Clifford A. Massie, PhD and Robert W. Hart, MD
Objective: Determine whether CPAP pressure requirements change during the first six months following bariatric surgery.
Bariatric surgery patients diagnosed with obstructive sleep apnea (OSA) were placed on Autoset therapy before surgery and were seen at 2-weeks, 3-months and 6-months after surgery. Compliance and pressure data were obtained at each time point and weight was recorded.
Results: Data were analyzed with repeated measures ANOVA using weight loss as a covariate and Tukey’s HSD for post hoc comparisons. A main effect for the 95th pressure centile (P95) and median pressure (PMed) were observed. The mean P95 decreased from 11.4 cm H2O pre-operatively to 9.5 cm H2O 2-weeks after surgery (95% CI 0.3–3.9). The PMed decreased from 8.8 cm H2O pre-operatively to 6.9 cm H2O 2-weeks after surgery (95% CI 0.4–4.1) and to 6.2 cm H2O at 3-months (95% CI 0.3–6.7). A trend was observed for a decrease in the maximum pressure over time (p=0.07). A main effect for weight loss was not observed.
Conclusions: Reduction in optimal CPAP pressure (P95) was observed 2 weeks after surgery and remained stable at 3-months and 6-months. The PMed was significantly lower at 2-weeks and 3-months, whereas no change in PMax was observed. Pressure changes were independent of weight loss. Rapid weight loss and changes in hormonal regulation of weight and metabolism may be responsible for the pressure reductions. Self-adjusting CPAP offers advantages over fixed pressure therapy by compensating for changes in optimal pressure requirements following surgery.
Obesity and obstructive sleep apnea (OSA) are prevalent and are associated with significant cardiovascular morbidity.1 Greater than 50% of American adults are overweight with nearly one quarter of those overweight obese (BMI = 30 kg/m2).2 Obesity itself is associated with decreased life expectancy and increased mortality.3,4 The prevalence of OSA (AHI = 5) in the general population is estimated at 24% for men and 9% for women.5,6 In the morbidly obese (BMI = 40 kg/m2), nearly 100% of men and 60–70% of women have OSA.6,7 Bariatric surgery is an intervention for morbid obesity that results in dramatic weight loss and improvement or resolution of medical co-morbidities, such as diabetes, hypertension and hyperlipidemia.8,9 Within the first post-operative week, patients with adult onset diabetes mellitus (NIDDM) show improvement in glycemic control.9 Several studies have shown improvement or resolution of OSA following surgically induced weight loss.10–14
Unrecognized and untreated OSA can impact the bariatric surgery patient peri-operatively and throughout the weight loss period. Obesity itself predisposes to hypoxemia following surgery, and anesthesia and narcotic analgesics exacerbate pre-existing OSA and further contribute to hypoxemic burden.15–17 Increased length of stay and serious post-operative complications following knee or hip replacement surgery are seen in patients with unrecognized or untreated OSA, compared to those with treated OSA or those without OSA.18
The greatest degree of relative weight loss occurs within the first 6 months after surgery; final weight is achieved within two years.19,20 Following surgery, OSA may improve or resolve, but the apnea/hypopnea index (AHI) does not normalize in all patients. Even after substantial weight loss one year after surgery, as many as 54% of patients continued to have OSA that warranted intervention with CPAP.11 A significant reduction in AHI may be seen after surgery, only to increase without concomitant weight gain years later.21
Weight loss following bariatric surgery alters CPAP pressure requirements. Lankford and colleagues24 evaluated 15 bariatric surgery patients post-operatively with the Autoset Spirit to determine the optimal pressure as measured by the 95th pressure centile (P95). This is the pressure which is not exceeded 5% of the night, and which is equivalent to the effective pressure obtained during manual laboratory titration.22,23 Lankford’s patients were able to lower their fixed pressure prescriptions by an average of 18%.24 Guardiano et al. reported that 3/8 (37.5%) of bariatric surgery patients who continued to have OSA at their goal weight required lower CPAP pressures.10 These studies examined changes in CPAP requirements at only one time point following bariatric surgery, typically 6 months or more after surgery and often at final weight. Inferential statistics were not reported in those studies.
