Abstract
We evaluated pulmonary function abnormalities associated with the sleep apnea syndrome (SAS) in 170 habitual snorers without SAS (n = 62, apnea-hypopnea index [AHI] < 10 per hour of sleep), with moderately severe SAS (n = 56, 10 ⩽ AHI < 30) or with severe SAS (n = 52, AHI ⩾ 30). The three groups were similar regarding obesity (BMI ∼ 30 kg · m− 2) and smoking history ( ∼ 20 pack-years). Pulmonary function was assessed by spirometry, forced oscillation mechanics, and gas exchange studies. Forced expiratory flows decreased as the SAS severity increased (p < 0.001, p < 0.02, and p < 0.05 for FEF50, FEV1, and FEV1/VC, respectively). Multiple regression analysis showed that the correlation between FEV50 and the AHI persisted when smoking history was taken into account (p < 0.05), suggesting that SAS may be an independent risk factor for small airway disease. A highly significant correlation was found between specific respiratory conductance (sGrs) and the AHI (p < 0.0001). In a multiple regression analysis (p < 0.0001), variables that influenced sGrs were distal airway obstruction as assessed by FEV50 (p < 0.05), morphological upper airway abnormalities as assessed by cephalometric parameters (p < 0.02), and the AHI (p < 0.0005). SAS appears to be highly correlated to lower and upper airway obstruction, as demonstrated by a reduction in specific respiratory conductance, which adds to the increase in breathing load due to obesity. Zerah-Lancner F, Lofaso F, Coste A, Ricolfi F, Goldenberg F, Harf A. Pulmonary function in obese snorers with or without sleep apnea syndrome.
In recent years, chronic obstructive pulmonary disease (COPD) and sleep apnea syndrome (SAS) have been found to coexist in many patients, who are at increased risk of respiratory insufficiency (1, 2). In a large series of consecutive sleep apnea syndrome (SAS) patients, an obstructive ventilatory defect defined by a FEV1/VC ratio (forced expiratory volume in one second/vital capacity) below 60% was found in 11% of cases, which is probably higher than the prevalence of COPD in the general population. Whether an association exists between SAS and early-stage COPD has not been evaluated.
Early-stage COPD is characterized by small airway disease, whose diagnosis relies mainly on flow-volume curve analysis. The value of the forced oscillation technique (FOT) for the diagnosis of early-stage COPD has recently been demonstrated (3). The detection of small airway disease may be difficult in SAS patients since this syndrome is associated with various pulmonary function abnormalities due to obesity and upper airway obstruction. Both these factors can be responsible for airway obstruction. In non-SAS subjects with various degrees of obesity, overweight has been shown to influence forced expiration and to increase respiratory resistance as measured using the FOT (4, 5). Patients with obstructive SAS have anatomic craniomandibular and upper airway abnormalities (6) that have been shown to correlate with forced expiratory indices (7).
A number of studies have been performed to explore pulmonary function in obese patients with SAS (8-12). To date, the only maximal expiratory flow-volume curve abnormality that has been found in SAS patients but not in non-SAS snorers has been a sawtooth-like appearance, with no decrease in forced expiratory flows. However, in these studies, greater severity of the SAS was consistently accompanied with greater severity of the obesity, and this last factor may have influenced pulmonary function. To evaluate the association between SAS and airway abnormalities, we studied obese snorers without SAS, with moderately severe SAS, or with severe SAS, taking care to obtain homogeneous groups regarding the degree of overweight as assessed by the body mass index (BMI), and to include only patients without chronic bronchitis.
To improve the airway abnormalities detection, we evaluated the mechanical impedance of the respiratory system using the FOT, which is a noninvasive and effort-independent test. The FOT has been shown to detect early airway abnormalities (3) and to be sensitive to central airway obstruction (13). The diagnostic efficacy of the FOT in differentiating obese snorers with and without obstructive sleep apnea, according to the value of the respiratory disturbance index, has not been evaluated.
One hundred seventy patients (136 males and 34 females) who attended the sleep clinic for snoring and a history of restless sleep were included in the study. All patients were overweight as defined by a Quetelet's index (or body mass index [BMI], wt/ht2, where wt is the body weight in kg and ht the height in meters) greater than 25 (14).
