Diabetes Spectrum
Volume 12 Number 2, 1999, Pages 84–88

These pages are best viewed with Netscape version 3.0 or higher or Internet Explorer version 3.0 or higher. When viewed with other browsers, some characters or attributes may not be rendered correctly.

From Research to Practice and Coronary Heart Disease

  In Brief

People with diabetes who suffer an acute myocardial infarction (MI) are at markedly increased risk of future cardiovascular morbidity and mortality. The DIGAMI study compared "conventional" anti-diabetic therapy to intensive insulin therapy consisting of acute insulin infusion during the early hours of MI and thrice-daily subcutaneous insulin injection for the remainder of the hospital stay and a minimum of 3 months thereafter. Although there was an overall reduction in adverse outcomes in patients receiving the intensive insulin regimen, it is unclear which component (the IV insulin infusion or the intensive chronic therapy) was responsible.

A Review of the DIGAMI Study: Intensive Insulin Therapy During and After Myocardial Infarctions in Diabetic Patients

Jennifer Cummings, MD
Kevin Mineo, MD
Richard Levy, MD
Richard A. Josephson, MS, MD

Despite many advances in modern medicine, diabetes mellitus continues to be associated with increased morbidity and mortality. The leading cause of death in people with diabetes continues to be myocardial infarction (MI). Although improvements have been made in the treatment of cardiac disease, diabetic patients with acute MI continue to have a poor prognosis. This increase in mortality is shown during both initial hospitalization and long-term follow up.1-4

The higher death and complication rates in these diabetic patients appear to be multifactorial. Diabetes may be associated with severe coronary artery disease, systolic left ventricular dysfunction, autonomic neuropathy, and larger infarct size. These processes not only leave these diabetic patients at higher risk of death when having an acute MI, but also increase their risk of recurrent cardiac events and other long-term complications.1-6

During an acute MI, multiple hormonal and physiological changes occur. Plasma catecholamines, glucagon, and cortisol increase, resulting in insulin resistance. Decreased insulin sensitivity causes impaired glucose utilization and increased free fatty acid turnover in cardiac muscle. People with diabetes are more sensitive to catacholamine stimulation, and thus they have a dramatic increase in plasma free fatty acids and a decrease in glucose utilization.1,7,8 As the blood levels of free fatty acids rise, the myocardium metabolizes more free fatty acids than glucose (up to 90% of energy supply in animal studies).9

Though the myocardium normally uses some free fatty acid (6070% of overall use), this metabolic process requires oxygen. Glucose does not require oxygen when metabolized (glycolysis), but this process, as described above, is impaired by the hormonal changes that take place during MI. This shift from glucose use to free fatty acid use increases the oxygen demand of the heart muscle. The heart's demand for more oxygen cannot be met by the blood supply because of the infarction process.1,10,11

This supply-demand imbalance creates an energy deficit that leads to myocardial ischemia. Insulin, either endogenous or exogenous, favors the use of glucose rather than free fatty acids as an energy source. By preferentially using glucose, myocardial oxygen demand decreases, and the supply-demand imbalance may be reduced.

Studies have shown that insulin may have a role in restoring other cardiac and metabolic dysfunctions common in diabetic patients. One such dysfunction is the increased platelet aggregation, which can be reduced with insulin administration.12 Also, tissue plasminogen activator inhibitor-1 (tPAi-1) activity, a coagulation factor that inhibits fibrinolysis, is increased in many people with diabetes. This impairment may potentiate ischemic heart disease by facilitating coronary artery occlusion and reocclusion. Insulin therapy has been shown to decrease tPAi-1 levels and possibly normalize the fibrinolytic process. These actions of insulin appear to reduce many of the biochemical obstacles diabetic patients face during and after MI.12-15

Based on these multifaceted benefits of insulin, a multicenter trial was recently performed in Sweden to evaluate the effects of intravenous insulin and glucose infusions in diabetic patients who were experiencing an acute MI. Titled The Diabetes Mellitus Insulin-Glucose Infusion in Acute Myocardial Infarction (DIGAMI) trial, this study attempted to evaluate the influence of intensive insulin therapy on morbidity and mortality of diabetic patients with MI.

Intensive insulin therapy included an insulin-glucose infusion during the initial 24 hours of hospitalization, followed by subcutaneous insulin four times daily for a minimum of 3 months. Morbidity and mortality were assessed in the acute, sub-acute, and chronic phases. Acute effects, determined during the first 1224 hours after MI, included assessment of initial infarct size. Sub-acute effects (230 days) and chronic effects (>30 days) included the incidence of significant arrhythmias, re-infarction, congestive heart failure, and death.

