h, respectively, over the study duration. Table 2 describes the summary measures for each of the anthropometric variables and shows that there were no significant differences in the estimated mean differences between treatment groups for either male or female subjects. Figure 1 illustrates that median weights remained stable for both CPAP and control participants, in both male and female subjects. There were no significant treatment time interactions.
382 383 384
Table 3 presents the results of the multivariate linear regression analysis for the change in weight between
baseline and the end of follow-up. Older age (P < .001), higher baseline weight (P < .001), and recruitment in China (P < .001) or India (P ¼ .007) vs Australia were associated with significant weight loss. Being male (P ¼ .003), having previous percutaneous coronary revascularization with or without a stent (P ¼ .001), and recruitment in New Zealand (P ¼ .006 compared with Australia) were associated with significant weight gain. Interactions between treatment status and other variables were not significant, indicating that there was no subgroup in which CPAP had a significant effect. Despite different patterns of weight change indicated by the regression coefficients according to country, there is no evidence that CPAP treatment had a differential effect on weight between countries. A sensitivity analysis compared good CPAP adherers ($ 4 h per night; n ¼ 516) with propensity-matched control subjects (n ¼ 531) (e-Table 2). Each individual contributed 6.1 1.5 weight measures (range, 1-8), including the baseline visit. Table 4 describes the summary measures for each anthropometric outcome as well as the estimated mean difference between groups. There was a significantly higher weight gain after CPAP treatment compared with control for male subjects: the mean (95% CI) baseline-adjusted difference was 0.38 kg (0.04-0.73) (P ¼ .031). No significant differences were found for any of the other anthropometric measures, either in male or female subjects.
385
407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440
4 Original Research
[
-#- CHEST - 2018
FLA 5.5.0 DTD CHEST1963_proof 23 October 2018 1:54 am EO: CHEST-18-1406
]
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
chestjournal.org
FLA 5.5.0 DTD CHEST1963_proof 23 October 2018 1:54 am EO: CHEST-18-1406
TABLE 2
] Anthropometric Variables for Each Treatment Group, Separated According to Sex CPAP Group
Variable
Control Group
Baseline
Follow-up
Baseline
Follow-up
Median (IQR) (No.)
Median (IQR) (No.)
Median (IQR) (No.)
Median (IQR) (No.)
Adjusted D in Change From Baseline (CPAP vs Control), Mean (95% CI)
P
Weight, kg Female
70 (64-80) (238)
70 (63-79) (1,049)
70 (62-77) (239)
69.5 (61-76) (1,084)
Male
80 (73-90) (1,010)
80 (73-90) (4,737)
80 (73-90) (996)
80 (74-89) (4,684)
0.07 (–0.40 to 0.54)
.773
–0.14 (–0.37 to 0.09)
.233
Mean SD (No.)
Mean SD (No.)
Mean SD (No.)
Mean SD (No.)
Female
29.58 5.78 (238)
29.35 5.62 (1,049)
28.71 4.99 (239)
28.34 4.94 (1,084)
0.035 (–0.13 to 0.20)
.68
Male
28.43 3.96 (1,010)
28.38 4.06 (4,737)
28.39 3.93 (996)
28.45 4.07 (4,684)
–0.05 (-0.13 to 0.030)
.22
0.124 (–0.104 to 0.352)
.29
BMI, kg/m2
Neck circumference, cm Female
37.5 3.8 (235)
37.3 3.8 (679)
36.9 3.9 (239)
36.8 4.0 (695)
Male
41.5 3.6 (1,004)
41.2 3.5 (3,008)
41.5 3.7 (995)
41.4 3.8 (2,989)
–0.108 (–0.218 to 0.0017)
.054
Waist circumference, cm 98.5 13.3 (236)
98.5 13.8 (681)
101.5 11.1 (1,009)
101.4 11.2 (3,007)
Female
0.92 0.08 (235)
0.93 0.09 (679)
0.90 0.08 (239)
0.92 0.08 (696)
Male
0.97 0.06 (1,003)
0.97 0.07 (3,005)
0.96 0.07 (995)
0.97 0.07 (2,990)
Female Male
96.2 13.2 (239)
96.1 13.6 (696)
101.3 11.1 (996)
101.4 11.6 (2,995)
0.013 (–0.58 to 0.61)
.97
–0.015 (–0.30 to 0.27)
.92
–0.002 (–0.007 to 0.004)
.55
–0.0004 (–0.003 to 0.002)
.78
WHR
No. indicates the total number of measurements performed during either baseline or follow-up. A single measure per subject was performed at baseline, but multiple measures were performed during follow-up. IQR ¼ interquartile range; WHR ¼ waist-to-hip ratio.
