Background: Epidemiology studies have shown a consistently increased risk of acute myocardial infarction (AMI) correlated with particulate matter (PM) exposure. However, little is known about the association with specific AMI subtypes. In this work, we investigated the association between short-term PM exposure and emergency department visits (EDVs) for AMI, ST-elevation myocardial infarction (STEMI), and non-ST-elevation myocardial infarction (NSTEMI). Methods: We based this case-crossover study on 2749 patients from Chaoyang District hospitalized with AMI in Anzhen Hospital during 2014. Meteorological and air pollution data were collected during this period. We used a time-stratified case-crossover design with lag model, adjusted for meteorological conditions and/or other gaseous pollutants, to estimate risk of EDVs for AMI, STEMI, and NSTEMI. We conducted stratified analyses by gender, age, season, and comorbid conditions to examine potential effect modification. Results: We found that each 10 mu g/m(3) increment of PM2.5 concentration (1-day lagged) was associated with an increased risk of EDVs for STEMI (OR 1.05; 95% CI, 1.00-1.11). We found no association of PM2.5 concentration with overall AMI or NSTEMI. No effect modification was found when stratified by gender, season, or comorbid conditions, even though the effect size was larger in patients who were male, smokers, and comorbid with hypertension. Patients aged >= 65 years showed a significantly increased risk of STEMI associated with PM2.5 in the previous day than those aged <65 years. Conclusions: Our study indicated a transient effect of short-term PM2.5 exposure on EDVs for STEMI. Patients aged >= 65 years appeared to be particularly susceptible. Our findings suggest that studies of the association between PM exposure and AMI should consider AMI subtypes, lag times, and individual characteristics.