Aims: To assess the association between past level of physical activity (PA) and risk for death during the acute phase of myocardial infarction (MI) in a pooled analysis of cohort studies. Methods and results: European cohorts including participants with a baseline assessment of PA, conventional cardiovascular (CV) risk factors, and available follow-up on MI and death were eligible. Patients with an incident MI were included. Leisure-time PA was grouped as sedentary (<7 MET-hours), low (7-16 MET-hours), moderate (16.1-32 MET-hours), or high (>32 MET-hours) based on calculated net weekly energy expenditure. The main outcome measures were instant and 28-day case fatality of MI. Pooled odds ratios (ORs) with 95% confidence intervals (CIs) were calculated using multivariate random-effects models. Adjustments for age, sex, CV risk factors, alcohol consumption, and socioeconomic status were made. From 10 cohorts including a total of 1 495 254 participants, 28 140 patients with an incident MI comprised the study population. A total of 4976 (17.7%) died within 28 days - of these 3101 (62.3%) were classified as instant fatal MI. Compared with sedentary individuals, those with a higher level of PA had lower adjusted odds of instant fatal MI: low PA [OR, 0.79 (95% CI, 0.60-1.04)], moderate PA [0.67 (0.51-0.89)], and high PA [0.55 (0.40-0.76)]. Similar results were found for 28-day fatal MI: low PA [0.85 (0.71-1.03)], moderate PA [0.64 (0.51-0.80)], and high PA [0.72 (0.51-1.00)]. A low-to-moderate degree of heterogeneity was detected in the analysis of instant fatal MI (I2 = 47.3%), but not in that of 28-day fatal MI (I2 = 0.0%). Conclusion: A moderate-to-high level of PA was associated with a lower risk of instant and 28-day death in relation to a MI.