The aim of this study was to evaluate the contribution of clinical, angiographic and haemodynamic findings in predicting the cardiorespiratory efficacy of thrombolytic therapy in acute massive pulmonary embolism. Haemodynamic measurements and pulmonary angiography were performed before (H0) and 12 h after (H12) initiating thrombolytic therapy in 23 patients with acute massive pulmonary embolism (Miller index > or =20/34), and free of prior cardiopulmonary disease. Patients were divided into two groups according to the variation in oxygen delivery (deltaDO2) between H0 and H12: deltaDO2 >20% (responders, n=10) and deltaDO2 < or =20% (nonresponders, n=13). Before thrombolysis, clinical and angiographic findings were similar in both groups. Mean right atrial pressure (RAP) and total pulmonary (vascular) resistance (TPR) were higher, while cardiac index (CI), DO2 and mixed venous oxygen saturation (Sv,O2) were lower in responders. DO2 and Sv,O2 were more closely correlated with deltaDO2 than RAP, TPR and CI. Eight out of the 10 responders and two out of the 13 nonresponders had an Sv,O2 <55%, while nine of the responders and two of the nonresponders had a DO2 <350 mL x min(-1) x m(-2). In conclusion, the initial oxygen delivery and mixed venous oxygen saturation may predict the cardiorespiratory efficacy of thrombolytic therapy in acute massive pulmonary embolism. When pulmonary angiography is performed, measurement of mixed venous oxygen saturation may be a simple method by which to select patients for thrombolytic therapy.