CHEST Recent Advances in Chest Medicine

CHEST Recent Advances in Chest Medicine Recent Advances in Obesity Hypoventilation Syndrome* Babak Mokhlesi, MD, MSc, FCCP; and Aiman Tulaimat, MD Obesity hypoventilation syndrome (OHS) consists of a combination of obesity and chronic hypercapnia accompanied by sleep-disordered breathing. During the last 3 decades, the prevalence of extreme obesity has markedly increased in the United States and other countries. With a global epidemic of obesity, the prevalence of OHS is bound to increase. Patients with OHS have a lower quality of life with increased health-care expenses and are at a higher risk for the development of pulmonary hypertension and early mortality compared to eucapnic patients with sleep-disordered breathing. Despite the significant morbidity and mortality associated with this syndrome, it is often unrecognized and treatment is frequently delayed. Clinicians must maintain a high index of suspicion since early recognition and treatment reduces the high burden of morbidity and mortality associated with this syndrome. In this review, we will discuss the definition and clinical presentation of OHS, provide a summary of its prevalence, review the current understanding of the pathophysiology, and discuss the recent advances in the therapeutic options. (CHEST 2007; 132:1322 1336) Key words: bilevel positive airway pressure; continuous positive airway pressure; hypercapnia; hypoventilation; obesity hypoventilation syndrome; pickwickian syndrome; sleep apnea; sleep-disordered breathing Abbreviations: AHI apnea-hypopnea index; AVAPS average volume-assured pressure support; BMI body mass index; CPAP continuous positive airway pressure; EPAP expiratory positive airway pressure; IPAP inspiratory positive airway pressure; NIPPV noninvasive positive-pressure ventilation; NREM non-rapid eye movement; OHS obesity hypoventilation syndrome; OSA obstructive sleep apnea; PAP positive airway pressure; REM rapid eye movement Before Burwell coined the term Pickwickian syndrome, 1 Auchincloss and colleagues 2 gave the first detailed description of a patient with obesity hypoventilation syndrome (OHS). Since then, our *From the Section of Pulmonary and Critical Care Medicine (Dr. Mokhlesi), The University of Chicago Pritzker School of Medicine, Chicago, IL; and the Division of Pulmonary and Critical Care Medicine (Dr. Tulaimat), Cook County Hospital and Rush University Medical Center, Chicago, IL. The authors have reported to the ACCP that no significant conflicts of interest exist with any companies/organizations whose products or services may be discussed in this article. Manuscript received January 4, 2007; revision accepted March 6, 2007. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (www.chestjournal. org/misc/reprints.shtml). Correspondence to: Babak Mokhlesi, MD, MSc, FCCP, Section of Pulmonary and Critical Care Medicine, The University of Chicago Pritzker School of Medicine, 5841 S Maryland Ave, MC 0999/Room L11B, Chicago, IL 60637; e-mail: bmokhles@ medicine.bsd.uchicago.edu DOI: 10.1378/chest.07-0027 knowledge about the epidemiology, pathophysiology, treatment, and outcomes of OHS has improved significantly. In the United States, a third of the adult population is obese, and the prevalence of extreme obesity (ie, body mass index [BMI] 40 kg/m 2 ) is increasing rapidly. From 1986 to 2000, the prevalence of BMI of 40 kg/m 2 has quadrupled, and that of BMI of 50 kg/m 2 has increased by fivefold. 3,4 The obesity epidemic is not only impacting adults in the United States, it is a global phenomenon affecting children and adolescents. 5 8 With such a global epidemic of obesity, the prevalence of OHS is likely to increase. In this review, we will discuss the definition and clinical presentation of OHS, provide a summary of its prevalence, attempt to give a cohesive and comprehensive review of its pathophysiology, and provide evidence that early recognition and treatment reduces the high burden of morbidity and mortality associated with this syndrome. 1322 Recent Advances in Chest Medicine

Definitions OHS is defined as a combination of obesity (ie, BMI 30 kg/m 2 ) and awake chronic hypercapnia (ie, Paco 2 45 mm Hg) accompanied by sleepdisordered breathing. 9,10 It is important to recognize that OHS is a diagnosis of exclusion and should be distinguished from other conditions that are commonly associated with hypercapnia (Table 1). In approximately 90% of patients with OHS, the sleepdisordered breathing consists of obstructive sleep apnea (OSA). 11 13 Due to this association, the term hypercapnic OSA has been interchangeably used with OHS. The remaining 10% of patients with OHS have an apnea-hypopnea index (AHI) 5. 11 13 The sleep-disordered breathing in this subset of patients has been labeled as sleep hypoventilation and is defined as an increase in Paco 2 during sleep by 10 mm Hg above wakefulness or a significant oxygen desaturation that is not explained by obstructive apneas or hypopneas. 9 Overlap syndrome is the term used to describe the association of COPD and OSA. The prevalence of overlap syndrome among consecutive patients with OSA has been reported to be between 10% and 15%. 12,14,15 The prevalence of COPD in patients with OSA, however, is similar to its prevalence in the general population. 16 Patients with the overlap syndrome have an obstructive pattern on spirometry and, in comparison with patients with simple OSA, are more likely to have hypoxemia, hypercapnia, and pulmonary hypertension. 14,15,17 Hypercapnia develops in patients with the overlap syndrome at a lower BMI and AHI than that of patients with OHS without obstructive defects seen on spirometry and at a higher FEV 1 than hypercapnic patients with pure COPD. The breathing pattern and hypercapnic ventilatory response in these patients is, however, similar to those in patients with OHS. 11,18 Congenital central hypoventilation syndrome is a disorder of ventilatory control that typically presents in newborns and results (in 90% of the cases) from a polyalanine repeat expansion mutation in the PHOX2B gene. 19 Symptomatic and asymptomatic children have survived to adulthood without ventilatory support. 20 These patients are heterozygous for the mildest of the PHOX2B polyalanine expansion mutations. 