Patients in the present study were prescribed self-adjusting CPAP (Autoset Spirit, ResMed, San Diego, CA) to compensate for pressure requirements that accompany weight loss. The Autoset Spirit (Autoset) self-adjusts to provide the minimum pressure at each time point during treatment to eliminate apneas, hypopneas and upper airway resistance, while preserving sleep architecture and continuity.25,26 The device provides statistics regarding pressure requirements during treatment, and patient use and efficacy data. The current study was designed to explore how CPAP pressure requirements changed during the first 6 months after bariatric surgery, and determine how those changes were related to weight loss.
Eligible patients were = 18 years of age with a BMI = 40 kg/m2 who were being evaluated for Roux-en-Y gastric bypass surgery. Patients had not received prior surgical intervention for weight loss and were CPAP naive. Institutional review board approval and written informed consent were obtained.
This was a prospective study. Per our institution’s protocol, patients seen pre-operatively by our practice for pulmonary and sleep clearance prior to surgery were invited to participate; patients were not necessarily seeking therapy for symptomatic OSA. At the initial visit, demographic and anthropometric data were recorded and patients completed the Epworth Sleepiness Scale (ESS). At the time of consent, patients were told that if they were diagnosed with OSA, they would need to use Autoset for at least 18 months after surgery, or until weight loss equaled or exceeded 70% of excess body weight.
Patients were scheduled for comprehensive laboratory based polysomnography before surgery. Outcome measures included the AHI and the oxygen desaturation index (ODI). Desaturations were defined as a drop in oxyhemoglobin saturation of >3%. An obstructive apnea was defined as cessation of airflow (airflow tracing between 0% and 20% of baseline) for =10 sec, accompanied by a desaturation. A hypopnea was defined as above, except that airflow tracing was between 20 and 50% of baseline. Clinically significant OSA requiring intervention was defined as an overall AHI =15 or a REM AHI =15. A REM stage marker of OSA severity was included to capture patients in whom OSA showed marked worsening or was seen exclusively during REM sleep.
Patients received either a full night diagnostic study followed by a full night CPAP titration, or a split-night study was performed according to AASM guidelines.27 Patients diagnosed with OSA were titrated manually in the laboratory by a skilled technician. The pressure was started at 4 cm H2O and increased in 1 cm increments to eliminate apneas, hypopneas and snoring. Effective pressure was defined as the fixed CPAP pressure that eliminated apneas, hypopneas, snoring and hypoxemia. Patients successfully titrated with fixed pressure CPAP in the laboratory were subsequently initiated on Autoset therapy at home with a range of 4–20 cm H2O. Patients were instructed to use Autoset on a daily basis, including during their hospital stay.
Patients were seen for an office visit 2 weeks after surgery, and again at 3-months and 6-months after surgery. At each visit weight was obtained. Compliance and pressure data were downloaded from the Autoset device using compliance software (AutoScan 5.4; ResMed). Actual use per 24 hour period was recorded; a pressure transducer recorded use only when the patient was breathing with the mask in place. Regular use was defined as mask on time =4 hrs/night for =70% of nights used.28 For each time point, pressure data was averaged for the 3 nights preceding the office visit, provided that use was =4 hrs for at least 2 of the 3 nights and leak values were within acceptable limits.
Data are presented as mean ± SD. Changes in CPAP pressure over time were analyzed with repeated measures ANOVA using weight as a covariate. Significant main effects were followed with Tukey’s HSD post hoc comparisons. All analyses were performed using JMP 5.1.2 (SAS Institute, Cary, NC).
