Between 1968 and 1988, 96 consecutive patients with acute massive pulmonary embolism underwent pulmonary embolectomy under cardiopulmonary bypass. The operative mortality rate was 37.5%. We analyzed 12 clinical and hemodynamic variables by univariate and multivariate analyses to assess the predictive factors of postoperative outcome. Multivariate analysis disclosed that cardiac arrest and associated cardiopulmonary disease were independent predictors of operative death. Long-term follow-up (range, 2 to 144 months; mean, 56 months) information was available for 55 of the 60 discharged patients: 6 had died, and 5 complained of persistent mild or severe exertional dyspnea (New York Heart Association class II). These results help assess the preoperative risk in patients undergoing pulmonary embolectomy. They also show that, in the few patients who do not benefit from optimal medical therapy, pulmonary embolectomy remains an acceptable procedure in view of the long-term results.