21,22 Clinical Presentation and Diagnosis In general, patients with OHS are middle-aged with a 2:1 male-to-female ratio. These patients tend to be extremely obese and experience significant sleep-disordered breathing. On presentation, the patients usually report the classic symptoms of OSA such as fatigue, hypersomnolence, loud habitual snoring, nocturnal choking episodes, and morning headaches. In contrast to patients with simple OSA, dyspnea, lower extremity edema, and low oxygen saturation measured by pulse oximetry during wakefulness are common. A restrictive defect seen on pulmonary function tests is common and is due to obesity. If left untreated, pulmonary hypertension and cor pulmonale can develop in patients with OHS. 23 Table 2 summarizes the clinical features of 631 patients with OHS reported in the literature. 11 13,17,24 34 Patients with OHS have an elevated serum bicarbonate level due to the metabolic compensation for the chronic respiratory acidosis. 12,13,28,35 Therefore, serum bicarbonate level is a reasonable test to screen for hypercapnia because it is readily available, physiologically sensible, and less invasive than an arterial puncture to measure blood gas levels. It was recently shown 13 that the serum bicarbonate level combined with the severity of OSA can be used as clinical predictors of OHS in patients with morbid obesity and OSA (Fig 1). Accordingly, arterial blood gas measurements should be obtained to confirm the presence and severity of daytime hypercapnia in patients with obesity and sleep-disordered breathing who have hypoxemia on pulse oximetry during wakefulness or elevated serum bicarbonate levels. 10,13 If hypercapnia is present, pulmonary function testing and chest imaging can be useful in excluding other causes of hypercapnia. Laboratory testing should also include thyroid function tests to exclude severe hypothyroidism and a CBC count to rule out second- Table 1 Definition of OHS Required Conditions Description Obesity BMI 30 kg/m 2 Chronic hypoventilation Awake daytime hypercapnia (Paco 2 45 mm Hg) Sleep-disordered breathing OSA (AHI 5 with or without sleep hypoventilation) present in 90% of cases; sleep hypoventilation (AHI 5) present in 10% of cases Exclusion of other causes Severe obstructive airways disease; severe interstitial lung disease; severe chest wall disorders (eg, of hypercapnia kyphoscoliosis); severe hypothyroidism; neuromuscular disease; and congenital central hypoventilation syndrome www.chestjournal.org CHEST / 132 / 4/ OCTOBER, 2007 1323

Table 2 Clinical Features of Patients With OHS* Variables Values Age, yr 52 (42 61) Male gender, % 66 (49 90) BMI, kg/m 2 44 (35 56) Neck circumference, cm 46.5 (45 47) ph 7.38 (7.34 7.40) Paco 2,mmHg 52 (47 61) Pao 2,mmHg 60 (46 74) Serum bicarbonate, meq/l 32 (31 33) Hemoglobin, g/dl 15 AHI 66 (20 100) Sao 2 nadir during sleep, % 65 (59 76) Percentage of TST with Sao 2 50 (46 56) 90%, % FVC, % predicted 73 (57 102) FEV 1, % of predicted 67 (53 92) FEV 1 /FVC ratio 77 (74 88) MRC dyspnea class 3 and 4, % 69 Epworth sleepiness scale score 14 (12 16) CPAP, cm H 2 O(n 86) 14 Bilevel PAP, cm H 2 O(n 55) 18/9 *Values are given as the mean (range) (includes a total of 631 patients with OHS). 11 13,17,24 34 MRC Medical Research Council; Sao 2 arterial oxygen saturation; TST total sleep time. ary erythrocytosis. An ECG in these patients could demonstrate signs of right heart strain, right ventricular hypertrophy, and right atrial enlargement. 36 patients with OSA. The prevalence ranges between 10% and 20% 11,13,25,26,29,31,32 and is higher in the subgroup of patients with extreme obesity (ie, BMI 40 kg/m 2 ) [Fig 2]. 13,26,30 Two studies 27,37 reported a much higher prevalence of OHS in patients with OSA. These studies were limited by the exclusive enrollment of Japanese men, a small sample size, or the inclusion of patients with COPD. 27,37 The prevalence of OHS among hospitalized adult patients with a BMI of 35 kg/m 2 has been reported at 31% of patients. 28 Although the prevalence of OHS tends to be higher in men, the male predominance is not as clear as in OSA (Table 2). In fact, three studies 17,28,38 had a higher proportion of women with OHS. Similarly, there is no clear racial or ethnic predominance. However, the prevalence of OHS might be higher in African Americans due to the high prevalence of extreme obesity. 3,39 Because of cephalometric differences, OHS associated with OSA occurs at a lower BMI in Asians compared to whites. 32,40,41 Taken together, these findings suggest that OHS may be more prevalent in the United States than in other nations because of its obesity epidemic. Up to a fifth of patients with OSA may have OHS; this high prevalence is closely related to obesity and is bound to increase with the rising incidence of extreme obesity around the globe. 5 8,42 Epidemiology As measurement of arterial blood gases is not a standard practice in patients with OSA or extreme obesity, the precise prevalence of OHS in the general population remains uncertain. Table 3 summarizes several studies from various geographical regions that estimated the prevalence of OHS among Morbidity and Mortality Compared to eucapnic patients with OSA, patients with OHS have a lower quality of life, higher healthcare expenses, and a greater risk of pulmonary hypertension. Even patients with mild OHS (ie, Paco 2 between 46 to 50 mm Hg) are more somnolent and have a lower quality of life than patients Figure 1. Decision tree to screen for OHS in patients with OSA (AHI 5) and BMI of 30 kg/m 2. The predictors were obtained from a sample of 163 patients with OSA and were validated prospectively in a sample of 359 patients with OSA. 13 1324 Recent Advances in Chest Medicine

Table 3 Prevalence of OHS in Patients With OSA* Study Patients, No. Design Country Age, yr BMI AHI OHS, % Verin et al 31 218 Retrospective France 55 34 51 10 Laaban and Chailleux 26 1,141 Retrospective France 56 34 55 11 Kessler et al 11 254 Prospective France 54 33 76 13 Resta et al 29 219 Prospective Italy 51 40 42 17 Golpe et al 25 175 Retrospective Spain NA 32 42 14 Akashiba et al 37 143 Retrospective Japan 48 30 53 38 Akashiba et al 32 611 Retrospective Japan 48 29 52 9 Leech et al 27 111 Prospective United States 47 0.71 58 37 Mokhlesi et al 13 359 Prospective United States 48 43 62 20 *OSA has been defined by a range of AHI from 5 to 15. NA not applicable. Age, BMI, and AHI values are given as the mean of all patients (ie, OSA and OHS patients) and were calculated from the data provided by the authors in the article. Did not exclude patients with obstructive defects found on spirometry. Value given as height/weight ratio reported in kilograms per centimeter. with OSA when matched for age, BMI, and lung function. 43 Compared to patients with similar degrees of obesity, patients with OHS have increased medical resource utilization and are more likely to be hospitalized and require intensive care monitoring. 