Background: From a registry of 249 confirmed pulmonary embolism (PE) patients submitted to thrombolytic therapy (TT), we analysed predictors of in-hospital course and long-term mortality.
Methods and results: The combined clinical end point of in-hospital course associated death, recurrent PE, repeat thrombolysis, surgical embolectomy or bleeding complications. The long-term follow-up included analysis of survival, and occurrence of PE-related events, defined as recurrent deep vein thrombosis, recurrent PE, occurrence of congestive heart failure or change of New York Heart Association functional class to class III or IV in patients who survived the acute phase.In-hospital clinical course was uneventful in 165 (66.3%) patients. Initial right ventricular (RV) dysfunction was reversible in 80% within 48 h following TT. Initial pulmonary vascular obstruction >70% (RR=5.3 [2.1; 13.6]); haemodynamic instability at presentation (RR=2.6 [1.1; 6]); persistence of septal paradoxical motion after TT (RR=5.9 [1.4; 25.9]); and insertion of intracaval filter (RR=3.7 [1.4; 9.4]) were independent predictors of poor in-hospital course. Mean follow-up was 5.3+/-2.6 years. Of the 227 patients alive after the hospital stay, the probability of survival was 92% at 1 year, 79% at 3 years and 56% at 10 years. Multivariate predictors of long-term mortality were age >75 years (RR=2.73 [2.18; 3.21]; P=0.0002), persistence of vascular pulmonary obstruction >30% after thrombolytic treatment (RR=2.22 [1.69; 2.74]; P=0.003), and cancer (RR=2.03 [1.40; 2.65]; P=0.04).
Conclusion: The recovery of RV function should be considered as a marker of thrombolysis efficacy, while residual pulmonary vascular obstruction and cancer are independent predictors of long-term mortality. These results advocate the identification of high-risk patients by means of systematic lung-scan and echocardiography pre- and post-thrombolysis, and raise the question of the need for thromboendarterectomy in patients with residual pulmonary vascular obstruction.