Background:

Optimization of the indications for nocturnal polysomnography in the diagnosis of obstructive sleep apnea syndrome (OSAS) could lead to significant reductions in health expenditure. To this end, we assessed the usefulness of truncal obesity measurements in the diagnosis of OSAS.

Methods and Procedures:

One hundred ninety-two patients (152 men and 40 women) referred to our clinic with suspected OSAS underwent a complete polysomnography study and their BMI and truncal obesity measurements were obtained.

Results:

Comparison of patients defined as snorers with OSAS patients, by means of the Student's t -test revealed significant differences in the truncal obesity parameters such as waist-to-hip ratio (0.94 vs. 0.98) and waist circumference (100.7 cm vs. 106.3 cm). We found no significant differences between BMI values in the two groups. Indices of truncal obesity were analyzed as predictive factors for OSAS using a multivariate logistic regression model that included variable sex, and in which waist-to-hip ratio, BMI, neck circumference, and age were analyzed as binary variables. According to this model, our patients' risk of suffering from OSAS was 2.6 times greater if the waist-to-hip ratio was >1 in men and >0.85 in women. The risk of OSAS was also increased if the patients were men or over 52 years of age.

Discussion:

Obtaining simple measurements such as those for truncal obesity may help prioritize the use of polysomnography in patients with a greater risk of OSAS. Our results suggest that BMI is not a good predictor of OSAS in a group of patients with a high BMI.

Introduction

Obstructive sleep apnea syndrome (OSAS) occurs in 2–4% of the population and is characterized by the repeated collapse of the upper airway at night with consequent oxygen desaturation and interruption of sleep (1). The disease can have major consequences, as indicated by epidemiological data showing that it is an independent risk factor contributing to morbidity and mortality in patients with cardiovascular disease (2,3,4,5,6,7).

Between 60 and 70% of patients diagnosed with OSAS are obese (8), and according to some authors, >50% of OSAS patients have a BMI of >40, thereby falling into the category of morbid obesity (9). Regional fat distribution is different in each sex, with fat tending to accumulate predominantly around the waist in men and around the hips in women (10). The predominance of truncal obesity in men may explain the difference in the prevalence of OSAS between the sexes (11).

Measurement of abdominal tissue using computed axial tomography (CAT) and magnetic resonance imaging (MRI) has shown that patients with OSAS have significantly greater amounts of abdominal fat (12,13), and waist measurement provides a good estimate of the amount of abdominal fatty tissue. Foster et al. (14) defined truncal obesity as corresponding to a waist circumference of at least 101.6 cm in men and 88.9 cm in women, and similar data relating to waist circumference have been obtained in Spain (15). A waist-to-hip ratio >1 in men and >0.85 in women is associated with higher cardiovascular risk (16). Waist circumference and the waist-to-hip ratio have been found to be better than BMI as predictors of cardiovascular disease (17,18). These indices for truncal obesity, i.e., the amount of intraabdominal fat, are also linked to a higher probability of these patients presenting with respiratory disorders during sleep (7,12,19,20), as has also been shown with intraabdominal fat measured using computed axial tomography scans and magnetic resonance imaging (12,13).

The aim of this study was to determine whether these indices of truncal obesity could aid in the diagnosis of OSAS in Spain. If this were the case, the simple measurement of waist circumference or the waist-to-hip ratio could help us to prioritize patients. Considering the large number of patients who attend clinics suffering from disruptive snoring and OSAS, these easily obtained measurements could be very helpful.

Methods and Procedures

We performed a cross-sectional study in which data relating to obesity indices and polysomnography were obtained in 192 patients who had been referred to our sleep respiratory disorders clinic with suspected OSAS. The patient group consisted of 40 women and 152 men. All patients were adults and had been referred for assessment of hypersomnolence, severe snoring, and/or evidence of apnea at night reported by their partners. Patients suffering from chronic obstructive pulmonary disease, cancer, advanced liver diseases, mental illness requiring treatment with antidepressants or benzodiazepines, drug addiction, or alcoholism were excluded from the study.