28,33 Pulmonary hypertension is more common (50% vs 15%, respectively) and more severe in patients with OHS than in those with OSA. 11,44 46 Other common comorbidities seen in patients with OHS are described in Table 4. Earlier case series of hospitalized patients with severe OHS reported a mortality rate approaching 50% including cases of sudden unexpected deaths. 47,48 In two more recent prospective studies, 28,49 however, there were no in-hospital deaths among a total of 64 consecutive patients with OHS. Compared to patients with a similar degree of obesity but without hypoventilation, patients with OHS had higher rates of ICU admission (6% vs 40%, respectively) and a greater need for invasive mechanical ventilation (0% vs 6%, respectively). 28 Observational and retrospective studies have demonstrated that the treatment of OHS is associated with lower long-term morbidity and mortality. Positive airway pressure (PAP) therapy reduces healthcare expenses and hospital readmission rates. 12,33,38 A retrospective study 12 reported that 7 of 15 patients with OHS (46%) who refused long-term noninvasive PAP therapy died during an average 50-month follow-up period. Similarly, in a prospective study, 28 47 patients with OHS were followed for 18 months after hospital discharge. The mortality of patients with OHS was 23% compared to 9% in patients with a similar degree of obesity but without hypoventilation (hazards ratio, 4.0), and most deaths occurred in the first 3 months after hospital discharge (Fig 3). 28 Only 13% were discharged from the hospital while receiving treatment for hypoventilation. In contrast, the 2-year to 4-year mortality rate in patients with OHS treated with PAP is 10%. 12,33,50 Two large observational studies 51,52 from 2005 have reported an increased mortality and cardiovascular morbidity in Table 4 Comorbidities Reported in Patients With OHS Conditions Prevalence, % Figure 2. The prevalence of OHS in patients with OSA by different categories of BMI in three countries. 13,26 The data from Italy were provided by Professor Onofrio Resta from the University of Bari, Italy. Hypertension 12,17,28 61 79 Heart failure 12,17,28 21 32 Pulmonary hypertension 59 88 (mean PAP 20 mm Hg) 11,46 Significant pulmonary hypertension 31 (mean PAP 40 mm Hg) 46 Type 2 diabetes mellitus 11,17 30 32 Asthma 12,17 18 24 Erythrocytosis* 12,17,46 8 15 *Hemoglobin: women, 16 g/dl; men, 18 g/dl. www.chestjournal.org CHEST / 132 / 4/ OCTOBER, 2007 1325

Figure 3. Survival curves for patients with OHS (n 47; mean BMI, 45 kg/m 2 ) vs simple obesity (n 103; mean BMI, 42 kg/m 2 ). All patients survived hospitalization, and only 13% of patients with OHS were discharged from the hospital while receiving therapy for hypoventilation. The hazards ratio for mortality in patients with OHS was 4.0 after adjustment for BMI, age, gender, electrolyte abnormalities, renal function, history of thromboembolic disease, and hypothyroidism. Reprinted from Nowbar et al 28 with permission from Elsevier Publishers. patients with severe OSA who are not adherent to PAP therapy. Although these reports did not exclusively include patients with OHS, the majority of patients with OHS do have severe OSA. In addition to adherence with PAP therapy, COPD, smoking, and FEV 1 were also predictors of mortality in patients with OSA. 52 54 Accordingly, we believe that identifying patients with OHS in a timely manner is important and treatment should be initiated without delay to avoid adverse outcomes such as readmission to the hospital, acute on chronic respiratory failure requiring intensive care monitoring, or death. More importantly, adherence with therapy should be emphasized and monitored objectively. 17 Pathophysiology Paco 2 is determined by the balance between CO 2 production and elimination (ie, minute ventilation and the fraction of dead space ventilation). In patients with OHS, short-term treatment with continuous PAP (CPAP) or bilevel PAP improves hypercapnia without any significant changes in body weight, CO 2 production, or the volume of dead space. Therefore, this disorder is entirely due to hypoventilation. 35 The exact mechanisms that lead to hypoventilation in obese individuals remain controversial. Since the initial description of the syndrome by Auchincloss et al 2 in 1955, three factors have been classically tested to explain the evolution of this disorder. These factors are the excessive mechanical load imposed on the respiratory system by excess weight, a blunted central respiratory drive, and sleep-disordered breathing (Fig 4). Recently, Norman and colleagues 55 proposed a model that combines sleepdisordered breathing, central respiratory dive, and renal buffering to explain the development of this condition. The Excessive Load on the Respiratory System Upper Airway Obstruction: The assessment of upper airway resistance by impulse oscillometry during wakefulness has demonstrated that patients with OHS have a higher upper airway resistance both in the sitting and supine position when compared to patients with moderate-to-severe OSA with similar degrees of obesity and control subjects. In contrast to patients with OHS, the upper airway resistance in patients with OSA is increased only in the supine position. 56 Respiratory Muscles: The maximal inspiratory and expiratory pressures are normal in eucapnic morbidly obese patients but are reduced in patients with OHS. 57 59 Patients with mild OHS, however, might have normal inspiratory and expiratory pressures. 34 A more accurate assessment of the diaphragmatic strength by cervical magnetic stimulation has not been performed in patients with OHS. 60 The role of 1326 Recent Advances in Chest Medicine

Figure 4. Potential mechanisms by which obesity can lead to chronic daytime hypercapnia. See the Pathophysiology section for a more detailed explanation. diaphragmatic weakness in the pathogenesis of this disorder remains uncertain because patients with OHS can generate similar transdiaphragmatic pressures at any level of diaphragmatic activation compared to eucapnic obese subjects. 58 Respiratory System Mechanics: Patients with OHS have lower respiratory system compliance when compared to eucapnic morbidly obese patients and to nonobese control subjects (0.045 vs 0.081 vs 0.104 L/cm H 2 O, respectively). This reduction is due to a 50% reduction in chest wall compliance. Patients with OHS also have a threefold increase in lung resistance that has been attributed to a low functional residual capacity. 61,62 The changes in lung mechanics are frequently demonstrated on spirometry by a low FVC and FEV 1 and a normal FEV 1 / FVC ratio. The spirometric abnormalities may be related to the combination of abnormal respiratory mechanics and weak respiratory muscles. 