5 497
496
499
498
501
500
502
503
505
504
507
506
509
508
511
510
512
513
515
514
517
516
519
518
521
520
522
523
525
524
527
526
529
528
531
530
532
533
535
534
537
536
539
538
541
540
542
543
545
544
547
546
548
550
549
551 552
140
553 555 556 557 558
*
*
120
*
*
*
*
Female
Weight (kg)
554
*
*
100 80 60
559 560
40
561 562
140
563 565 566 567 568 570
575 576 577 578 579 580
En
d
of
St
ud
m
y
o
o m
60
m
o
48
m
o 36
o
24
m 12
m
o
e 6
se
lin Q21 Q18
Ba
print & web 4C=FPO
574
80
40
571 573
100
60
569
572
120 Male
Weight (kg)
564
Control
CPAP
Figure 1 – Boxplots of observed weight at baseline and during follow-up according to treatment group and sex. The end-of-study visit indicates the last measurement for each individual. Each box shows median and 25th to 75th percentiles.
581
Discussion
582
The principal finding of this large clinical trial was that there was no clinically significant weight change
583 584 585
TABLE 3
586
Variable
587 588 589
b (95% CI)a
Sex (1 ¼ female, 2 ¼ male)
591
Treatment group (1 ¼ CPAP, 2 ¼ control)
592
Duration in study (y)
593
PCR (yes vs no)
594
Country Australia
609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638
641 642 643
.003
–0.049 (–0.065 to –0.032)
607 608
639
< .001
0.789 (0.268 to 1.310)
606
640 Pa
–0.058 (–0.084 to –0.031)
Age (y)
Baseline weight (kg)
596
Only one other randomized controlled trial of CPAP vs usual care in overweight or obese patients with moderate to severe OSA recorded weight and BMI over a similar period (median follow-up, 4 years)21 as our study. Patients allocated to the control group were found to have had a significant decrease in BMI, whereas those allocated to CPAP treatment experienced no significant change in BMI over time. Q10 The reported between-group mean difference in change in BMI for CPAP vs control in that trial was 0.25 kg/m2 (95% CI, 0.04-0.46), and the corresponding difference in change in weight (F. Barbe, personal communication) was 0.51 kg (95% CI, 0.13-1.25). Neither the study by Barbe et al21
] Regression Coefficients From a Multivariate Linear Regression for Change in Weight
590
595
in patients with established CV disease and comorbid OSA over several years, and there was no differential effect of CPAP on other anthropometric measures. In a subanalysis, male subjects, but not female subjects, who achieved high use of CPAP each night had a small (< 400 g) but statistically significant increase in weight compared with matched control patients but without changes in other anthropometric indices such as BMI or neck or waist circumference. Taken together, these results indicate that long-term CPAP treatment is unlikely to exacerbate the problems of overweight and obesity that are common among patients with OSA. Such a small change in weight, even with good adherence over several years, is highly unlikely to have any serious clinical ramifications.
