Current issues of ACP Journal Club are published in Annals of Internal Medicine


Intensive insulin regimen reduced long-term mortality after MI in diabetes mellitus

ACP J Club. 1997 Nov-Dec;127:60. doi:10.7326/ACPJC-1997-127-3-060

Source Citation

Malmberg K, for the DIGAMI (Diabetes Mellitus, Insulin Glucose Infusion in Acute Myocardial Infarction) Study Group. Prospective randomised study of intensive insulin treatment on long term survival after acute myocardial infarction in patients with diabetes mellitus. BMJ. 1997 May 24;314:1512-5.



To determine the long-term effect of intensive insulin treatment initiated at the time of an acute myocardial infarction (MI) in patients with diabetes mellitus.


Randomized controlled trial with mean follow-up of 3.4 years.


Coronary care units (CCUs) of 19 hospitals in Sweden.


620 patients (mean age 68 y, 63% men) who had had an acute MI in the previous 24 hours and had a blood glucose concentration > 11 mmol/L. Exclusion criteria were poor health, residence outside the hospital catchment area, enrollment in other studies, or previous participation in the Diabetes Mellitus, Insulin Glucose Infusion in Acute Myocardial Infarction (DIGAMI) study. Follow-up was complete.


Patients were allocated to intensive insulin treatment (n = 306) or to standard treatment (n = 314). In addition to standard treatment in the CCU, patients in the intensive insulin group received an insulin-glucose infusion for 24 hours followed by subcutaneous insulin, 4 times daily for ≥ 3 months. Patients in the standard treatment group did not receive insulin unless it was clinically indicated.

Main outcome measure

All-cause mortality.

Main results

At 1 year, fewer deaths had occurred in the intensive insulin group than in the standard treatment group (P = 0.027) (Table). During the follow-up period, the difference in mortality between groups persisted (P = 0.011) (Table). The greatest reduction in mortality was seen in low-risk patients who had not been receiving insulin; the absolute reduction between groups was 15% (P = 0.004).


Insulin-glucose infusion followed by at least 3 months of multiple-dose insulin reduced long-term mortality in patients with diabetes mellitus who had had an acute myocardial infarction.

Sources of funding: Swedish Heart-Lung Foundation and Hoechst Marion Roussel Sweden.

For article reprint: Dr. K. Malmberg, Department of Cardiology, Karolinska Hospital, S-171 76 Stockholm, Sweden. FAX 46-8-344964.

Table. Intensive insulin treatment vs standard treatment*

Outcomes Intensive insulin EER Standard treatment CER RRR (95% CI) ARR |EER-CER| NNT (CI)
Death at 1 y 19% 26% 27% (2.5 to 46) 7% 15 (7 to 172)
Death at mean 3.4 y 33% 44% 24% (7.3 to 38) 11% 10 (6 to 34)

*Abbreviations defined in Glossary; RRR, ARR, NNT, and CI calculated from data in article.


The role of glycemic control in reducing the risk for large-vessel disease in diabetic patients remains undefined. In terms of primary cardiovascular prevention for non-insulin-dependent diabetes mellitus, the possible benefits of tight control await the results of the U.K. Prospective Diabetes Study (1), which will be reported next year. The DIGAMI study is about secondary prevention. The distinction is important for 2 reasons. First, the process of reinfarction may be a consequence of procoagulant, rather than proatherogenic, mechanisms. Second, studies in high-risk populations provide evidence about interventions that generally require small NNTs to prevent adverse outcomes (in this instance, 10 patients to prevent 1 death at 3.4 years).

The intervention consisted of insulin and glucose infusion in the hospital followed by multiple-injection insulin treatment for at least 3 months. Nearly three quarters of the infusion group, and less than half the controls, still received insulin at 12 months, with continuing improvement in glycemic control. The authors have no information on subsequent treatment of glycemia. It is unknown whether the benefit accrued from insulin infusion during the acute event or from improved glycemic control after discharge.

Similarly, the observation that mortality curves in the 2 groups continued to diverge after discharge could imply less damage during the acute event or reduced risk for rethrombosis. However, the fact that the greatest benefit was seen in low-risk patients not previously receiving insulin raises the possibility of a pharmacologic benefit of withdrawing oral hypoglycemic agents, some of which have potentially harmful effects in the postinfarction setting (2).

The ongoing DIGAMI 2 study may enable the component parts of the intervention to be separated; however, in the interim, all diabetic patients with acute MI should begin receiving insulin and glucose on admission and should continue to receive multiple-injection insulin therapy indefinitely after discharge.

John S. Yudkin, MD
University College LondonLondon, England, UK


1. Turner R, Cull C, Holman R. Ann Intern Med. 1996;124:136-45.

2. Engler RL, Yellon DM. Circulation. 1996;94:2297-301.