Exclusion criteria were previous evaluation or treatment for sleep apnea, a BMI greater than 34, a history of cardiopulmonary disease, a history of airway obstruction due to asthma or chronic obstructive pulmonary disease (COPD), with chronic bronchitis defined as cough and sputum production for at least 3 mo per year during 2 yr or more, alcohol abuse, regular use of hypnotic medication, a history of upper respiratory tract disorders, and a history or clinical evidence of neuromuscular disease.
Informed consent was obtained from all subjects before inclusion into the study.
Overnight sleep studies were performed in all patients with a clinical suspicion of SAS, and consisted in full polysomnography including electroencephalography (C4-A1, C3-A2), electrooculography, chin electromyography, oro-nasal airflow, rib cage and abdominal movements, and arterial oxygen saturation monitored via a finger probe (Nellcor N200; Nellcor Inc., Hayward, MN).
Sleep staging was performed using standard criteria (15). An abnormal breathing event during objectively measured sleep was defined according to commonly used clinical criteria as either complete cessation of airflow lasting 10 s or more (apnea) or an at least 50% decrease in oro-nasal airflow lasting 10 s or more (hypopnea). The average number of apneas and hypopneas per hour of sleep (AHI) was calculated from the sum of sleep-disordered events. For categorical analysis, an AHI cutoff of 10 per hour of sleep was used. The severity of sleep-disordered breathing was quantified based on the number of apneas and hypopneas per hour of sleep, and an AHI cutoff of ⩾ 30 per hour of sleep was used to define severe SAS.
Spirometry measurements and flow-volume curves were obtained using a spirometer (PF/DX 1085D; MedGraphics, St. Paul, MN). The highest values of three technically satisfactory forced expirations were used. All these values were expressed as percentages of predicted values (16). The spirometry technique met international standards (16).
The forced oscillation method used in our study has been described elsewhere (17). The technique met international standards (18). Briefly, a random noise signal of 4 to 32 Hz was generated by loudspeakers and superimposed on the spontaneous breathing of the subject, who was equipped with a mouthpiece and a nose-clip. The subject's cheeks were held firmly. The measurements were performed in the sitting position, with the head in the neutral position. Three technically acceptable measurements were performed. Mouth flow was measured using a screen pneumotachograph (Jaeger, Wurzburg, Germany) connected to a differential pressure transducer (Sensym LX 0600ID; Sensym, Sunyvale, CA). An identical transducer was used to measure mouth pressure. The signals were lowpass-filtered (Butterworth, 8th order, cutoff frequency = 32 Hz) to prevent aliasing (i.e., to eliminate a possible influence of high on low frequencies) and sampled at a frequency of 128 Hz. The signals were fed into a microcomputer on which spectral analysis was performed using a 512 point fast fourier transform (FFT) algorithm. The real component (which is related to the resistive properties of the system) and the imaginary component (which corresponds to inertance and compliance properties) of respiratory impedance were computed every 0.25 Hz from 4 to 32 Hz and displayed as a function of frequency. For each of these frequencies, a coherence function (a function that evaluates the reproducibility of impedance measurements) ranging from 0 to 1 was calculated, and 0.9 was chosen as the lower limit of data acceptance.
The real component of impedance was subjected to linear regression analysis over the 4 to 16 Hz frequency range to obtain the zero-intercept resistance (Rrs) and the slope (S) of the linear relationship of resistive impedance versus frequency, which reflects both the shunt impedance of the upper airways and the distribution of flow among intrathoracic parallel inhomogeneities. Respiratory conductance (Grs) was calculated as the reciprocal of Rrs. Quality of fit was assessed by calculating, for each patient in every condition, the relative distance (RD) between the response of the linear model and that of the patient (19). The relative distance averaged 3.5%, with a standard deviation of 0.8%, demonstrating that the linear model was adequate for describing resistive impedance over the 4-16 Hz frequency range.
Arterial blood was drawn from the radial artery with the patient awake and semi-supine. The blood sample was analyzed for pH, Pco 2, and Po 2 (Blood gas analyzer Radiometer ABL 30; Copenhagen, Denmark). Oxygen saturation was determined using a spectrometer (Radiometer OSM3, Hemoximeter).
