Obstructive sleep apnea (OSA) is a risk factor for a first episode of pulmonary embolism (PE), although its impact on the risk of thromboembolism recurrence is uncertain. Our objective was to explore the prognostic value of OSA after discontinuing oral anticoagulation (OAC) in patients with a first episode of PE.
In 120 consecutive patients who had stopped OAC for a first episode of PE, we performed a home respiratory polygraphy and recorded sleep characteristics, classical risk factors for PE, blood pressure, spirometric parameters, physical activity and levels of D-dimer and prothrombin fragment 1+2. Patients were followed for 5 to 8 years, and the main end-point was PE recurrence. Restarting OAC for any thromboembolic event was evaluated as a secondary end-point.
During the follow-up period, 19 patients had a PE recurrence, and 16 of them had an apnea-hypopnea index (AHI) ≥10 h-1. In a multivariate Cox regression model, an AHI ≥10 (hazard ratio [HR], 20.73; 95%CI, 1.71-251.28), mean nocturnal oxygen saturation (SaO2) (HR, 0.39; 95%CI, 0.20-0.78), time with SaO2<90% (HR, 0.90; 95%CI, 0.82-0.98) and D-dimer level (HR 1.001; 95%CI, 1.00-1.002) were identified as independent risk factors for recurrent PE. 24 patients resumed OAC, and AHI ≥10 (HR, 20.66; 95%CI, 2.27-188.35), mean nocturnal SaO2 (HR, 0.54; 95%CI, 0.32-0.94) and Epworth sleepiness score (HR, 0.73; 95%CI, 0.56-0.97) were retained as independent risk factors to resume OAC.
After a first episode of PE, OSA is an independent risk factor for PE recurrence or restarting OAC for a new thromboembolic event.