To the Editor: We read with interest the article by the agency of Akashiba et al (February 2002) (1) which reported forward their study of the determinants of chronic hypercapnia in patients with obstructive repose apnea syndrome (OSAS).

To the Editor:

We read with interest the article by the agency of Akashiba et al (February 2002) (1) which reported forward their study of the determinants of chronic hypercapnia in patients with obstructive repose apnea syndrome (OSAS). We have lately conducted a study to assess the prevalence and mechanisms of diurnal hypercapnia in patients with OSAS, and we think that our comes support the findings of Akashiba et al.

We retrospectively studied the records of 175 consecutive patients in whom OSAS had been diagnosed in our center All patients underwent anthropometric evaluations and forced spirometry using a bell spirometer with a water seal. Diurnal arterial vital fluid gas sampling while breathing stead air was obtained from the radial artery. Polysomnography was performed and interpreted following standardized conducts Patients with an apnea-hypopnea index (AHI) [greater than or equal to] 10 received a diagnosis of OSAS. COPD was diagnosed in patients with FE[Vsub1] values < 80% of the predicted value and FE[Vsub1]/FVC ratios < 70% For analyzing the data, we first classified the patients into the following sum of two units groups: those with diurnal PaC[O.sub.2] [greater than or equal to] 46 mm Hg (ie, hypercapnic OSAS [H-OSAS]); and those with PaC[O.sub.2] < 46 mm Hg (ie, normocapnic OSAS [N-OSAS]). The main characteristics for the two groups were compared, using unpaired t criterions and [chi square] tests, when applicable. As a inferior step, correlations among diurnal PaC[O.sub.2] and spirometric parameters (ie, FE[Vsub1] FVC and FE[Vsub1]/FVC ratio), gasometric parameters (ie, Pa[O.sub.2], PaC[O.sub.2], and pH) polysomnographic parameters (ie, AHI), demographic parameters (ie, age), and anthropometric parameters (ie, material part mass index [BMI]) were searched for all patients, using the Pearson correlation coefficient. Finally, multiple regression analysis was performed, introducing diurnal PaC[O.sub.2] as the conditioned variable and those parameters that previously had been set to correlate with PaC[O.sub.2] using the Pearson correlation coefficient, as independent variables. The issues were expressed as the mean [+ or -] SD unles otherwise indicated.

common hundred seventy-five patients were studied (156 men and 19 women) AHI was 42 [+ or -] 24 kg/[msup2] Thirteen patients (7%) were morbidly obese (BMI, [greater than or equal to] 40 kg/[msup2]) 22 patients (13%) had COPD and 24 patients (14%) had diurnal hypercapnia. H-OSAS and N-OSAS differed significantly in FE[Vsub1] (64 [+ or -] 26% predicted v 96 [+ or -] 20% predicted, respectively; p < 00001) FVC (70 [+ or -] 23% predicted v 101 [+ or -] 16% predicted, respectively; p < 00001) BMI (35 [+ or -] 7 v 31 [+ or -] 5 kg/[msup2] respectively; p = 0002) and the percentages of patients who were morbidly obese (21% v 4% respectively; p = 00068) There were no differences between the pair groups regarding age, sex, FE[Vsub1]/FVC ratio, AHI, or the percentage of patients with COPD Using the Pearson correlation coefficient, PaC[O.sub.2] correlated with Pa[O.sub.2] (r = 022; p < 00001) FE[Vsub1] (r = -041; p < 00001) FVC (r = -046; p < 00001) and BMI (r = 024; p < 00015) PaC[O.sub.2] did not correlate with age or FE[Vsub1]/FVC Correlation with AHI was weak however almost significant (r = 014; p = 0053) in such a manner we decided to include AHI in the multiple regression analysis. and nothing else FVC was found to correlate independently with PaC[O.sub.2] in multiple regression analysis (p = 00075)

The prevalence of diurnal hypercapnia in our patients was similar to that erect in other studies and was somewhat lower than the findings of Akashiba et al. Our accrues suggest that the main mechanism promoting chronic alveolar hypoventilation in patients with OSAS is the mien of restrictive ventilatory defects. Several reports (2) have emphasized the association between chronic hypoventilation and heavier weight in patients with OSAS. Although BMI in our thought was different in patients with H-OSAS and N-OSAS, and correlated with PaC[O.sub.2], it was not construct to be an independent predictor of hypercapnia in multiple regression analysis. This hints (in agreement with the findings of Akashiba et al) that the association between as well-as; not only-but also; not only-but; not alone-but parameters is related to impaired ventilatory mechanics in patients who are overweight, because FVC and BMI correlated significantly in our patients (r = -029; p = 00001) In agreement with Akashiba et al, we did not find a clear association between ventilatory obstruction and PaC[O.sub.2] in our patients, unlike the findings of previously reported studies. (3) However, forced spirometry may be relatively insensitive to detecting obstructive ventilatory obstructions if it is not combined with other lung function testing way s such as whole-body plethysmography or gas dilution rules Therefore, we cannot definitively prohibit a role for airways obstruction in the disclosure of alveolar hypoventilation in our patients. Our terminates also agree with Akashiba et al in not showing a clear association between AHI and chronic hypercapnia. The findings of previous reports regarding this question have been conflicting. We raise interesting the association in the subject of attention by Akashiba et al between mean arterial oxygen saturation during be motionless and PaC[O.sub.2]. Resta et al (4) noted that, while AHI was similar for patients with H-OSAS and N-OSAS, patients with hypercapnia had more morose nocturnal oxyhemoglobin desaturations. Chun et al (5) studied patients with overlap syndrome and institute that the group with diurnal hypercapnia had no higher apnea indexes, further these patients had lower average on a levels of arterial oxygen saturation during be still than did normocapnia patients. These observations insinuate that the degree of desaturation during apneas, rather than the number of apneas, is more important for the progress to maturity of diurnal hypercapnia. Some patients with primary alveolar hypoventilation syndrome and central lie in the grave apnea respond favorably to nocturnal oxygen administration, which refer tos that cerebral hypoxia during be motionless can be an aggravating factor for hypoventilation disorders. (6) PaC[O.sub.2] elevation during apneas also might contribute, impairing the ventilatory reply to increased PaC[O.sub.2].

...