Key words: hypoxemia; obesity; obstructive sleep apnea-hypopnea syndrome; subcutaneous fat accumulation; visceral fat accumulation

Sleep Oxygen Desaturation and Circulating Leptin in Obstructive Sleep Apnea-Hypopnea Syndrome* Koichiro Tatsumi, MD, FCCP; Yasunori Kasahara, MD; Katsushi Kurosu, MD; Nobuhiro Tanabe, MD, FCCP; Yuichi Takiguchi, MD, FCCP; and Takayuki Kuriyama, MD, FCCP Background: Obstructive sleep apnea-hypopnea syndrome (OSAHS) is characterized by repeated oxygen desaturation. Obesity and visceral fat accumulation (VFA) are risk factors for the development of OSAHS. Circulating leptin increases in accordance with body mass index (BMI), and under experimental conditions intermittent hypoxia stimulates leptin production. Methods: The primary objective of this study was to investigate whether hypoxemia during sleep influences the levels of circulating leptin and whether the location of body fat deposits, ie, the distribution of VFA and subcutaneous fat accumulation (SFA), affects circulating leptin levels in patients with OSAHA who are not obese. We assessed VFA and SFA by abdominal CT scan and measured circulating levels of leptin in 96 male patients with OSAHS and 52 male patients without OSAHS matched for BMI. To be matched for BMI in the two groups, patients whose BMIs were < 27 were selected for the OSAHS group. Results: In the whole study group, circulating leptin levels correlated with BMI (r 0.30), VFA (r 0.44), SFA (r 0.28), apnea-hypopnea index (AHI) [r 0.48], sleep mean arterial oxygen saturation (SaO 2 )[r 0.59], and sleep lowest SaO 2 (r 0.37). Multiple regression analysis showed that average SaO 2 (p < 0.01) and lowest SaO 2 (p 0.03) were explanatory variables for serum leptin values, but AHI (p 0.054), BMI (p 0.33), VFA (p 0.11), and SFA (p 0.36) were not. Conclusions: These results suggest that sleep hypoxemia may be the main determinant of circulating leptin levels, although the location of body fat deposits could contribute to the elevated circulating leptin levels in patients with OSAHS who are not obese. (CHEST 2005; 127:716 721) Key words: hypoxemia; obesity; obstructive sleep apnea-hypopnea syndrome; subcutaneous fat accumulation; visceral fat accumulation Abbreviations: AHI apnea-hypopnea index; BMI body mass index; OSAHS obstructive sleep apnea-hypopnea syndrome; Sao 2 arterial oxygen saturation; SFA subcutaneous fat accumulation; VC vital capacity; VFA visceral fat accumulation Ostructive sleep apnea-hypopnea syndrome (OS- AHS) is characterized by recurrent collapse of the upper airway during sleep leading to periods of intermittent hypoxia and fragmentation of sleep. *From the Department of Respirology, Graduate School of Medicine, Chiba University, Chiba, Japan. This study was supported by a Grant-in-Aid for Scientific Research (C)(16590735) from the Ministry of Education, Science, Sports and Culture, and grants awarded from the Ministry of Health, Labour and Welfare of Japan to the Respiratory Failure Research Group. Manuscript received May 26, 2004; revision accepted September 15, 2004. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (e-mail: permissions@chestnet.org). Correspondence to: Koichiro Tatsumi, MD, FCCP, Department of Respirology, Graduate School of Medicine, Chiba University, 1 8-1 Inohana, Chuou-ku, Chiba 260-8670, Japan; e-mail: tatsumi@faculty.chiba-u.jp Obesity and visceral fat accumulation (VFA) are known to be risk factors for the development of OSAHS. In addition, obstructive sleep apnea may affect VFA, 1 which increases the risk for obesityrelated disorders. Leptin was first described as an adipose-derived hormone, which induces a complex response including control of body weight and energy expenditure after interaction with specific receptors located in the CNS and in peripheral tissues. 2 Leptin receptors are found in the hypothalamus, particularly in the arcuate nucleus, where leptin is thought to exert its primary feedback signaling. 3 Circulating levels of leptin reflect the amount of energy stored in adipose tissue and are reported to correlate with the body mass index (BMI) in humans. 4,5 In experimental animals, intermittent hypoxia re- 716 Clinical Investigations

