Features of REM-related Sleep Disordered Breathing in the Japanese Population

Transcription

1 ORIGINAL ARTICLE Features of Sleep Disordered Breathing in the Japanese Population Seiichiro Sakao, Takayuki Sakurai, Misuzu Yahaba, Yoriko Sakurai, Jiro Terada, Nobuhiro Tanabe and Koichiro Tatsumi Abstract Objective Rapid eye movement (REM)-related sleep disordered breathing () is an entity in which the cessation or reduction of breathing occurs primarily during the REM period. Most studies have shown that more frequently affects women, younger people and patients with mild or moderate. The aim of this study was to prospectively investigate the prevalence and features of in Japanese subjects compared with the findings of previous reports. Methods A total of 468 patients were evaluated in this study. The diagnosis of was established using polysomnographic monitoring. The patient variables included age, gender, body characteristics, comorbidities, etc. Results was more prevalent in women than non- (male ratio; 66.3% vs. 79.5%, p=0.03). Moreover, the patients with had lower body mass indexes (25.9±6.9 vs. 28.5±7.7; p=0.003), arousal indexes (31.8±10.7 vs. 61.0±29.1; p<0.001), apnea hypopnea indexes (15.0± 8.0 vs. 54.9±35.9) and glycosylated hemoglobin (HbA1c) levels (5.5±0.9 vs. 5.9±2.6; p=0.02) than the patients with non-. However, the overall and female gender prevalence of among the Japanese subjects was lower than that shown in previous reports. The finding that was not prevalent in younger individuals or severely obese patients was not consistent with the results of previous studies. Conclusion The present findings suggest that may have different clinical characteristics in the Japanese population than that observed in previous reports. Key words: sleep apnea syndrome, rapid eye movement (REM), body mass index (BMI), polysomnography (PSG), sleep disordered breathing (), glycosylated hemoglobin (HbA1c) (Intern Med 54: , 2015) () Introduction Obstructive sleep apnea syndrome (OSAS) is the most common type of sleep disordered breathing (), affecting between 2% and 4% of adults in developed nations (1, 2). It is generally acknowledged that rapid eye movement (REM)- related is an entity involving the partial or complete cessation of breathing mainly during the REM period (3). However, it remains to be elucidated why some patients experience breathing cessation primarily during REM period. REM sleep has been shown to induce a rise in the sympathetic nerve activity and instability of the cardiovascular system, processes that are accelerated by obstructive apnea and hypopnea during the REM period in patients with OSAS, thus leading to increased cardiovascular risks (4, 5). Indeed, a recent study demonstrated a cross-sectional and longitudinal correlation between REM OSAS and hypertension (6). Most studies have shown that more often affects women, younger people and patients with mild or moderate, rather than severe, airway obstruction (3, 7-9). However, in daily clinical practice, appears to also affect men and older patients, suggesting that may have different clinical features in the Department of Respirology (B2), Graduate School of Medicine, Chiba University, Japan Received for publication October 9, 2014; Accepted for publication January 27, 2015 Correspondence to Dr. Seiichiro Sakao, sakaos@faculty.chiba-u.jp 1481

2 Japanese population from that seen in other ethnic groups. Despite the findings of numerous studies concerning REMrelated, little is currently known about this type of in Asian populations, since most reports have focused on the features of only in Caucasian and African-American populations. It has been demonstrated that obesity is less prevalent and less epidemic in Far-East Asian men than in Caucasians, despite the presence of severe OSAS in the former group (10). Moreover, a cephalometric analysis showed that the cranial base dimensions of Far-East Asian men are significantly smaller than those of Caucasians (10). These results suggest that racial differences in the susceptibility to obesity and the craniofacial anatomy may differentiate the features of REMrelated in the Japanese population from that noted in non-asian populations. As mentioned above, we believe that is clinically important and should be assessed separately. The aim of this study was therefore to investigate the prevalence and features of in