Patients accepted therapy by agreeing to a home visit from a respiratory therapist for Autoset delivery and instruction. Twenty-five women and six men (n=31) diagnosed with OSA used Autoset regularly until the first follow-up visit at 2-weeks post surgery. Six women and two men (n=8) continued as regular users for the entire follow-up period of 6 months. There were no differences in AHI, ODI, ESS, BMI, gender or weight loss at 2-week follow-up between the 23 patients who stopped using Autoset regularly 2 weeks after surgery and the 8 patients who remained regular users for 6 months (all p-values = 0.58). The 8 patients who continued with therapy were younger (mean age 43 vs. 50, p=0.03).
The mean age for the 8 patients was 43 ± 9, range 29–55 yrs. Mean weight was 161 ± 27 kg, range 118.2 to 201.8 kg. The mean BMI was 57.5 ± 12 kg/m2, range 47.6 to 81.3 kg/m2. The mean ESS score was 11.9 ± 7.4, range 4–22. Six of the 8 patients (75%) had either hypertension (HTN) and/or NIDDM, and were being treated pharmacologically for these conditions. Patients 1 and 4 discontinued antihypertensive medication at the post-op visit; all other patients were maintained on the same pharmacological agents during the 6 months. Patient characteristics are presented in Table 1. The adherence rate for regular Autoset use 6 months after surgery was 26%.
Diagnostic and Treatment Parameters
The mean AHI was 54.6 ± 48, range 11 to 130.8. Patient #2 had an overall AHI of 11 and a REM AHI of 62. The mean ODI was 51.3 ± 40.6, range 14.8 to 121.9. Substantial hypoxemic burden was observed during NREM and/or REM sleep in all patients. Percent of time with oxygen saturation levels below 90% in either NREM or REM sleep was =16.6%. The optimal CPAP pressure determined by manual laboratory titration ranged from 7 to 15, mean pressure 10.5 ± 2.6. These data are presented in Table 2.
Pressure Requirements and Weight Change
The mean value for PMax, P95 and PMed were calculated at each time point using the previous 3 nights’ data. Paired sample t-tests were used to compare PMax, P95 and PMed from pre-surgery to 2-weeks post-surgery in the sample of 31 patients who remained regular users for the first assessment at 2-weeks. Significant decreases were observed for all variables. The PMax decreased from 12.8 to 10.8 (CI 1.1–3.0, p < 0.0001), the P95 from 11.4 to 9.8 (CI 0.8–2.5, p=0.0001) and the PMed from 8.8 to 7.3 (CI 0.6–2.2, p=0.0004).
Repeated measures ANOVA with weight as a covariate was used to compare changes in PMax, P95 and PMed over time for the 8 patients who remained compliant at 6 months. A main effect for time was observed for P95 (F=3.9, p=0.02) and PMed (F=5.0, p=0.009), and a trend was seen for PMax (F=2.7, p=0.07). There was a significant decrease in the P95 from pre-Surgery to two weeks post-surgery (11.4 to 9.5 cm H2O; 95% CI 0.3–3.9). The PMed decreased significantly from pre-surgery to two weeks post surgery (8.8 to 6.9 cm H2O; 95% CI 0.4–4.1), and from pre-surgery to 3-months post-surgery (8.8 to 6.2 cm H2O; 95% CI 0.3 – 6.6). Two weeks after surgery, patients lost an average of 9.0 kg (range 4.1–15.5). Three months after surgery weight loss averaged 23 kg (range 14.5–33.2) and at 6 months average weight loss was 39.4 kg (range 29.4–48.2). A main effect for weight loss was not observed. Pressure requirements and weight changes for individual patients are presented in Table 3. Figure 1 illustrates these pressure changes over time.