Lateral cephalometric roentgenograms were obtained using the technique described by Riley and colleagues (20). The following angles in degrees and dimensions in millimeters were measured on the flat film: SNA, angle measurement from the sella (S) to the nasion (N) to point A (subspinale); SNB, angle measurement from the sella (S) to the nasion (N) to point B (supramentale); MP-H, distance between the mandibular plane (MP) to the hyoid bone (H); PAS, posterior airway space, defined as the space located behind the base of the tongue and limited by soft tissues; and PNS-P, distance between the posterior nasal spine and the tip of the soft palate.
Data were expressed as means ± SEMs.
Comparison of the three groups according to the apnea-hypopnea index (AHI) was done using analysis of variance (ANOVA). When a significant difference was found, individual means were compared using the modified t test. We also used the chi-square test to compare the distribution of males and females according to the value of the AHI (three groups).
Correlations between variables were analyzed using least-square linear regression techniques, and multiple regression analysis was also performed.
For all comparisons, p values < 0.05 were considered significant.
Anthropometric, pulmonary function, and polysomnographic data of the 170 study patients are shown in Table 1 and Figures 1 and 2.
| AHI < 10 (n = 62) | AHI = 10–30 (n = 56) | AHI ⩾ 30 (n = 52) | ANOVA | |||||
|---|---|---|---|---|---|---|---|---|
| AHI (n/ h) | 4.5 ± 0.4 | 18.3 ± 0.7 | 57.2 ± 3.4 | |||||
| Sex | 44 M | 44 M | 48 M | NS | ||||
| Age, yr | 52 ± 2 | 57 ± 2 | 57 ± 2 | NS | ||||
| Smoking history, pack yr | 17 ± 4 | 17 ± 4 | 18 ± 3 | NS | ||||
| BMI, kg · m−2 | 28.5 ± 0.3 | 28.5 ± 0.3 | 29.9 ± 0.4 | NS | ||||
| Neck circumference, cm | 42.2 ± 0.5 | 41.9 ± 0.4 | 43.1 ± 0.7 | NS | ||||
| O2 saturation (nadir), % | 88.2 ± 0.6 | 85.4 ± 1.0 | 73.5 ± 3.4 | p < 0.0001 | ||||
| CT90, min | 3.8 ± 1.6 | 8.9 ± 2.9 | 95.3 ± 24.9 | p < 0.0001 | ||||
| CT90 % tst | 1.0 ± 0.4 | 2.4 ± 0.8 | 22.5 ± 5.6 | p < 0.0001 | ||||
| Mean sleep SaO2 , % | 95.8 ± 0.2 | 96.0 ± 0.2 | 92.9 ± 0.9 | p < 0.0001 | ||||
| TLC % | 95 ± 2 | 95 ± 2 | 92 ± 2 | NS | ||||
| VC % | 100 ± 2 | 99 ± 2 | 94 ± 3 | NS | ||||
| FRC % | 86 ± 2 | 84 ± 3 | 87 ± 3 | NS | ||||
| ERV % | 69 ± 6 | 69 ± 5 | 67 ± 6 | NS |
Fig. 1. Changes in maximal expiratory flow rates in the three groups of obese snorers. The apnea-hypopnea index (AHI) was used to separate subjects without SAS (AHI < 10), with moderate sleep apnea (10 ⩽ AHI < 30), and with severe sleep apnea (AHI ⩾ 30). Expiratory flow rates at 50% and 25% of the vital capacity (FEV50 and FEF25) and the forced mid-expiratory flow rate decreased significantly as the AHI increased. The result of the analysis of variance is given for each panel, together with the between-group comparisons. Regression analysis with AHI as a continuous dependent variable and expiratory flow rates as independent variables yielded similar results (p < 0.001), with correlation coefficients (r) = 0.30.