27,29,30,63,64 The changes in respiratory system mechanics in subjects with OHS imposes a significant load on the respiratory muscles and leads to a threefold increase in the work of breathing. 62 As a result, morbidly obese patients dedicate 15% of their oxygen consumption to the work of breathing compared to 3% in nonobese individuals. 65 Taken together, the data suggest that obesity imposes a significant load on the respiratory system in patients with OHS. Obesity is not, however, the only determinant of hypoventilation since hypercapnia develops in less than one third of morbidly obese individuals. 13,26,63 Central Respiratory Drive Hypercapnic Ventilatory Response: Obese eucapnic subjects and patients with OHS have a similar respiratory drive measured by mouth occlusion pressure over the first 100 ms of inspiration against an occluded airway (P 0.1 technique), and these levels are higher than those seen in nonobese individuals. 34,58 Patients with OHS are unable to increase their respiratory drive as much as obese eucapnic subjects in response to a hypercapnic challenge. The slope of the ventilatory response to hypercapnia is 1 L/min/mm Hg in patients with obesity hypoventilation, 2 L/min/mm Hg in eucapnic obese subjects, and 3 L/min/mm Hg in healthy subjects. 34,35,58,66 The response of the timing components in the breathing pattern to hypercapnia (ie, duration of respiratory cycle, inspiratory time, and duty cycle ratio) is similar among the three subject groups. Accordingly, the ventilatory response is diminished due to an inadequate increase in the tidal volume as a result of a blunt neural response to hypercapnia. 58,67,68 Obesity, genetic predisposition, sleep-disordered breathing, and leptin resistance have been proposed as mechanisms for the blunt response to hypercapnia. The weight load was suggested as a mechanism www.chestjournal.org CHEST / 132 / 4/ OCTOBER, 2007 1327

behind the blunt respiratory drive because weight loss improves Paco 2 levels in patients with OHS. But this is unlikely to be related directly to weight because, if anything, weight loss blunts the response of eucapnic morbidly obese subjects to hypercapnia. 69 The blunt respiratory response to hypercapnia is also unlikely to be familial because the ventilatory response to hypercapnia is similar between firstdegree relatives of patients with OHS and control subjects. 70 The treatment of sleep-disordered breathing with PAP therapy might improve the response to hypercapnia. 34,66,71 The P 0.1 response to hypercapnia improves as early as after 2 weeks of therapy with PAP and reaches normal levels after 6 weeks of therapy with PAP in patients with mild OHS (Paco 2 between 46 to 50 mm Hg). The response of minute ventilation to hypercapnia improves by the sixth week of therapy but does not normalize. 34 These findings are not universal, as other investigators have reported a left and upward shift in the ventilatory response curve to hypercapnia without any change in the slope 66 or even no improvement in the ventilatory response after treatment despite improvement in Paco 2. 35,72 Leptin: The product of the adipocyte-specific ob gene, leptin primarily regulates food intake and energy expenditure. Mutations in the ob gene that result in a lack of leptin lead to obesity in ob/ob mice. The treatment of these mice with leptin leads to weight loss. The breathing pattern of ob/ob mice is similar to that of patients with OHS (ie, a rapid breathing frequency and an attenuated hypercapnic ventilatory response). These mice fail to generate adequate minute ventilation during severe hypercapnia due to an attenuated increase in the tidal volume, but not in the breathing frequency. It is important to note that these changes in the control of breathing are independent of weight gain as they occur before the onset of profound obesity in ob/ob mice. 73,74 The total lung capacity and the lung compliance of these mice are half of that in wild-type mice. The proportion of the diaphragmatic myosin heavy chain type I is increased, and the proportion of the type II myosin chain is decreased, conferring resistance to fatigue. The long-term replacement of leptin in these mice prevents or attenuates these changes in the breathing pattern, lung mechanics, and myosin level. In contrast to ob/ob mice, ventilation is appropriately compensated in wild-type mice with diet-induced obesity that leads to high endogenous leptin levels. 75 In humans, however, the leptin serum level and leptin messenger RNA levels in adipocytes are strongly related to the percentage of body fat. 76 Patients with OSA have high serum leptin levels that are mostly associated with obesity and are unrelated to OSA. 77 Dietary restrictions reduce serum leptin levels and attenuate the hypercapnic ventilatory response, suggesting that leptin acts to maintain minute ventilation in response to obesity. 78 80 Patients with OHS have a higher serum leptin level than eucapnic subjects with OSA matched for percentage of body fat and AHI, and their serum leptin level drops after treatment with PAP. 81 83 These observations suggest that patients with OHS might be resistant to leptin. For leptin to affect the respiratory center and increase minute ventilation, it has to penetrate into the cerebrospinal fluid. The mean ( SE) leptin CSF/serum ratio is fourfold higher in lean individuals compared to obese subjects (0.047 0.01 vs 0.011 0.002, respectively; p 0.05). 84 This variability among individuals could lead to variability in ventilatory control among obese individuals and could explain the leptin resistance seen in patients with hypercapnia. A recent pilot study 85 of six patients with OHS without concomitant OSA (ie, AHI 5) reported that serum leptin levels were lower in patients with OHS without OSA compared to matched obese subjects without OSA. The serum leptin levels increased after successful long-term treatment with bilevel PAP without any significant change in BMI. The increase in serum leptin level correlated with the improvement in central CO 2 chemosensitivity. 85 These findings suggest that chronic hypoxia may suppress serum leptin levels, and the low leptin levels can lead to a blunt response to hypercapnia. Successful treatment with bilevel PAP was associated with an increase in the hypercapnic ventilatory response and correlated strongly with the increase in serum leptin levels. The exclusion of patients with OSA may explain the contrast between this study and a previous study 83 that reported a reduction in serum leptin levels after the successful treatment of OHS with PAP therapy. Hypoxic Ventilatory Response: Ventilatory response to hypoxia is also blunted in patients with obesity hypoventilation. This abnormality is not familial and improves with therapy. 34,70 Sleep-Disordered Breathing In patients with OHS, sleep-disordered breathing can occur in the following three forms: obstructive apneas and hypopneas; obstructive hypoventilation due to increased upper airway resistance; and central hypoventilation. 11,24 The role of OSA in the pathogenesis of hypoventilation has been established by the resolution of hypercapnia in the majority of patients with OHS with either PAP therapy or tracheostomy. 12,17,24,34,35,38,86,87 But how would OSA lead to chronic daytime hypercapnia? In patients with OSA, minute ventila- 1328 Recent Advances in Chest Medicine