This is the first study to systematically examine CPAP pressure changes during the first 6 months after bariatric surgery. Significant decreases in Pmax, P95 and Pmed were observed in 31 patients who were regular Autoset users 2 weeks after surgery. Eight of those 31 patients continued to be regular users for 6 months following surgery. In that group of 8, changes in optimal pressure (P95) were seen two weeks post-surgery. Optimal pressure remained stable at 3-months and 6-months. The median pressure was lower at 2-weeks and at 3-months after surgery compared to pre-surgery values. A trend was observed for maximum pressure. Weight was used as a covariate in the analyses to determine how pressure requirements changed as a function of weight loss. A main effect for weight loss was not observed, suggesting that CPAP pressure changed independent of weight.
The observation that CPAP pressures decreased with modest weight reduction was not anticipated. Previous studies demonstrated lower CPAP pressure requirements following substantial weight loss.24,29 Observations in those studies were made at varying time points following surgery, ranging from 6 months to over 2 years. In the present study, optimal pressure decreased within weeks after surgery and remained stable at 6 months. It is possible that a further decrease in pressure would be observed with additional weight loss and perhaps at final weight. Patients in the Lankford study lost an average of 44.6 kg (range 19.1–88.6) at the time of evaluation, but most patients remained obese (BMI = 30 kg/m2). One patient required an optimal pressure < 5 cm H2O (BMI < 30 kg/m2), which likely indicates a mild degree or an absence of sleep-disordered breathing. Over 50% of patients in the Lankford study (8/15) required an optimal pressure =10 cm H2O. The average weight loss at 6 months in the current investigation was 39.4 kg, and none of the patients had a BMI < 30 kg/m2.
The immediate change in optimal CPAP pressure may be explained by rapid weight loss and changes in hormone levels that regulate weight and metabolism. At 2 weeks post-surgery, patients had lost an average of 9.0 kg or nearly 6% of their body weight. Anthropometric measures and AHI may account for up to 76% of the variance in the CPAP pressure required to abolish OSA.30 Using the prediction equation CPAP pressure = -5.12 + 0.13(BMI) + 0.16(Neck) + 0.04(AHI), the predicted change in CPAP pressure in the present sample would be a decrease of 0.5 cm H2O 2 weeks after surgery. At 3 months the predicted change would be 1.4 cm H2O and at 6 months 2.5 cm H2O. The actual mean drop in P95 at 2-weeks post surgery was 1.9 cm H2O, close to a value intermediate to 3- and 6-months if one considered only the change in BMI and neck circumference. A similar prediction equation using BMI and AHI, but not neck circumference [(CPAP pressure = 0.52 + 0.174(BMI) + 0.042(AHI)], produ ced an even smaller reduction in P95 at each time point.31 Rapid initial weight loss may have altered fat deposition in the vicinity of the upper airway, accounting for a small portion of the variance in P95 and PMed.
Hormonal regulation of weight and metabolism may have accounted for a larger portion of the variance in CPAP pressure changes. Within weeks after surgery there is a significant decrease in blood glucose, insulin and leptin in bariatric surgery patients with and without NIDDM.32 The AHI is independently associated with these hormones after controlling for obesity and other risk factors.33,34 Treatment with CPAP produces a decrease in leptin that is observed within the first week.35,36 Furthermore, patients with effectively treated OSA, defined as an AHI less than 5, show a reduction in leptin, whereas patients on therapy but with incomplete control of OSA (AHI > 5) show no change in leptin.33 If OSA represents a metabolic disorder involving regulation of breathing during sleep, then endocrine changes may act on respiratory control mechanisms. Immediate and dramatic changes in hormone levels, coupled with rapid weight loss, may alter upper airway collapsibility. The result is a lower CPAP pressure required to maintain upper airway patency.
The change in CPAP pressure shortly after surgery does not necessarily reflect a significant or clinically relevant decrease in the AHI. In a sample of 14 patients studied pre-opreatively and at 4.5 months after surgery, the AHI decreased from a mean of 40 to 11 events/hr. The BMI in those patients decreased from 45 to 33 kg/m2.21 In a sample of 11 patients re-evaluated between 3–21 months after surgery, the AHI decreased from a mean of 56 events/hr pre-operatively to 23 events/hr. Two thirds of the patients in that sample still had moderate OSA, and most patients remained morbidly obese.13 Although not systematically evaluated, it appears that patients who continue to exhibit moderate OSA are those with a higher BMI.11,13,21 Patients remain at risk for cardiovascular morbidity if OSA is left untreated.
