More Bad News About Diabetes


Synopsis: The hazard of diabetes in post-MI patients may be due to a greater reocclusion rate of initially patent arteries and possibly due to an increased prevalence of congestive heart failure.

Source: Woodfiled SL, et al. J Am Coll Cardiol 1996; 28:1661-1669.

Diabetics are known to have an increased risk of death and adverse events following acute myocardial infarction. Patients with coronary disease who have diabetes have more advanced atherosclerosis, greater left ventricular dysfunction (frequently due to myocardial infarction), are often older, and more likely to be female. The GUSTO-I angiographic study sought to identify whether there are angiographic differences between diabetic and non-diabetic patients who received thrombolytic therapy for acute myocardial infarction, and whether LV function, coronary infarct artery patency, restenosis rates, and mortality were different in this large cohort of individuals. The authors were interested in evaluating possible mechanisms responsible for the excess mortality associated with diabetes. The angiographic substudy consisted of 2431 patients who underwent early (90 minutes) and late angiography (5-7 days). Of the entire cohort, 13% were diabetic, and three-fourths were on oral agents only. As expected, the diabetics were older, more often female, and presented later from pain onset to thrombolytic therapy than non-diabetics. Diabetics had a greater incidence of previous MI and a higher incidence of congestive heart failure. The angiographic findings demonstrated no difference in patency rates at 90 minutes between the two cohorts. The number of subjects with TIMI grade 2-3 flow at 90 minutes were comparable, and five- to seven-day angiographic findings were also similar between the groups. Reocclusion, although uncommon, was almost twice as likely in diabetics. Quantitative angiography documented that diabetic subjects had a smaller minimum lumen diameter of the infarct-related artery as well as smaller reference segments throughout the coronary bed. Although the rate of subsequent angioplasty in diabetics was comparable to non-diabetics, the mortality rate in diabetics who underwent angioplasty was almost three times higher (15.2% vs 4.5%); much of this increase in death was related to rescue angioplasty in the diabetic group. However, 30-day mortality in diabetics undergoing elective angioplasty was 8% compared to 2.2% in the non-diabetics. Left ventricular function was not different between the diabetics and non-diabetics, including global and regional assessment; non-infarct zone hyperkinesia was less in the diabetics. Overall, mortality was twice as high in diabetics, with an unadjusted 30-day mortality rate of 11.3% vs. 5.9% in the larger non-diabetic cohort. For patients with TIMI grade 2-3 flow at 90 minutes, diabetics still had a two-fold increase in mortality at 30 days. In subjects with TIMI-3 flow at 90 minutes, diabetics had an 8.6% one-month mortality rate vs 3.4% in non-diabetics. Insulin treatment was associated with a doubling of mortality rate within the diabetic cohort. Diabetes remained an independent predictor of 30-day mortality after adjustment for angiographic as well as clinical variables.

In this study, as in many previous reports, the diabetic cohort had more extensive and more severe coronary atherosclerosis. Fewer than half had single-vessel disease vs. 60% of the non-diabetic patients, and, as mentioned, the "normal" reference segments in the diabetics were smaller, suggesting more extensive epicardial disease. The blunting of the hyperkinetic response in the non-infarct zone in the diabetics was attributed to a greater likelihood of prior myocardial infarction as well as more multivessel disease. Congestive heart failure was believed to be related to diastolic dysfunction, as systolic performance was not different between the two groups. No clear cut explanation was identified for the more than three-fold increase in mortality in the diabetic patients who underwent elective angioplasty within one week after thrombolysis. The authors conclude that the hazard of diabetes may be in part due to greater reocclusion rate of initially patent arteries and possibly due to an increased prevalence of congestive heart failure, but, basically, the poor outcomes remain inadequately characterized.


The GUSTO angiographic study comparing diabetic to non-diabetic individuals provides additional information to the vast database documenting a substantially increased morbidity and mortality in diabetic patients with coronary disease. Of interest, differences in left ventricular function do not explain the adverse outcome in the diabetics. More severe coronary atherosclerosis is a possible explanation, and the decreased hyperkinetic response suggests that left ventricular functional reserve (as well as coronary flow reserve) may be impaired in diabetics. An increased risk of death after angioplasty in the diabetic cohort, both in the rescue angioplasty group as well as in the elective group, is concordant with other recent reports that angioplasty carries a substantially higher risk in the diabetic patient. The authors’ conclusion that angioplasty in diabetics following myocardial infarction should only be done for documented ischemia seems reasonable, but this recommendation also is appropriate for non-diabetics. There is little good news for the clinician in this report other than to further define the high-risk state of diabetic patients. Aggressive prevention strategies would appear to be a critical component to care of the diabetic subject. Lipid lowering, control of hypertension (more common in these diabetic patients), smoking cessation, and hormone replacement are all strategies that are relatively low tech, yet may have substantial payoff by decreasing the likelihood of an acute coronary event, which is so much more lethal in the diabetic. (For those interested in this subject, the October 1996 issue of Coronary Artery Disease [Vol. 7, No. 10] has seven review articles on diabetes and the cardiovascular system.)