Purpose. The physiological and clinical data for using angiotensin-converting- enzyme (ACE) inhibitors to prevent new-onset type 2 diabetes mellitus are reviewed.
Summary. ACE inhibitors have established their role in hypertension, primary and secondary prevention of cardiovascular events, and prevention of progression to and worsening of renal function. However, their ability to preserve pancreatic function and prevent new-onset diabetes is also coming to the forefront. Secondary analyses of large-scale clinical trials, such as the Captopril Prevention Project, the Heart Outcomes Prevention Evaluation, and the Studies of Left Ventricular Dysfunction trial, are revealing the potential benefits of these agents in diabetes prevention. However, the results of such studies have limited application because they are secondary analyses and, in some cases, were conducted 10 or more years after the original study. Enrollees were evaluated using different diagnostic guidelines for diabetes or not formally evaluated at all. Even in the most recent of the trials, the validity of the results is questionable because researchers coadministered a disease-modifying drug with the ACE inhibitor, potentially blunting the results. While intense lifestyle modifications are still superior in the prevention of new-onset diabetes, patients and providers will continue to investigate new options for preventing the progression of impaired fasting glucose to diabetes, though this delay does not correlate with a decrease in morbidity and mortality.
Conclusion. ACE inhibitors may preserve pancreatic function and prevent new-onset diabetes, especially for patients who are hypertensive with impaired glucose tolerance. Large studies investigating the effect of ACE inhibitors on the prevention of diabetes as a primary outcome are needed to determine the use for this indication.
- Angiotensin-converting-enzyme inhibitors
- Clinical studies
- Diabetes mellitus
- Mechanism of action
Diabetes affects more than 20.8 million adults throughout the United States and an estimated 180 million adults worldwide.1,2 Ninety percent of patients diagnosed with diabetes have type 2 diabetes mellitus, and one third of persons with type 2 diabetes remain undiagnosed.3,4 In some instances, the development and appearance of diabetes-associated complications precede the formal diagnosis of diabetes.3 Various strategies for preventing progression from impaired fasting glucose to diabetes have been evaluated, including lifestyle modifications, biguanides, and thiozolidinediones.5–8 Translational research implicates a correlation between angiotensin II and progression to type 2 diabetes and the potential role of angiotensin-converting-enzyme (ACE) inhibitors for preventing type 2 diabetes.
Free fatty acid, renin-angiotensin system, and insulin
Insulin resistance leads to increased lipolysis and decreased lipoprotein lipase activity, thereby creating excess free fatty acids (FFAs) and further insulin resistance and β-cell dysfunction.9,10 Elevated FFA levels and hyperglycemia also increase the production of reactive oxygen species, resulting in excess oxidative stress on the pancreas and leading to eventual islet cell fibrosis and decreased insulin gene transcription.11,12 In addition to impaired insulin production, “glucotoxicity” impairs peripheral glucose uptake.13,14 The reninangiotensin system (RAS) is a key physiological pathway that factors into the control of the cardiovascular and renal systems.6,7 The RAS plays an indirect role in FFA metabolism, as increased RAS activity is linked to syndromes of hypertension and inadequate FFA metabolism, leading to impaired glucose tolerance.15
ACE inhibitors work to block the conversion of angiotensin I to angiotensin II, which is a potent vasoconstrictor that regulates blood pressure and may be implicated in cardiac and vascular hypertrophy.16 Initially, ACE inhibitors were used to lower blood pressure and protect the kidneys. Within the past 10 years, clinical trials have shown that ACE inhibitors have cardioprotective benefits for high-risk patients, including patients with diabetes.7 ACE inhibitors may also affect the pancreas.1,7 The exact physiology of angiotensin II in the pancreas and endocrine system is still not fully understood, though research has demonstrated a link among the RAS, FFA metabolism, endogenous kinin activity, first-phase insulin response, and islet-cell perfusion.17–21
Blocking the conversion of angiotensin I to angiotensin II leads to an increase in bradykinin.1 In animals, there appears to be improved insulin sensitivity with increased endogenous kinins, which are associated with some ACE inhibitors.22 The increased kinin levels lead to an increased production of both prostaglandins and nitric oxide, thereby improving muscle sensitivity to insulin and exercise-induced glucose metabolism. In addition, ACE inhibitors prevent the formation of angiotensin II and vasoconstriction, thereby improving skeletal muscle blood flow and increasing pancreatic islet-cell perfusion; the end result is improved glucose uptake and insulin release.7,22 In short, inhibiting RAS will decrease insulin resistance, improve peripheral glucose handling and uptake, and improve insulin release.15
Knowler et al.5 evaluated intense lifestyle modifications versus metformin or placebo for the prevention or delay of new-onset diabetes in the Diabetes Prevention Program. The frequency of type 2 diabetes was reduced by 58% (95% confidence interval [CI], 48–66%; p < 0.001) in the lifestyle modification group and 31% (95% CI, 17–43%; p < 0.001) in the metformin group compared with placebo. The frequency of diabetes with lifestyle modifications was reduced by 39% (95% CI, 24–51%) when compared with metformin. The results demonstrated that lifestyle modifications were more effective than drug therapy in reducing the onset of type 2 diabetes.
