Impact on Survival of a Decade of Change in the Management of Patients Who Have Sustained a Myocardial Infarction

Publisher Website
Google Scholar
Abstract
The survival of 906 consecutive patients who had sustained an acute myocardial infarction was monitored between the beginning of week 2 and the end of week 52. Deaths which occurred during this period were successfully predicted in terms of just 6 features; namely the value of left ventricular systolic time intervals measured during the 1st week, the occurrence of a myocardial infarction prior to the current hospitalisation, the patient's age, the coexistence of diabetes mellitus, the presence of left or right bundle branch blocks, and the administration of a diuretic agent while the patient was in hospital. By means of stepwise linear discriminant analysis predictions of mortality among the first 302 patients based upon these 6 features were 80% specific and 63% sensitive. Prognostic features derived from the first 302 patients were then used to predict mortality among the remaining 604 patients. Observed and predicted mortalities were similar for both sets of patients. Although mortality was similar, the 604 latter patients had received a considerably more interventional style of management than patients in the former group. Possible reasons for this failure of a change of management to improve prognosis are discussed.

This publication has 0 references indexed in Scilit: