|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Association Nationale pour le Traitement à Domicile de lInsuffisance Respiratoire Chronique, Paris, France.
Correspondence to: Jean-Pierre Laaban, MD, Department of Pneumology, Hôtel-Dieu, 1 place du Parvis Notre-Dame, 75004 Paris, France; e-mail: j-pierre.laaban{at}htd.ap-hop-paris.fr
| Abstract |
|---|
|
|
|---|
Study objectives: To assess the prevalence of daytime hypercapnia in a very large population of adult patients with OSAS, free of associated COPD, and with a wide range of body mass index (BMI), and to evaluate the relationship between daytime hypercapnia and the severity of obesity and obesity-related impairment in lung function.
Design: Retrospective analysis of prospectively collected data.
Methods: The database of the observatory of a national nonprofit network for home treatment of patients with chronic respiratory insufficiency (Association Nationale pour le Traitement à Domicile de lInsuffisance Respiratoire Chronique) was used. Collected data at treatment initiation were age, apnea-hypopnea index, BMI, FEV1, vital capacity (VC), and arterial blood gases. The study included 1,141 adult patients with OSAS treated in France with nocturnal nasal continuous positive airway pressure (CPAP), FEV1
80% predicted, FEV1/VC
70%, and absence of restrictive respiratory disease other than related to obesity.
Results: The prevalence of daytime hypercapnia (PaCO2
45 mm Hg) before initiating CPAP therapy was 11% in the whole study population. The prevalence of daytime hypercapnia was 7.2% (27 of 377 patients) with BMI < 30, 9.8% (58 of 590 patients) with BMI from 30 to 40, and 23.6% (41 of 174 patients) with BMI > 40. Patients with daytime hypercapnia had significantly higher BMI values and significantly lower VC, FEV1, and PaO2 values than the normocapnic patients. Stepwise multiple regression showed that PaO2, BMI, and either VC or FEV1 were the best predictors of hypercapnia, but these variables explained only 9% of the variance in PaCO2 levels.
Conclusion: Daytime hypercapnia was observed in > 1 of 10 patients with OSAS needing CPAP therapy and free of COPD, and was related to the severity of obesity and obesity-related impairment in lung function. However, other mechanisms than obesity are probably involved in the pathogenesis of daytime hypercapnia in OSAS.
Key Words: alveolar hypoventilation carbon dioxide hypercapnia obesity obesity-hypoventilation syndrome Pickwickian syndrome respiratory insufficiency sleep apnea syndrome
| Introduction |
|---|
|
|
|---|
| Materials and Methods |
|---|
|
|
|---|
Patients
The inclusion criteria were as follows: (1) patients with OSAS needing home treatment with nasal CPAP, in whom FEV1,VC, and arterial blood gas values were performed before starting CPAP therapy; (2) patient data collected in the ANTADIR observatory from January 1, 1985, until January 1, 2000; (3) AHI
10/h8; (4) age
18 years; (5) absence of restrictive respiratory disease, other than that related to obesity, namely pulmonary fibrosis, sequels of pulmonary tuberculosis, chest wall diseases, and neuromuscular disorders; (6) FEV1/VC ratio
70%; therefore, patients with an associated COPD defined by a FEV1/VC ratio < 70% were excluded9; and (7) FEV1
80% of normal European values predicted; therefore, we excluded patients with severe restriction which could account in itself for hypercapnia.
