BACKGROUND The aim of this study was to assess the significance of ST-segment depressions (ST-SD) detected during exercise test or Holter monitoring and to determine which parameters of ST-SD are the most important prognostic factors in patients after myocardial infarction. METHODS The study group consisted of 164 patients (126 men and 38 women) who survived their first uncomplicated myocardial infarction. Twenty-four-hour Holter monitoring on day 10 (+/- 2) after infarction and a treadmill exercise test 1 or 2 days later were performed. The following parameters of ST-SD were taken into consideration: amplitude, localization according to the area of infarction and presence or absence of concomitant angina. Patients were observed for 24 months to assess the occurrence of new cardiac events. RESULTS In 78 patients (group I) ST-SD were detected in both Holter monitoring and the exercise test, and in 32 patients (group II) in the exercise test only. Fifty-four patients (group III) were without ST-SD. During follow-up there were 83 cardiac events in group I, 24 in group II and 16 in group III (P < 0.01, group I versus II; P < 0.0001, group I versus III; P < 0.05, group II versus III). In multivariate analysis the presence of ST-SD during Holter monitoring or the exercise test, or both, appeared to be of most prognostic significance (P < 0.0001). The number of new cardiac events was significantly higher in patients with painful ST-SD greater than 3 mm, detected outside the area of infarction (distant ischaemia). CONCLUSIONS This study shows a significant prognostic value of early post-infarction ischaemia detected by Holter monitoring and an exercise test. Distant, painful ST-SD greater than 3 mm were more powerful determinants of poor prognosis than others. Electrode placement during Holter monitoring appears to be very important, particularly in post-infarction patients.