All patients underwent conventional nocturnal polysomnography that included the following signals: two electroencephalogram channels, one submental electromyography channel, two electrooculography channels, one thermistor for nasal and oral flow, thoracic and abdominal belts for respiratory effort, a tracheal microphone to measure snoring intensity, and a pulse oximeter attached to the finger to measure oxygen saturation (SpO₂). All the variables were evaluated using the KnightScan polysomnography software package (version 1.1996, Nellcor Puritan Bennet, Minneapolis, MN). Sleep was staged according to the criteria of Rechtschaffen and Kales (21). Apnea was defined as cessation of oronasal airflow for >10 s. Hypopnea was defined as a reduction in flow amplitude of 50% for 10 s accompanied by a 4% reduction in SpO₂, or by arousal. The total number of episodes of apnea and hypopnea was divided by the total sleep time to give the apnea hypopnea index (AHI) and this index was used to classify the subjects as snorers or patients with OSAS. The cutoff point for OSAS was an AHI 10. Other cutoff points, such as AHI 5 and 15 were also used.

The patients were weighed and their heights measured, and the BMI for each one was calculated by dividing the weight in kilograms by the square of the height in meters (kg/m²). The circumference of the neck was measured in cm at the level of the cricothyroid membrane. Waist circumference was measured at the narrowest point between the last costal arch and the iliac crest. Hip circumference was measured as the greatest circumference at the level of the trochanters. The waist-to-hip ratio was calculated by dividing the value for the waist measurement by that for the hip measurement.

A statement was obtained from the Institutional Review Board indicating that approval of the study was not necessary, since it did not involve interventional procedures affecting the course of the disease, and since diagnostic procedures did not differ from those normally used. The study was carried out in accordance with the Declaration of Helsinki and informed consent was obtained from all patients.

Statistical analysis. Results are shown as mean s.d. for quantitative variables and as relative frequency and percentage for qualitative variables. Comparisons between quantitative variables were performed with the Student's t -test for independent samples. Where applicability requirements were not fulfilled, the Mann–Whitney U -test was used. Graphical analysis of possible trends in AHI (that were potentially associated with factors such as BMI and truncal obesity, indicated by waist measurement, waist-to-hip ratio, and neck circumference) was performed by distribution of the variables into quartiles. After converting the AHI score to OSAS according to the previously mentioned criteria, these variables were analyzed as binary variables in a logistic regression model in order to determine whether they were associated with the appearance of OSAS. After that we calculated the odds ratios of the different probable variables, as a function of other cutoff points such as AHI 5 and 15. Data were analyzed using the software package SPSS version 13 for Windows. Values of P 0.05 were considered to be statistically significant.

Results

One hundred ninety-two patients with suspected OSAS were chosen from our sleep disorder clinic. The mean (SD) age was 51.9 9.5 years and 22.8% of the patients (40 subjects) were women. The mean BMI was 31.9 5.9. The rest of the descriptive data for the sample are shown in Table 1. Sixty-eight of the one hundred and ninety-two subjects were snorers (AHI < 10) (Table 2).

Table 1 - Characteristics of the study group.

Full table

Table 2 - Comparison of several parameters between patients with and without obstructive sleep apnea syndrome.

Full table

There were no significant differences in BMI between snorers (31.4 5.5) and subjects with OSAS (31.02 6.6). However, the parameters for truncal obesity determined as the waist-to-hip ratio and waist circumference did show significant differences (P < 0.001 and P = 0.009, respectively; Student's t- test for independent samples). Thus, the snorers had a waist-to-hip ratio of 0.94 0.06 cm and a waist circumference of 100.7 14.9 cm, while in the OSAS group the figures were 0.98 0.06 cm and 106.3 12.4 cm, respectively. Neck circumference was also significantly greater (P = 0.01) in the OSAS group than in the snorers: 42.6 3.6 cm in the OSAS group and 40.4 4.6 cm in the snorers group.

Of the patients with OSAS, 85% (105 patients) were men. Prevalence of OSAS was 61% (105 patients) among men and 47% (19 patients) among women. This clear male predominance was statistically significant (P = 0.010) (Table 2).

We subsequently categorized the following obesity and truncal obesity variables in quartiles: BMI, waist circumference, waist-to-hip ratio, and neck circumference. A general tendency was observed for AHI to increase in line with the value for the studied factor, especially the neck circumference, where there was also a tendency toward increased dispersion of AHI with increasing circumference (Figures 1–3). However, we did not observe this tendency between BMI and severity of OSAS (Table 3 and data not shown).

Figure 1.

Boxplot of the relationship between apnea-hypopnea index (AHI) and waist-to-hip ratio.

Full figure and legend (9K)

Figure 3.

Boxplot of the relationship between apnea-hypopnea index (AHI) and neck circumference.