Summary of Methods
The DIGAMI study considered all patients admitted to the coronary care units (CCUs) of 16 Swedish hospitals between January 1990 and December 1993. Inclusion criteria were 1) a previous diagnosis of diabetes or an admission blood glucose of >198 mg/dl (11 mM), and 2) probable MI.

The diagnosis of MI was divided into probable or possible using conventional criteria (Table 1). Using these criteria, 1,240 patients were potentially eligible for the study, but 620 (50%) of the patients were immediately excluded. Exclusion criteria were 1) inability to participate for reasons of health (45%), 2) refusal (38%), 3) residence outside the hospital area (7%), and 4) prior enrollment in DIGAMI or other studies (10%).

The remaining 620 patients were then randomized to either a control group (314 patients) or an insulin infusion group (306 patients). Patients then received insulin-glucose infusions or standard diabetes care according to the study's protocol (Figure 1).

Figure 1. A schematic presentation of the patient population included in the Diabetes Mellitus Insulin-Glucose Infusion in Acute Myocardial Infarction (DIGAMI) study. Adapted from reference 1.

The protocols for those in the infusion group included using an insulin-glucose infusion and frequent blood glucose checks to bring and maintain patients' serum glucose to a range between 126 and 180 mg/dl. This range was selected as a balance between providing strict control and minimizing the hypoglycemic events that occurred in previous studies, which had used a lower target of 90144 mg/dl. The infusion was continued for a minimum of 24 hours, at which point the patients were changed to subcutaneous insulin injections 4 times daily. This dosing of both long/intermediate- and short-acting insulin was continued for a minimum of 3 months and possibly indefinitely.

Table 1. Diagnosis of Patient Population of the Diabetes Mellitus Insulin-Glucose Infusion in Acute Myocardial Infarction (DIGAMI) Study

     Definite Myocardial Infarction
     Two of the following:

  1. Chest pain of at least 15 minutes' duration

  2. Two values of serum creatine kinase and
    serum creatine kinase-B above the normal
    range 1016 hours after onset of symptoms
    or two serum lactate dehydrogenase values
    above the normal range 4872 hours after symptoms.

  3. Development of new Q waves in 2 of
    the 12 standard electrocardiogram leads.

      Possible Myocardial Infarction
      Chest pain combined with at least one of
       the following:

     1.  One serum creatine kinase or serum
            lactate dehydrogenase above the
            normal range.

       2.  New Q waves in 1 of the 12 standard
            electrocardiogram leads.


Patients who were randomized to the control group were managed according to standard coronary care practice without insulin-glucose infusion. Subcutaneous insulin was used only if determined to be necessary by a CCU physician, particularly if the patient was on insulin before admission.

In addition to the diabetes treatment, all patients received thrombolytic therapy (streptokinase), beta-blockers, aspirin, heparin, nitroglycerin, percutaneous transluminal coronary angioplasty (PTCA), and coronary artery bypass graft (CABG) surgery acutely and chronically as deemed appropriate by their physicians.

All patients in the study were classified into four pre-stratified groups according to their previous anti-diabetic management and initial cardiac risk classification. High cardiac risk was determined by two or more of the following: 1) age > 70 years, 2) history of previous MI, 3) history of congestive heart failure, and 4) current treatment with digitalis. The pre-stratified risk groups were 1) no previous insulin; low cardiac risk, 2) previous insulin; low cardiac risk, 3) no previous insulin; high cardiac risk, and 4) previous insulin; high cardiac risk.1,16

Summary of Results
Group characteristics
The control group and the infusion group had similar demographics and pre-hospitalization characteristics. Both groups consisted of approximately 60% men, who were slightly overweight by body mass index (BMI), typically not on insulin at presentation, and had an average hemoglobin A1c of approximately 8.0%. (The average BMI was 27 for both control and insulin groups with no one excluded because of BMI.) In the overall population, over 80% of patients were not using insulin upon presentation. Of the entire study, 13% of patients were newly diagnosed with diabetes by elevated blood glucose upon presentation at the CCU.

Overall, concomitant therapy was similar between both groups of the study. More than 80% of patients were on aspirin, and 70% of patients were on beta-blockers at the time of discharge. The groups also had similar percentages of thrombolytics, PTCA, and CABG.1,15-17

Both groups had an approximately equal number of in-hospital complications. Complications considered when comparing the infusion group to the control group were 1) ventricular fibrillation (5% vs. 3%, P = NS), 2) re-infarction (4% vs. 5%, P = NS), 3) high degree of atrioventricular conduction disturbances (3% vs. 7%, P = NS), and 4) congestive heart failure (48% vs. 50% P = NS). Thus, regardless of the treatment protocol received, the in-hospital courses were similar.