the Japanese population in comparison with that observed in previous reports. Study subjects Materials and Methods From September 2006 to March 2013, 472 suspected OSAS patients (361 men and 111 women) referred for clinical sleep evaluations at Chiba University Hospital were enrolled in this study. The diagnosis of OSAS in all patients was established using polysomnographic monitoring based on a documented apnea hypopnea index (AHI) of 5 events per hour. The study inclusion criteria were as follows: 1) an age of 18 years; 2) an AHI of 5/h and 3) a total sleep time of 100 minutes and REM sleep time of 10 minutes (9). A total of 468 patients were finally included and evaluated in this study. The patient variables evaluated in the analysis included age, gender, height, weight, body mass index [BMI (in kilograms per meter squared)], Epworth sleepiness scale (ESS) (11) and comorbidities, with a focus on cardiovascular, psychiatric and lung diseases as well as metabolic disorders. Venous blood samples were collected for the biochemical analyses on the day of hospital admission. According to Japanese legislation, written informed consent is not required for retrospective data collection in this type of research. However, the patient database was anonymized and managed according to the restrictive requirements of the Ministry of Health, Labour and Welfare, dedicated to privacy, information technology and civil rights in Japan. The Ethics Committee of Chiba University Hospital approved the study protocol. Polysomnography Overnight polysomnography (PSG) was performed using a digital polysomnographic monitor (E Series, Compumedics, Victoria, Australia). The PSG device monitors many biophysiological changes and includes the findings of electroencephalography (EEG) (C4-A2, C3-A1, O2-A1, O1-A2), bilateral electrooculography (EOG), submental and bilateral anterior tibial electromyography (EMG), ECG, thoracoabdominal piezoelectric belts for respiratory effort, a nasal pressure cannula for the nasal airflow, a thermistor for the nasal and oral flow, finger pulse oximetry, a neck microphone to record snoring and a sensor on the thoracic belt to assess body posture. The 3% oxygen desaturation index is the number of times per hour of sleep involving oxygen desaturation of 3% from baseline, as measured with finger pulse oximetry. The biophysiological changes on the PSG device were evaluated using the 2007 American Academy of Sleep Medicine Manual for the Scoring of Sleep and Related Events system (12). Apnea was recorded as the cessation of airflow for at least 10 seconds, and hypopnea was defined as a 50% reduction in airflow for at least 10 seconds associated with either oxygen desaturation of 3% measured with finger pulse oximetry or a microarousal (12). The AHI was defined as the number of apneas and hypopneas divided by the total sleep time. The AHI-NREM and AHI-REM were defined as the number of apneas and hypopneas during NREM sleep (stages I to IV) divided by the total NREM sleep time and during REM sleep divided by the total REM sleep time, respectively. According to the definitions of previously reported in the literature (3), we defined three different criteria for this condition, as follows. The common criterion was an overall AHI of 5 and AHI-REM/AHI-NREM of 2. Definition #I satisfies the common criterion. Definition #II satisfies the criterion of an AHI-NREM of <15 in addition to the common criterion. Definition #III satisfies the criterion of an AHI-NREM of <8 and at least 10.5 minutes of REM sleep in addition to the common criterion. Statistical analysis The data were analyzed using the JMP software program (Japanese version, SAS Institute Japan, Tokyo, Japan) and the Excel-Toukei 2010 software program (Social Survey Research Information, Tokyo, Japan). All results are expressed as the number or percentage for categorical variables and the mean ± SD for continuous variables. Continuous variables that were normally distributed and not normally distributed were compared using Student s t-test and the Mann-Whitney U-test, respectively. A p value of less than 0.05 was considered to be statistically significant. Results Of the 472 patients referred for clinical sleep evaluations, 468 subjects with (99.2%) were initially enrolled in this study. As shown in a previous report (3), we applied definition #II to compare the features of the patients with to those of the patients with non-rem- 1482