This is the first study to report CPAP compliance in a population of bariatric surgery patients. The adherence rate of 26% at 6 months is considerably lower than what is typically reported for other populations of CPAP users.28 Bariatric surgery patients represent a unique group with many barriers to CPAP compliance. These patients may have been symptomatic for OSA, but were not seeking treatment for their sleep disorder. Instead, the sleep evaluation was one component of a thorough diagnostic work up prior to surgery. Many patients perceive the diagnosis of OSA to be irrelevant and without consequence. Patients are not motivated to adhere to a treatment that is perceived as temporary since many of them believe weight loss will cure OSA. Patients may accept CPAP therapy to comply with the requirements of pre-surgery evaluation, but abandon therapy prematurely. Another factor contributing to the low adherence rate may be the marked improvement in quality of life that occurs within weeks after surgery; patients reported less depression, greater self-esteem, more energy and an improvement in overall health.37 Medical co-morbidities were not reported in that quality of life study, so it is not known if any patients had been diagnosed with OSA or were on CPAP therapy. A dramatic improvement in quality of life shortly after surgery may influence a patient’s decision to abandon CPAP therapy.
Patients in this study received extensive education regarding the diagnosis and treatment of OSA and the importance of Autoset use during weight loss. All polysomnograms were performed at an AASM-accredited hospital based sleep laboratory with a skilled technician in attendance. Respiratory therapists with extensive clinical experience reviewed Autoset use instructions with all patients during home set up. Despite intensive educational efforts, only a small percentage of patients used Autoset regularly for 6 months after surgery. This highlights the difficult task of treating OSA in this patient population. Physicians and surgeons need to emphasize the importance of treating OSA post-operatively and the associated risks of non-compliance with therapy. Bariatric surgery will not cure OSA in all patients. A post-operative polysomnogram is required to re-assess OSA severity.
The primary limitation of this study was the small sample size. An important assumption of inferential statistics is that the sample be representative of the population. The changes in PMax, P95 and PMed seen at 2-weeks for the larger group of 31 patients were similar in magnitude and absolute value to the subset of 8 patients that remained compliant for 6 months. The 23 patients who became irregular users or who abandoned treatment after 2-weeks were compared to the 8 compliant patients. The patients who remained compliant were younger, but the two groups did not differ on BMI, AHI, ODI, gender or the amount of weight loss at 2-weeks. Further support for the use of inferential statistics and repeated measures ANOVA is provided by an examination of the plot of the residuals, which shows a random distribution. This indicates that the model fits the data well. An important assumption for a repeated measures design is homogeneity of variance, which was confirmed with Levene’s test. These points justify the use of a repeated measures design to explore how CPAP pressures changed over time with weight loss.
Another potential limitation of the study was that we did not assess metabolic function, nor did we assess OSA severity after modest weight reduction. The study was designed to permit statistical evaluation of changes in weight loss and CPAP over time, which previously have not been reported. Analysis of insulin and leptin levels were not part of the research protocol, nor was it part of routine clinical evaluation. Follow-up polysomnography was scheduled to occur when the patient had lost 70% or more of excess body weight or after 18 months. A reassessment of OSA severity several weeks after surgery may have demonstrated a modest reduction in the AHI, but it probably would not have been statistically significant or clinically relevant. The degree of sleep-disordered breathing and hypoxemia would still be severe enough to warrant continued intervention with CPAP.
