Effect of the ACE Inhibitor Lisinopril on Mortality in Diabetic Patients With Acute Myocardial Infarction
Data From the GISSI-3 Study
Background Mortality of diabetic patients with acute myocardial infarction (MI) remains high despite recent improvement in their management. There is a need to evaluate efficacy and safety of novel treatments of MI in this high-risk population. We evaluated whether treatment with an ACE inhibitor begun within 24 hours from the onset of symptoms is able to decrease mortality and morbidity of diabetic patients with acute MI.
Methods and Results A retrospective analysis of the data of the GISSI-3 study in patients with and without a history of diabetes was performed. Patients with suspected acute MI were randomized to treatment with lisinopril (2.5 to 5 up to 10 mg/d) with or without nitroglycerin (5 to 20 μg IV then 10 mg/d) begun within 24 hours and continued for 6 weeks. The main end point was mortality at 6 weeks, and the secondary end point was a combined evaluation of mortality and severe left ventricular dysfunction. Information on diabetic status was available for 18 131 patients (≈94% of the total population enrolled), of whom 2790 patients had a history of diabetes. Treatment with lisinopril was associated with a decreased 6-week mortality in diabetic patients (8.7% versus12.4%; OR, 0.68; 95% CI, 0.53 to 0.86; 37±12 lives saved per 1000 treated patients), an effect that was significantly (P<.025) higher than that observed in nondiabetic patients. The survival benefit in diabetics was mostly maintained at 6 months despite withdrawal from treatment at 6 weeks (12.9% versus 16.1%; OR, 0.77; 95% CI, 0.62 to 0.95).
Conclusions Early treatment with the ACE inhibitor lisinopril in diabetic patients with acute MI is associated with a decreased 6-week mortality. This beneficial effect supports a widespread and early use of ACE inhibitors in diabetic patients with acute MI. The burden of mortality plus morbidity for ventricular dysfunction in diabetics remains clinically important and warrants further testing of novel therapeutic approaches.
Several studies performed before the introduction of fibrinolysis as a routine treatment of patients with MI consistently showed that diabetic patients have a higher mortality in hospital and after discharge.1–4 In the past decade, the widespread use of fibrinolytic agents and aspirin led to a marked improvement in the prognosis of MI patients.5,6 However, diabetic patients remain a subgroup at high risk for mortality during evolving MI.5–7 Data from the TIMI II trial7 indicated that diabetic patients also have a high mortality when an invasive approach such as primary percutaneous transluminal coronary angioplasty is used. These data stress the importance of testing novel approaches in the treatment of these patients during the acute phase. ACE inhibitors may be of particular value in diabetic patients with acute MI. In fact, these drugs have been shown to be very effective in delaying some of the complications of diabetes mellitus, including chronic renal insufficiency,8 and they exert an early beneficial effect in patients with acute MI,9 presumably by counteracting the adrenergic10 and renin-angiotensin activation11 during acute MI that may be particularly important in diabetics with autonomic neuropathy. On the other hand, because diabetic patients have more extensive coronary atherosclerosis than nondiabetic patients,12 they may be more vulnerable to the effect of ACE inhibitors on the systemic and renal hemodynamics.13 Therefore, uncertainty exists whether the use of an ACE inhibitor during the acute phase of MI might be particularly beneficial in diabetics or, alternatively, would place them at an increased risk of hypotension and acute renal dysfunction with deleterious consequences on survival.
The purpose of this study was to evaluate whether treatment with lisinopril begun within 24 hours would be associated with a significant impact on mortality and morbidity in diabetic patients with acute MI.
Data obtained from the population enrolled in the GISSI-3 study were analyzed. The details of the design and the main results of the study were provided in the original report.14 Briefly, the GISSI-3 trial was a controlled multicenter randomized open trial in patients with acute MI with a 2×2 factorial design and four treatment groups: lisinopril (2.5 to 5 mg at randomization, 2.5 to 5 mg after 24 hours, 2.5 to 10 mg after 48 hours, then 10 mg/d for 6 weeks), nitroglycerin (5 to 20 mg/min IV infusion during the first 24 hours followed by 10 mg/d transdermally, combined therapy (lisinopril plus nitroglycerin), and no trial therapy. Patients entered the study if (1) they had chest pain with ST-segment elevation of ≥1 mm in any limb lead of the ECG and/or of ≥2 mm in any precordial lead; (2) they were admitted to the coronary care unit within 24 hours from the onset of symptoms; and (3) they were considered hemodynamically stable, with a systolic blood pressure >100 mm Hg. Exclusion criteria were Killip class 4, high risk of further serious hemodynamic deterioration after treatment with vasodilators (systolic blood pressure ≤100 mm Hg), history of clinically relevant renal failure (creatinine >2 mg/dL) or presence of life-threatening conditions in the short term other than acute MI, or prior randomization within the trial.