644
< .001
645
0.105 (–0.281 to 0.491)
.594
–0.103 (–0.278 to 0.072)
.249
647
0.698 (0.270 to 1.126)
.001
648
646
649 650
1.00
651
Brazil
–0.82 (–1.82 to 0.18)
598
China
–1.60 (–2.38 to –0.808)
< .001
599
India
–1.81 (–3.12 to –0.502)
.007
600
New Zealand
1.75 (0.51 to 3.00)
.006
655
601
Spain
0.16 (–0.84 to 1.15)
.758
656
602
United States
3.73 (–2.74 to 10.21)
.258
597
603 604 605
.107
652 653 654
657
The b-coefficients are adjusted for all other variables in the table. Age, sex, treatment group, and duration in study were included in the model a priori. PCR ¼ percutaneous coronary revascularization with or without a stent. a Using a multivariate linear regression for change in weight between baseline and end of follow-up.
6 Original Research
[
-#- CHEST - 2018
FLA 5.5.0 DTD CHEST1963_proof 23 October 2018 1:54 am EO: CHEST-18-1406
]
658 659 660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
chestjournal.org
FLA 5.5.0 DTD CHEST1963_proof 23 October 2018 1:54 am EO: CHEST-18-1406
TABLE 4 ]
Change in Anthropometric Variables in Patients With Good CPAP Compliance and Propensity-Matched Control Subjects, Separated According to Sex CPAP Group
Variable
Adjusted D in Change From Baseline (CPAP vs Control), Mean (95% CI)
Control Group
Baseline
Follow-up
Baseline
Follow-up
Median (IQR) (No.)
Median (IQR) (No.)
Median (IQR) (No.)
Median (IQR) (No.)
P
Weight, kg –0.217 (–0.97 to 0.53)
Female
70.0 (65.0-79.50) (92)
70.0 (63.0-78.0) (445)
70.0 (61.0-76.0) (99)
70.0 (60.0-76.0) (415)
Male
82.0 (74.0-91.0) (439)
82.0 (74.0-92.0) (2,196)
82.0 (74.0-90.0) (417)
82.0 (74.0-90.0) (1,955)
Mean SD (No.)
Mean SD (No.)
Mean SD (No.)
Mean SD (No.)
BMI, kg/m
.569
0.383 (0.035 to 0.730)
.031a
–0.059 (–0.324 to 0.207)
.665
2
Female
29.5 5.0 (92)
29.3 5.1 (445)
29.0 4.9 (99)
28.7 4.9 (415)
Male
28.8 4.1 (439)
28.9 4.2 (2,196)
28.8 4.2 (417)
28.8 4.3 (1,955)
0.123 (–0.001 to 0.246)
.052
Female
37.4 3.8 (90)
37.1 3.6 (285)
36.3 3.6 (99)
36.2 3.7 (272)
0.122 (–0.239 to 0.483)
.507
Male
41.8 3.6 (438)
41.6 3.6 (1,387)
41.8 3.6 (416)
41.6 3.7 (1,267)
0.043 (–0.123 to 0.209)
.610
98.5 12.7 (91)
98.9 14.2 (285)
97.0 12.9 (99)
95.9 13.2 (272)
102.0 11.1 (439)
102.5 11.3 (1,386)
102.4 11.8 (417)
102.5 12.2 (1,269)
Female
0.92 0.08 (90)
0.93 0.09 (285)
0.91 0.09 (99)
0.91 0.09 (272)
Male
0.97 0.06 (438)
0.97 0.07 (1,386)
0.97 0.07 (416)
0.97 0.07 (1,268)
Neck circumference, cm
Waist circumference, cm Females Male
0.89 (–0.06 to 1.84)
.068
0.290 (–0.154 to 0.733)
.200
0.004 (–0.005 to 0.013)
.357
–0.000 (–0.004 to 0.004)
.862
WHR
No. indicates the total number of measurements performed during either baseline or follow-up. A single measure per subject was performed at baseline, but multiple measures were performed during follow-up. See Table 1 legend for expansion of abbreviations. a P < .05.
