Insulin infusion followed by multiple daily insulin injections reduced mortality after MI in diabetes
ACP J Club. 1996 Jan-Feb;124:1. doi:10.7326/ACPJC-1996-124-1-001
Malmberg K, Rydén L, Efendic S, et al., on behalf of the DIGAMI study group. Randomized trial of insulin-glucose infusion followed by subcutaneous insulin treatment in diabetic patients with acute myocardial infarction (DIGAMI Study): effects on mortality at 1 year. J Am Coll Cardiol. 1995 Jul;26:57-65. [PubMed ID: 7797776]
To evaluate the effectiveness of insulin-glucose infusion in the hospital followed by 4 daily injections of subcutaneous insulin after discharge in patients with diabetes mellitus and a suspected acute myocardial infarction (MI).
Randomized controlled trial with 1-year follow-up.
Coronary care units (CCUs) of 19 hospitals in Sweden.
620 patients (mean age 68 y, 63% men) with a suspected acute MI within the previous 24 hours who had a blood glucose level > 11 mmol/L at admission either with or without a previous diagnosis of diabetes. More than 80% of patients were classified as having non-insulin-dependent diabetes mellitus (NIDDM). Exclusion criteria were inability to participate for health reasons, residence outside the hospital catchment area, or enrollment in other studies.
All patients received standard CCU therapy. 306 patients were assigned to insulin-glucose infusion for ≥ 24 hours followed by subcutaneous insulin 4 times daily for ≥ 3 months. 314 patients were assigned to usual care.
Main Outcome Measures
Mortality, length of hospital stay, and hemoglobin A1C (HbA1C) level.
Fewer patients assigned to insulin-glucose infusion died within 1 year compared with those assigned to usual care (P = 0.025) (Table). Patients in the infusion group without previous insulin treatment who had a low cardiovascular risk profile benefited the most (1-year mortality 8.6% vs 18.0% in the usual care group, P = 0.02). The hospital stay was 11.3 days in the infusion group and 9.5 days in the usual care group (P = 0.043). A difference existed for HbA1C level by 3 months (7.0% [SD 1.6%] in the infusion group vs 7.5% [SD 1.9%] in the usual care group, P < 0.01); the 1-year value had decreased by 0.9% in the treatment group and 0.35% in the control group (P < 0.05).
Insulin-glucose infusion followed by multiple daily subcutaneous insulin injections reduced 1-year mortality in patients with diabetes and acute myocardial infarction.
Source of funding: Swedish Heart and Lung Foundation.
For article reprint: Dr. K. Malmberg, Department of Cardiology, Karolinska Hospital, 171 76 Stockholm, Sweden. FAX 46-8-344964.
Table. Insulin infusion followed insulin injections (insulin group) after myocardial infarction in diabetes*
|Outcome at 1 y||Insulin group||Usual care||RRR (95% CI)||NNT(CI)|
|Mortality||18.6%||26.1%||29% (4 to 47)||13% (7 to 108)|
*Abbreviations defined in Glossary; RRR, NNT, and CI calculated from data in article.
Middle-aged patients with both IDDM and NIDDM have a high risk for cardiovascular disease and a poor prognosis after MI. Epidemiologic studies suggest that this risk is correlated with the degree of hyperglycemia in patients with either diabetes or impaired glucose tolerance (1, 2). This study by Malmberg and colleagues clearly shows that reducing hyperglycemia by an insulin-glucose infusion followed by ambulatory multiple daily injections of insulin decreased the 1-year mortality rate by > 25%.
Because the nature of the intervention precluded blinding, the treatment group may have received more attention, including more positive feedback about the level of glycemia; this may have led to better compliance with other cardioprotective therapies during the year after MI. Although such a bias may magnify the isolated effect of insulin, it does not detract from the conclusion that the tested therapeutic strategy improved mortality.
The Diabetes Control and Complications Trial showed that tight glycemic control prevents microvascular complications of IDDM. This study by Malmberg and colleagues has several messages. First, identification and insulin-mediated glycemic control of both established and previously undiagnosed diabetes for patients with an acute MI is beneficial. Second, improved mortality occurred despite modest differences in HbA1C 3 months after MI and despite prerandomization HbA1C values (8%) that many would consider acceptable for patients who have had an MI. This suggests that clinicians should not be complacent about diabetes care in the wake of an MI. Third, the perils of improving control, such as severe hypoglycemia and weight gain, cannot be neglected and require that patients be informed and willing to monitor their glucose levels and that there be close medical supervision. Although the benefits outweighed the risks in this study, future research to confirm and expand on these important observations is keenly anticipated.
Hertzel C. Gerstein, MD
McMaster UniversityHamilton, Ontario, Canada