Hospital stay was slightly different in that the average length of stay was 11.3 (13.3) days for the infusion group and 9.5 (9.4) days for the control group (P = 0.043).

Overall, patients had a mortality of 10.2% in the hospital, 14.0% at 3 months, and 22.4% at 1 year. Though the infusion group had a slightly lower mortality than the control group in the hospital (9.1% vs. 11.1%, P = NS), at 3 months post-MI (12.4% vs. 15.6%, P = NS), and at 1 year (18.6% vs. 26.1%, P = 0.0273), only the 1-year mortality was statistically significant.

When examining mortality in the pre-stratified risk groups, the greatest mortality reduction is noted for patients who had never been on insulin before and were classified as low cardiac risk. At 3 months, this low-risk group's relative mortality reduction for those placed on insulin infusion versus control was 52% (P = 0.046). This reduction persisted, and at the 1-year mark, mortality continued to be reduced by 52% (P = 0.020).

Patients with diabetes not only are at increased risk of having an MI, but also are more likely to have complications, increased mortality, and recurrent infarctions. Though other studies have evaluated the relationship between improved diabetic control and post-MI morbidity and mortality, the DIGAMI study is the first to show such a large reduction in mortality for diabetic patients with MIs at 1 year. Also, the DIGAMI population, though modest in size, was larger than had been examined before.

To place the DIGAMI results in perspective, it is important to note the decrease in overall mortality in diabetic patients with MI in all treatment groups. The DIGAMI study showed overall 1-year post-MI mortality to be 22.4%. This differs from previously reported data showing that diabetic patients with MIs have a 1-year post-MI mortality of 53%.3 Many factors are suspected in influencing this decrease in the mortality of diabetic patients with MI. Some credit the overall improvement to widespread use of medications, such as beta-blockers, aspirin, and thrombolytics.1,10,17,18

In the past, beta-blockers were not used in diabetic patients for fear of masking and possibly prolonging of hypoglycemic episodes. But studies now show that there is a distinct beneficial effect of beta-blockade in these patients (up to a 50% reduction in mortality).18,19

The benefits of beta-blockade, specifically in diabetic patients, are multifaceted in nature. There are indications in experimental settings that propranolol may shift myocardial metabolism from free fatty acid utilization to glucose utilization. This change in metabolism would decrease myocardial muscle oxygen requirements, shifting the supply-demand imbalance and possibly reducing infarction size.18,20 Another benefit of beta-blockers is their role in improving the autonomic imbalance that many diabetic patients possess.18,21 Diabetic patients have also been found to have a higher heart rate on average than nondiabetic patients when presenting with an MI. Thus, some researchers believe diabetic patients may show a greater benefit when heart rate, and thus myocardial oxygen demand, is decreased.18,19 In the DIGAMI study, 70% of patients were discharged on beta-blockers, which may have helped decrease overall mortality, independent of the study group in which the patient was enrolled.

Other cardiac medications now widely used that may have had an effect on overall diabetic MI mortality are aspirin and thrombolytics. Both therapies were used similarly and equally in the control and infusion groups of the DIGAMI trial.

Thromboxane A production and platelet aggregability have been shown to be increased in diabetic patients, fostering a hypercoagulable state.1,12-15,17,18 Consistent with this fact, diabetic patients have been shown to benefit more than nondiabetic patients from streptokinase in the International Study of Infarct Survival (ISIS) II trial.22 More than 80% of patients enrolled in the DIGAMI study were discharged on aspirin therapy, and many patients (50%) received thrombolytic therapy during their hospitalization. It is reasonable to propose that increased anti-platelet and fibrinolytic therapy may have aided in the overall decreased mortality in the diabetic patients with cardiac events included in this study.

Another point to emphasize is that diabetic patients who received intensive insulin therapy during their acute event as well as through the following year did show a significant decrease in mortality at 1 year (P = 0.0273) when compared to the control group. The change in mortality during hospitalization and at 3 months post-MI was not significant.

In this group, insulin was given in both the acute setting (insulin-glucose infusion) as well as throughout the year following the MI (subcutaneous insulin). This makes determining which intervention was responsible for the decrease in mortality impossible: the acute intervention (insulin-glucose infusions) or the long-term intervention (subcutaneous insulin).