7 717
716
719
718
721
720
722
723
725
724
727
726
729
728
731
730
732
733
735
734
737
736
739
738
741
740
742
743
745
744
747
746
749
748
751
750
752
753
755
754
757
756
759
758
761
760
762
763
765
764
767
766
768
770
769
771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823
nor the present study found evidence of clinically significant weight gain among CPAP-treated patients with long-term follow-up. The small differences in weight change noted between the CPAP and control groups in the two studies may relate to differences in the patient characteristics and levels of CPAP adherence achieved; for example, mean age (52 vs 61 years), prevalence of preexisting CV disease (0% vs 100%), and median CPAP use (5.00 h [interquartile range, 2.18-6.25 h] vs 3.26 h [interquartile range, 1.26-5.08 h]) per night. The meta-analysis by Drager et al17 had raised concern that CPAP may cause an increase in weight gain, although the scale of the effect was small (0.17 kg; 95% CI, 0.10-0.24). It is possible that this difference between our observations and that of the previous shorter studies in the meta-analysis may be related to differences in study populations such as baseline obesity, ethnicity, age, and chronic comorbidities. Encouragingly, the relatively small short-term changes seen in the meta-analysis17 do not seem to have been compounded during long-term follow-up. It is reassuring that we found no clinically significant weight gain in CPAP-treated patients with OSA who, in the main, were overweight or obese and had proven (mainly atherosclerotic) CV disease. Conversely, it is noteworthy and arguably disappointing from a clinical perspective that weight did not decrease in the male or female subjects in either the CPAP or the control group. All SAVE participants had their CV risk factors managed according to national clinical guidelines, including advice on weight loss for secondary CV prevention in overweight/obese patients.22-25 They were also given verbal lifestyle advice regarding weight control to minimize the severity of OSA. The absence of weight loss likely reflects the widely acknowledged dual problems of suboptimal adherence to clinical guidelines by practitioners26 and the difficulties that patients experience in adopting the lifestyle changes that are necessary to achieve sustained weight loss, as discussed elsewhere.25,27 Most people with OSA are overweight or obese and have a high frequency of CV risk factors (eg, hypertension,28 diabetes mellitus29) and disease.30
Results of the Wisconsin Sleep Cohort Study highlight the potential benefit of weight loss in the management of OSA: a 10% reduction in weight was associated with a 26% drop in AHI,31 and weight loss is recommended in OSA management guidelines32,33 and for CV risk reduction.23,25,27 Our study findings indicate that CPAP, and the symptom benefits that flow from it, cannot be relied upon to lead to meaningful weight loss.17 Enhanced weight management is therefore needed for overweight patients with OSA receiving long-term CPAP treatment and is ideally delivered as part of a personalized chronic disease management program.34 Indeed, combining weight loss with CPAP therapy for OSA seems necessary to improve CV risk markers.35 The importance of encouraging weight loss in those with OSA and CV disease living in affluent societies is emphasized in our multivariable regression model, which showed that male subjects were less likely to lose weight than female subjects, as were those from Australia and New Zealand compared with India and China, independent of CPAP status. These differences observed irrespective of CPAP treatment may relate to national differences in dietary patterns, health-care systems, socioeconomic status, or other unknown lifestyle factors.36 We recognize that our study has several limitations. First, the study population was restricted to older patients with moderate to severe OSA and established CV disease but with very severe cases (nocturnal oxygen desaturation < 80% for > 10% of recording time) and marked self-reported daytime sleepiness (ESS score > 15) excluded. Thus, our results may not be generalizable to all patients with OSA. However, the large size of the study and prospective recording of weight, CPAP adherence, and other measures over several years allowed sensitivity analyses to be undertaken according to levels of adherence to CPAP.