sulted in an increase in fasting blood glucose levels and an increase in serum leptin levels. Microarray messenger RNA analysis of adipose tissue revealed that leptin was the only up-regulated gene affecting glucose uptake. Leptin may play an important role in mitigating the metabolic disturbances that accompany intermittent hypoxia. 6 The purpose of the present study was to examine whether repeated hypoxemia during sleep influences the levels of circulating leptin and, in addition, whether the location of body fat deposits, ie, the distribution of VFA and subcutaneous fat accumulation (SFA), affects the circulating levels of leptin in OSAHS patients who are not obese (BMI 27). We hypothesized that repeated hypoxemia itself, with little influence of obesity, promotes leptin production in patients with OSAHS. Materials and Methods The study population consisted of 96 male patients with OSAHS and 52 male patients without OSAHS matched for BMI, who were examined using polysomnography from April 2001 to March 2004. The patients were recruited from the sleep clinic where they had been referred for investigation for snoring or possible OSAHS. Presenting symptoms were either snoring or daytime sleepiness or both. The subjects were all Japanese, and no American, European, African, Asian, or other ethnic groups were included to avoid the ethnic effects in the present study. The BMI in non-osahs patients was 24.7 0.3 (mean SEM). Therefore, to match BMI between OSAHS and non-osahs patients, a population with a BMI 27 was selected for the OSAHS patients group. All patients were free from respiratory infection, heart failure, or other respiratory problems, such as COPD, at the time of polysomnography. They were asked to complete a questionnaire on sleep symptoms, medical history, and medications. OSAHS was established on the basis of clinical and polysomnographic criteria. The average number of episodes of apnea and hypopnea per hour of sleep (the apnea-hypopnea index [AHI]) was calculated as the means of sleep-disordered breathing. In addition to clinical symptoms, an AHI 5 events per hour was also used as a selection criterion. Pulmonary function tests were performed to determine the vital capacity (VC) and FEV 1 using a standard spirometer (Fudac-60; Fukuda Denshi; Tokyo, Japan). Arterial blood for the analysis of gases during room air breathing was drawn with the patient in the supine position, and Pao 2 and Paco 2 were measured in a blood gas analyzer (Model 1312; Instrumentation Laboratory; Milano, Italy). Overnight polysomnography (Compumedics; Melbourne, Australia) was performed between 9 pm and 6 am. The polysomnography consisted of continuous polygraphic recording from surface leads for EEG, electro-oculography, electromyography, ECG, thermistors for nasal and oral airflow, thoracic and abdominal impedance belts for respiratory effort, pulse oximetry for oxyhemoglobin level, tracheal microphone for snoring, and sensor for the position during sleep. Polysomnography records were staged manually according to standard criteria. 7 Respiratory events were scored according to American Academy of Sleep Medicine criteria 8 : apnea was defined as complete cessation of airflow lasting 10 s; hypopnea was defined as either a 50% reduction in airflow for 10sora 50% but discernible reduction in airflow accompanied either by a decrease in oxyhemoglobin saturation of 3% or an arousal. Severity of OSAHS was determined by the AHI and mean and lowest arterial oxygen saturation (Sao 2 ). At 7 am on the morning after the sleep study, venous blood was obtained in the fasting state to measure leptin. Serum levels of leptin were determined by radioimmunoassay (Linco Research; St. Louis, MO) with intra-assay and interassay coefficients of variation of 2.8 to 3.8% (n 10) and 0.4 to 4.6% (n 10), respectively. 