Self-adjusting CPAP was chosen over fixed pressure therapy for several reasons. First, the purpose of this study was to obtain pressure settings at several time points following surgery. It is highly unlikely that bariatric surgery patients would consent to several additional laboratory titration studies. Second, additional laboratory studies would be cost-prohibitive. Third, fixed time points would not permit examination of trends and patterns in pressure changes. Fourth, averaged data for several days’ use at home would provide a more representative sample of pressure requirements compared to a single night in the laboratory.
Future research will need to replicate and extend the findings of this study to a larger sample of bariatric surgery patients who remain compliant with therapy. Improving CPAP adherence in bariatric surgery patients will require intensive educational efforts directed at both physicians and patients regarding the high prevalence of OSA in morbid obesity. Aggressive efforts should continue at identifying and treating those patients with a high BMI who are at risk for OSA long before the pre-surgical evaluation. Bariatric surgery does not cure OSA in all patients, and continued use of CPAP may be needed.
This study was funded by a grant from the ResMed Foundation.The authors wish to thank Health Management, Inc. for their support.
1. Lavie P, Hoffstein V. Sleep apnea syndrome: A possible contributing factor to resistant hypertension. Sleep 2001; 24:721–725.
2. Flegal KM, Carroll MD, Kucamarski RJ et al. Overweight and obesity in the United States: prevalence and trends. Int J Obes 1998; 22:39–47.
3. Peeters A, Barendregt JJ, Willekens F et al. MEDCOM, the Netherlands Epidemiology and Demography Compression of Morbidity research group. Obesity in adulthood and its consequences for life expectancy: a life-table analysis. Ann Intern Med 2003; 138:24–32.
4. Fontaine KR, Redden DT, Wang C et al. Years of life lost due to obesity. JAMA 2003;289:187–193.
5. Young T, Palta M, Dempsey J et al. The occurrence of sleep-disordered breathing among middle-aged adults. N Engl J Med 1993;328:1230–1273.
6. O’Keeffe T, Patterson EJ. Evidence supporting routine polysomnography before bariatric surgery. Obes Surg 2004; 14:23–26.
7. Frey WC, Pilcher J. Obstructive sleep-related breathing disorders in patients evaluated for bariatric surgery. Obes Surg 2003; 13:676–683.
8. Buchwald H, Avidor Y, Braunwald E et al. Bariatric surgery: A systematic review and meta-analysis. JAMA 2004; 292:1724–1737.
9. Dabawala A, Cannan RJ, Stubbs RS. Improvements in co-morbidities following weight loss from gastric bypass surgery. Obes Surg 2000;10:428–435.
10. Guardiano SA, Scott JA, Ware JC et al. The long-term results of gastric bypass on indexes of sleep apnea. Chest 2003; 124:1615–1619.
11. Scheuller M, Weider D. Bariatric surgery for treatment of sleep apnea syndrome in 15 morbidly obese patients: Long-term results. Otolaryngol Head and Neck Surg 2001;125:299–302.
12. Charuzi I, Lavie P, Peiser J et al. Bariatric surgery in morbidly obese sleep-apnea patients: short- and long-term follow-up. Am J Clin Nutr 1992;55:594S–596S.
13. Rasheid S, Banasiak M, Gallagher SF et al. Gastric bypass is an effective treatment for obstructive sleep apnea in patients with clinically significant obesity. Obes Surg 2003;13:58–61.
14. Charuzi I, Ovnat A, Peiser J et al. The effect of surgical weight reduction on sleep quality in obesity-related sleep apnea syndrome. Surgery 1985;97:535–538.
15. Brown LT, Purcell GJ, Traugott FM. Hypoxemia during postoperative recovery using continuous pulse oximetry. Anaesth Intens Care 1990;18:509–516.
16. Clyburn PA, Rosen M, Vickers MD. Comparison of the respiratory effects of IV infusions of morphine and regional analegesia by extradural block. Br J of Anaesth 1990;64:446–449.
17. Eastwood PR, Szollosi I, Platt PR et al. Comparison of upper airway collapse during general anaesthesia and sleep. Lancet 2002; 359:1207–1209.
