Purpose: We performed nocturnal polysomnography in patients with pulmonary hypertension (PH) of varying etiologies to determine the association of metrics describing sleep-disordered breathing (SDB) with measures of PH severity.
Methods: Consecutive patients referred for evaluation of dyspnea on exertion and elevated pulmonary arterial pressure >30 mmHg on echocardiography, who underwent right and left heart catheterization and polysomnography, were included. Patients were not pre-selected for symptoms of sleep-disordered breathing.
Results: Twenty-eight patients including 22 females and six males with a mean age of 55.2 ± 11.9 years were evaluated. Etiologies of PH were idiopathic (32%) and PH associated with other diseases (68%). Most were World Health Organization (WHO) Functional class II (39%) and III (39%). The group mean pulmonary arterial pressure (mPAP) was 40.9 ± 15.1 mmHg. Diurnal resting and exercise arterial oxygen saturations (SaO(2)) were 94.9 ± 3.7% and 88.3 ± 8.9%. The mean apnea-hypopnea index (AHI) was 11.4 ± 19.8/h; 50% of all patients had an AHI ≥ 5/h; 30.6 ± 36.0% of total sleep time was spent with SaO(2) < 90% (T90%); 66% of subjects with an AHI ≥ 5/h of sleep reported snoring, and 60% noted daytime somnolence; however, only 29% had an Epworth Sleepiness Scale ≥10. Right atrial pressure and mPAP were significantly correlated with AHI and T90%. The best predictive model relating PH severity to metrics of SDB was a highly significant association (p = 0.005) between mPAP and a linear combination of AHI and T90%.
Conclusions: SDB comprised of obstructive apneas, hypopneas, and nocturnal hypoxemia is prevalent in PH and cannot be accurately predicted by sleep apnea signs and symptoms or diurnal rest and exercise SaO(2). The association of AHI and T90% with mPAP suggests a potential relationship between the pathophysiology of sleep-disordered breathing and PH.