Objective To analyze the changes in the number and severity of sleep-disordered breathing (SDB) from the acute to the stable phase of pulmonary embolism (PE), as well as the factors associated with these changes.

Methods Prospective and multicenter study in consecutive normotensive patients hospitalized as a result of acute PE. In the acute phase (first 48 hours from diagnostic) recruited patients undergo an overnight sleep study and transthoracic echocardiography; and three months after hospital discharge, a new sleep study was performed. Right ventricular dysfunction (RVD) was defined as the presence of right ventricular (RV) free wall hypokinesia, RV dilatation (diameter>30mm or RV/LV ratio?0.9) or RV-right atrium (RA) gradient>30 mmHg. The factors associated with changes in the number of SDB were analyzed by multivariable logistic regression.

Results 111 patients (44.5% women) were recruited. Mean (±SD) age 63(±15) years, BMI 28.4(±4.7) kg/m2 and Epworth Sleepiness Scale 5.5(±3.3). At diagnosis, the mean oxygen saturation was 92.8% (±2.6), 69% had RVD. Apnoea-hypopnoea index (AHI) decreased from the acute to the stable phase of PE (24.4/h vs 15.7/h; p=0.013). This decrease was associated with proximal PE, RV dilatation, elevated cardiac biomarkers, rectification of the interventricular septum, and an elevated RV-RA gradient. The greatest decrease was observed in patients with RVD compared to those without RVD (AHI: 12.3 vs 0.43 events/h; p<0.001). Moreover, there was a positive correlation between RVD and AHI decrease (r=0.41)

Conclusion There is a transient increase in the number of SDB in PE patients especially in those with RVD and higher initial AHI