Randomized treatments were withdrawn at 6 weeks in the absence of specific indications, and then patients were followed up until 6 months after randomization.
Information on history of diabetes, including type, ie, IDDM or NIDDM, was collected by the attending physician during the in-hospital stay.
Quality control of the collection of epidemiological and clinical data was performed centrally. Clinically relevant events (death, reinfarction, cerebrovascular accidents, renal dysfunction, persistent hypotension, and shock) were carefully collected in the study forms.
The main end point of the study was all-cause mortality at 6 weeks, and the secondary end point was the combined end point, ie, the number of patients who died plus the number of survivors who had late (beyond day 4 of hospital stay) clinical CHF or extensive left ventricular damage in the absence of clinical heart failure.
LVD was defined by the presence of late clinical CHF (defined as at least two of the following: a third sound, rales, dyspnea, evidence of pulmonary congestion at chest radiograph) or, in patients without clinically evident CHF, by the presence of extensive left ventricular damage, defined as either an echocardiographic LVEF of ≤35% or ≥45% injured myocardial segments (of the left ventricle evaluated by two-dimensional echocardiography and divided into 11 segments). A total of 1645 diabetic patients underwent echocardiography with quantification of LVEF and wall motion abnormalities, representing 66% of the total diabetic population alive at follow-up. The summation of mortality plus documented LVD was defined as combined end point.
For the purpose of this analysis, because the data indicated that the systemic use of nitroglycerin did not have any influence on mortality in this subgroup of patients, analysis was focused on the comparison between patients receiving lisinopril versus open control patients. The effect of lisinopril on mortality in the different subgroups was assessed by a test for heterogeneity. Such a statistic is approximately distributed as a χ2 with N−1 degrees of freedom, where N is the number of subgroups.15 The Neyman-Pearson χ2 statistic was used to test the statistical significance of differences in the baseline characteristics of the different groups of patients according to their diabetic status and in the incidence of events. The effect of lisinopril was evaluated by an “intention-to-treat” analysis. The significance of the difference observed was defined as a two-tailed value of P<.05. Results are also presented in terms of Mantel-Haenszel-Peto ORs and their 95% CIs for the overall nondiabetic and diabetic populations, 99% CI for subgroups of diabetic patients. Survival estimates were calculated by the product-limit method, and comparison between treated and control patients was done with the log-rank test.
Information on diabetic status was available for 18 131 patients (≈94% of total population enrolled in GISSI-3), of whom 2790 (15.2%) were defined as diabetics: 2.7% of patients had a history of IDDM (n=496) and 12.5% had a history of NIDDM (n=2294).
Baseline Characteristics of Patients
Main demographic and clinical characteristics of diabetic and nondiabetic patients enrolled in GISSI-3 are shown in Table 1⇓, together with the characteristics of patients with missing information for diabetes (6% of total population). In general, diabetic patients had a worse baseline risk profile than nondiabetics. For example, 20.6% of diabetic patients presented in Killip class >1, compared with 13.3% of nondiabetics. Baseline characteristics were well balanced overall between control and lisinopril-treated patients, without significant differences in any of the subgroups analyzed.
Effect of Lisinopril on Mortality
Treatment with lisinopril was associated with a decreased 6-week mortality in diabetic patients (8.7% versus 12.4%; OR, 0.68; 95% CI, 0.53 to 0.86). This difference was significantly (χ2 for heterogeneity=5.85, P<.025) greater than that observed in nondiabetic patients (5.6% versus 5.9%; OR, 0.95; 95% CI, 0.83 to 1.09) (Fig 1⇓). These figures correspond to a 29.8% reduction, or 37±12 lives saved per 1000 in the diabetic population, and to a 5.0% reduction, or 3±4 lives saved per 1000 in the nondiabetic population. The advantage was present in both IDDM (11.8% versus 21.1%; OR, 0.51; 95% CI, 0.32 to 0.82) and NIDDM (8.0% versus 10.6%; OR, 0.73; 95% CI, 0.55 to 0.97) patients, corresponding to a 44.1% and 24.5% reduction, respectively. However, because the test for heterogeneity did not show a significant difference in this primary end point of the study between IDDM and NIDDM patients, analysis was thereafter focused on the diabetic population overall. Most but not all the survival gain in diabetic patients was maintained at 6 months, despite withdrawal of the ACE inhibitor treatment at 6 weeks: total mortality in lisinopril-treated and control patients at 6 months was 12.9% versus 16.1% (OR, 0.77; 95% CI, 0.62 to 0.95). Analysis of 6-week mortality in different subgroups of diabetic patients is shown in Fig 2⇓. It appears that within the diabetic population, the effect is consistently present in subgroups at different risk based on demographic and clinical characteristics, with a nonsignificant value for heterogeneity of the effect of lisinopril within each subgroup.