[More] [Minimize]
Fig. 2. Changes in oscillatory mechanics in the three groups of obese snorers. The apnea-hypopnea index (AHI) was used to separate subjects without SAS (AHI < 10), with moderate sleep apnea (10 ⩽ AHI < 30), and with severe sleep apnea (AHI ⩾ 30). Significant decreases in respiratory conductance (Grs) and specific respiratory conductance (sGrs) and a significant increase in respiratory resistance (Rrs) were observed with increasing AHI. The result of the analysis of variance is given for each panel, together with the between-group comparisons. Regression analysis with AHI as a continuous dependent variable and sGrs as an independent variable yielded similar results (p < 0.0001), with the correlation coefficient (r) = 0.50.
[More] [Minimize]Of the 170 study patients, 62 were found to be free of SAS (group 1) and 108 to have SAS (mean AHI, 39.1 ± 3.5 per hour of sleep). Of the 108 SAS patients, 56 had moderate obstructive sleep apnea (10 ⩽ AHI < 30) (group 2), and 52 had severe obstructive sleep apnea (AHI ⩾ 30) (group 3). The sex-ratio was not significantly different in the three groups (chi-square test, NS). All patients were overweight, and the body mass index (BMI) was not significantly different in the three groups (Table 1). Regression analysis confirmed the absence of any relationship between BMI, as a continuous variable, and AHI, in our patients. No significant difference in neck circumference was found between groups (Table 1). Neither was the smoking history significantly different in the three groups (Table 1); 25% of patients were nonsmokers, 35% were ex-smokers, and 40% were smokers.
To examine whether the severity of SAS was correlated with pulmonary function alterations, we compared the different parameters in the three groups. Regression analysis was also used with AHI as a continuous variable.
Lung volumes and flows. Total lung capacity and vital capacity (VC) were similar in the three groups. As compared to predicted values, all three groups showed significant decreases (p < 0.0001) in functional residual capacity and, most markedly, in expiratory residual volume (30% reduction). These decreases were of similar magnitude in the three groups, as shown in Table 1 (ANOVA, NS).
FEV1, expressed as the percentage of predicted values, was near-normal in the three groups (99 ± 3, 98 ± 3, 89 ± 4%), but decreased significantly as the AHI increased (ANOVA, p < 0.02). Same results were found for the FEV1/VC ratio, expressed as the percentage of predicted values, (100 ± 2, 99 ± 2, 94 ± 2%) with a significant decrease as the AHI increased (ANOVA, p < 0.05). A significant and considerably larger decrease with increasing AHI was observed for expiratory flow rates at 50% and 25% of the vital capacity (FEV50 and FEV25) and for the forced mid-expiratory flow (ANOVA, p < 0.001 for all three parameters) (Figure 1). A significant relationship between FEV50 (expressed as the percentage of the predicted value) and AHI was also found (r = 0.3, p < 0.001). No significant differences were found between the three groups when the slopes of the regression lines were compared using analysis of variance.
Oscillatory parameters. Significant decreases in respiratory conductance (Grs) and the specific respiratory conductance (sGrs) and a significant increase in respiratory resistance (Rrs) were observed with increasing AHI (ANOVA, p < 0.001, p < 0.0001, and p < 0.001, respectively) (Figure 2). A highly significant relationship was found between sGrs and AHI (r = 0.5, p < 0.0001).
When oscillatory parameters and expiratory flow rates were compared, a significant relationship (r = 0.3, p < 0.003) was found between sGrs and FEV50 (expressed as the percentage of the predicted value).
Gas exchange. Analysis of variance showed significant decreases with increasing severity of SAS for arterial Po 2 (85.0 ± 1.5, 83.7 ± 1.4, and 75.5 ± 1.2 mm Hg, for groups 1, 2, 3, respectively, ANOVA p < 0.0001), arterial O2 saturation (96.8 ± 0.2, 96.2 ± 0.2 and 95.2 ± 0.2%, for groups 1, 2, and 3, respectively, ANOVA p < 0.0001), as well as a significant increase in PaCO2 (38.0 ± 0.4, 38.7 ± 0.6 and 40.4 ± 0.7 mm Hg, for groups 1, 2, 3, respectively, ANOVA p < 0.002). Identical conclusions can be drawn from regression analysis between AHI and arterial Po 2, arterial O2 saturation or PaCO2 (p < 0.0001, r = 0.35; p < 0.0001, r = 0.50; and p < 0.002, r = 0.30, respectively).