This is an important consideration especially when contemplating implementing this protocol. This study did not provide data showing a direct relationship between insulin-glucose infusion and the decreased mortality at 1 year. Instead, it showed a decreased mortality in diabetic patients admitted with MIs when given an intensive insulin regimen extending from admission up to 1 year.

When considering treating diabetic patients in such an aggressive manner using insulin, other social needs should be considered. In implementing this protocol, a patient should be willing to commit to both an acute and a long-term intervention program to fully improve their mortality. This type of commitment may not be possible in a portion of the population due to either inability or unwillingness to administer insulin.

In reviewing this study's population, patients who were not willing or able to commit to insulin were excluded. Of the 1,240 patients who met inclusion criteria, 620 patients (50%) were excluded because of unwillingness or inability to participate. This in itself may have created a bias because the patients studied were required to agree to aggressive insulin therapy for an extended period of time. Though unavoidable, this bias should be taken into consideration before making insulin therapy the standard of care. Characteristics such as compliance and willingness may affect a patient's success in following intensive insulin therapy for any period of time.

There is also a possibility that in the course of implementing prolonged intensive insulin therapy, patients had increased follow-up appointments and continuing care visits with their outpatient physicians. The authors of the DIGAMI study addressed this possible discrepancy, stating that any increased follow-up should be considered part of comprehensive aggressive diabetic management.1,17 It may be this comprehensive approach combined with aggressive initial therapy that benefits diabetic patients most. Though insulin may help reduce mortality in theory, a treatment plan that patients are willing and able to perform may achieve more long-term success.

A final point to emphasize in reviewing the DIGAMI results is the significant mortality reduction noted in patients who were never previously on insulin and who had low cardiac risk factors. A 52% mortality reduction is noted at 3 months and at 1 year (P = 0.046 and P = 0.020, respectively). Interestingly, the patients who appeared to have the least baseline disease benefited most from aggressive therapy.

This is consistent with other studies that found early aggressive management may protect patients from diabetic complications. The Diabetes Control and Complications Trial (DCCT) revealed that intensive insulin therapy delayed the onset and slowed the progression of microangiopathies (nephropathy, neuropathy, and retinopathy). These benefits of intensive treatment were greater in the primary prevention group, whose subjects had no symptoms at baseline.23 Similarly, the DIGAMI trial revealed increased benefit to a diabetic patient's mortality when aggressive treatment is begun in the face of mild cardiac and diabetic disease.

Diabetes continues to adversely affect the prognosis in individuals presenting with MI. The purpose of the DIGAMI study was to evaluate whether aggressive initial insulin therapy continued for 3 months would improve these patients' morbidity and mortality following an acute MI. Though the DIGAMI study showed decreased mortality in diabetic patients 1 year after MI, the modest sample size and lack of positive findings in other endpoints make this finding less robust.

Overall, this study was successful in showing the feasibility and potential advantage of aggressive long-term insulin management in diabetic patients with MIs. This advantage likely cannot be accounted for by the insulin-glucose infusion protocol alone. The design of the DIGAMI study was such that the decrease in mortality could not be specifically attributed to either the insulin-glucose infusion or the 4/day subcutaneous insulin alone.

The researchers acknowledged this and emphasized that it was the comprehensive care, including insulin and frequent physician visits both acutely and chronically, that might have benefited these patients most. This is in accord with the DCCT, which recommended treating diabetic patients aggressively early in their disease process to provide the greatest benefit and possibly prevent long-term complications.23 A recent article in the Journal of the American Medical Association has also shown increased quality of life in diabetic patients with early intensive therapy and better glucose control.24 Thus, an intensive and comprehensive care plan both acutely and chronically may help diabetic patients with MIs reduce serious and often fatal complications.

1Malmberg K, Ryden L, Hamsten A, Herlitz J, Waldenstrom A, Wedel H, Welin L: 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 26:56-65, 1995.

2Czyzk A, Kvolewski A, Szablowska S, Alot A, Kopcznski J: Clinical course of myocardial infarction among diabetic patients. Diabetes Care 3:526-29, 1980.

3Malmberg K, Ryden L: Myocardial infarction in patients with diabetes mellitus. Eur Heart J 9:256-64, 1988.

4Herlitz J, Malmberg K, Karlson B, Ryden L, Hjalmarson A: Mortality and morbidity during five year follow-up of diabetics with myocardial infarction. Acta Med Scand 224:31-38, 1988.

5Granger C, Califf R, Young S, Condela R, Samaha J, Worley S, Kereiakes D, Topol E: Outcome of patients with diabetes mellitus and acute myocardial infarction treated with thrombolytic agents: the Thrombolysis and Angioplasty in Myocardial Infarction (TAMI) study group. J Am Coll Cardiol 21(4):920-25, 1993.

