In 60 patients with acute myocardial infarction (pain less-than-or-equal-to 4 h), we examined the value of ST segment monitoring in predicting early reperfusion, resulting left ventricular damage, and complications during hospitalization. Two criteria were determined by observation of the ST segment elevation during the first 4 h following initiation of thrombolysis. Early reperfusion was assessed by an early increase of the creatine phosphokinase (CK) with measurements taken in 15-min intervals. Cardiac catheterization was performed on days 11 +/- 5. According to the CK measurements, a reduction of the ST elevation greater-than-or-equal-to 50 % within 1 h of serial ECG follow-up (ST criterion A) was the best indicator of early reperfusion (sensitivity 84 %, specificity 80 % positive predictive value 93 %, negative predictive value 67 %). Simple comparison of the ST segment in the initial ECG and an ECG recorded 3 h later (ST criterion B) was less accurate according to the detection of early reperfusion (sensitivity 68 %, specitivity 93 %, positive predictive value 97 %, negative predictive value 50 %). However, contrary to ST criterion A, criterion B was useful in predicting subsequent left ventricular damage. Patients with a resolution of the initial ST elevation greater-than-or-equal-to 70 %/3 h showed smaller regional wall motion abnormalities (dyssynergic area 21.3 +/- 20.3 vs 33.8 +/- 18.4, p < 0.01) and a better left ventricular ejection fraction (57.7 +/- 11.6 vs 50.2 +/- 12.6, p < 0.05). Patients with early reduction of the ST elevation following either criterion experienced fewer critical events (reinfarction, reischemia, death). In conclusion, the investigated criteria are useful in assessing reperfusion of the infarcted artery following thrombolysis. Simple comparison of the initial ST elevation and the ST elevation after 3 h gives acute information according to patient outcome. This ST criterion could be useful in selecting candidates who may profit from an early, more aggressive therapeutical approach.