Sleep-related breathing disorders in patients with pulmonary hypertension

Abstract

Background: Cheyne-Stokes respiration (CSR) and central sleep apnea (CSA) are common in patients with left-heart failure. We investigated the hypothesis that sleep-disordered breathing is also prevalent in patients with right ventricular dysfunction due to pulmonary hypertension (PH).

Methods: We studied 38 outpatients (median age, 61 years; quartiles, 51 to 72) with pulmonary arterial hypertension (n = 23) or chronic thromboembolic PH (n = 15). New York Heart Association (NYHA) class was II to IV, and median 6-min walk distance was 481 m (quartiles, 429 to 550). In-laboratory polysomnography (n = 22) and ambulatory cardiorespiratory sleep studies (n = 38) including pulse oximetry were performed. Quality of life and sleepiness by the Epworth sleepiness score were assessed.

Results: The median apnea/hypopnea index was 8 events/h (quartiles, 4 to 19), with 8 central events (quartiles, 4 to 17), and 0 obstructive events (quartiles, 0 to 0.3) per hour. Seventeen patients (45%) had > or = 10 apnea/hypopnea events/h. Comparison of 13 patients with > or = 10 CSR/CSA events/h with 21 patients with < 10 CSR/CSA events/h (excluding 4 patients with > or = 10 obstructive events/h from this analysis) revealed no difference in regard to hemodynamics, NYHA class, and Epworth sleepiness scores. However, patients with > or = 10 CSR/CSA events/h had a reduced quality of life in the physical domains. Ambulatory cardiorespiratory sleep studies accurately predicted > or = 10 apnea/hypopnea events/h during polysomnography in patients who underwent both studies (area under the receiver operating characteristic curve, 0.93; SE +/- 0.06; p = 0.002). The corresponding value for pulse oximetry was 0.63 +/- 0.14 (p = not significant).

Conclusions: In patients with PH, CSR/CSA is common, but obstructive sleep apnea also occurs. Sleep-related breathing disorders are not associated with excessive sleepiness but affect quality of life. They should be evaluated by polysomnography or cardiorespiratory sleep studies because pulse oximetry may fail to detect significant sleep apnea.