tion during sleep does not decrease due to the large increase in the minute ventilation between the obstructive respiratory events. Obstructive respiratory events can, however, lead to acute hypercapnia if the duration of the interevent hyperventilation is inadequate to eliminate the accumulated CO 2. 88 This acute hypercapnia causes a small increase in serum bicarbonate level that is not corrected before the next sleep period if the time constant of bicarbonate excretion is longer than that of CO 2. 89 The elevated bicarbonate level blunts the ventilatory response to CO 2 from its initial value by reducing the change in hydrogen ions for a given change in CO 2 and would ultimately result in a higher waking CO 2 level. 55,90 92 In the subgroup of patients with OHS who have an AHI of 5, minute ventilation decreases by 25% during non-rapid eye movement (NREM) sleep and by 40% during rapid eye movement (REM) sleep. 59 This reduction in minute ventilation is due to a drop in the tidal volume. Hypercapnia would subsequently trigger metabolic compensation that would ultimately result in chronic hypercapnia, as mentioned above. Although short-term total sleep deprivation has been associated with a decreased hypercapnic ventilatory response, 93,94 the contribution of sleep fragmentation associated with OSA to the pathogenesis of OHS remains to be elucidated. Treatment Figure 5. Therapeutic algorithm during PAP titration in patients with OHS. See PAP Therapy section for a more detailed explanation. The optimal management of patients with OHS remains uncertain. Several studies have reported improvement in chronic daytime hypercapnia and hypoxia with PAP therapy (CPAP or bilevel PAP). Approximately half of patients with OHS require oxygen therapy in addition to PAP therapy upon initiation of treatment. Although PAP is the mainstay of therapy in both OSA and OHS patients, there is no standard protocol for its titration. 95 Figure 5 provides a therapeutic algorithm during polysomnography in order to address the variety of respiratory events that are observed in patients with OHS. 24 Even though autoadjusting PAP technology can be used in patients with simple OSA to bypass laboratory-based titration studies, this technology cannot be recommended in patients with OHS because it does not have the ability to recognize hypoventilation and hypoxemia. As a result, patients with OHS require a laboratory-based PAP therapy and oxygen titration. The use of phlebotomy has not been systematically studied in patients with OHS in whom secondary erythrocytosis develops. Secondary erythrocytosis is a physiologic response to tissue hypoxia in order to enhance oxygen-carrying capacity. However, hyperviscosity impairs oxygen delivery and can counteract the beneficial effects of erythrocytosis. In adult patients with congenital cyanotic heart disease, phlebotomy has been recommended if the hematocrit is 65% only if symptoms of hyperviscosity are present. 96 However, it is difficult to extrapolate this recommendation to patients with OHS because many symptoms of hyperviscosity are similar to the symptoms of OHS. Reversing hypoventilation and hypoxemia with PAP therapy eventually improves secondary erythrocytosis, and phlebotomy is rarely needed in patients with OHS. 50 PAP Therapy CPAP Therapy: Given that the majority of patients with OHS have concomitant severe OSA, treatment with CPAP seems reasonable. 97,98 CPAP therapy www.chestjournal.org CHEST / 132 / 4/ OCTOBER, 2007 1329

alone is effective in most patients with OHS, particularly those who have concomitant OSA. A recent prospective controlled study 99 compared the impact of a full night of CPAP titration without supplemental oxygen therapy or bilevel PAP therapy between 23 patients with OHS and 23 patients with eucapnic OSA who were matched for BMI, AHI, and lung function. Both patient groups were extremely obese and had severe sleep-disordered breathing, and those with OHS had significant daytime hypercapnia. In more than half of patients with OHS (57%), CPAP therapy resolved sleep-disordered breathing and nocturnal hypoxemia. The optimal mean CPAP pressure of 13.9 3.1 cm H 2 O was reached within 1 h of sleep onset. Ten patients (43%) with OHS, however, had refractory hypoxemia during CPAP titration. Compared to the OHS patients who were successfully titrated, they had a higher BMI, more severe nocturnal hypoxemia on the baseline polysomnogram, and a higher residual AHI during the night of CPAP titration. 99 The fact that more than half of the patients with stable but extreme cases of OHS (based on BMI, AHI, and the level of daytime hypercapnia) were successfully titrated with CPAP, without requiring bilevel PAP or supplemental oxygen, suggests that the majority of patients with milder forms of OHS can be successfully titrated with CPAP as well. However, given the lack of intermediate and long-term follow-up, it is difficult to establish whether outcomes with CPAP therapy would be similar to those with bilevel PAP therapy. 100 The improvement in hypercapnia and hypoxia is directly related to the daily dose of PAP therapy, and maximum improvement in blood gas levels is achieved as early as 1 month after the start of therapy. 17,34,101 In a study of 75 ambulatory patients with OHS, 17 Paco 2 decreased by 1.8 mm Hg and Pao 2 increased by 3 mm Hg per hour of daily CPAP or bilevel PAP use. Moreover, patients who used PAP therapy for 4.5 h/d experienced larger improvement in Paco 2 and Pao 2 compared to less adherent patients ( Paco 2 7.7 mm Hg vs 2.4 mm Hg, respectively [p 0.001]; Pao 2 9.2 mm Hg vs 1.8 mm Hg, respectively [p 0.001]). Similarly, the need for daytime home oxygen therapy decreased from 30 to 6% in adherent patients. There was no significant difference in improvement of hypercapnia and hypoxemia between patients receiving CPAP therapy (n 48) and patients receiving bilevel PAP therapy (n 27). The improvement in chronic daytime hypercapnia in patients who are adherent with PAP therapy is neither universal nor complete because up to 25% of patients who are adherent to PAP therapy do not become eucapnic (Table 5). 