9 The amount of abdominal and visceral fat deposition was assessed using CT. The computer determined the area of all pixels with attenuation values between 150 Hounsfield units and 50 Hounsfield units for adipose tissue. Corrections for beam-hardening artifacts were done manually as necessary. The areas of subcutaneous fat and visceral fat were measured in a single cross-sectional scan at the level of the umbilicus. An image histogram was computed for the subcutaneous fat layers to determine the range of CT numbers for the fat tissue. The total fat area was then calculated by counting the pixels that had intensities within the selected range of CT numbers. The intraperitoneal space was defined by tracing its contour on the scan image. The total area with the same CT numbers was considered to represent the visceral fat area. Subtraction of the visceral fat area from the total fat area was defined as the subcutaneous fat area. 10 The study protocol was approved by the Research Ethics Committee of Chiba University School of Medicine, and all patients gave their informed consent prior to the study. Statistics The results are expressed as mean SEM. Age, BMI, serum parameters, sleep parameters, and CT scan parameters were compared between OSAHS and non-osahs patients using the Mann-Whitney U test. Linear regression analysis was performed to examine the association of two parameters. Analysis of covariance was used to compare the influence of sleep mean Sao 2, BMI, and VFA on circulating leptin levels between OSAHS and non-osahs patients. Multiple regression analysis was performed with serum leptin levels as the dependent variable and the degree of OSAHS and obesity as explanatory variables; p 0.05 was considered statistically significant. Table 1 Clinical Characteristics of OSAHS and Non- OSAHS Patients* Characteristics OSAHS (n 96) Non-OSAHS (n 52) p Value Age, yr 51.9 1.3 46.7 2.0 0.05 FVC, % predicted 96.8 1.5 96.3 2.6 NS FEV 1 /FVC, % 83.0 0.7 85.8 0.9 0.05 Pao 2,mmHg 87.2 1.1 91.9 1.3 0.01 Paco 2,mmHg 42.3 0.4 42.5 0.5 NS BMI 24.7 0.2 24.7 0.3 NS VFA/SFA ratio 1.17 0.04 1.08 0.05 NS VFA, cm 2 140 5 124 7 0.07 SFA, cm 2 137 11 130 11 NS AHI, events/h 32.5 2.0 2.3 0.2 0.01 Sleep average Sao 2,% 93.2 0.4 96.5 0.2 0.01 Sleep lowest Sao 2,% 77.7 1.0 88.7 0.5 0.01 Leptin, ng/ml 6.0 0.4 3.8 0.2 0.01 *Data are presented as mean SEM. NS not significant. www.chestjournal.org CHEST / 127 / 3/ MARCH, 2005 717

Table 2 Correlation Coefficients of Several Parameters With the Levels of Serum Leptin Variables OSAHS (n 96) Results OSAHS vs Non-OSAHS Patients Non-OSAHS (n 52) Total (n 148) Sleep average Sao 2 0.54 0.25 0.59 Sleep lowest Sao 2 0.24* 0.12 0.37 AHI 0.38 0.04 0.48 BMI 0.33 0.48 0.30 VFA 0.50 0.37 0.44 SFA 0.29 0.58 0.28 *p 0.05. p 0.01. Baseline characteristics of the study population are shown in Table 1. None of the patients had obstructive airway disease (FEV 1 /FVC 70%). OSAHS patients were older than non-osahs patients. FEV 1 / FVC ratio and Pao 2 were higher in the non-osahs group, while FVC percentage of predicted and Paco 2 did not differ between the two groups. BMI, VFA, SFA, and VFA/SFA ratio did not differ significantly between OSAHS and non-osahs patients, although VFA tended to be higher in the OSAHS group. Mean and lowest Sao 2 levels during sleep were significantly lower, and AHI was higher in OSAHS patients; circulating leptin levels were significantly higher in OSAHS patients. Factors Influencing Serum Levels of Leptin Circulating leptin levels correlated with all variables, including BMI, VFA, SFA, sleep mean Sao 2, and sleep lowest Sao 2 in OSAHS patients, while circulating leptin levels correlated with BMI, VFA, and SFA in non-osahs patients. In the whole study group, circulating leptin levels correlated with all parameters, which was similar in OSAHS patients (Table 2). OSAHS patients had higher leptin levels than non-osahs patients when compared relative to BMI, VFA, or sleep mean Sao 2. The linear regression line between leptin levels and BMI in OSAHS patients was shifted upward compared with that in non-osahs patients (p 0.01) [Fig 1]. When leptin values were plotted against VFA, they were distributed at a higher level in OSAHS patients (p 0.01) [Fig 2]. Similarly, OSAHS patients had higher leptin levels than non-osahs patients when compared relative to sleep mean Sao 2 (p 0.01) [Fig 3]. Factors that influence the levels of serum leptin in OSAHS patients were examined. A significant correlation was observed between serum leptin levels and sleep average Sao 2, sleep lowest Sao 2, AHI, Figure 1. Relationship between BMI and serum levels of leptin in OSAHS patients and non-osahs patients. The solid and dashed regression lines describe the relationship between serum leptin levels and BMI in OSAHS and non-osahs patients, respectively. The slopes of the regression lines with SEM were 0.73 0.21 and 0.30 0.08 in OSAHS and non-osahs patients, respectively. The regression line between serum levels of leptin and BMI was located upward in patients with OSAHS compared with that non-osahs patients (p 0.0002, analysis of covariance). 718 Clinical Investigations