826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872
Conclusions The present study showed that patients with moderate to severe OSA and co-existing CV or cerebrovascular disease had no significant changes in BMI or neck or waist circumference from the use of CPAP over several years. We found no evidence that patients
873 874 875 876 877 878
824
879
825
880
8 Original Research
[
-#- CHEST - 2018
FLA 5.5.0 DTD CHEST1963_proof 23 October 2018 1:54 am EO: CHEST-18-1406
]
881
with moderate to severe OSA and coexisting CV disease are susceptible to clinically concerning
882 883
936
long-term weight change from the use of CPAP therapy.
937 938
884
939
885
940
886
Acknowledgments
887
Author contributions: R. J. W. and K. A. L. had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Q. O., B. C., C. S. A., R. D. M., and R. J. W. planned the study. Q. O., Y. L., X. Z., R. C., Q. W., L. F. D., G. L.-F., M. H., N. M., S. M., O. M., F. B., C. S. A., and R. D. M. recruited participants and collected data and samples. R. J. W. and K. A. L. conducted statistical analyses. Q. O., B. C., R. D. M., R. J. W., K. A. L., and C. S. A. interpreted data and drafted the manuscript. All authors contributed to drafting the manuscript, and read and approved the manuscript for submission.
888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903
Q12
904 905 906 907 908 909 910 911
Q13
912 913 914 915 916
Q14
917
Q11
918 920 921 922 923 925 926 927 928 929 930 931 932 933 934 935
Role of sponsors: The sponsor had no role in the design of the study, the collection and analysis of the data, or the preparation of the manuscript. Collaborators: The SAVE trial was conducted by a large team of investigators. Details and affiliations are given in eAppendix 2.
919
924
Financial/nonfinancial disclosures: The authors have reported to CHEST the following: L. F. D. received research study equipment from Philips Respironics. M. H. is a shareholder of Cansleep Ltd. N. M. has received research funding support from Oventus Pty Ltd (Brisbane, Australia), Nyxoah Pty Ltd (Mont-Saint Guibert, Belgium), and Zelda Therapeutics Pty Ltd (Australia). F. B. received research grants from ResMed and Philips. R. D. M. reports grants and nonfinancial support from Philips Respironics, nonfinancial support from ResMed, grants from the National Health and Medical Research Council of Australia, and grants from Fisher & Paykel, during the conduct of the study; nonfinancial support from Air Liquide; and speaker fees from ResMed. None declared (Q. O., B. C., K. A. L., Y. L., X. Z., R. C., Q. W., G. L.-F., S. M., O. M., C. S. A., R. J. W.).
Q19
Additional information: The e-Appendixes and e-Tables can be found in the Supplemental Materials section of the online article.
References 1. Genta PR, Schorr F, Eckert DJ, et al. Upper airway collapsibility is associated with obesity and hyoid position. Sleep. 2014;37(10):1673-1678. 2. Ryan S, Crinion SJ, McNicholas WT. Obesity and sleep-disordered breathing— when two ’bad guys’ meet. QJM. 2014;107(12):949-954. 3. Wheatley JR, Mezzanotte WS, Tangel DJ, White DP. Influence of sleep on genioglossus muscle activation by negative
pressure in normal men. Am Rev Respir Dis. 1993;148(3):597-605. 4. Shechter A. Obstructive sleep apnea and energy balance regulation: a systematic review. Sleep Med Rev. 2017;34:59-69. 5. Harsch IA, Konturek PC, Koebnick C, et al. Leptin and ghrelin levels in patients with obstructive sleep apnoea: effect of CPAP treatment. Eur Respir J. 2003;22(2): 251-257. 6. Cao Y, Wittert G, Taylor AW, Adams R, Shi Z. Associations between macronutrient intake and obstructive sleep apnoea as well as self-reported sleep symptoms: results from a cohort of community dwelling Australian men. Nutrients. 2016;8(4):207. 7. Basta M, Lin HM, Pejovic S, Sarrigiannidis A, Bixler E, Vgontzas AN. Lack of regular exercise, depression, and degree of apnea are predictors of excessive daytime sleepiness in patients with sleep apnea: sex differences. J Clin Sleep Med. 2008;4(1):19-25. 