Oxyhemoglobin nadir during sleep (O2 nadir %), time spent at SaO2 below 90% (CT90), expressed either as absolute values (min) or as the percentage of total sleep time, and the mean sleep SaO2 , were measured from the polysomnograms. All these parameters were shown to be significantly different between groups, with significant relationships with the AHI (Table 1).
As seen in Table 2, a significant increase in MP-H and significant decreases in SNA and SNB were observed with increasing AHI.
| AHI < 10 | AHI 10–30 | AHI ⩾ 30 | ANOVA | |||||
|---|---|---|---|---|---|---|---|---|
| SNA angle, degrees | 82.1 ± 0.5 | 81.6 ± 0.6 | 80.3 ± 0.5 | < 0.05 | ||||
| SNB angle, degrees | 79.7 ± 0.5 | 77.9 ± 0.5 | 77.4 ± 0.6 | < 0.01 | ||||
| PAS, mm | 11.7 ± 0.5 | 11.9 ± 0.5 | 11.7 ± 0.7 | NS | ||||
| PNS-P, mm | 42.3 ± 0.8 | 41.9 ± 0.8 | 43.3 ± 1.3 | NS | ||||
| MP-H, mm | 17.4 ± 0.7 | 19.3 ± 0.9 | 22.7 ± 1.3 | < 0.001 |
Comparison of cephalometric parameters and oscillatory parameters demonstrated a significant relationship (r = 0.25, p < 0.006) between SNB and sGrs.
Conventional pulmonary function tests are generally considered unhelpful for the diagnosis of SAS. Flow-volume curve alterations have been reported in SAS, including an FEF50/ FIF50 ratio greater than unity, a sawtooth appearance of the expiratory segment of the flow-volume loop, and plateauing of the inspiratory flow rate (21, 22). Other studies found that the sensitivity and specificity of oral flow-volume loops data for the diagnosis of SAS were lower than initially reported and that flow-volume curve analysis was not useful as a screening test for SAS (9). Maximal inspiratory and expiratory flow-volume curves have been found unhelpful for differentiating SAS patients from controls (23). None of the studies reported to date have found a decrease in expiratory flow rates in SAS patients as compared to controls. On the other hand an association between chronic obstructive pulmonary disease (COPD) and SAS has been suggested in several studies, and COPD has been shown to be the determining factor in hypercapnia and daytime pulmonary hypertension (24-26). Weitzenblum and coworkers (2) and Chaouat and colleagues (1) found an obstructive ventilatory defect (FEV1/VC < 60%) in 30 of 264 consecutive SAS patients with similar BMIs. Our data are from obese subjects who differed regarding the severity of SAS but were comparable regarding the severity of obesity as well as smoking histories. Our goal was to evaluate the effects of SAS in such a way that our data would not be confounded by potential adverse effects of obesity on lung function. We found that, although smoking history was comparable between groups and none of the subjects had chronic bronchitis, significant decreases in expiratory flow rates and in sGrs occurred with increasing severity of SAS. Interestingly, sGrs was more closely linked to SAS than forced expiratory flows.