17 Bilevel PAP Therapy: Bilevel PAP therapy, also known as noninvasive positive-pressure ventilation (NIPPV), should be considered if during PAP titration the oxygen saturation remains persistently below 90% after the resolution of apneas, hypopneas, and flow limitation with CPAP therapy (Fig 5). During PAP titration, the expiratory PAP (EPAP) should be increased until there is resolution of obstructive respiratory events (eg, apneas, hypopneas, and flow limitation). If the oxygen saturation remains persistently below 90% after the resolution of apneas, hypopneas, and flow limitation, inspiratory PAP (IPAP) should be added to the final EPAP to improve ventilation. IPAP was at least 8 to 10 cm H 2 O above EPAP in the studies that showed successful long-term resolution of hypercapnia and hypoxia with bilevel PAP therapy. 12,24,85,87 In patients with OHS who do not have OSA, EPAP can be set at 5 cm H 2 O and IPAP can be titrated to improve ventilation. 85,87 Bilevel PAP should also be considered if the Paco 2 does not normalize after 1 to 2 months of therapy with CPAP. A preliminary report of a randomized controlled trial 102 of 20 patients with OHS has suggested that 3 months of therapy with bilevel PAP may be more effective than CPAP therapy in improving awake hypercapnia ( Paco 2, 7.8 vs 5.3 mm Hg). Whether the difference in Paco 2 of 2.5 mm Hg is clinically relevant remains to be elucidated. Many patients with OHS who initially require bilevel PAP therapy can be switched to CPAP therapy after the resolution of hypercapnia. 12,101 Bilevel PAP therapy might prevent endotracheal intubation and invasive mechanical ventilation in patients with OHS during acute-on-chronic respiratory failure. 12,49 The impact of long-term NIPPV on vital capacity and lung volumes is contradictory. Several studies 12,38,72 have reported no change in lung volumes or FVC after successful treatment of OHS patients with bilevel PAP therapy. In contrast, two more recent studies 50,87 of patients with OHS reported significant improvements in vital capacity and expiratory reserve volume after 12 months of NIPPV therapy without any significant changes in BMI or FEV 1 / FVC ratio. Table 5 Reasons for Lack of Improvement of Hypercapnia With PAP Therapy in Patients With OHS Inadequate adherence with PAP therapy Inadequate PAP titration Sleep-disordered breathing other than OSA (central hypoventilation) Unidentified respiratory disease (ie, COPD and interstitial lung disease) Unidentified hypothyroidism or neuromuscular disease Metabolic alkalosis (ie, due to high doses of loop diuretics) 1330 Recent Advances in Chest Medicine

Average Volume-Assured Pressure Support: Nocturnal treatment with volume-limited and pressurelimited NIPPV are equally effective in patients with chronic respiratory failure. 103,104 Although pressurelimited NIPPV (bilevel) is easier to tolerate, volumelimited NIPPV provides a more stable tidal volume. 105 Average volume-assured pressure support (AVAPS), a hybrid mode that delivers a more consistent tidal volume with the comfort of pressure support ventilation, has been compared to bilevel PAP in a randomized crossover trial 106 in patients with OHS who were unable to achieve a transcutaneous CO 2 level 45 mm Hg and an AHI 10 during CPAP titration. Six weeks of therapy with AVAPS was more successful than bilevel PAP therapy in improving nocturnal and daytime ventilation (mean transcutaneous CO 2 during sleep, 45 3vs 52 4 mm Hg, respectively; mean daytime Paco 2, 42 5vs46 4 mm Hg, respectively). Changes in sleep quality and quality of life, however, were similar between the two modes of ventilation. 106 Oxygen Therapy: Approximately half of patients with OHS require supplemental nocturnal oxygen in addition to PAP therapy. 12,17,38,50 The need for nocturnal and daytime oxygen therapy decreases significantly in patients who are adherent to PAP therapy. 12,17,38 Supplemental oxygen without PAP therapy is inadequate and does not improve hypoventilation. 107 Taken together, the data suggest that CPAP therapy is effective in the majority of stable patients with OHS, particularly in the subgroup of patients who have severe OSA. Bilevel PAP therapy should be strongly considered in patients who do not respond to CPAP therapy, patients with OHS who experience acute-on-chronic respiratory failure, and in patients who have OHS without OSA. Whether AVAPS therapy has long-term benefits over bilevel PAP therapy remains uncertain. The treatment of OHS with PAP improves blood gas levels, morning headaches, excessive daytime sleepiness and vigilance, dyspnea, pulmonary hypertension, and leg edema. 12,38,72 Improvements in symptoms and blood gas levels are directly related to adherence with therapy, and maximum improvement in blood gas levels can be achieved as early as after 2 to 4 weeks of therapy. Therefore, early follow-up is imperative and should include repeat measurement of arterial blood gases and objective assessment of adherence with PAP therapy as patients frequently overestimate adherence. 108 110 Changes in serum bicarbonate level and pulse oximetry could be used as a less invasive measure of ventilation. Changing bilevel PAP to CPAP therapy and discontinuing oxygen therapy when no longer indicated can decrease the cost of therapy in patients with OHS. Surgery Weight Reduction Surgery: Weight loss after bariatric surgery is effective in treating OSA. A drop in BMI from 56 to 36 kg/m 2 was associated with a decrease in the AHI from 72 to 19. 111 This improvement can be enough to allow adequate ventilation between obstructive events, therefore improving ventilation during sleep. The weight loss after bariatric surgery has been associated with long-term improvement in arterial blood gas levels in patients with OHS. In a series of 12 patients who lost 45% of excess body weight and in whom arterial blood gas levels were available 5 years after surgery, Pao 2 increased from 54 to 68 mm Hg and Paco 2 decreased from 53 to 47 mm Hg. At the same time, FVC, FEV 1, expiratory reserve volume, and total lung capacity improved, and hemoglobin levels normalized in the one patient who had secondary erythrocytosis. 112,113 Gastric bypass, in the form of a Roux-en-Y procedure, produces greater weight loss and weight loss maintenance than purely restrictive approaches such as gastric banding. The average excess weight loss in patients who have undergone gastric bypass ranges from 65 to 75%, corresponding to a loss of approximately 35% of initial weight. 