3 Table 1. Comparison between Non- and Based on Definition #II. Variables Entire group Non- p value (n=468) (n=385) (definition #II) (n=83) Age 54.9± ± ± Gender, % female * Body mass index, kg/m ± ± ± * Primary complaint Snoring, % Witnessed apnea, % Nocturnal choking, % Excessive daytime sleepiness, % Medical history Cardiovascular, % Psychiatric, % Lung disease, % Metabolic, % Biochemical parameters Glucose, mg/dl 123.9± ± ± HbA1C, % 5.8± ± ± * Triglycerides, mg/dl 178.5± ± ± Cholesterol, mg/dl 191.3± ± ± Medications Antihypertensive, % Antidepressant, % Benzodiazepine-hypnotic, % Epworth Sleepiness Scale 8.9± ± ± %ODI, no/h 31.4± ± ±8.8 <0.001* 4%ODI, no/h 25.2± ± ±6.3 <0.001* TST, min 332.0± ± ± * Stage 1, min 71.2± ± ±43.6 <0.001* Stage 2, min 175.2± ± ±65.9 <0.001* Stage 3, min 15.9± ± ±15.7 <0.001* Stage 4, min 9.8± ± ± * REM, min 59.5± ± ± * Sleep latency, min 55.1± ± ± Sleep efficiency, % 59.9± ± ± Arousal index, no/h 55.9± ± ±10.7 <0.001* AHI 47.8± ± ±8.0 <0.001* AHIREM 38.3± ± ±15.6 <0.001* AHINREM 33.6± ± ±4.7 <0.001* * Non- vs. definition #II. related (Table 1). In the entire group, the mean age was 54.9±15.2 years (range, 18 to 82), with a 1:3.25 female to male ratio. The overall patient population had the following characteristics (mean values): BMI, 28.0±7.6 kg/m 2 ; AHI, 47.8±34.1/h; AHI-REM, 38.3±25.5/h; AHI-NREM, 33.6±27.7/h; 3%ODI, 31.4±23.7/h; 4%ODI, 25.2±22.5/h; TST, 323.2±92.7 min; sleep efficiency, 59.9±15.3% and arousal index, 55.9±29.0/h (Table 1). was more prevalent in women (female ratio; 33.7% vs. 20.5%, p=0.029), although both genders showed a similar age, sleep latency and sleep efficacy. In addition, the patients with had lower body mass indexes (BMI) (25.9±6.9 vs. 28.5±7.7; p=0.003), lower 3%ODI values (10.7±8.8 vs. 35.9±23.5; p<0.001), lower 4%ODI values (6.0±6.3 vs. 29.2±22.6; p<0.001), lower arousal indexes (31.8±10.7 vs. 55.9±29.0; p<0.001) and lower AHI values (15.0±8.0 vs. 54.9±35.9; p<0.001) than the patients with non-. Moreover, lower HbA1c levels were confirmed in the patients with (5.5±0.9 vs. 5.9±2.6; p=0.018), although there were no statistically significant differences in the glucose, triglyceride or cholesterol levels between the two groups (Table 1). The prevalence of based on the different criteria (3) in the studied Japanese patient population ranged from 11.0% to 24.8%. When using definition #II, 83 patients were identified as having. The features of this group were as follows: age, 53.0±17.0 with a 1:1.96 female to male ratio; BMI, 25.9±6.9 kg/m 2 ; AHI, 15.0±8.0/h; AHI-REM, 31.5±15.6/h; AHI-NREM, 7.2±4.7/h; 3%ODI, 10.7±8.8/h; 4%ODI, 6.0±6.3/h; TST, 351.8±102.1 min; sleep efficiency, 62.5±16.1% and arousal index, 31.8± 10.7/h (Table 2). There are various definitions of hypopnea as well as cited in the previous literature. In fact, these different definitions of hypopnea have been shown to result in significantly different AHI values (13). Therefore, the definitions of both and hypopnea should be considered when comparing the features of patients with among different races and even among different studies. In the current study, notwithstanding the differences in the hypopnea definitions, the 1483

4 Table 2. Comparison among the Definitions. Variables definition #I (n=116) definition #II (n=83) definition #III (n=52) p value Percentage of patients, % Age 54.1± ± ± Gender, % female Body mass index, kg/m ± ± ± Primary complaint Snoring Witnessed apnea Nocturnal choking Excessive daytime sleepiness Medical history Cardiovascular Psychiatric Lung disease Metabolic Biochemical parameters Glucose, mg/dl 122.0± ± ± HbA1C, % 5.5± ± ± Triglycerides, mg/dl 186.3± ± ± Cholesterol, mg/dl 193.7± ± ± Medications Antihypertensive Antidepressant Benzodiazepine-hypnotic Epworth Sleepiness Scale 8.8± ± ± %ODI, No/h 15.4± ± ±7.3 <0.001* 4%ODI, No/h 9.8± ± ±4.8 <0.001* TST, min 344.1± ± ± Stage 1, min 40.8± ± ± * Stage 2, min 197.9± ± ± Stage 3, min 22.6± ± ± Stage 4, min 16.9± ± ± REM, min 65.9± ± ± Sleep latency, min 60.7± ± ± Sleep efficiency, % 61.5± ± ± Arousal index 35.5± ± ±9.5 <0.001* AHI 21.8± ± ±3.5 <0.001* AHIREM 40.2± ± ±13.1 <0.001* AHINREM 11.9± ± ±1.9 <0.001* * definition #I vs. definition #III. overall and female gender prevalence of was lower than that shown in previous reports, as summarized in Table 3. Furthermore, the