In our population of 170 obese snorers, a significant decrease in respiratory conductance was observed, which was more marked when the apnea-hypopnea index (AHI) increased. Statistical analysis showed that sGrs was influenced neither by the age nor by the BMI. We previously demonstrated that, since the decrease in conductance observed in obese subjects was due primarily to decrease in lung volumes, specific conductance (i.e., the ratio of conductance over volume) was independent from the BMI (5). In addition, in the current study we were careful to select patients with similar degrees of obesity with the goal of eliminating a confounding effect of BMI differences on lung function test results. Indeed, regression analysis confirmed the absence of any relationship between BMI, as a continuous variable, and pulmonary function parameters. The significant decrease in sGrs with increasing AHI in our study could be due to obstruction of the upper and/or peripheral airways. To define the role of peripheral obstruction, we further analyzed our data using multiple regression with sGrs as the dependent variable and with FEV50 and AHI as the explanatory variables. This significant multiple regression (r = 0.52, p < 0.0001) demonstrated a stronger partial correlation between sGrs and AHI (p < 0.0001) than between sGrs and FEF50 (p < 0.02). The correlation between sGrs and FEV50 suggests that the decrease in sGrs with increasing AHI was due in part to distal airway obstruction. However, it can also be speculated that upper airway obstruction explained the stronger correlation between sGrs and AHI. Mechanical abnormalities of the upper airways in SAS patients promote recurrent upper airway narrowing and occlusion during sleep (6, 27, 28). We performed stepwise multiple regression analysis (29) with sGrs as the dependent variable and with cephalometric parameters, FEV50, and AHI as the independent variables. This significant multiple regression (r = 0.5, p < 0.0001) demonstrated a significant partial correlation between sGrs and SNB (p < 0.02), as well as a partial correlation between sGrs and FEF50 (p < 0.05); however, persistence of a strong partial correlation between sGrs and AHI (p < 0.0005) was also found. Surprisingly, sGrs was not significantly correlated neither with cephalometric indices related to soft tissue abnormalities, such as PAS or PNS-P, nor with neck circumference. It has been suggested that the adverse effects of obesity on breathing during sleep may be due to accumulation of fat in the neck (30). Using magnetic resonance imaging (MRI), Shelton and coworkers (31) demonstrated that fat was deposited adjacent to the pharyngeal airway in patients with SAS, and that the volume of fat was related to the presence and degree of SAS but was not correlated with the BMI. In this study we used conventional cephalometry which may not be sufficiently sensitive to provide an accurate quantification of soft tissue abnormalities, and it cannot be ruled out that soft tissue abnormalities may have contributed to the correlation between sGrs and AHI in the multiple regression analysis.
Smoking has been shown to be a risk factor for sleep-disordered breathing, and it has been reported that current cigarette smokers were at greater risk than were never smokers (32). To evaluate the possibility that cigarette smoking with its associated peripheral airway obstruction may contribute to the reduction in expiratory flow rates observed in our SAS groups, we compared mean smoking histories, which we found to be similar in the three groups. However, since there were different smoking histories in each group we further analyzed our data using multiple regression with the FEF50 as the dependent variable and with the AHI and the smoking history (pack-years) as the explanatory variables. This significant multiple regression (r = 0.45, p < 0.0001) demonstrated a highly significant correlation (p < 0.0001) between FEF50 and cigarette smoking and a weaker but significant correlation (p < 0.05) between FEF50 and AHI. In other words, even when smoking history was taken into account, FEF50 was still influenced by the severity of the SAS.
We also considered the possibility that asthma secondary to gastroesophageal reflux may have contributed to the decrease in expiratory flow rates. An increased frequency of gastroesophageal reflux has been reported in SAS patients (33). In addition, obesity is associated with an elevation in intraabdominal pressure that increases the likelihood of gastroesophageal reflux. Snoring has been shown to be more common in asthmatics than in nonasthmatics, a difference that was not ascribable to a difference in BMI between the groups (34). It has been suggested that recurrent upper airway obstruction and snoring may be important triggering mechanisms of nocturnal asthma attacks (35). However, none of our subjects complained of sleep-related heartburn or regurgitation of gastric contents into the pharynx, and no asthmatics were included into the study.
We found a strong correlation between the apnea-hypopnea index and the Pco 2, with more severe SAS being associated with higher Pco 2 values. Both PaO2 and SaO2 decreased significantly when the AHI increased. Disturbances in gas exchange have been shown to be largely determined by alterations in ventilatory mechanics, and to be related to obesity and diffuse airway obstruction (24). Bradley and coworkers (24) suggested that obese apneic patients with day time hypercapnia have mild to moderate diffuse airway obstruction that distinguishes them from equally obese patients without hypercapnia. The presence of diffuse airway obstruction may be an important predisposing factor for the development of chronic CO2 retention in such patients (1).