114,115 However, weight gain and significant increase in AHI can occur between 3 and 7 years after gastric bypass surgery. 116 Bariatric surgery is, however, associated with significant risk. The perioperative mortality rate is between 0.5% and 1.5%. OHS may be associated with higher operative mortality. 113 The independent risk factors associated with mortality are as follows: intestinal leak; pulmonary embolism; preoperative weight; and hypertension. Depending on the type of surgery, intestinal leak occurs in 2 to 4% of patients, and pulmonary embolism occurs in 1% of patients. 117 We believe that patients with OHS should be treated with PAP therapy, or with tracheostomy in patients who do not respond to PAP therapy, before undergoing surgical intervention in order to decrease perioperative morbidity and mortality. We also believe that PAP therapy should be initiated immediately after extubation to avoid postoperative respiratory failure. 118 120 Moreover, there is no evidence that PAP therapy initiated postoperatively leads to anastomotic disruption or leakage. 119,121 Tracheostomy: There have been no large studies evaluating long-term outcome after tracheostomy in patients with OHS. Three of seven patients normalized their Paco 2 2 weeks after undergoing tracheostomy for the treatment of OHS. 35 Minute ventilation remained at the pretreatment level in the four patients in whom Paco 2 did not normalize. The www.chestjournal.org CHEST / 132 / 4/ OCTOBER, 2007 1331

responders were similar to the nonresponders in FVC, FRC, FEV 1 /FVC ratio, and physiologic dead space. 35 Tracheostomy can lead to the complete resolution of obstructive respiratory events in patients with simple OSA. However, sleep-disordered breathing can persist in patients with OHS after tracheostomy. In a retrospective study 121 of 13 patients with OHS, tracheostomy was associated with a significant improvement in OSA. With the tracheostomy closed, the mean NREM AHI and REM AHI were 64 and 46, respectively; with the tracheostomy open, the mean NREM AHI and REM AHI decreased to 31 and 39, respectively. In seven patients, the AHI remained at 20. These residual respiratory events were associated with persistent respiratory effort, suggesting that disordered breathing was caused by hypoventilation through an open tracheostomy rather than central apneas. However, the overall improvement in the severity of sleep-disordered breathing after tracheostomy led to the resolution of hypercapnia in the majority of the patients. 122 Pharmacotherapy Medroxyprogestrone: The respiratory response to progesterone is mediated at the hypothalamus through an estrogen-dependent progesterone receptor, a mechanism that is similar to that mediating its reproductive effects. 123 Medroxyprogesterone could have a role in the treatment of patients with OHS because it reduces the AHI in patients with OSA and increases the ventilatory response to hypercapnia. 124 The results of treatment in patients with OHS have been contradictory. In a series of 10 patients who were treated with medroxyprogestrone, Paco 2 decreased from 51 to 38 mm Hg, and the Pao 2 increased from 49 to 62 mm Hg. 125 In contrast, medroxyprogestrone did not improve Paco 2, minute ventilation, and ventilatory response to hypercapnia in three patients who remained hypercapnic after undergoing tracheostomy. 35 Most but not all patients with OHS can normalize their Paco 2 with voluntary hyperventilation. 126 The inability to eliminate CO 2 with voluntary hyperventilation may be due to mechanical impairment. In one study, 127 the ability to drop the Paco 2 by at least 5 mm Hg with voluntary hyperventilation was the main predictor of a favorable response to medroxyprogesterone. Therefore, it is probably reasonable to evaluate the ability of patients to lower Paco 2 by at least 5 mm Hg with voluntary hyperventilation before starting treatment with respiratory stimulants. Clinicians, however, should be aware that medroxyprogesterone can increase the risk of venous thromboembolism. 128,129 Acetazolamide: In contrast to loop diuretics that cause metabolic alkalosis, the carbonic anhydrase inhibitor acetazolamide causes metabolic acidosis. Within 24 h of dosing, the serum bicarbonate level drops by 4 to 6 meq/l and the ph drops by 0.05 to 0.1; this metabolic acidosis increases the minute ventilation by 15% and reduces the Paco 2 level by 5 to 6 mm Hg. The decrease in Paco 2 is not due to a change in the slope of ventilatory response to hypercapnia but to a left shift in the CO 2 response curve by 7.3 mm Hg. 130,131 Acetazolamide could have a role in the treatment of OHS for three reasons. First, the insight into the role of metabolic compensation for acute respiratory acidosis during sleep in the development of hypercapnia could make it a suitable agent for preventing the development of metabolic alkalosis in patients with severe OSA. 55 Second, it corrects the right shift of the CO 2 response curve in patients with OHS. 66 Third, it can reduce the frequency of obstructive events in patients with moderate-to-severe OSA. 132,133 In fact, three patients who remained hypercapnic after undergoing tracheostomy became eucapnic after treatment with acetazolamide, 250 mg daily for 2 weeks. 35 In summary, the treatment options other than PAP therapy have been poorly studied. It is, therefore, essential to aggressively encourage adherence with PAP therapy to prevent the serious adverse outcomes of OHS. If PAP therapy fails to achieve the desired results, physicians should consider weight reduction surgery, tracheostomy, and pharmacotherapy with respiratory stimulants. These therapeutic modalities will not completely eliminate hypoventilation, and the patient might require a combination of treatments such as tracheostomy combined with either mechanical ventilation or acetazolamide administration. Conclusion With such a global epidemic of obesity, the prevalence of OHS is likely to increase. Despite the significant morbidity and mortality associated with OHS, it is often unrecognized, and treatment is frequently delayed. It is essential for clinicians to maintain a high index of suspicion, particularly because early recognition and treatment improve outcomes. Further research is needed to better understand the pathophysiology and long-term treatment outcomes of patients with OHS. References 1 Burwell CS, Robin ED, Whaley RD, et al. Extreme obesity associated with alveolar hypoventilation: a Pickwickian syn- 1332 Recent Advances in Chest Medicine