6Jacoby R, Nesto R: Acute myocardial infarction in the diabetic patient: pathophysiology, clinical course and prognosis. J Am Coll Cardiol 20:736-44, 1992.

7Vetter N, Strange R Sr., Adams W, Oliver M: Initial metabolic and hormonal response to acute myocardial infarction: relationship to acute MI. Lancet 2:284-49, 1974.

8Opie L: Metabolism of free fatty acids, glucose, and catacholamines in acute myocardial infarction: relation to myocardial ischemia and infarct size. Am J Cardiol 36:938-50, 1975.

9Morgan H, Cadenas E, Regen D, Park C: Regulation of glucose uptake in muscle. II. Rate limiting step and effects of insulin and anoxia in heart muscle from diabetic rats. J Biol Chem 236:262-68, 1961.

10Davey G, McKeigue P: Insulin infusion in diabetic patients with acute myocardial infarction. Br Med J 313:7058:639-40,1996.

11Oliver M, Kurien V, Greenwood T: Relation between serum free fatty acids and arrhythmias and death after acute myocardial infarctions. Lancet 1:710-15, 1968.

12Davi G, Catalang I, Averna M, Notarbartolo A, Strano A, Ciabattoni G, Patrono C: Thromboxane biosynthesis and platelet function in type II diabetes mellitus. N Engl J Med 322:1769-74, 1990.

13Vague P, Raccah D, Juhan-Vague I: Hemobiology, vascular disease, and diabetes with special reference to impaired fibrinolysis. Metabolism 41 (Suppl 1):2-6, 1992.

14Hamsten A, deFaire U, Walldius G, Szamosi A, Dahlen G, Landou C, Bomback M, Wiman B: Plasminogen activator in plasma, risk factor for recurrent myocardial infarction. Lancet 2:3-9, 1987.

15Malmberg K, Ryden L, Hamsten A, Herlitz J, Waldenstrom A, Wendel H: Effects of insulin treatment on cause specific one-year mortality and morbidity in diabetic patients with acute myocardial infarctions. Eur Heart J 17:1337-44, 1996.

16Malmberg K, Efendic S, Ryden L: Feasibility of insulin glucose infusion in diabetic patients with acute myocardial infarction. Diabetes Care 17:1007-14, 1994.

17Malmberg K: Prospective randomized study of intensive insulin treatment of long-term survival after acute myocardial infarction in patients with diabetes mellitus. Br Med J 314:1512-15, 1997.

18Malmberg K, Ryden L, Hamsten A, Herlitz J, Waldenstrom A, Wedel H: Mortality prediction in diabetic patients with myocardial infarction: experiences from the DIGAMI study. Cardiovasc Res 34:248-53, 1997.

19Malmberg K, Herlitz J, Hjalmarsson A, Ryden L: Effects of metoprolol on mortality and late infarction in diabetics with suspected acute myocardial infarction: retrospective data from two large scale studies. Eur Heart J 10:423-28, 1989.

20Opie L, Thomas M: Propranalol and experimental myocardial infarction: substrate effects. Postgrad Med J 52 (Suppl 4):124-32, 1976.

21Ewing D: Cardiac autonomic neuropathy. In Diabetes and Heart Disease. Jarret R, ed. Amsterdam, Elsevier, 1981, p. 99-132.

22ISIS-2 Collaborative Group: Randomized trial of intravenous streptokinase, oral aspirin, both or neither among 17,187 cases of suspected acute myocardial infarction: ISIS-2. Lancet 2:349-60, 1988.

23The DCCT Research Group: The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med 329(14):977-86, 1993.

24Testa M, Simonson D: Health economic benefits and quality of life during improved glycemic control in patients with diabetes mellitus: a randomized, controlled, double-blind trial. JAMA 280(17):1490-96, 1998.

The authors are all employed at the Summa Health System (Akron City Hospital) in Akron, Ohio. Jennifer Cummings, MD, is an internal medicine resident. Kevin Mineo, MD, is is an internal medicine intern. Richard Levy, MD, is an attending physician at Summa and a professor of medicine at Northeast Ohio Universities College of Medicine. Richard A. Josephson, MS, MD, is director of Cardiology, Research, and Education at Summa and a professor of medicine at Northeast Ohio Universities College of Medicine.

Return to Issue Contents

Copyright 1999 American Diabetes Association

Last updated: 5/99
For Technical Issues contact