Figure 2. Relationship between VFA and serum levels of leptin in OSAHS and non-osahs patients. The solid and dashed regression lines describe the relationship between serum leptin levels and VFA in OSAHS and non-osahs patients, respectively. The slopes of the regression lines with SEM were 3.48 0.68 10-4 and 1.10 0.39 10-4 in OSAHS and non-osahs patients, respectively. The regression line between serum levels of leptin and VFA was located upward in patients with OSAHS compared with that in non-osahs patients (p 0.0001, analysis of covariance). BMI, VFA, and SFA (Table 2), indicating that both repeated hypoxemia and obesity may influence the serum levels of leptin. Multiple regression analysis was performed with serum leptin levels as the dependent variable and the degree of OSAHS (sleep average Sao 2, sleep lowest Sao 2, and AHI) and obesity (BMI, VFA, and SFA) as explanatory variables. The results showed that average Sao 2 (p 0.01) and lowest Sao 2 (p 0.03) were explanatory variables for serum leptin values, but AHI (p 0.054), BMI (p 0.33), VFA (p 0.11), and SFA (p 0.36) were not. Discussion The serum levels of leptin were correlated with the degree of OSAHS, such as average Sao 2, lowest Sao 2, and AHI, and the degree of obesity, such as BMI, VFA, and SFA. The regression lines between serum levels of leptin and BMI, VFA, and average Sao 2 were shifted upward in patients with OSAHS compared with those in patients without OSAHS, suggesting that both the degree of OSAHS and obesity may influence the production of leptin. The results of the multiple regression analysis suggested that repeated sleep hypoxemia with apnea-hypopnea may mainly regulate serum leptin levels, independent of the degree of obesity, although its precise mechanisms have not been elucidated. Obesity is known to be the major factor regulating circulating leptin, which is also influenced by gender and age. 11,12 Circulating leptin concentrations were higher in obese subjects than in normal-weight subjects, although several factors other than the amount of body fat may contribute to the elevation of circulating leptin concentrations. 5 The mechanism of the increase in circulating leptin involves the induction of the ob gene. 13 Circulating leptin concentrations seem to be regulated by changes in body fat at the level of ob gene expression. 4 If leptin acts as supposed it does, when adipocytes send the signal to the brain about the amount of adipose tissue, appetite should decrease and energy expenditure should increase, resulting in weight loss. Considering the fact that circulating leptin levels are elevated in most overweight individuals, obesity may be associated with leptin resistance. 14 16 In the present study, circulating leptin concentrations increased in parallel with BMI in both OSAHS and non-osahs patients, although the range of BMI was almost between 20 and 30, and this relationship was not as constant as observed in inbred mice. 4 Whether leptin resistance is present in patients with BMIs 30 remains to be defined. Whether the increased levels of leptin was a cause www.chestjournal.org CHEST / 127 / 3/ MARCH, 2005 719

Figure 3. Relationship between sleep mean Sao 2 and serum levels of leptin in OSAHS and non-osahs patients. The solid and dashed regression lines describe the relationship between serum leptin levels and sleep mean Sao 2 in OSAHS and non-osahs patients, respectively. The slopes of the regression lines with SEM were 0.55 0.09 and 0.33 0.18 in OSAHS and non-osahs patients, respectively. The regression line between serum levels of leptin and sleep mean Sao 2 was located upward in patients with OSAHS compared with that in non-osahs patients (p 0.0001, analysis of covariance). or a result of fat deposition was unclear. The close correlation between serum leptin levels and VFA may reflect an increased leptin resistance, contributing to central deposition of