drome. Am J Med 1956; 21:811 818 2 Auchincloss JH Jr, Cook E, Renzetti AD. Clinical and physiological aspects of a case of obesity, polycythemia and alveolar hypoventilation. J Clin Invest 1955; 34:1537 1545 3 Freedman DS, Khan LK, Serdula MK, et al. Trends and correlates of class 3 obesity in the United States from 1990 through 2000. JAMA 2002; 288:1758 1761 4 Sturm R. Increases in clinically severe obesity in the United States, 1986 2000. Arch Intern Med 2003; 163:2146 2148 5 Prentice A, Webb F. Obesity amidst poverty. Int J Epidemiol 2006; 35:24 30 6 Skidmore PM, Yarnell JW. The obesity epidemic: prospects for prevention. Q J Med 2004; 97:817 825 7 Spritzer DA. Obesity epidemic migrates east [letter]. Can Med Assoc J 2004; 171:1159 8 Miech RA, Kumanyika SK, Stettler N, et al. Trends in the association of poverty with overweight among US adolescents, 1971 2004. JAMA 2006; 295:2385 2393 9 American Academy of Sleep Medicine. Sleep-related breathing disorders in adults: recommendations for syndrome definition and measurement techniques in clinical research: the report of an American Academy of Sleep Medicine task force. Sleep 1999; 22:667 689 10 Olson AL, Zwillich C. The obesity hypoventilation syndrome. Am J Med 2005; 118:948 956 11 Kessler R, Chaouat A, Schinkewitch P, et al. The obesityhypoventilation syndrome revisited: a prospective study of 34 consecutive cases. Chest 2001; 120:369 376 12 Perez de Llano LA, Golpe R, Ortiz Piquer M, et al. Short-term and long-term effects of nasal intermittent positive pressure ventilation in patients with obesity-hypoventilation syndrome. Chest 2005; 128:587 594 13 Mokhlesi B, Tulaimat A, Faibussowitsch I, et al. Obesity hypoventilation syndrome: prevalence and predictors in patients with obstructive sleep apnea. Sleep Breath 2007; 11:117 124 14 Resta O, Foschino Barbaro MP, Brindicci C, et al. Hypercapnia in overlap syndrome: possible determinant factors. Sleep Breath 2002; 6:11 18 15 Chaouat A, Weitzenblum E, Krieger J, et al. Association of chronic obstructive pulmonary disease and sleep apnea syndrome. Am J Respir Crit Care Med 1995; 151:82 86 16 Bednarek M, Plywaczewski R, Jonczak L, et al. There is no relationship between chronic obstructive pulmonary disease and obstructive sleep apnea syndrome: a population study. Respiration 2005; 72:142 149 17 Mokhlesi B, Tulaimat A, Evans AT, et al. Impact of adherence with positive airway pressure therapy on hypercapnia in obstructive sleep apnea. J Clin Sleep Med 2006; 2:57 62 18 Radwan L, Maszczyk Z, Koziorowski A, et al. Control of breathing in obstructive sleep apnoea and in patients with the overlap syndrome. Eur Respir J 1995; 8:542 545 19 Berry-Kravis EM, Zhou L, Rand CM, et al. Congenital central hypoventilation syndrome: phox2b mutations and phenotype. Am J Respir Crit Care Med 2006; 174:1139 1144 20 Doherty LS, Kiely JL, Deegan PC, et al. Late-onset central hypoventilation syndrome: a family genetic study. Eur Respir J 2007; 29:312 316 21 Weese-Mayer DE, Berry-Kravis EM, Zhou L. Adult identified with congenital central hypoventilation syndrome: mutation in PHOX2b gene and late-onset CHS [letter]. Am J Respir Crit Care Med 2005; 171:88 22 Antic NA, Malow BA, Lange N, et al. PHOX2B mutationconfirmed congenital central hypoventilation syndrome: presentation in adulthood. Am J Respir Crit Care Med 2006; 174:923 927 23 Ahmed Q, Chung-Park M, Tomashefski JF Jr. Cardiopulmonary pathology in patients with sleep apnea/obesity hypoventilation syndrome. Hum Pathol 1997; 28:264 269 24 Berger KI, Ayappa I, Chatr-Amontri B, et al. Obesity hypoventilation syndrome as a spectrum of respiratory disturbances during sleep. Chest 2001; 120:1231 1238 25 Golpe R, Jimenez A, Carpizo R. Diurnal hypercapnia in patients with obstructive sleep apnea syndrome [letter]. Chest 2002; 122:1100 1101 26 Laaban JP, Chailleux E. Daytime hypercapnia in adult patients with obstructive sleep apnea syndrome in France, before initiating nocturnal nasal continuous positive airway pressure therapy. Chest 2005; 127:710 715 27 Leech JA, Onal E, Baer P, et al. Determinants of hypercapnia in occlusive sleep apnea syndrome. Chest 1987; 92:807 813 28 Nowbar S, Burkart KM, Gonzales R, et al. Obesity-associated hypoventilation in hospitalized patients: prevalence, effects, and outcome. Am J Med 2004; 116:1 7 29 Resta O, Foschino Barbaro MP, Bonfitto P, et al. Hypercapnia in obstructive sleep apnoea syndrome. Neth J Med 2000; 56:215 222 30 Resta O, Foschino-Barbaro MP, Bonfitto P, et al. Prevalence and mechanisms of diurnal hypercapnia in a sample of morbidly obese subjects with obstructive sleep apnoea. Respir Med 2000; 94:240 246 31 Verin E, Tardif C, Pasquis P. Prevalence of daytime hypercapnia or hypoxia in patients with OSAS and normal lung function. Respir Med 2001; 95:693 696 32 Akashiba T, Akahoshi T, Kawahara S, et al. Clinical characteristics of obesity-hypoventilation syndrome in Japan: a multi-center study. Intern Med 2006; 45:1121 1125 33 Berg G, Delaive K, Manfreda J, et al. The use of health-care resources in obesity-hypoventilation syndrome. Chest 2001; 120:377 383 34 Han F, Chen E, Wei H, et al. Treatment effects on carbon dioxide retention in patients with obstructive sleep apneahypopnea syndrome. Chest 2001; 119:1814 1819 35 Rapoport DM, Garay SM, Epstein H, et al. Hypercapnia in the obstructive sleep apnea syndrome: a reevaluation of the Pickwickian syndrome. Chest 1986; 89:627 635 36 Levin DL, Muster AJ, Pachman LM, et al. Cor pulmonale secondary to upper airway obstruction: cardiac catheterization, immunologic, and psychometric evaluation in nine patients. Chest 1975; 68:166 171 37 Akashiba T, Kawahara S, Kosaka N, et al. Determinants of chronic hypercapnia in Japanese men with obstructive sleep apnea syndrome. Chest 2002; 121:415 421 38 Masa JF, Celli BR, Riesco JA, et al. The obesity hypoventilation syndrome can be treated with noninvasive mechanical ventilation. Chest 2001; 119:1102 1107 39 McTigue K, Larson JC, Valoski A, et al. Mortality and cardiac and vascular outcomes in extremely obese women. JAMA 2006; 296:79 86 40 Sakakibara H, Tong M, Matsushita K, et al. Cephalometric abnormalities in non-obese and obese patients with obstructive sleep apnoea. Eur Respir J 1999; 13:403 410 41 Yu X, Fujimoto K, Urushibata K, et al. Cephalometric analysis in obese and nonobese patients with obstructive sleep apnea syndrome. Chest 2003; 124:212 218 42 Mokdad AH, Ford ES, Bowman BA, et al. Prevalence of obesity, diabetes, and obesity-related health risk factors, 2001. JAMA 2003; 289:76 79 43 Hida W, Okabe S, Tatsumi K, et al. Nasal continuous positive airway pressure improves quality of life in obesity hypoventilation syndrome. Sleep Breath 2003; 7:3 12 44 Atwood CW Jr, McCrory D, Garcia JG, et al. Pulmonary www.chestjournal.org CHEST / 132 / 4/ OCTOBER, 2007 1333