INFLUENCE OF DIABETES ON MORTALITY IN ACUTE MYOCARDIAL-INFARCTION - DATA FROM THE GISSI-2 STUDY

Objectives. This study was conducted to determine the role of insulin dependent and noninsulin-dependent diabetes in the prognosis of patien ts after myocardial infarction and treatment with fibrinolytic agents. Background. Several studies have shown that diabetic patients have a high mortality rate after acute myocardial infarction. However, the im pact of diabetes on survival in patients treated with fibrinolytic age nts is still undefined. It is also not known whether the type of diabe tes or gender affects prognosis. Methods. We analyzed prevalence and p rognostic significance of a history of diabetes in patients enrolled i n the GISSI-2, study, all of whom received fibrinolytic agents. The in cidence of deaths in the hospital and at 6 months after study entry wa s computed for patients without diabetes and for insulin-dependent and noninsulin-dependent diabetic patients; relative risks were evaluated by univariate and multivariate analysis. Results. Information on diab etic status was available for 11,667 patients, 94.2% of those randomiz ed in the GISSI-2 study. The prevalence of diabetes was higher in wome n than in men (8.75% vs. 1.85%, p<0.01 for insulin-dependent and 23.7% vs. 13.8%, p<0.01 for noninsulin-dependent diabetic patients), The ty pe of fibrinolytic agent did not affect mortality rates; the increase in in-hospital mortality of diabetic patients was moderate and similar for men with insulin- and noninsulin-dependent diabetes (8.7% and 10. 1%, respectively, vs. 5.8% in nondiabetic patients); in women, mortali ty was markedly higher for insulin-dependent and only slightly higher for noninsulin-dependent diabetic patients (24.0% and 15.8%, respectiv ely, vs. 13.9% for nondiabetic patients). The adjusted relative risks were 1.9 (95% confidence interval 1.2 to 2.9) for insulin dependent di abetic women and 1.4 (95% confidence interval 1.1 to 1.8) for noninsul in-dependent diabetic men. The mortality rate after discharge showed a similar gender difference, and in insulin-dependent diabetic women, p rognosis was ominous even in the absence of left ventricular damage be fore discharge. Conclusions. A history of diabetes is associated with a worse prognosis after myocardial infarction, even in patients treate d with fibrinolytic agents. Gender and type of diabetes appear to be c ritical in affecting survival. In men, both insulin-dependent and noni nsulin dependent diabetes are associated with a moderately higher mort ality rate; in women, insulin-dependent diabetes is, in itself, a stro ng risk factor for death after myocardial infarction.