fat and predisposing the patients to upper airway obstruction. On the contrary, the increased visceral and surface fat deposition, reflected by higher leptin levels, may predispose the patients to the development of OSAHS. This study is fundamentally based on a two-hit theory. In addition to fat deposition (the first hit ), whether repeated hypoxemia associated with sleep apnea-hypopnea (second hit ) accentuates the leptin production was the main question. BMI, SFA, and VFA were similar in OSAHS and non-osahs patients, although the amount of visceral fat tended to be higher in OSAHS than in non-osahs patients. The increase in BMI, especially VFA, may accentuate leptin production, even though BMI of the patients was 31, since circulating levels of leptin tightly correlated with VFA compared with BMI. In the present study, the circulating levels of leptin relative to BMI and VFA were higher in OSAHS patients compared with those in non-osahs patients, suggesting that other mechanisms apart from the degree of obesity could have influenced the circulating levels of leptin. It seems that repeated hypoxemia in the presence of body fat deposition promoted leptin production in patients with OSAHS. Although multiple regression analysis showed that average Sao 2 and lowest Sao 2 values were the explanatory variables for serum leptin values, serum leptin levels may be influenced by hypoxemia as well as VFE in patients with OSAHS. It is believed that increased deposition of fat or soft tissue in the neck and upper airway region predisposes the subject to upper airway collapse and apnea during sleep. 17 Central obesity, partly reflected by VFA, may be a better predictor of OSAHS than BMI. 18 Central obesity of patients with OSAHS may be a pathogenetic risk for OSAHS due to its mechanical effect on the upper airway, in addition to the vascular implications of central obesity. 19 Our data that BMI and SFA were similar, but VFA tended to be higher in patients with OSAHS, suggest that even a slight degree of central obesity may predispose the subjects to the development of OS- AHS. Insulin resistance and glucose intolerance are positively associated with OSAHS, independent of the degree of obesity. 20,21 In addition, the extent of insulin resistance may be related to the severity of hypoxic stress in OSAHS. 20,21 Then it is likely that intermittent hypoxia exacerbates insulin resistance and may influence metabolic function, associated with obesity in patients with OSAHS. Treatment with continuous positive airway pressure resulted in 720 Clinical Investigations

a reduction of leptin levels in OSAHS patients. 22,23 Polotsky et al 6 reported that exposure to intermittent hypoxia resulted in a decrease of fasting blood glucose levels, improvement of glucose tolerance without a change in serum insulin levels, and an elevation of leptin messenger RNA expression and protein level. These findings suggest that leptin plays an important role in mitigating the metabolic disturbances that accompany intermittent hypoxia, and that the elevation of leptin levels caused by hypoxic stress associated with OSAHS may represent an important compensatory response that acts to minimize metabolic dysfunction. In this regard, our data that hypoxic stress was a primary determinant of circulating levels of leptin may suggest that the increased levels of leptin were neither a cause nor a result of fat deposition, but that intermittent hypoxia may play an important role in determining leptin production in patients with OSAHS who are not obese (BMI 30). Limitations of this study include a relatively small number of non-osahs patients and possible patient selection bias toward a cohort of OSAHS patients whose BMI was 27 to match the two groups for BMI. Ideally, we should have performed the study in all subjects with and without OSAHS matched for BMI. However, it would have been difficult to match OSAHS subjects with non-osahs subjects for BMI, if all subjects had been included in this study, although East Asian subjects are more likely to acquire OSAHS at a lower BMI than Western subjects. 