Hypercapnia in obstructive sleep apnoea syndrome

Background: The reports on the prevalence of hypercapnia in Obstructive Sleep Apnoea Syndrome (OSAS) are conflicting. We studied the prevalence of hypercapnia in a population of OSAS patients referred to a Department of Respiratory Medicine and the mechanism of the respiratory failure in OSAS associated with Obesity Hypoventilation Syndrome (OHS) or with Chronic Obstructive Pulmonary Disease (COPD) (Over-lap syndrome). Methods: We studied 219 consecutive OSAS patients during a period of 3 years. We recorded age and anthropomorphic data and performed polysomnography and pulmonary function tests. In relation to the value of PaCO(2), the patients were divided in hypercapnic (PaCO(2) > 45 mmHg) patients and normocapnic patients. They were also divided into three groups in relation to the presence of "simple" or "pure" OSAS, to the presence of OSAS associated with COPD, to the presence of OSAS associated with OHS. Results: Seventeen per cent of the patients were hypercapnic. They were significantly heavier, had more severe lung function test abnormalities and more severe nocturnal oxyhemoglobin desaturations than the normocapnic ones, while Forced Expiratory Volume in one second as a percentage of Forced Vital Capacity (FEV1/FVC %) and Apnoea/Hypopnoea Index (AHI) were similar. OHS patients (13%) were significantly younger and heavier, had lower PaO(2) and higher PaCO(2) than "simple" OSAS patients (77%) and Overlap patients (10%) and had more severe restrictive defect. There was no difference in terms of AHI among the three groups, but nocturnal oxyhemoglobin desaturations were more severe in OHS group. In OHS group hypercapnia was correlated to FVC% of predicted and FEV1% of predicted and to the mean nocturnal oxyhemoglobin saturation; in Overlap patients PaCO(2) was correlated to Forced Expiratory Flow rate at low Vital Capacity. Conclusion: Seventeen per cent of OSAS patients referred to a Department of Respiratory Medicine were hypercapnic. Hypercapnia in OHS patients correlates to the restrictive ventilatory defect whereas in Overlap patients it seems to correlate to peripheral airways obstruction. The distinction between patients with "simple" or "pure" OSAS and patients affected by OSAS associated with OHS or COPD could be important not only for clinical and prognostic implications, but also for the consequences in terms of ventilatory treatment. (C) 2000 Elsevier Science B.V. All rights reserved.