finding that was not prevalent in younger individuals or severely obese patients (with an increased BMI) in the present study is not consistent with previous studies (3, 7, 9, 14) (Table 3). Discussion The current study demonstrated that is more prevalent in women than non- (female ratio; 33.7% vs. 20.5%, p=0.029), which is consistent with the results of previous reports (3, 8). However, the difference in age between the two groups did not reach a level of statistical significance (53.0±17.0 vs. 55.3±14.8; p=0.25). In addition, our patients with had lower BMI values than the patients with non- (25.9±6.9 vs. 28.5±7.7; p=0.003). These findings do not agree with those of previous reports for non-asian patients (3, 8) (Table 1). Previous reports have suggested that may be the initial finding of early-stage in younger individuals and women, as body weight gain and aging are associated with a decrease in the prevalence of, in accordance with the increased prevalence of non- (3, 9). However, the present findings indicate that may have different clinical characteristics in the Japanese population from that observed in non-asian populations and may not be the initial step in sequentially progressive. Additionally, our patients with had lower HbA1c levels than the patients with non- (5.5±0.9 vs. 5.9±2.6; p=0.018), despite the fact that there were no statistically significant differences in the glucose, triglyceride or cholesterol levels (Table 1). It has been demonstrated in several reports that indexes of OSAS severity are associated with the HbA1c levels in patients with or without diabetes (15-18). In fact, our patients with REM- 1484

5 Table 3. The Prevalence of Based on Various Definitions. Reference definition Hypopnea definition 7 I Chicago criteria 14 I AASM This study I AASM Percentage of patients, % Age / ±11.6/ 54.1± ±13.7/ 52.2± ±15.8/ 54.9±15.2 Gender, %Female / Body mass index, kg/m 2 / 46.4/ ±6.8/ 31.3±6.3 36/ ±5.2/ 29.3± / ±7.2/ 28.0±7.6 AHI TST / 15.7±9.2/ 34.1± ±8.7/ 25.3± ±15.7/ 47.8±36.1 Ethnicity Caucasian Caucasian Asian 9 II Chicago criteria II AASM This study II AASM ±0.8/ NA ±15.0/ 52.0± ±17.0/ 54.9± /32.9 NA/ 30.8± / ±8.8/ 35.9± / ±6.9/ 28.0±7.6 NA/ 29.7±0.5 12(8-17)/ 42(25-27) 15.0±8.0/ 47.8±36.1 NA Caucasian 35%, African- American 65% Asian 3 III AASM This study III AASM ±14/ 52.0± ±18.1/ 54.9± / ±9.3/ 35.9± / ±6.1/ 28.0±7.6 9(7-12) / 42(25-27) 10.0±3.5/ 47.8±36.1 Chicago criteria: 50% drop in accurate flow OR discernible drop in flow + > 3% desaturation or arousal. AASM manual A (recommended): 30% or greater drop in flow associated with a > 4% oxygen desaturation. AASM (alternative): 50% or greater drop in flow associated with a > 3% oxygen desaturation or arousal. Caucasian 33%, African- American 67% Asian related had less severe OSAS, with lower 3%ODI, 4% ODI, arousal index and AHI values (Table 1). Therefore, these previous reports likely support our results; however, it is also plausible that the higher HbA1c levels in the patients with non- are attributable to the higher BMI values observed in these patients, which may be related to insulin resistance and potentially result in postprandial hyperglycemia. In order to compare the features of in the Japanese population to those noted in previous reports, we applied three definitions described in the Methods (3). We applied these definitions because it would be impossible to compare the features between different races using a single definition due to the variety of definitions cited in the previous literature. In the present study, the prevalence of varied from 11.0% to 24.8% based on the three different definitions (3) (Table 2). This range of prevalence rates appears to be lower than that seen in previous studies (3, 7, 9, 13) (Table 3). Indeed, a two-fold decrease in the prevalence of was observed in the analysis using the strictest definition (definition #III) (Table 2). However, Conwell et al. suggested that this strict definition, which minimizes the contribution of AHI-NREM, appears to have a disadvantage in classifying patients according to clinical variables and/or symptomatology (3). Moreover, they indicated that definition #I may not be effective for excluding patients significantly affected by during NREM sleep (3). Therefore, definition #II is considered to be the most clinically useful for defining. The current study demonstrated a