Effect of Lisinopril on Combined End Point in Diabetic Patients
The effect of lisinopril on the secondary end point of mortality plus morbidity is shown in Table 2⇓. The treatment was associated with a lower combined end point incidence at 6 weeks both in diabetics and in nondiabetics: 21.6% versus 24.5% (OR, 0.85; 95% CI, 0.71 to 1.01) and 14.3% versus 15.5% (OR, 0.91; 95% CI, 0.83 to 1.00), respectively. The less marked effect than that observed for mortality in diabetic patients is due to the presence of a slightly higher fraction of patients with late clinical CHF among survivors in lisinopril-treated patients (7.2% versus 4.8%). To further evaluate this point, data on occurrence up to 6 weeks of clinical CHF, low (<35%) LVEF, and/or marked (≥45% of segment areas affected) akinesia/dyskinesia of ventricular motion at 6-week echocardiogram were analyzed. No significant difference was found in any of these three indexes of left ventricular function between lisinopril and control patients. More precisely, CHF was present in 8.5% of lisinopril-treated and 7.3% of control patients; LVEF <35% was present in 10.2% and 12.7% and akinesia and/or dyskinesia ≥45% was present in 5.3% and 5.4%, respectively.
At 6 months, after treatment had been withdrawn at 6 weeks, in diabetic patients the difference in combined end point was 25.6% versus 28.9%, a nonsignificant 15±8% reduction.
Safety of Lisinopril in Diabetic Patients
One of the major issues related to the use of ACE inhibitors during the acute phase of MI is the incidence and clinical significance of adverse drug reactions.
The main adverse events recorded in the diabetic and nondiabetic patients are shown in Table 3⇓. There was no difference in the incidence of stroke, shock, postinfarction angina, and reinfarction, whereas treatment with lisinopril was associated with an increased incidence of persistent hypotension and renal dysfunction, similar to what was observed in the nondiabetic population. Furthermore, even in the subgroup of patients who developed persistent hypotension and/or renal dysfunction, the absolute number of deaths was slightly lower in the lisinopril-treated patients (6 versus 11 for renal dysfunction and 24 versus 32 for persistent hypotension). There was no difference in the duration of hospital stay of diabetic and nondiabetic patients discharged alive in control and lisinopril-treated patients (15.3±7.1 versus 15.3±7.4 days for diabetics and 14.3±6.1 versus 14.2±6.1 days for nondiabetics).
Studies performed after the advent of fibrinolysis indicate that diabetic patients continue to have a higher mortality after MI than nondiabetic patients. For example, data from the GISSI-2 study5 show a 30% to 60% higher in-hospital mortality in diabetic patients of both sexes compared with nondiabetic patients, similar to the GUSTO trial,6 in which mortality was 10.6% in diabetics and 6.2% in nondiabetics (OR, 1.79; 95% CI, 1.64 to 1.97). The search for novel therapeutic approaches in this high-risk population appears to be relevant for both its clinical and its socioeconomic implications.
Mechanisms for Increased Mortality in Diabetic Patients With Acute MI and Potential for the Beneficial Effect of ACE Inhibitors
Several mechanisms may contribute to the increase of in-hospital mortality in diabetic patients16,17: (1) more extensive coronary atherosclerosis, (2) clinical or subclinical diabetic cardiomyopathy with systolic and/or diastolic dysfunction, (3) alterations in the fibrinolytic system that impair reperfusion and facilitate reocclusion after fibrinolysis, (4) diabetic neuropathy with sympathetic/parasympathetic imbalance, and (5) endothelial dysfunction leading to impaired myocardial perfusion. It is impossible to identify the relative roles of these mechanisms in mediating the increased mortality; however, ACE inhibitors have the potential to interfere with most of these mechanisms either directly (ie, altering the effect of the renin-angiotensin and bradykinin systems) or indirectly (ie, through their hemodynamic effects). These drugs also appeared to be a promising therapeutic approach in diabetic patients on the basis of their documented beneficial effect in different clinical settings.8
Effect of ACE Inhibitors in Diabetic Patients Post-MI: Long-term Studies
Several recent trials used ACE inhibitors in the attempt to reduce mortality and morbidity in patients with LVD and/or overt heart failure after acute MI. All these studies have shown a significant benefit of long-term ACE-inhibitor therapy, with a risk reduction in mortality of 19% to 27% over 2.5 to 4 years of follow-up. Subgroup analyses of the SAVE18 and TRACE studies19 indicate that the beneficial effect documented in the overall population is present also when the analysis is limited to patients with a history of diabetes, with a similar reduction in mortality. This suggests that ACE inhibitors can be safely administered in patients with LVD post-MI and that their beneficial effect on mortality is similar in diabetic and nondiabetic patients.
Efficacy and Safety of ACE Inhibitors During the Acute Phase: Present Study
The data of this study confirm that diabetic patients have a high mortality during the acute phase of MI. They also indicate that treatment with the ACE inhibitor lisinopril is associated with an ≈30% proportional reduction of 6-week mortality in diabetics (8.7% versus 12.4%). This decrease of mortality is significantly higher than that observed in nondiabetics (≈5% reduction) and consistent among subgroups at different risk, on the basis of demographic and clinical characteristics. It may be speculated that because the effect of ACE inhibitors in the acute setting is generally more pronounced in high-risk patients,9 the worse baseline risk profile of diabetic patients could have positively influenced the effect of lisinopril. However, the statistical analysis showed a significant positive interaction between presence of diabetes and effect of the ACE inhibitor; also, for each subgroup, the beneficial effect of lisinopril was consistently higher in diabetics than in nondiabetics. Interestingly, the beneficial effect of lisinopril in diabetics appears to persist for the most part even after withdrawal from treatment at 6 weeks, on the basis of the study protocol. Among survivors, on the other hand, there was a higher percentage of patients with severe LVD among diabetic patients treated with lisinopril (7.2% versus 4.8%). A detailed analysis of the incidence of symptomatic or asymptomatic CHF in hospital did not show any difference between lisinopril and control patients. This suggests that the treatment might be able to decrease the pathophysiological mechanisms leading to mortality during the acute phase and leaves a burden of morbidity for LVD among survivors that decreases the impact of the treatment on the combined end point.
The incidence of most adverse events, such as stroke and reinfarction, was similar in treated and nontreated patients, and the increase in the incidence of renal dysfunction and persistent hypotension due to lisinopril, like that observed in nondiabetics, was not associated with a worsened outcome.
The overall analysis of these data clearly indicates that the use of an ACE inhibitor during the acute phase of MI in diabetic patients has a favorable risk/benefit ratio.
Limitations and Implications of the Present Study
The main limitation of this study lies in its retrospective nature: the effect of ACE-inhibitor treatment in diabetic patients was not a predefined analysis of GISSI-3.20 A second limitation is that, at variance with ad hoc studies such as DIGAMI,21 no data on indexes of glycemic control were available. On the other hand, it has to be acknowledged that this retrospective type of information is and will be the only kind available for evaluating the effect of ACE inhibitors in diabetic patients during MI, because no other prospective studies on this topic are either ongoing or planned. In this respect, the size of the diabetic population enrolled in GISSI-3 is by far the largest available for analysis. The question remains whether this marked effect of lisinopril on mortality in diabetic patients documented in GISSI-3 is representative of a class effect of ACE inhibitors. In this context, it is interesting to observe that even in CONSENSUS II,22 a smaller study that did not demonstrate a beneficial effect of ACE-inhibitor treatment in the overall population, the subgroup of diabetic patients experienced a trend toward a lower mortality (13.5% versus 16.7%; OR, 0.78; 95% CI, 0.51 to 1.19). A preliminary analysis of the pooled data of CONSENSUS-II and CCS-1 trials with those of GISSI-3 recently performed by the ACE Inhibitors in MI Collaborative Group (unpublished data, 1997) shows a trend toward a more beneficial effect of ACE inhibitors in diabetic versus nondiabetic patients (17.3 versus 3.2 lives saved per 1000 treated patients). Current guidelines on the treatment of acute MI consider mandatory the use of fibrinolysis and aspirin in all eligible patients with acute MI.23 According to these guidelines, early ACE-inhibitor treatment is recommended only for selected subgroups of patients, such as those with anterior MI or documented signs of CHF. A document from a consensus conference9 has pointed out that the early use of ACE inhibitors may lead to a further beneficial effect, particularly during the early, high-risk phase. The data from the present analysis indicate that the benefit from early ACE inhibition appears to be particularly striking in diabetic patients, whose overall mortality is confirmed to be higher than the nondiabetic population. Although the quest for a specific intervention able to reduce the increased risk associated with a preexisting diabetic condition remains open, acute ACE inhibition appears to be specifically recommended for the routine treatment of acute MI patients with diabetes.
Selected Abbreviations and Acronyms
|CHF||=||congestive heart failure|
|IDDM||=||insulin-dependent diabetes mellitus|
|LVD||=||left ventricular dysfunction|
|LVEF||=||left ventricular ejection fraction|
|NIDDM||=||non–insulin-dependent diabetes mellitus|
Dr Zuanetti is a recipient of a Maurelio Caniato award fellowship. The GISSI-3 was endorsed by the Associazione Nazionale Medici Cardiologi Ospedalieri and by the Istituto di Ricerche Farmacologiche Mario Negri and was supported by a research grant from Zeneca and Schwarz-Pharma.
Presented in part at the 45th American College of Cardiology Scientific Session, Orlando, Fla, March 24-27, 1996.
↵1 A complete list of collaborators and participating centers was published in Reference 14.
- Received April 30, 1997.
- Revision received June 17, 1997.
- Accepted July 3, 1997.
- Copyright © 1997 by American Heart Association
Singer DE, Moulton AW, Nathan DM. Diabetic myocardial infarction: interaction of diabetes with other preinfarction risk factors. Diabetes. 1989;38:350–357.
Stone PH, Muller JE, Hartwell T, York BJ, Rutherford JD, Parker CB, Turi ZG, Strauss W, Willerson JT, Robertson T, Braunwald E, Jaffe AS, and the MILIS Study Group. The effect of diabetes mellitus on prognosis and serial left ventricular function after acute myocardial infarction: contribution of both coronary disease and diastolic left ventricular dysfunction to the adverse prognosis. J Am Coll Cardiol. 1989;14:49–57.
Lee KL, Woodlief LH, Topol EJ, Weaver D, Betriu A, Col J, Simoons M, Aylward P, Van de Werf F, Califf RM, for the GUSTO-I Investigators. Predictors of 30-day mortality in the era of reperfusion for acute myocardial infarction: results from an international trial of 41 021 patients. Circulation. 1995;91:1659–1668.
Mueller HS, Cohen LS, Braunwald E, Forman S, Feit F, Ross A, Schweiger M, Cabin H, Davison R, Miller D, Solomon R, Knatterud GL, for the TIMI Investigators. Predictors of early morbidity and mortality after thrombolytic therapy of acute myocardial infarction: analyses of patient subgroups in the thrombolysis in myocardial infarction (TIMI) trial, phase II. Circulation. 1992;85:1254–1264.
Laffer LMB, McGill JB, Gans DJ, on behalf of the North American Micro Albuminuria Study Group. The beneficial effect of angiotensin-converting enzyme inhibition with captopril on diabetic nephropathy in normotensive IDDM patients with microalbuminuria. Am J Med. 1995;99:497–504.
Latini R, Maggioni AP, Flather M, Sleight P, Tognoni G. ACE-inhibitor use in patients with myocardial infarction: summary of evidence from clinical trials. Circulation. 1995;92:3132–3137.
Schwartz PJ, Priori SG. Sympathetic nervous system and cardiac arrhythmias. In: Zipes DP, Jalife J, eds. Cardiac Electrophysiology: From Cell to Bedside. Philadelphia, Pa: WB Saunders Co; 1990:330–343.
Woodfield SL, Lundergan CF, Reiner JS, Greenhouse SW, Thompson MA, Rorhbeck SC, Deychak Y, Simoons ML, Califf RM, Topol EJ, Ross A, for the GUSTO-I Angiographic Investigators. Angiographic findings and outcome in diabetics treated with thrombolytic therapy for acute myocardial infarction: the GUSTO-I experience. J Am Coll Cardiol. 1996;28:1661–1669.
Moyé LA, Pfeffer MA, Wun CC, Davis BR, Geltman E, Hayes D, Farnham DJ, Randall OS, Dinh H, Arnold JMO, Kupersmith J, Hager D, Glasser SP, Biddle T, Hawkins CM, Braunwald E, for the SAVE Investigators. Uniformity of captopril benefit in the SAVE study: subgroup analysis. Eur Heart J. 1994;15:2–8.
Malmberg K, Rydén L, Efendic S, Herlitz J, Nicol P, Waldenström A, Wedel H, Welin L, 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;26:57–65.
ACC/AHA Guidelines for the management of patients with acute myocardial infarction: a report of the American College of Cardiology/American Heart Association task force on practice guidelines (Committee on Management of Acute Myocardial Infarction). J Am Coll Cardiol. 1996;28:1328–1428.