Full figure and legend (8K)

Table 3 - Apnea-hypopnea index corresponding to truncal obesity variables divided into quartiles.

Full table

We then studied the characteristics of the patients according to sex. As shown in Table 4, the men had a significantly higher AHI despite also having a significantly lower BMI. The AHI score was 29.5 25.7 for men and 16 17.8 for women, whereas the BMI was 33.5 8.4 for women and 30.7 4.9 for men. Of the truncal obesity indices, only the waist-to-hip ratio showed significant differences: 0.91 0.7 in women and 0.98 0.5 in men (P < 0.0001).

Table 4 - Comparison of variables according to sex.

Full table

Finally, we categorized BMI, neck circumference, waist-to-hip ratio, waist circumference, and age into binary variables. According to their BMI, we divided the patients into obese and non-obese groups with the cutoff point being at 30. The neck circumference was assessed as an overall binary variable, taking sex into account. Hence, women with a neck measurement of > 37 cm and men with a neck measurement of >42 cm were given the same value. Those figures corresponded to the respective mean of each population and coincide with those of the mean for the group reported by Dancey et al. (22)

The waist-to-hip ratio and waist circumference were also categorized as overall binary variables, taking sex into account. Based on data from the literature, truncal obesity was defined as a waist-to-hip ratio of >1 in men and 0.85 in women (16) and a waist circumference of >102 cm in men and >90 cm in women. (15) Age was categorized according to the mean for the group of 192 patients (52 years).

Using these independent variables, we constructed a multivariate logistic regression model in which OSAS was considered as a dependent variable based on an AHI of >10. The results of that analysis showed that BMI did not play a role as a risk factor for OSAS and that the truncal obesity variable that represented the greatest risk for OSAS was the waist-to-hip ratio. In the model that included waist-to-hip ratio, age, sex, BMI, and neck circumference, we observed that OSAS was 2.6 times more likely to occur if the waist-to-hip ratio was >1 in men and >0.85 in women. That model also showed that there was a fourfold increase in the probability of suffering from OSAS if the patient was male and a twofold increase if the patient was >52 years of age (Table 5). Subsequently we wanted to see what would happen if we used other cutoff points such as an AHI 5 or >15. We observed an inverse progression of the BMI and waist/hip ratio values with respect to the different cutoffs. As the AHI cutoff was increased, the BMI lost its usefulness as a screening tool, while the waist/hip ratio gained in usefulness. The sex of the patient was found to be always useful as a screening tool (Table 6).

Table 5 - Odds ratio of the indicated variables for development of OSAS according to a multivariate logistic regression model.

Full table

Table 6 - Odds ratio of the indicated variables for development of OSAS according to a different ^a multivariate logistic regression models.

Full table

Discussion

In this study, we observed significant differences in the truncal obesity parameters of waist and neck circumference and waist-to-hip ratio between patients who were snorers and those with OSAS (AHI > 10). From studying these preliminary data, we noted that the BMI was not related to the diagnosis of OSAS in our patients. This finding, which was subsequently confirmed by multivariate logistic regression analysis, contrasts with a large number of studies that have shown a direct relationship between BMI and AHI. A positive relationship between AHI and BMI was observed in the Sleep Heart Health Study, a large population-based study conducted on a group of 4991 subjects taken from the general population (7). That study is clearly not comparable to ours, either in the number of patients or their origin. In our study the patients had previously been referred to a sleep disorder clinic and therefore already had an initial BMI bias. Schäfer et al. (23) obtained similar data to those of the Sleep Heart Health Study, but with a group of 81 male patients who had previously been referred to a sleep disorder clinic. Using a simple linear regression analysis, they observed a relationship between the variables AHI and BMI and the waist-to-hip ratio. However, other studies similar to ours but with a smaller number of patients also found no clear relationship between BMI and AHI. For instance, in a study conducted on 44 male patients who had also been referred to a sleep disorder clinic, Levinson et al. (8) found no significant relationship between AHI and BMI. Another study by O'Keeffe and Patterson (24) using a group of 170 patients with a BMI >35, found no relationship between a diagnosis of OSAS and BMI. By contrast, a study by Namyslowski et al. (25) undertaken with 106 subjects found no relationship between BMI and AHI in patients with a BMI of between 25 and 30 but did observe a positive correlation in patients with a BMI >30. Thus, the literature is inconsistent regarding the role of BMI, as a general obesity index, in the diagnosis of OSAS.