lower prevalence of in Japanese patients (Table 3). Although the reason(s) for the lower prevalence of in the Japanese population remains unknown, it is important to recognize that the patients in our study had lower BMI values than Caucasian or African-American populations (Table 3). It has been demonstrated that a higher BMI is associated with the severity of oxygen desaturation. Moreover, oxygen desaturation resulting from a higher BMI is more obvious during REM sleep than non-rem sleep (19). It may therefore be possible that the lower number of obese patients in the Japanese population is related to the observation of less severe oxygen desaturation, which may have resulted in a lower hypopnea index, thus accounting for the lower prevalence of in the Japanese population (Table 3). However, because previous studies (3, 9) had higher female ratios than the current study (around 30-40% vs. 22.9%), the higher prevalence of REMrelated in those studies may be attributable to the differences in the male to female ratio. Although some features of in this study were different from those described in previous reports, as summarized in Table 3, the previous studies by Haba-Rubio et al. and Koo et al. cited different definitions 1485

6 of hypopnea (Table 3). One study that evaluated the effects of the use of these different criteria in patients with suspected showed that different percentages of patients were diagnosed with obstructive sleep apnea (OSA) based on the varying criteria (19). In that study, approximately 25% of the patients who were previously evaluated to have OSA using the Chicago criteria would not have been diagnosed with OSA using the 2007 AASM definition (alternative) (19). Therefore, the overall prevalence of considered in the studies by Haba-Rubio et al. and Koo et al. would likely be lower if the 2007 AASM definition (alternative) had been used in their reports. The current findings demonstrated that is less prevalent in the Japanese population than in Caucasian and African-American populations. The AHI values during the total sleep time (TST), however, tended to be higher in our Japanese patients than in Caucasians or African-Americans (Table 3), indicating that Japanese patients with have more severe, despite the fact that there are fewer obese patients in the Japanese population. The severity of in Japanese patients may be attributed to the characteristics of the craniofacial anatomy, rather than obesity. In fact, a cephalometric analysis recently showed that the maxillomandibular dimensions of Asian OSA patients are significantly smaller than those of white OSA patients (10, 20), which may be a possible cause of pharyngeal airway obstruction during sleep. However, it should be considered that the higher total AHI values noted in the patients with non- in this study may be attributable to the differences in the proportion of the REM/NREM time between the present and previous studies (3, 9). In previous studies, was shown to be more prevalent in women than in men (3, 7-9), as also observed in the present study. However, the female ratio among the patients with in our Japanese population was much lower than that reported in previous studies, especially when using criteria II and III (Table 3). Koo et al. (9) speculated that the higher prevalence of women may be attributed to the effects of sex hormones, which increase the tensioning force of the genioglossus, thereby preventing airway collapse (21) and possibly driving the ventilatory response toward hypoxia and hypercapnia (22). REM sleep-dependent atonia is thought to decrease the effects of these hormones on the muscles of the upper airway, thus resulting in a predisposition to develop upper airway obstruction among women, especially during REM sleep (9). In the Japanese population, however, the smaller maxillomandibular dimensions, rather than the actions of female hormones, may account for the unique gender prevalence of, as the upper airway in Japanese patients, with smaller mandibles, may be more prone to collapse in the atonic REM period, and the mandible is smaller inwomenthaninmen. There are several limitations to the present study that should be considered when interpreting the results. First, this study was an observational study at a single center. Second, the number of patients was almost certainly too small to draw sweeping conclusions about