8. Lopes C, Esteves AM, Bittencourt LR, Tufik S, Mello MT. Relationship between the quality of life and the severity of obstructive sleep apnea syndrome. Braz J Med Biol Res. 2008;41(10):908-913. 9. Quan SF, O’Connor GT, Quan JS, et al. Association of physical activity with sleepdisordered breathing. Sleep Breath. 2007;11(3):149-157. 10. West SD, Kohler M, Nicoll DJ, Stradling JR. The effect of continuous positive airway pressure treatment on physical activity in patients with obstructive sleep apnoea: a randomised controlled trial. Sleep Med. 2009;10(9): 1056-1058. 11. Ryan CF, Love LL, Buckley PA. Energy expenditure in obstructive sleep apnea. Sleep. 1995;18(3):180-187. 12. Stenlof K, Grunstein R, Hedner J, Sjostrom L. Energy expenditure in obstructive sleep apnea: effects of treatment with continuous positive airway pressure. Am J Physiol. 1996;271(6 pt 1): E1036-E1043. 13. Ucok K, Aycicek A, Sezer M, et al. Resting metabolic rate and anthropometric measurements in male sleep apnea patients. Intern Med. 2011;50(8):833-838. 14. Loube DI, Loube AA, Erman MK. Continuous positive airway pressure treatment results in weight loss in obese and overweight patients with obstructive sleep apnea. J Am Diet Assoc. 1997;97(8): 896-897. 15. Quan SF, Budhiraja R, Clarke DP, et al. Impact of treatment with continuous positive airway pressure (CPAP) on weight in obstructive sleep apnea. J Clin Sleep Med. 2013;9(10):989-993.
16. Redenius R, Murphy C, O’Neill E, AlHamwi M, Zallek SN. Does CPAP lead to change in BMI? J Clin Sleep Med. 2008;4(3):205-209. 17. Drager LF, Brunoni AR, Jenner R, Lorenzi-Filho G, Bensenor IM, Lotufo PA. Effects of CPAP on body weight in patients with obstructive sleep apnoea: a meta-analysis of randomised trials. Thorax. 2015;70(3):258-264. 18. Antic NA, Heeley E, Anderson CS, et al. The Sleep Apnea cardioVascular Endpoints (SAVE) trial: rationale, ethics, design, and progress. Sleep. 2015;38(8): 1247-1257. 19. McEvoy RD, Antic NA, Heeley E, et al. CPAP for prevention of cardiovascular events in obstructive sleep apnea. N Engl J Med. 2016;375(10):919-931. 20. ClinicalTrials.gov. NCT00738179; ACTRN 126080000409370. https:// clinicaltrials.gov/ct2/show/ NCT0073817 9. 21. Barbe F, Duran-Cantolla J, Sanchez-de-laTorre M, et al. Effect of continuous positive airway pressure on the incidence of hypertension and cardiovascular events in nonsleepy patients with obstructive sleep apnea: a randomized controlled trial. JAMA. 2012;307(20):2161-2168. 22. National Vascular Disease Prevention Alliance. Guidelines for the Management of Absolute Cardiovascular Disease Risk. National Stroke Foundation; 2012. 23. National Heart Foundation of Australia; Cardiac Society of Australia and New Zealand. Reducing Risk in Heart Disease: An Expert Guide to Clinical Practice for Secondary Prevention of Coronary Heart Disease. Melbourne, Australia: National Heart Foundation of Australia; 2012. 9781-74345-016-1. 24. Piepoli MF, Hoes AW, Agewall S, et al. 2016 European guidelines on cardiovascular disease prevention in clinical practice: the Sixth Joint Task Force of the European Society of Cardiology and Other Societies on Cardiovascular Disease Prevention in Clinical Practice (constituted by representatives of 10 societies and by invited experts) developed with the special contribution of the European Association for Cardiovascular Prevention & Rehabilitation (EACPR). Eur Heart J. 2016;37(29):2315-2381. 25. Smith SC Jr, Benjamin EJ, Bonow RO, et al. AHA/ACCF secondary prevention and risk reduction therapy for patients with coronary and other atherosclerotic vascular disease: 2011 update: a guideline from the American Heart Association and American College of Cardiology Foundation. Circulation. 2011;124(22): 2458-2473. 26. Barnes ER, Theeke LA, Mallow J. Impact of the Provider and Healthcare team
9
chestjournal.org
FLA 5.5.0 DTD CHEST1963_proof 23 October 2018 1:54 am EO: CHEST-18-1406
941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990
991
Adherence to Treatment Guidelines (PHAT-G) intervention on adherence to national obesity clinical practice guidelines in a primary care centre. J Eval Clin Pract. 2015;21(2): 300-306.