Comparison of the variables between the sexes revealed that women had a mean AHI that represents mild OSAS (AHI, 16) and men had moderate OSAS (AHI, 30). The women had a lower AHI than the men, although they were more obese. This finding implies that men require a lower general obesity index in order to develop apnea. The waist-to-hip ratio was significantly different between the sexes—higher in men than in women—and was closer to 1 in men than in women. Truncal obesity is a known cardiovascular risk factor (17,26,27,28) and has, in turn, been correlated with OSAS (7,12,19,20). The results of this study add to those that show a relationship between OSAS and truncal obesity and also suggest that women have factors that protect them against developing OSAS and truncal obesity. Millman et al. (19) observed that the disease was less severe for women, despite the fact that the BMI was similar in both sexes. They suggested that the difference in severity was partly attributable to the truncal distribution of fat in men. Mohsenin (29) obtained similar results, which in that case were attributed to a protective effect of mechanisms in the upper airway. Resta et al. (30) showed in a group of 230 obese patients that values for AHI, prevalence of OSAS, neck circumference, and waist-to-hip ratio were lower in women than in men. This favors the view that obesity as a general index is of less importance. Instead, we should pay greater attention to the "nature" or "type" of obesity of our OSAS patients and, specifically, place greater value on truncal obesity indices than on BMI as a general index. Several studies have assessed the relationship between truncal obesity and OSAS, and some studies have observed a correlation between this type of obesity and the severity of OSAS. For example, the study of 85 men by Shäfer et al. (23), showed that AHI could be correlated with intraabdominal and subcutaneous abdominal fat measured by MRI. A correlation was also found in the Sleep Heart Health Study (7). Using a multiple linear regression model, adjusted for age and BMI, that study found a relationship between AHI and the waist-to-hip ratio.

It would be very useful to have a tool that, using data easily obtained in the clinic, allowed us to detect those patients at greater risk of OSAS. With this in mind, we constructed a multivariate logistic regression model that included the waist-to-hip ratio, sex, age, BMI, and neck circumference. We used different AHIs as cutoff points (AHI >5, AHI >10 and AHI >15) and we observed that the effectiveness of the waist-to-hip ratio as a screening tool increased with the value of the AHI being used. The BMI was found useful for detecting minor cases or snorers, but this variable lost its effectiveness as an indicator when the cutoff reached an AHI >10 or AHI >15. Therefore, we found that the waist-to-hip ratio, sex and age were independent predictive factors for OSAS (AHI > 10). And this relationship was maintained for an AHI >15, except in the case of age. Similar studies have been published in this vein, although based on different populations and methods. For example, Teculescu et al. (31) compared patients who had been referred for intense snoring, with light snorers and non-snorers. They found significant differences in the waist-to-hip ratios but not for BMI. Polysomnography was not performed in that study and so their data could be called into question on this point. Sharma et al. (32) carried out a study where polysomnography was performed on 150 subjects. They used a multivariate logistic regression model to show that a BMI > 25 corresponded to an odds ratio of 4.98 for OSAS and that a waist-to-hip ratio >0.95 in men and >0.88 in women corresponded to an odds ratio of 12.94 for OSAS. Sharma et al. (33) had previously carried out a study only on patients with a BMI greater than 25 (corresponding to obesity in Asia) where they constructed a multivariate logistic regression model in which the variables that maintained a significant odds ratio for OSAS were the waist-to-hip ratio (1.07), male sex (3.97), and neck circumference (1.23). The BMI was not maintained in that model. This data can be extrapolated to our study, as we worked with patients with a high BMI. This suggests that the indices for truncal obesity, which are already good predictors of OSAS in populations with a normal BMI, are better predictors in obese patients. We therefore believe that truncal obesity should always be measured in outpatient clinics and particularly in obese patients.

Based on our results, a male patient with suspected OSAS who is over 52 years of age and has a waist-to-hip ratio >1 would be very likely to produce a positive result with polysomnography. For this reason, we believe that such simple measurements as waist and hip circumference should be made routine in a sleep disorder clinic.