In conclusion, we found, in a population of obese habitual snorers with no symptoms of chronic bronchitis, a significant correlation between the severity of the apnea-hypopnea index and the presence of lower and upper airway obstruction responsible for decreases in expiratory flow rates and in specific respiratory conductance. Earlier studies have found that COPD often coexists with SAS, and that patients with both disorders frequently have marked hypoxemia, hypercapnia, and pulmonary hypertension. Our study supports and adds to these results by showing that, even in a population of obese SAS patients with mild airway obstruction, the severity of the airway obstruction was significantly correlated with the severity of the sleep-related breathing disorder. Our data also confirm earlier studies demonstrating the significant role of the upper airways in SAS patients. Finally, our findings demonstrate the abnormality of a forced oscillation technique parameter, namely specific respiratory conductance, in patients with SAS.
The authors thank Estelle Dahan for technical assistance, A. M. Lorino for stimulating discussions, and H. Lorino for help in the data processing of the forced oscillation technique.
| 1. | Chaouat A., Weitzenblum E., Krieger J., Ifoundza T., Oswald M., Kessler R.Association of chronic obstructive pulmonary disease and sleep apnea syndrome. Am. J. Respir. Crit. Care Med15119958286 |
| 2. | Weitzenblum E., Krieger J., Oswald M., Chaouat A., Bachez P., Kessler R.Chronic obstructive pulmonary disease and sleep apnea syndrome. Sleep1519923335 |
| 3. | Brochard L., Pelle G., De Palmas J., Brochard P., Carré A., Lorino H., Harf A.Density and frequency dependence of resistance in early airway obstruction. Am. Rev. Respir. Dis1351987579584 |
| 4. | Yap J. C. H., Watson R. A., Gilbey S., Pride N. B.Effects of posture on respiratory mechanics in obesity. J. Appl. Physiol79199511991205 |
| 5. | Zerah F., Harf A., Perlemuter L., Lorino H., Lorino A. M., Atlan G.Effects of obesity on respiratory resistance. Chest103199314701476 |
| 6. | Rivlin J., Hoffstein V., Kalbfleisch J., McNicholas W., Zamel N., Bryan A. C.Upper airway morphology in patients with idiopathic obstructive sleep apnea. Am. Rev. Respir. Dis1291984355360 |
| 7. | Stauffer J. L., Zwillich W., Cadieux R. J., Bixler E. O., Kales A., Varano L. A., White D. P.Pharyngeal size and resistance in obstructive sleep apnea. Am. Rev. Respir. Dis1361987623627 |
| 8. | Önal E., Leech J. A., Lopata M.Relationship between pulmonary function and sleep-induced respiratory abnormalities. Chest871985437441 |
| 9. | Hoffstein V., Wright S., Zamel N.Flow-volume curves in snoring patients with and without obstructive sleep apnea. Am. Rev. Respir. Dis1391989957960 |
| 10. | Gold A. R., Schwartz A. R., Wise R. A., Smith P. L.Pulmonary function and respiratory chemosensitivity in moderately obese patients with sleep apnea. Chest103199313251329 |
| 11. | Shepard J. W., Burger C.Nasal and oral flow-volume loops in normal subjects and patients with obstructive sleep apnea. Am. Rev. Respir. Dis142199012881293 |
| 12. | Stauffer J. L., White D. P., Zwillich C. W.Pulmonary function in obstructive sleep apnea: relationships to pharyngeal resistance and cross-sectional area. Chest971990302307 |
| 13. | Harf A., Decramer M., Zin W., Milic J., Emili, Chang H. K.Respiratory resistance in dogs by the single breath and the forced oscillation methods. J. Appl. Physiol591985262265 |
| 14. | Jequier E.Energy, obesity, and body weight standards. Am. J. Clin. Nutr45198710351047 |
| 15. | Rechtschaffen, A., and A. Kales. 1968. A manual of standardized terminology, techniques and scoring system for sleep stages of human sleep. National Institutes of Health, U.S. Government Printing Office, Washington, DC, Publication No. 204. |