24 In addition, whether increased circulating levels of leptin decreased after continuous positive airway pressure therapy was not determined in our subjects. Moreover, whether leptin affects visceral adiposity has not been determined in OSAHS, although it has been reported that leptin selectively decreases visceral adiposity in rats. 25 In conclusion, this study suggested that sleep hypoxemia may be the main determinant of circulating leptin levels, although the location of body fat deposits could contribute to the elevated circulating leptin levels observed in patients with OSAHS who are not obese. References 1 Chin K, Shimizu K, Nakamura T, et al. Changes in intraabdominal 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 2 Tartaglia LA. The leptin receptor. J Biol Chem 1997; 272: 6093 6096 3 Hakansson ML, Brown H, Ghilardi N, et al. Leptin receptor immunoreactivity in chemically defined target neurons of the hypothalamus. J Neurosci 1998; 18:559 572 4 Maffei M, Halaas J, Ravussin E, et al. Leptin levels in human and rodent: measurement of plasma leptin and ob RNA in obese and weight-reduced subjects. Nature Med 1995; 1:1155 1161 5 Considine RV, Sinha MK, Heiman ML, et al. Serum immunoreactive leptin concentrations in normal-weight and obese humans. N Engl J Med 1996; 334:292 295 6 Polotsky VY, Li J, Punjabi NM, et al. Intermittent hypoxia increases insulin resistance in genetically obese mice. J Physiol 2003; 552:253 264 7 Rechtschaffen A, Kales AA, eds. A manual of standardized terminology, techniques and scoring for sleep stages of human subjects. Washington, DC: Government Printing Office, 1968; National Institutes of Health publication No. 204 8 The report of an American Academy of Sleep Medicine task force. Sleep-related breathing disorders in adults: recommendations for syndrome definition and measurement techniques in clinical research. Sleep 1999; 22:667 689 9 Hosoda K, Masuzaki H, Ogawa Y, et al. Development of radioimmunoassay for human leptin. Biochem Biophys Res Commun 1996; 221:234 239 10 Tokunaga K, Matsuzawa Y, Ishikawa K, et al. A novel technique for the determination of body fat by computed tomography. Int J Obes 1983; 7:437 445 11 Ostlund RE, Yang JW, Klein S, et al. Relation between plasma leptin concentration and body fat, gender, diet, age, and metabolic covariates. J Clin Endocrinol Metab 1996; 81:3909 3913 12 Rosenbaum M, Nicolson M, Hirsch J, et al. Effects of gender, body composition, and menopause on plasma concentration of leptin. J Clin Endocrinol Metab 1996; 81:3424 3427 13 Campfield LA, Smith FJ, Guisez Y, et al. Recombinant mouse OB protein: evidence for a peripheral signal linking adiposity and central neural networks. Science 1995; 269:546 549 14 Campfield LA, Smith FJ, Burn P. The ob protein (leptin) pathway: a link between adipose tissue mass and central neural networks. Horm Metab Res 1996; 28:619 632 15 Lönnqvist F, Arner P, Nordfors L, et al. Overexpression of the obese (ob) gene in adipose tissue of human obese subjects. Nat Med 1995; 1:950 953 16 Friedman JM, Halaas JL. Leptin and the regulation of body weight in mammals. Nature 1998; 395:763 770 17 Stradling JR, Crosby JH. Predictors and prevalence of obstructive sleep apnea and snoring in 1001 middle-aged men. Thorax 1991; 46:85 90 18 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 1235 19 Bjorntorp P. Obesity. Lancet 1997; 350:423 426 20 Ip MSM, Lam B, Ng MM, et al. Obstructive sleep apnoea is independently associated with insulin resistance. Am J Respir Crit Care Med 2002; 165:670 676 21 Punjabi NM, Sorkin JD, Katzel LI, et al. Sleep-disordered breathing and insulin resistance in middle-aged and overweight men. Am J Respir Crit Care Med 2002; 165:677 682 22 Ip MSM, Lam KSL, Ho C-M, et al. Serum leptin and vascular risk factors in obstructive sleep apnea. Chest 2000; 118:580 586 23 Chin K, Shimizu K, Nakamura T, et al. Changes in intraabdominal visceral fat and serum leptin levels in patients with obstructive sleep apnoea syndrome following nasal continuous positive pressure therapy. Circulation 1999; 100:706 712 24 Li KK, Powell NB, Kushida C, et al. A comparison of Asian and white patients with obstructive sleep apnea syndrome. Laryngoscope 1999; 109:1937 1940 25 Barzilai N, Wang J, Massilon D, et al. Leptin selectively decreases visceral adiposity and enhances insulin action. J Clin Invest 1997; 100:3105 3110 www.chestjournal.org CHEST / 127 / 3/ MARCH, 2005 721