We would like to briefly discuss the finding that neck circumference was not a risk factor for OSAS in our logistic regression model. In the study by Shäfer et al. (23) the authors observed that subcutaneous fat in the neck region and parapharyngeal fat did not correlate with AHI. However, the literature is inconsistent on this, as several articles show neck circumference as a predictor of OSAS in men (34) and even as a factor responsible for airway obstruction (35,36). Our finding may indicate that neck circumference and BMI are measures of general obesity like subcutaneous fat or skin folds and it confirms that waist-to-hip ratio is a measure of visceral obesity, which is directly involved in insulin resistance or metabolic syndrome (37).

To conclude, our study shows that indices of truncal obesity are better than BMI for evaluating obesity as a related or risk factor for OSAS. Furthermore, we provide data supporting a protective effect for the female sex—in our study, women had a lower AHI despite being more obese. Considering the large number of patients who are snorers with suspected OSAS who attend our clinics every day, we believe that measures of truncal obesity could be of considerable use in discriminating between patients with a higher or lower risk for OSAS.

DISCLOSURE

The authors declared no conflict of interest.

References

REFERENCES

1. 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 | Article | PubMed | ISI | ChemPort |
2. He J, Kryger MH, Zorick FJ. et al. Mortality and apnea index in obstructive sleep apnea. Chest 1988;94:9–14 | Article | PubMed | ChemPort |
3. Partinen M, Jamieson A, Guilleminault C. Long-term outcome for obstructive sleep apnea syndrome patients. Chest 1988;94:1200–1204 | Article | PubMed | ChemPort |
4. Lavie P, Herer P, Peled R. et al. Mortality in sleep apnea patients: a multivariate analysis of risk factors. Sleep 1995;18:149–157 | PubMed | ChemPort |
5. Martí S,Sampol G, Muñoz X. et al.Mortality in severe sleep apnoea/hypopnoea syndrome patients: impact of treatment.Eur Respir J 2002;20:1511–1518 | Article |
6. Marín JM, Carrizo SJ, Vicente F. et al. Long-term cardiovascular outcomes in men with obstructive sleep apnoea-hypopnoea with or without treatment with continuos positive airway pressure: an observational study. Lancet 2005;365:1046–1053 | PubMed | ISI |
7. Newman AB, Nieto FJ, Guidry U. et al. Relation of sleep-disordered breathing to cardiovascular disease risk factors. The Sleep Heart Health Study. Am J Epidemiol 2001;154:50–59 | Article | PubMed | ISI | ChemPort |
8. Levinson PD, McGarvey ST, Carlisle CC. et al. Adiposity and cardiovascular risk factors in men with obstructive sleep apnea. Chest 1993;103:1336–1342 | Article | PubMed | ISI | ChemPort |
9. Resta O, Foschino-Barbaro MP, Legari G. et al. Sleep-related breathing disorders, loud snoring and excessive daytime sleepiness in obese subjects. Int Obes Relat Metab Disord 2001;25:669–675 | Article | ChemPort |
10. Folsom AR,Stevens J, Schreiner PJ. et al. Body mass index, waist/hip ratio, and coronary heart disease incidence in African Americans and whites. Atherosclerosis Risk in Communities Study Investigators. Am J Epidemiol 1998;148:1187–1194 | PubMed | ISI | ChemPort |
11. Kissebah AH, Vydelingum N, Murray R. et al. Relation of body fat distribution to metabolic complications of obesity. J Clin Endocrinol Metab 1982;54:254–260 | PubMed | ISI | ChemPort |
12. Vgontzas AN, Papanicolaou DA,Bixler EO. et al. Sleep apnea and daytime sleepiness and fatigue: relation to visceral obesity, insulin resistance, and hypercytokinemia. J Clin Endcrinol Metab 2000;85:1151–1158 | ChemPort |
13. Shinohara E, Kihara S, Yamashita S. et al. Visceral fat accumulation as an important risk factor for obstructive sleep apnoea syndrome in obese subjects. J Intern Med 1997;241:11–18 | Article | PubMed | ISI | ChemPort |
14. Foster GD. Principles and practices in the management of obesity. Am J Respir Crit Care Med 2003;168:274–280 | Article | PubMed |
15. Sociedad Española par el Estudio de la Obesidad (SEEDO).Consenso SEEDO'2000 para la evaluación del sobrepeso y la obesidad y el establecimiento de criterios de intervención terapéutica.Med Clin (Barc) 2000;115:587–597
