ST‐Segment depression on ambulatory electrocardiography in the early in‐hospital period after acute myocardial infarction predicts early and late mortality: A short‐term and a 3‐year follow‐up study

Abstract

A surveillance study was conducted to determine the in‐hospital and long‐term prognostic value of ST‐segment depression assessed by ambulatory electrocardiographic monitoring (AEM) during the early in‐hospital period after acute myocardial infarction (AMI). ST‐segment depression (STD) was determined by computer analysis of 24‐h ECG tapes as a horizontal or downsloping change in ST level by ≥ 0.1 mV from the reference base line. The ST level was measured 80 ms after the J point of all normally conducted complexes for ≥ 1 min. All computer‐detected ST events were verified by one trained reader. Tapes corresponding to 74 patients were analyzed. In addition, 23 tapes corresponding to age‐ and gendermatched controls were also analyzed. Patients were divided into two groups: 22 patients (30%) showed STD (Group A), and 52 patients (70%) had no episode of STD (Group B). Among controls, 1 person (4%) showed STD. During the early follow‐up period (14 ± 11 days after hospital admission), cardiac events occurred in 11 patients [7 (32%) in Group A and 4 (8%) in Group B, p < 0.01], including 6 cardiac deaths [5 (23%) in Group A and 1 (2%) in Group B, p < 0.01], 3 acute coronary artery bypass surgeries [2 (9%) in Group A and 1 (2%) in Group B, p = NS], and 2 nonfatal myocardial infarctions (both in Group A, p = NS). During a mean follow‐up period of 3 years (36 ± 15 months), 18 patients died [10 (45%) in Group A and 8 (15%) in Group B, p = 0.01]. Eleven deaths were sudden [7 (32%) in group A and 4 (8%) in Group B, p <0.01 ]. Eighteen AMI occurred [11 (50%) in Group A and 7 (13%) in Group B, p < 0.005]. Twenty patients underwent revascularization procedures [7 (32%) in Group A and 13 (25%) in Group B, p=NS]. Thirty‐eight patients [18 (82%) in Group A and 20 (38%) in Group B, p<0.001] suffered at least one cardiac event during the follow‐up period (death, myocardial infarctions, and revascularization therapy). Survival analysis using Kaplan‐Meier curves showed that patients with STD (Group A) had shorter survival times (p < 0.001, Log rank test) than those without STD (Group B). The same analysis showed that patients in Group A had shorter survival times free of cardiac events (myocardial infarction, p < 0.001; sudden death, p < 0.001; revascularization therapy, p < 0.05; all cardiac events, p < 0.001) than those in Group B. No coronary events were reported in control persons. A multivariate stepwise regression analysis (Cox's proportional hazards model) of a number of clinical variables, including STD, showed that its presence was the most important factor predicting mortality (p < 0.001, hazards ratio 5.09), followed by the presence of diabetes (p < 0.01, hazards ratio 4.67) and hypertension (p < 0.01, hazards ratio 3.52). The findings of this survey illustrate both the short‐ and long‐term prognostic value regarding mortality and morbidity of STD recorded on AEM during the early in‐hospital period after AMI. Copyright © 1995 Wiley Periodicals, Inc.