18. Gupta RM, Parvizi J, Hanssen AD et al. Postoperative complications in patients with obstructive sleep apnea syndrome undergoing hip and knee replacement: A case-control study. Mayo Clin Proc 2001; 76:897–905.
19. Moose D, Lourie D, Powell W et al. Laparoscopic Roux-en-Y gastric bypass: minimally invasive bariatric surgery for the superobese in the community hospital setting. Am Surg 2003; 69:930–932.
20. Luyckx FH, Scheen AJ, Desaive C et al. Effects of gastroplasty on body weight and related biological abnormalities in morbid obesity. Diabetes Metab 1998; 24:355–361.
21. Pillar G, Peled R, Lavie P. Recurrence of sleep apnea without concomitant weight increase 7.5 years after weight reduction surgery. Chest 1994; 106:1702–1704.
22. Lloberes P, Ballester E, Montserrat M et al. Comparison of manual and automatic CPAP titration in patients with sleep apnea/hypopnea syndrome. Am J Respir Crit Care Med 1996; 154:1755–1758.
23. Gagnadoux G, Rakotonanahary D, Martins de Araujo MT et al. Long-term efficacy of fixed CPAP recommended by Autoset for OSAS. Sleep 1999; 22:1095–1099.
24. Lankford DA, Proctor CD, Richard R. Continuous positive airway pressure (CPAP) changes in bariatric surgery patients undergoing rapid weight loss. Obes Surg 2005; 15:336–341.
25. Berthon-Jones M. Feasibility of a self-setting CPAP machine. Sleep 1993; 16:S120–S123.
26. Teschler H, Berthon-Jones M, Thompson AB et al. Automated continuous positive airway pressure titration for obstructive sleep apnea syndrome. Am J Respir Crit Care Med 1996; 154:734–740.
27. American Sleep Disorders Association. American Sleep Disorders Association Report: practice parameters for the indications for polysomnography and related procedures. Sleep 1997; 20:406–422.
28. Kribbs NB, Pack AI, Kline LR et al. Objective measurement of patterns of nasal CPAP use by patients with obstructive sleep apnea. Am Rev Respir Dis 1993; 147:887–895.
29. Guardiano SA, Scott JA, Ware JC et al. The long-term results of gastric bypass on indexes of sleep apnea. Chest 2003; 124: 1615–1619.
30. Miljeteig H, Hoffstein V. Determinants of continuous positive airway pressure level for treatment of obstructive sleep apnea. Am Rev Respir Dis 1993; 147:1526–1530.
31. Lin I, Chuang M, Liao Y et al. Predicting effective continuous positive airway pressure in Taiwanese patients with obstructive sleep apnea syndrome. J Formos Med Assoc 2003; 102:215–221.
32. Rubino F, Gagner M, Gentileschi P et al. The early effects of the Roux-en-Y gastric bypass on hormones involved in body weight regulation and glucose metabolism. Ann Surg 2004; 240(2):236–242.
33. Sanner BM, Kollhosser P, Buechner N et al. Influence of treatment on leptin levels in patients with obstructive sleep apnea. Eur Respir J 2004; 23:601–604.
34. Ip MSM, Lam B, Ng MMT et al. Obstructive sleep apnea is independently associated with insulin resistance. Am J Respir Crit Care Med 2002; 165:670–676.
35. Shimizu K, Chin K, Nakamura T et al. Plasma leptin levels and cardiac sympathetic function in patients with obstructive sleep apnea-hypopnea syndrome. Thorax 2002; 57:429–434.
36. Chin K, Shimizu K, Nakamura T et al. Changes in intra-abdominal visceral fat and serum leptin levels in patients with obstructive sleep apnea syndrome following nasal continuous positive airway pressure therapy. Circulation 1999; 100:706–712.
37. Dymek MP, Grange D, Neven K et al. Quality of life after gastric bypass surgery: a cross-sectional study. Obes Res