16. Heymsfield SB, Allison DB, Wang ZM. et al. Evaluation of total and regional body composition.In:Bray GA,Bouchard C,James WPT. (eds).Handbook of Obesity.Marcel Dekker: New York,1998, pp 41–78
17. Larsson B, Svardsudd K, Welin L. et al.Abdominal adipose tissue distribution, obesity, and risk of cardiovascular disease and death: 13 year follow-up of participants in the study of men born in 1913. Br Med J (Clin Res Ed) 1984;288:1401–1404 | ISI | ChemPort |
18. Lakka HM, Lakka TA, Tuomilehto J. et al. Abdominal obesity is associated with increased risk of acute coronary events in men.Eur Heart J 2002;23:706–713 | Article | PubMed | ISI |
19. Millman RP, Carlisle CC, McGarvey ST. et al.Body fat distribution and sleep apnea severity in women. Chest 1995;107:362–366 | PubMed | ChemPort |
20. Grunstein R, Wilcox I, Yang T. et al. Snoring and sleep apnoea in men: association with central obesity and hypertension. Int J Obes 1993;17:533–540 | ChemPort |
21. Rechtschaffen A, Kales A. A manual of standardised terminology, techniques and scoring system for sleep stages of human subjects.Public Health Service: Washington DC,1968 (National Institutes of Health Publication no. 204)
22. Dancey RD, Hanley PJ, Soong Ch. et al. Gender differences in sleep apnea: the role of neck circumference. Chest 2003;123:1544–1550 | Article | PubMed |
23. Schäfer H, Pauleit D, Sudhop T. et al. Body fat distribution, serum leptin, and cardiovascular risk factors in men with obstructive sleep apnea. Chest 2002;122:829–839 | Article | PubMed |
24. O'Keeffe T, Patterson EJ. Evidence supporting routine polysomnography before bariatric surgery. Obes Surg 2004;14:23–26 | Article | PubMed |
25. Namyslowski G, Scierski W,Mrowka-Kata K . et al.Sleep study in patients with overweight and obesity. J Physiol Pharmacol 2005;56:59–65 | PubMed |
26. Hubert HB, Feinleib M, McNamara PM. et al. Obesity as an independent risk factor for cardiovascular disease: a 26-year follow-up of participants in the Framingham Heart Study. Circulation 1983;67:968–977 | PubMed | ISI | ChemPort |
27. Despres JP. Abdominal obesity as important component of insulin-resistance syndrome. Nutrition 1993;9:452–459 | PubMed | ChemPort |
28. Higgins M, Kannel W, Garrison R. et al. Hazards of obesity—the Framingham experience. Acta Med Scand Suppl 1989;723:23–36
29. Mohsenin V. Gender differences in the expression of sleep-disordered breathing: role of upper airway dimensions. Chest 2001;120:1442–1447 | Article | PubMed | ISI | ChemPort |
30. Resta O, Caratozzolo G, Pannacciulli N. et al. Gender, age and menopause effects on the prevalence and the characteristics of obstructive sleep apnea in obesity. Eur J Clin Invest 2003;33:1084–1089 | Article | PubMed | ChemPort |
31. Teculescu D, Benamghar L, Hannhart B. et al. Habitual loud snoring. A study of prevalence and associations in 850 middle-aged French males. Respiration 2006;73:68–72 | Article | PubMed |
32. Sharma SK, Kumpawat S, Banga A. et al. Prevalence and risk factors of obstructive sleep apnea syndrome in a population of Delhi, India. Chest 2006;130:149–156 | Article | PubMed |
33. Sharma SK, Kurian S, Malik V. et al. A stepped approach for prediction of obstructive sleep apnea in overtly asymptomatic obese subjects: a hospital based study. Sleep Med 2004;4:351–357 | Article |
34. Davies RJO,Ali AJ, Stradling JR. Neck circumference and other clinical features in the diagnosis of the obstructive sleep apnea syndrome. Thorax 1992;47:101–105 | PubMed | ChemPort |
35. Koenig JE, Thach BT. Effects of mass loading on the upper airway. J Appl Physiol 1988;64:2294–2299 | PubMed | ChemPort |
36. Horner RL, Mohiaddin RH, Lowell DG. et al. Sites and sizes of fat deposits around the pharynx in obese patients with obstructive sleep apnea and weight matched controls. Eur Resp J 1989;2:613–622 | ChemPort |
37. Vgontzas AN, Bixler EO, Chrousos GP. Metabolic disturbances in obesity versus sleep apnoea: the importance of visceral obesity an insulin resistance. J Intern Med 2003;254:32–44 | Article | PubMed | ISI | ChemPort |