| 16. | Quanjer, P. H., G. J. Tammeling, J. E. Cotes, O. F. Pedersen, R. Peslin, and J. C. Yernault. 1993. Lung volumes and forced ventilatory flows. Report working parting. Standardization of lung function tests. European community for steel and coal. Eur. Respir. J. 6(Suppl. 16):5–40. |
| 17. | Pelle G., Lorino A. M., Lorino H., Mariette C., Harf A.Microcomputer-based system to calculate respiratory impedance from forced random noise data. Med. Biol. Eng. Computer241986541544 |
| 18. | van de Woestijne K., Desager K., Duiverman E., Marchal F.Recommendations for measurement of respiratory input impedance by means of the forced oscillation method. Eur. Respir. Rev41994235237 |
| 19. | Oostveen E., Peslin R., Gallina C., Zwart A.Flow and volume dependence of respiratory mechanical properties studied by forced oscillation. J. Appl. Physiol67198922122218 |
| 20. | Riley R., Guilleminault C., Herran J., Powell N.Cephalometric analyses and flow-volume loops in obstructive sleep apnea patients. Sleep61983303311 |
| 21. | Neukirch F., Weitzenblum E., Liard R., Korobaeff M., Henry C., Orvoën-Frija E., Kauffmann F.Frequency and correlates of the saw-tooth pattern of flow-volume curves in an epidemiological survey. Chest1011992425431 |
| 22. | Tammelin B. R., Wilson A. F., De Berry B., Borowiecki, Sassin J.Flow-volume curves reflect phyaryngeal airway abnormalities in sleep apnea syndrome. Am. Rev. Respir. Dis1261983712715 |
| 23. | Katz I., Zamel N., Slutsky S., Rebuck A. S., Hoffstein V.An evaluation of flow-volume curves as a screening test of obstructive sleep apnea. Chest981990337340 |
| 24. | Bradley T. D., Rutherford R., Lue F., Moldofsky H., Grossman R. F., Zamel N., Phillipson E. A.Role of diffuse airway obstruction in the hypercapnia of obstructive sleep apnea. Am. Rev. Respir. Dis1341986920924 |
| 25. | Krieger J., Sforza E., Apprill M., Lampert E., Weitzenblum E., Ratomaharo J.Pulmonary hypertension, hypoxemia and hypercapnia in obstructive sleep apnea patients. Chest961989729737 |
| 26. | Weitzenblum E., Krieger J., Apprill M., Vallée E., Ehrhart M., Ratomaharo J., Oswald M., Kurtz D.Daytime pulmonary hypertension in patients with obstructive sleep apnea syndrome. Am. Rev. Respir. Dis1381988345349 |
| 27. | Yildirim N., Fitzpatrick M. F., Whyte K. F., Jalley R., Wightman A. J. A., Douglas N. J.The effect of posture on upper airway dimensions in normal subjects and in patients with the sleep apnea/ hypopnea syndrome. Am. Rev. Respir. Dis1441991845847 |
| 28. | Brown I. G., Bradley T. D., Phillipson E. A., Zamel N., Hoffstein V.Pharyngeal compliance in snoring subjects with and without obstructive sleep apnea. Am. Rev. Respir. Dis1321985211215 |
| 29. | Altman, D. G. 1991. Practical Statistics for Medical Research. Chapman and Hall, London. 340–350. |
| 30. | Ferguson K. A., Ono T., Lowe A. A., Ryan F., Fleetham J. A.The relationship between obesity and craniofacial structure in obstructive sleep apnea. Chest1081995375381 |
| 31. | Shelton K. E., Woodson H., Gay S., Suratt P. M.Phyaryngeal fat in obstructive sleep apnea. Am. Rev. Respir. Dis1481993462466 |
| 32. | Wetter D. W., Young T. B., Bidwell T. R., Badr S., Palta M.Smoking as a risk factor for sleep-disordered breathing. Arch. Intern. Med154199422192224 |
| 33. | Kerr P., Shoenut J. P., Millar T., Buckle P., Kryger M. H.Nasal CPAP reduces gastroesophageal reflux in obstructive sleep apnea syndrome. Chest101199215391544 |
| 34. | Fitzpatrick M. F., Martin K., Fossey E., Shapiro C. M., Elton R. A., Douglas N. J.Snoring, asthma and sleep disturbance in Britain: a community-based survey. Eur. Respir. J61993531535 |
| 35. | Shu Chan, C., A. J. Woolcock, and C. E. SullivanNocturnal asthma: role of snoring and obstructive sleep apnea. Am. Rev. Respir. Dis137198815021504 |