Statistical Analysis
Daytime hypercapnia was defined by PaCO2
45 mm Hg. This cut-off was chosen because daytime chronic alveolar hypoventilation is usually defined by a PaCO2
45 mm Hg.610 The anthropometric, polysomnographic, and functional characteristics of the patients with daytime hypercapnia (PaCO2
45 mm Hg) and of those with PaCO2 < 45 mm Hg were compared using the
2 test and the Student t test. The correlations between PaCO2 and anthropometric, polysomnographic, and functional data were studied by linear correlation and displayed graphically by dividing BMI, PaO2, VC, and FEV1 values into classes of 5 mm Hg, 10 mm Hg, 10% predicted, and 5% predicted, respectively. Multivariate analysis was performed using multiple linear regression and logistic regression, in order to find the best association of predictive factors for hypercapnia. Significance was recognized at p < 0.05.
| Results |
|---|
|
|
|---|
10 events/h, and free of a restrictive respiratory disease other than that related to obesity. Age, sex ratio, BMI, and AHI did not significantly differ between the patients with OSAS in whom pulmonary function tests and arterial blood gas analyses were available, and those in whom they were not. After the exclusion of 614 patients with FEV1/VC ratio < 70% and the exclusion of 462 patients with FEV1 < 80% predicted, the study population included 1,141 patients (943 men and 198 women). The demographic, functional, polysomnographic, and anthropometric data for the study population according to sex are shown on Table 1
.
|
45 mm Hg) was 11% (126 of 1,141 patients) in the study population. The prevalence of daytime hypercapnia was 10.3% in men and 14.6% in women (not significant [NS] difference). The prevalence of daytime hypercapnia according to BMI is shown on Table 2
. The prevalence of daytime hypercapnia was 7.2% (27 of 377 patients) in the nonobese (BMI < 30), 9.8% (58 of 590 patients) in moderate or severe obesity (BMI 30 to 40), and 23.6% (41 of 174 patients) with massive obesity (BMI > 40). The prevalence of daytime hypercapnia associated with daytime hypoxemia (PaO2 < 70 mm Hg) was 5.6% (63 of 1,141 patients).
|
|
|
|
| Discussion |
|---|
|
|
|---|
The presence of associated COPD is a major element of confusion in the study of daytime alveolar hypoventilation in OSAS. The prevalence of daytime hypercapnia in OSAS is higher when associated with COPD. In the study of Chaouat et al,4 including 265 patients with OSAS, the prevalence of daytime hypercapnia was 27% in the 30 patients with an associated COPD and only 8% in the 235 patients without COPD. Similar results were reported by Resta et al.3 In studies12 that did not exclude patients with COPD, apneic and hypercapnic patients had a lower FEV1 and FEV1/VC ratio than apneic and normocapnic patients. More often than not, this denotes a moderate obstructive ventilatory abnormality. In multivariate analyses, PaCO2 is inversely correlated in a significant fashion with FEV1 and the FEV1/VC ratio, independent of BMI.12 In our study, we excluded patients with an FEV1/VC ratio < 70%, and the mean value of the FEV1/VC ratio was 80% in hypercapnic and normocapnic patients. Hypercapnia observed in 11% of the patients with OSAS included in our study cannot be blamed on an obstructive ventilatory disorder, even when moderate in nature.
In our study, the AHI did not significantly differ between the hypercapnic and the normocapnic patients. Studies25111213 published on hypercapnia in OSAS have shown no relation between the severity of the OSAS and the presence of daytime hypercapnia: the apnea index, the AHI, and the duration of apneas do not differ between hypercapnic and normocapnic patients. Nighttime hypoxemia in apneic and hypercapnic patients is usually more serious than in apneic and normocapnic patients since they have a lower daytime PaO2 and therefore a lower preapneic arterial oxygen saturation.
In this study, we have shown that the patients with OSAS associated with daytime hypercapnia have a significantly higher BMI than the patients with an OSAS free of associated daytime hypercapnia. Similar results have been reported by various authors.2514 Furthermore, our study has formally demonstrated, by stepwise logistic regression, that the increase in BMI is an independent predictive factor of daytime hypercapnia in a very large group of patients in various degrees of obesity. This relationship between hypercapnia and obesity was observed, although the exclusion of patients with FEV1 < 80% predicted presumably excluded patients with severe restriction due to massive obesity. Some studies have found no link between the importance of the obesity and daytime hypercapnia in OSAS, but there are methodologic biases in these studies: either a predominant number of patients were moderately obese1 or a too small group of patients was included in the study.1112
In our study, the stepwise logistic regression showed that the decrease in VC is an independent predictive factor of daytime hypercapnia. The total lung capacity, functional residual capacity, and expiratory reserve volume were not systematically measured in our study, but the decrease in VC observed in the hypercapnic patients is very likely the consequence of more severe obesity, since the patients had neither chronic airflow obstruction nor any restrictive pulmonary problem other than obesity. Various studies51415 have shown that the daytime hypercapnia of OSAS is associated with a more serious restrictive pattern associated with more severe obesity.