992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005
Q15
30. Molnar MZ, Mucsi I, Novak M, et al. Association of incident obstructive sleep apnoea with outcomes in a large cohort of US veterans. Thorax. 2015;70(9):888-895.
27. World Heart Federation. Secondary Cardiovascular Disease Prevention and Control. Switzerland: 2014.
31. Peppard PE, Young T, Palta M, Dempsey J, Skatrud J. Longitudinal study of moderate weight change and sleepdisordered breathing. JAMA. 2000;284(23):3015-3021.
28. Marin JM, Agusti A, Villar I, et al. Association between treated and untreated obstructive sleep apnea and risk of hypertension. JAMA. 2012;307(20):21692176.
32. Epstein LJ, Kristo D, Strollo PJ Jr, et al. Clinical guideline for the evaluation, management and long-term care of obstructive sleep apnea in adults. J Clin Sleep Med. 2009;5(3):263-276.
29. Botros N, Concato J, Mohsenin V, Selim B, Doctor K, Yaggi HK. Obstructive sleep apnea as a risk factor for type 2 diabetes. Am J Med. 2009;122(12):11221127.
33. Qaseem A, Holty JE, Owens DK, et al. Management of obstructive sleep apnea in adults: a clinical practice guideline from the American College of Physicians. Ann Intern Med. 2013;159(7):471-483.
34. Heatley EM, Harris M, Battersby M, McEvoy RD, Chai-Coetzer CL, Antic NA. Obstructive sleep apnoea in adults: a common chronic condition in need of a comprehensive chronic condition management approach. Sleep Med Rev. 2013;17(5):349-355. 35. Chirinos JA, Gurubhagavatula I, Teff K, et al. CPAP, weight loss, or both for obstructive sleep apnea. N Engl J Med. 2014;370(24):2265-2275.
1046 1047 1048 1049 1050 1051 1052 1053 1054
36. Kim R, Kawachi I, Coull BA, Subramaniam SV. Contribution of socioeconomic factors to the variation in body-mass index in 58 low-income and middle-income countries: an econometric analysis of multilevel data. Lancet Glob Health. 2018;6(7):e777e786.
1055 1056 1057 1058 1059 1060
1006
1061
1007
1062
1008
1063
1009
1064
1010
1065
1011
1066
1012
1067
1013
1068
1014
1069
1015
1070
1016
1071
1017
1072
1018
1073
1019
1074
1020
1075
1021
1076
1022
1077
1023
1078
1024
1079
1025
1080
1026
1081
1027
1082
1028
1083
1029
1084
1030
1085
1031
1086
1032
1087
1033
1088
1034
1089
1035
1090
1036
1091
1037
1092
1038
1093
1039
1094
1040
1095
1041
1096
1042
1097
1043
1098
1044
1099
1045
1100
10 Original Research
[
-#- CHEST - 2018
FLA 5.5.0 DTD CHEST1963_proof 23 October 2018 1:54 am EO: CHEST-18-1406
]