REM in the Japanese population as a whole. Therefore, our results must be confirmed in multicenter studies in the future. Conclusion The findings of the current study suggest that REMrelated may have different clinical characteristics in the Japanese population than that observed in previous reports. Author s disclosure of potential Conflicts of Interest (COI). Seiichiro Sakao: Honoraria, Actelion, Pfizer and Nippon Shinyaku. Nobuhiro Tanabe: Honoraria, Actelion and Pfizer. Koichiro Tatsumi: Honoraria, GlaxoSmithKline and Pfizer. Acknowledgement All of the authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation Financial Support This study was supported by research grants from the Respiratory Failure Research Group from the Ministry of Health, Labour and Welfare, Japan. References 1. Young T, Palta M, Dempsey J, Skatrud J, Weber S, Badr S. The occurrence of sleep-disordered breathing among middle-aged adults. N Engl J Med 328: , Bixler E, Vgontzas A, Ten Have T, Tyson K, Kales A. Effects of age on sleep apnea in men: I. Prevalence and severity. Am J Respir Crit Care Med 157: , Conwell W, Patel B, Doeing D, et al. Prevalence, clinical features, and CPAP adherence in sleep-disordered breathing: a cross-sectional analysis of a large clinical population. Sleep Breath 16: , Somers VK, Dyken ME, Mark AL, Abboud FM. Sympatheticnerve activity during sleep in normal subjects. N Engl J Med 328: , Mokhlesi B, Punjabi NM. obstructive sleep apnea: an epiphenomenon or a clinically important entity? Sleep 35: 5-7, Mokhlesi B, Finn LA, Hagen EW, et al. Obstructive sleep apnea during REM sleep and hypertension. results of the Wisconsin Sleep Cohort. Am J Respir Crit Care Med 190: , Haba-Rubio J, Janssens JP, Rochat T, Sforza E. Rapid eye movement-related disordered breathing: clinical and polysomnographic features. Chest 128: , O Connor C, Thornley KS, Hanly PJ. Gender differences in the polysomnographic features of obstructive sleep apnea. Am J Respir Crit Care Med 161: , Koo BB, Patel SR, Strohl K, Hoffstein V. Rapid eye movementrelated sleep-disordered breathing: influence of age and gender. Chest 134: , Li KK, Kushida C, Powell NB, Riley RW, Guilleminault C. Obstructive sleep apneasyndrome: a comparison between Far-East Asian and white men. Laryngoscope 110: , Johns MW. A new method for measuring daytime sleepiness: the Epworth sleepiness scale. Sleep 14: ,

7 12. Iber C, Ancoli-Israel S, Chesson A, Quan SF. The AASM manual for the of sleep and associated events: rules, terminology and technical specifications. 1st ed. American Academy of Sleep Medicine, Westchester, Berry RB, Budhiraja R, Gottlieb DJ, et al. Rules for respiratory events in sleep: update of the 2007 AASM Manual for the Scoring of Sleep and Associated Events. Deliberations of the Sleep Apnea Definitions Task Force of the American Academy of Sleep Medicine. J Clin Sleep Med 8: , Oksenberg A, Arons E, Nasser K, Vander T, Radwan H. REMrelated obstructive sleep apnea: the effect of body position. J Clin Sleep Med 6: , Okada M, Takamizawa A, Tsushima K, Urushihata K, Fujimoto K, Kubo K. Relationship between sleep-disordered breathing and lifestyle-related illnesses in subjects who have undergone healthscreening. Intern Med 45: , Papanas N, Steiropoulos P, Nena E, et al. HbA1c is associated with severity of obstructive sleep apnea hypopnea syndrome in nondiabetic men. Vasc Health Risk Manag 5: , Aronsohn RS, Whitmore H, Van Cauter E, Tasali E. Impact of untreated obstructive sleep apnea on glucose control in type 2 diabetes. Am J Respir Crit Care Med 181: , Shpirer I, Rapoport MJ, Stav D, Elizur A. Normal and elevated HbA1C levels correlate with severity of hypoxemia in patients with obstructive sleep apnea and decrease following CPAP treatment. Sleep Breath 16: , Ruehland WR, Rochford PD, O Donoghue FJ, Pierce RJ, Singh P, Thornton AT. The new AASM criteria for hypopneas: impact on the apnea hypopnea index. Sleep 32: , Liu Y, Lowe AA, Zeng X, Fu M, Fleetham JA. Cephalometric comparisons between Chinese and Caucasian patients with obstructive sleep apnea. Am J Orthod Dentofacial Orthop 117: , Popovic R, White D. Upper airway muscle activity in normal women: influence of hormonal status. J Appl Physiol 84: , Zwillich C, Natalino M, Sutton F, Weil J. Effect of progesterone on chemosensitivity in normal man. J Lab Clin Med 92: , The Japanese Society of Internal Medicine