Principal Findings:

A total of 1,216 patients were enrolled in the registry for the time periods of interest, 842 patients from 1978-85 (Group 1) and 374 patients from 1986-88 (Group 2). In terms of baseline characteristics, there were no significant differences between groups for age (mean 59.7 for both groups), gender, or smoking status. Significant differences did occur between groups 1 and 2 for hypertension (51.8% vs. 65.2%), diabetes (19% vs. 21.9%, p<0.001), history of prior angina (29.3% vs. 22.8%), and presence of Killip class III or IV during hospitalization (13.7% vs. 6.1%, p<0.001).

During the 28 days after symptom onset, Group 2 had a higher incidence of coronary angiography (0.6% vs. 4%, p<0.001) and revascularization (0.4% vs. 2.4%, p<0.004) via either angioplasty or coronary artery bypass grafting (CABG). Mortality at 28 days was significantly improved in the later time period (group 2) from 14.6% to 8.8%, p=0.007. After adjusting for gender, age, hypertension, and diabetes in a logistic regression model, the risk of death was 35% lower in 1986-88 than 1978-85 (relative risk 0.65, 95% confidence interval 0.42-0.99). When the model went further to adjust for Killip class, the time period effect completely disappeared.

Mortality at three years (for patients surviving the first 28 days) was similar between groups (8.2% vs. 8.3%, p=NS). In a logistic regression model, three-year mortality was not affected by time period even when adjusted for age, gender, diabetes, hypertension, and Killip class.