One could even make the assumption that daytime hypercapnia in patients with OSAS is not secondary to sleep apneas, but is directly related to the consequences of obesity on lung function. Obesity is associated with a restrictive ventilatory deficit and a decrease in forced expiratory flow rates at low lung volumes.16171819 In fact, a certain number of arguments plead in favor of a direct link between sleep apneas and daytime hypercapnia. Most patients with an "obesity hypoventilation syndrome" have associated OSAS, in 88% of cases (23 of 26 patients) in the series by Kessler et al.14 Conversely, in our study, daytime hypercapnia was observed in 7.2% (27 of 377 patients) with OSAS and without obesity (BMI < 30), whereas these patients were free of COPD and restrictive pulmonary abnormalities, suggesting a relationship between sleep apneas and daytime hypercapnia. The major argument in favor of a cause and effect between sleep apneas and daytime hypercapnia is the fact that the disappearance of sleep apneas obtained with nasal CPAP can lead to a regression in daytime hypercapnia, even in the absence of weight loss or improved ventilatory function.720
Various hypotheses have been suggested to explain the relation between obstructive apneas and daytime hypercapnia, and more particularly, a dysfunction of the respiratory centers induced by the apneas.821 It has been demonstrated that the ventilatory responses to hypoxia and hypercapnia are lower in apneic and hypercapnic patients than in apneic and normocapnic patients, matched for BMI, ventilatory function, and age, and that they increase after suppression of sleep apneas with CPAP, along with a regression in daytime hypercapnia.20 These data therefore suggest that the daytime alveolar hypoventilation of the OSAS can result from an alteration of the ventilatory drive, which could be secondary to repeated episodes of nighttime hypoxemia and hypercapnia and sleep fragmentation. Nevertheless, some authors2223 have not observed a decreased ventilatory response to hypercapnia in patients with OSAS, and others7 have reported that the suppression of sleep apneas can bring about a normalization in daytime PaCO2, in absence of any variation in ventilatory response to hypercapnia. Other hypotheses have been proposed to explain daytime hypercapnia of OSAS: diaphragmatic fatigue secondary to repeated inspiratory efforts associated with obstructive apneas,10 decrease in the postapneic ventilatory response,824 marked decrease in the upper airway size,25 and leptin resistance.2627
| Conclusion |
|---|
|
|
|---|
| Footnotes |
|---|
Received for publication November 26, 2003. Accepted for publication July 15, 2004.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
N. Kawata, K. Tatsumi, J. Terada, Y. Tada, N. Tanabe, Y. Takiguchi, and T. Kuriyama Daytime Hypercapnia in Obstructive Sleep Apnea Syndrome Chest, December 1, 2007; 132(6): 1832 - 1838. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. Mokhlesi and A. Tulaimat Recent Advances in Obesity Hypoventilation Syndrome Chest, October 1, 2007; 132(4): 1322 - 1336. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. G. Norman, R. M. Goldring, J. M. Clain, B. W. Oppenheimer, A. N. Charney, D. M. Rapoport, and K. I. Berger Transition from acute to chronic hypercapnia in patients with periodic breathing: predictions from a computer model J Appl Physiol, May 1, 2006; 100(5): 1733 - 1741. [Abstract] [Full Text] [PDF] |
||||
Read all eLetters
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |