Hypertensive patients with diabetes are at increased risk for stroke and coronary events compared to their nondiabetic counterparts (1). Even among those who do not have clinically diagnosed diabetes, impaired glucose control, as indicated by increased levels of HbA1c, is a risk factor for atherosclerosis (2). Therefore, adequate control of hypertension in patients with diabetes or impaired glucose control is important for prevention of cardiovascular events. Large long-term randomized trials comparing active treatment to placebo have established that the treatment of hypertension is effective in preventing stroke, coronary events, heart failure, and death (3). The strongest evidence comes from trials in which low-dose diuretics were used as first-line agents. In hypertensive patients with diabetes, the relative benefit of the treatment of hypertension with low-dose diuretics is the same as in their nondiabetic counterparts, but the absolute reduction in cardiovascular events is about twice as great because the rate of events is so high in the diabetic population (1).
Several newer antihypertensive agents, including angiotensin-converting enzyme (ACE) inhibitors and calcium antagonists, are recommended for treating patients with diabetes, but the relative risk/benefit ratio in terms of prevention of cardiovascular events has not been established for these agents until recently. New data from both randomized clinical trials and observational studies among hypertensive patients with diabetes or impaired glucose control show striking differences in the risk for major cardiovascular events according to the class of antihypertensive treatment. This article reviews the data from three recently published randomized clinical trials of hypertensive patients with diabetes or pre-diabetes: the Fosinopril versus Amlodipine Cardiovascular Events Trial (FACET), the Appropriate Blood pressure Control in Diabetes (ABCD) trial, and the Multicenter Isradipine Diuretic Atherosclerosis Study (MIDAS). We also compare the event rates in the Systolic Hypertension in the Elderly Program (SHEP) and FACET. The findings of these trials have important implications for the treatment of hypertension in patients with diabetes.
The FACET was a randomized clinical trial designed to compare the effect of an ACE inhibitor, fosinopril, to that of the long-acting calcium antagonist amlodipine in patients with type 2 diabetes and hypertension (4). Participants were excluded if they had a history of coronary artery disease (CAD), stroke, congestive heart failure (CHF), or other relevant co-morbid conditions. Microal-buminuria >40 μg/min or serum creatinine >1.5 mg/dl were also exclusion criteria. A total of 380 patients were randomized and treated for up to 3.5 years (mean follow-up 2.8 years) to assess the effects of the drugs on serum lipids and glucose control. Prospectively defined cardiovascular events, blood pressure, and renal function were monitored as secondary outcomes. The randomized treatment was open-label fosinopril 20 mg/day or amlodipine 10 mg/day in a fixed dose. If blood pressure was not controlled the other study drug was added. A total of 108 patients, 58 in the fosinopril group and 50 in the amlodipine group, received the combination treatment. At the end of follow-up there were no significant differences between the two groups in total cholesterol, high-density lipoprotein (HDL) cholesterol, triglycerides, fasting glucose, insulin levels, HbA1c, serum creatinine, and microalbuminuria. Both treatment regimens were effective in significantly reducing systolic and diastolic blood pressures compared to baseline, but treatment with amlodipine led to a significantly greater reduction in systolic blood pressure than treatment with fosinopril (−19 mm Hg; 95% CI −22 to −15 mm Hg vs. −13 mm Hg, 95% −16 to −10 mm Hg). In an intention-to-treat analysis, patients randomized to fosinopril experienced a significantly lower rate of the combined end point of stroke, acute myocardial infarction (MI), or hospitalization for angina (RR 0.49; 95 CI 0.26-0.95; p = 0.030) compared to those allocated to amlodipine. The number of events and rates in each group are shown in Fig. 1.
To assess whether the increased risk for cardiovascular events found in the amlodipine group in FACET was associated with an increased pharmacologic effect of the drug, we have analyzed the maximum blood pressure reduction during follow-up according to treatment and the occurrence of cardiovascular events. Paradoxically, those who experienced a cardiovascular event and were randomized to amlodipine had, on average, a greater reduction in systolic blood pressure compared to those in the amlodipine group who did not experience events and compared to the entire fosinopril group (Fig. 2). The diastolic blood pressure reductions were similar in those with and without events in both the fosinopril and the amlodipine groups.
The ABCD trial was a prospective, randomized clinical trial in patients with type 2 diabetes, and was conducted to determine the effects of intensive and moderate blood pressure control and to compare the effects of a long-acting calcium antagonist, nisoldipine, and the ACE inhibitor enalapril (5). The primary outcome measure was glomerular filtration rate assessed by 24-h creatinine clearance. Secondary outcome measures were urinary albumin excretion, left ventricular (LV) hypertrophy, retinopathy, and neuropathy. Cardiovascular morbidity and mortality were also monitored as secondary end points. The study included two populations aged 40-74 years: 470 hypertensive patients who had a diastolic blood pressure ≥90 mm Hg and 480 normotensive patients. Patients were randomized to receive either intensive antihypertensive drug therapy or moderate antihypertensive drug therapy. Patients were also randomized to double-blind nisoldipine or enalapril. After an average of 5 years of treatment, the independent Data Safety Monitoring Board (DSMB) recommended early termination of the hypertensive arm of the trial because of a fivefold increase in fatal and nonfatal acute myocardial infarction (MI) in the nisoldipine group compared to the enalapril group (25/235 vs. 5/235; RR 5.5; 95% CI 2.1-14.6; p < 0.001) (6). The DSMB also recommended that hypertensive patients who had been receiving the long-acting nisoldipine be switched to enalapril. At baseline the enalapril group had a significantly lower mean level of HDL cholesterol and higher rates of angina and abnormal ankle-arm blood pressure indices than the nisoldipine group. After adjustment for these imbalances in baseline characteristics, the difference in risk for MI between the two groups was even greater than in the crude analyses (RR 7.0; 95% CI 2.3-21.4; Table 1). The risk for cerebrovascular events was also higher in the nisoldipine group, but the difference from the enalapril group did not reach statistical significance. The combined end point of all major cardiovascular events in ABCD has not yet been reported.
The MIDAS was a prospective randomized clinical trial designed to compare the 3-year effects of the intermediate-acting calcium antagonist isradipine to hydrochlorothiazide on the progression of carotid intimal medial thickness as assessed by carotid ultrasound (7). A total of 883 hypertensive patients were randomized and followed for up to 3 years. Patients with known diabetes were excluded. At the end of follow-up there were no significant differences between the two groups in the primary outcome, but those randomized to isradipine had a modestly higher risk for major cardiovascular events compared to those randomized to the diuretic (RR 1.78; 95% CI 0.94-3.38; p = 0.07). When the patients were stratified according to the median level of HbA1c, a measure of homeostatic glucose control, the relative risk for major cardiovascular events with isradipine compared to hydrochlorothiazide was 2.81 (95% CI 1.09-7.26; p = 0.04) among those with HbA1c ≥6.7% and was 1.11 (95% CI 0.43-2.90) among those with HbA1c <6.7% (8).
EVENTS IN SHEP AND FACET
To explore the issue of the risk/benefit ratio of ACE inhibitors, calcium antagonists, and diuretics compared to placebo in patients with diabetes, we have calculated the adjusted treatment-specific rates of cardiovascular events in the SHEP and in the FACET. The SHEP was a prospective, randomized clinical trial to assess whether the diuretic-based treatment of isolated systolic hypertension in the elderly was effective in decreasing the risk for stroke (9). A total of 4,736 older persons with isolated systolic hypertension were randomized to either placebo or chlorthalidone-based active treatment and were followed for 5 years. The active treatment was effective in decreasing the risk for stroke, coronary events, CHF, and all cardiovascular events. Further subanalyses in SHEP have shown that the active treatment was equally effective in decreasing the relative risk for cardiovascular events in patients with and without diabetes (1). For the present analyses, the diabetic population in SHEP was identified according to the methods described by Curb et al. (1). The patients were selected if they reported a history of diabetes, were using antidiabetic drugs, or had a fasting serum glucose >140 mg/dl. The data from 583 patients with type 2 diabetes from SHEP were compared to those of the FACET. Proportional hazards regression models were used to calculate the relative risks for events according to treatment after adjustment for baseline differences in age, gender, smoking habits, serum creatinine, systolic and diastolic blood pressure, and history of heart attack and stroke. The primary outcome of these analyses was the first occurrence of the combined end point of stroke, transient ischemic attack, MI, CHF, coronary artery bypass surgery, angioplasty, aneurysm, endarterectomy, sudden death, rapid cardiac death, or hospitalization for angina. These cardiovascular events were evaluated in a similar fashion in both SHEP and FACET.
All three treatments were associated with a lower rate of all cardiovascular events than placebo (Table 2). The greatest and only statistically significant reduction in risk compared to placebo was found in the fosinopril group. In the analyses of the combined end point of stroke or acute MI, both the diuretic and the ACE inhibitor were associated with the lowest risk for events. For this combined outcome, the rate in the calcium antagonist group was similar to that in the placebo group and, compared with placebo, the confidence interval for amlodipine was compatible with both a 50% reduction and a twofold increase in risk.
INTERPRETATION OF THE FINDINGS
Our exploratory analyses in SHEP and FACET should be interpreted with caution because the two trials took place in different settings and had different lengths of follow-up. Although it is difficult to compare event rates across trials even after adjustment for potential confounding factors, these data suggest that, among diabetic patients, low-dose diuretics and ACE inhibitors are associated with similar reductions in the risk for the combined end point of stroke and MI, and calcium antagonists and placebo are also associated with similar rates of these major cardiovascular events.
The FACET and ABCD trials have several common features (4,6). Both trials were single-site, prospective, randomized trials that enrolled patients with type 2 diabetes and hypertension. The randomized interventions included an ACE inhibitor and a long-acting calcium antagonist. The primary outcome was a surrogate measure in both trials, serum lipids in FACET and glomerular filtration rate in ABCD trial. Cardiovascular events were secondary outcomes in both trials. The excess in risk of events in the calcium antagonist group compared to the ACE inhibitor group was unexpected and cannot be explained by a better blood pressure control in patients receiving the ACE inhibitor or by differences in other measured risk factors. The notable difference between the two studies was that in FACET the medications were given open-label and in ABCD the intervention was double-blind. Another difference is that, in FACET when blood pressure was not controlled with monotherapy the patients (28%) received the other study drug in addition to the randomized treatment, whereas in ABCD other drugs were used as step-up treatment. Because 26% of the patients randomized to amlodipine were also taking fosinopril in FACET, this difference might explain the smaller relative risk for major cardiovascular events between the calcium antagonist and the ACE inhibitor in FACET than in ABCD (RR 2.0 in FACET and RR 5.5 in ABCD). These two trials used comparable interventions in similar patient populations, and the results in terms of cardiovascular outcomes were comparable.
The FACET, ABCD trial, and MIDAS raise the question of whether the cardiovascular findings in patients with diabetes or prediabetes are explained by a special benefit of the ACE inhibitors and diuretics, an adverse effect of calcium antagonists, or a combination of the two. Several potential mechanistic explanations are plausible. In patients with diabetes, the plasma levels of plasminogen activator inhibitor-1 (PAI-1), an inhibitor of the fibrinolytic system, are increased compared to nondiabetic counterparts, and this difference may account in part for an increased risk for stroke and MI (10). ACE inhibitors have been shown in several randomized trials to decrease PAI-1, an action that may explain how this class of drugs might provide special benefits in hypertensive patients with diabetes (11-13). Another potential special benefit of ACE inhibitors may be linked to their ability to decrease endothelin, a factor related to the severity of atherosclerosis. In small randomized trials of patients with heart failure (14) and in patients undergoing thrombolysis after an acute MI (15), fosinopril and captopril decreased the circulating levels of endothelin.
Other data suggest that potential adverse effects of calcium antagonists may be magnified in patients with diabetes. In vitro studies have shown that the binding of lipophilic drugs, such as amlodipine and other calcium antagonists, to the cell membrane depends largely on the membrane composition (16,17). A decrease in the cholesterol/phospholipid ratio in the membrane bilayer results in increased binding of the drug, a greater pharmacologic effect, and possibly an increased risk for toxic effects. Diabetes is characterized by changes in the cholesterol and phospholipid content of cell membranes that may increase the membrane partition coefficient of calcium antagonists (18), so that adverse effects of calcium antagonists might be more likely to occur in patients with diabetes (19). The data from FACET appear to support the hypothesis that the occurrence of events in the amlodipine group was associated with a pronounced pharmacologic effect of the drug, as reflected by the reduction in systolic blood pressure. Similar associations of a greater blood pressure reduction with an increased risk for cardiovascular events were found in the isradipine group of MIDAS (Robert Byington, personal communication). The positive association of a greater blood pressure reduction with cardiovascular events in the amlodipine group of FACET and in the isradipine group of MIDAS supports the intriguing hypothesis that blood pressure reduction alone may be an inadequate marker of drug efficacy.
In conclusion, FACET, ABCD, MIDAS, and the secondary analyses in SHEP suggest that ACE inhibitors and low-dose diuretics are superior to calcium antagonists for preventing cardiovascular events in hypertensive patients with diabetes or prediabetes. These results are supported by the published findings of two case-control (20,21) and two cohort (19,22) observational studies, and these observations apply to both the newer long-acting formulations and the short-acting calcium antagonists. The findings are strong and consistent, with risk ratios ranging from 2 to 7. Importantly, the conclusions are limited by the relatively limited number of events in each trial. Additional information is needed from trials that are already comparing calcium antagonists with ACE inhibitors and diuretics.
The SYST-EUR trial has shown that the intermediate-acting calcium antagonist nitrendipine was more effective than placebo in preventing stroke in older patients with isolated systolic hypertension (23). In contrast to active comparison studies, this was a placebo-controlled trial. To our knowledge, separate analyses in SYST-EUR according to diabetes vs. nondiabetes have not been published. Until data are available from the Antihypertensive and Lipid Lowering treatment to prevent Heart Attack Trial (ALLHAT) (24), clinicians should make treatment decisions based on the available data. Given the currently available data, clinicians and scientists are facing the following primary questions.
- Should patients with diabetes continue to be enrolled in new long-term clinical trials of calcium antagonists and ACE inhibitors?
- Should the ongoing long-term trials that compare calcium antagonists, ACE inhibitors, and other agents continue?
- In the clinical setting, how should the patients with hypertension and diabetes be treated?
Answers to these questions can be addressed from different perspectives.
The scientist's position
New long-term randomized trials that compare ACE inhibitors to calcium antagonists in hypertensive patients with diabetes should perhaps not be started at this time. There are already several ongoing long-term, randomized trials in hypertension, such as the ALLHAT, that examine a large number of patients with diabetes and are designed to assess the comparative effects of the individual antihypertensive agents in the prevention of cardiovascular events. Under intensive safety monitoring, these ongoing trials should be completed to provide the definitive evidence.
The clinician's position
An independent Committee, the Data Safety Monitoring Board of ABCD, made a clinical decision and recommended that hypertensive patients enrolled in ABCD who were receiving the calcium antagonist be reassigned to the ACE inhibitor. Until we have conclusive evidence in terms of prevention of cardiovascular events, it appears prudent to use ACE inhibitors and low-dose diuretics as the preferred first-line agents in hypertensive patients with diabetes or impaired glucose control.
1. Curb JD, Pressel SL, Cutler JA, et al. Effect of diuretic-based antihypertensive treatment on cardiovascular disease risk in older diabetic patients with isolated systolic hypertension
. Systolic Hypertension
in the Elderly Program Cooperative Research Group. JAMA
2. Vitelli LL, Shahar E, Heiss G, et al. Glycosylated hemoglobin level and carotid intimal-medial thickening in nondiabetic individuals. The Atherosclerosis Risk in Communities Study. Diabetes Care
3. Psaty BM, Smith NL, Siscovick DS, et al. Health outcomes associated with antihypertensive therapies used as first-line agents. A systematic review and meta-analysis. JAMA
4. Tatti P, Pahor M, Byington RP, et al. Outcome results of the fosinopril
cardiovascular events randomized trial in patients with hypertension
and NIDDM. Diabetes Care
5. Schrier RW, Estacio RO, Jeffers B. Appropriate Blood Pressure Control in NIDDM (ABCD) Trial. Diabetologia
6. Estacio RO, Jeffers BW, Hiatt WR, Biggi SL, Gifford N, Schrier RW. The effect of nisoldipine
as compared with enalapril on cardiovascular events in patients with non-insulin-dependent diabetes and hypertension
. N Engl J Med
7. Borhani NO, Mercuri M, Borhani PA, et al. Final outcome results of the Multicenter Isradipine Diuretic Atherosclerosis Study (MIDAS): a randomized controlled trial. JAMA
8. Byington RP, Craven T, Furberg CD, Pahor M. Isradipine, raised glycosylated hemoglobin, and risk of cardiovascular events. Lancet
9. SHEP Cooperative Research Group. Prevention of stroke by antihypertensive drug treatment in older persons with isolated systolic hypertension
. Final results of the Systolic Hypertension
in the Elderly Program (SHEP). JAMA
10. Juhan Vague I, Alessi MC, Vague P. Thrombogenic and fibrinolytic factors and cardiovascular risk in non-insulin-independent diabetes mellitus. Ann Med
11. Vaughan DE, Rouleau JL, Ridker PM, Arnold JM, Menapace PJ, Pfeffer MA. Effects of ramipril on plasma fibrinolytic balance in patients with acute anterior myocardial infarction. HEART Study Investigators. Circulation
12. Moriyama Y, Ogawa H, Oshima S, et al. Captopril reduced plasminogen activator inhibitor activity in patients with acute myocardial infarction. Jpn Circ J
13. Oshima S, Ogawa H, Mizuno Y, et al. The effects of the angiotensin-converting enzyme inhibitor imidapril on plasma plasminogen activator inhibitor activity in patients with acute myocardial infarction. Am Heart J
14. Galatius Jensen S, Wroblewski H, Emmeluth C, Bie P, Haunso S, Kastrup J. Plasma endothelin in congestive heart failure: effect of the ACE inhibitor, fosinopril
. Cardiovasc Res
15. Di Pasquale P, Valdes L, Albano V, et al. Early captopril treatment reduces plasma endothelin concentrations in the acute and subacute phases of myocardial infarction: a pilot study. J Cardiovasc Pharmacol
16. Mason RP. Probing membrane bilayer interactions of 1,4-dihydropyridine calcium channel blockers. Implications for aging and Alzheimer's disease. Ann NY Acad Sci
17. Mason RP, Moisey DM, Shajenko L. Cholesterol alters the binding of Ca2+ channel blockers to the membrane lipid bilayer. Mol Pharmacol
18. Nehal M, Venugopal P, Baquer NZ. Changes in the lipid composition of red blood cells in hyperglycemic rats. Biochem Int
19. Pahor M, Kritchevsky SB, Zuccalá G, Guralnik JM. Diabetes and risk of adverse events with calcium antagonists. Diabetes Care
20. Alderman MH, Madhavan S, Cohen H. Calcium antagonists and cardiovascular events in patients with hypertension
and diabetes. Lancet
21. Heckbert SR, Psaty BM, Kaplan RC, et al. ACE inhibitors and MI risk in diabetics with hypertension
22. Pahor M, Furberg CD, Psaty BM. Treatment of hypertensive patients with diabetes. Lancet
23. Staessen JA, Fagard R, Thijs L, et al. Randomised double-blind comparison of placebo and active treatment for older patients with isolated systolic hypertension
24. Davis BR, Cutler JA, Gordon DJ, et al. Rationale and design for the Antihypertensive and Lipid Lowering treatment to prevent Heart Attack Trial (ALLHAT). Am J Hypertens
Proceedings of a satellite symposium held at the 17th Scientific Meeting of the International Society of Hypertension June 7, 1998; Amsterdam, The Netherlands
Publication of this supplement was supported by a grant from Bristol-Myers Squibb