The Finnish Diabetes Prevention Study Group evaluated intense lifestyle modifications on new-onset type 2 diabetes mellitus in patients at high risk for developing diabetes.23,24 The intervention group was statistically superior to the control group in remaining physically active (83% versus 71%, p < 0.05), losing more than 5% of their body weight (46% versus 14%, p < 0.05) and meeting fat and fiber intake goals (37% versus 20%, p < 0.05, and 37% versus 23%, p < 0.05, respectively). At the end of the study, there was a 58% relative risk (RR) reduction in the frequency of diabetes with intense lifestyle modifications compared with the control group (hazard ratio [HR], 0.4; 95% CI, 0.3–0.7; p < 0.001).
Even though the results with oral medications were not as noteworthy as lifestyle changes versus placebo, oral therapies still had a significant effect. Despite the reduction of the incidence of type 2 diabetes resulting from lifestyle modifications, these options may not be an ideal regimen for specific patient populations, such as the elderly and persons with limited physical ability. In these cases, the use of pharmacologic agents for diabetes prevention may be an acceptable option.3,7,25
ACE inhibitors and diabetes prevention
Currently, ACE inhibitors and angiotensin II-receptor blockers are recommended to prevent cardiovascular disease and nephropathy in patients with type 2 diabetes.3,7,26 Clinical data from randomized controlled trials have validated the physiological effects of angiotensin II on the pancreas. Though none of the trials involving ACE inhibitors used new-onset diabetes as a primary endpoint and each trial had its own limitations, the reproducibility of the results is encouraging.
The Captopril Prevention Project (CAPPP) was a prospective, randomized, open-label trial with a blinded endpoint evaluation.27 The objective of this study was to compare cardiovascular morbidity and mortality in hypertensive patients using an ACE inhibitor or “conventional antihypertensive” treatment, which included diuretics (most commonly hydrochlorothiazide and bendrofluazide) and β-blockers (most commonly atenolol and metoprolol). A total of 10,985 patients age 25–66 years with a diastolic blood pressure of ≥100 mm Hg on two occasions were enrolled and randomly assigned to receive either captopril (50–200 mg daily) or conventional antihypertensive treatment for 6.1 years. New-onset diabetes mellitus, defined per 1985 World Health Organization criteria,28 was set as a secondary outcome for this study. At baseline, diabetes was more prevalent in the captopril group (5.6%) than the conventional therapy group (4.8%). However, by the study’s end, the prevalence of diabetes mellitus in the captopril-treated group was significantly less than that in the conventional treatment group (RR, 0.86; 95% CI, 0.74–0.99; p = 0.039).
Two major concerns noted about the trial were the randomization method and the control medications used. Patients were randomized using a computer-generated number sequence and then assigned using sealed envelopes. Baseline height, weight, systolic and diastolic blood pressures, and diabetes status significantly differed between groups, which may have created a bias that favored the intervention group. Responses after publication of the CAPPP questioned if researchers unsealed the envelopes before enrolling patients, thereby influencing study inclusion decisions.29 The primary researchers denied such claims. The control medications used in the CAPPP, thiazide diuretics and metoprolol, may be considered diabetogenic and the cause of increased new-onset diabetes.15 The secondary results of the recently published Anglo-Scandinavian Cardiac Outcomes Trial—Blood Pressure Lowering Arm (ASCOT-BPLA) further support this metabolic effect in that the group treated with a β-blocker (with or without a thiazide) had more patients develop diabetes compared with the amlodipine-treated group (with or without an ACE inhibitor).30
The reduced rate of diabetes in high-risk patients was further assessed as a secondary analysis in the Heart Outcomes Prevention Evaluation (HOPE).31,32 This was a double-blind, randomized, placebo-controlled trial assessing the use of ramipril in preventing cardiovascular death in high-risk patients (history of coronary artery disease, stroke, peripheral vascular disease, or diabetes plus at least one other cardiovascular disease risk factor). The rate of new-onset diabetes was included as a part of a secondary analysis. A total of 9297 patients age 55 years or older with evidence of vascular disease or diabetes in addition to other cardiovascular risk factors were assigned to receive either ramipril 10 mg daily or placebo for a mean of 5 years. Of these patients, 2837 (61.1%) in the ramipril-treated group and 2883 (62%) in the control group did not have a formal diagnosis of diabetes at the beginning of the study. Fewer patients in the ramipril-treated group received a new diagnosis of diabetes (n = 102 [3.6%] versus n = 155 [5.4%] in the control group) (RR, 0.66; 95% CI, 0.51–0.85; p < 0.001).
As with most of the clinical literature about the prevention of new-onset diabetes with ACE inhibitors, the HOPE was not designed to investigate this as a primary outcome. While this evidence is similar to data found in other clinical trials, further evaluation will define its role in clinical practice.15,33
The Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT), a randomized, double-blind, active-control trial, evaluated the effects of chlorthalidone versus lisinopril, amlodipine, or doxazosin on the composite outcome of fatal coronary heart disease (CHD) and nonfatal myocardial infarction (MI).34 Doxazosin treatment was discontinued early due to a lack of efficacy and a significant increase in primary outcomes versus treatment with chlorthalidone.35 All remaining agents had similar effects on the composite endpoint of fatal CHD and nonfatal MI when compared with chlorthalidone (amlodipine versus chlorthalidone: RR, 0.98; 95% CI, 0.90–1.07; and lisinopril versus chlorthalidone: RR, 0.99; 95% CI, 0.91–1.08). Patients using lisinopril had a decreased risk of new-onset diabetes compared with patients treated with chlorthalidone. Of the patients beginning the trial without diabetes and a blood glucose concentration of <126 mg/dL, the frequency of new-onset diabetes in the chlorthalidone, amlodipine, and lisinopril groups was 11.6%, 9.8%, and 8.1%, respectively, at four years. The amlodipine and lisinopril groups were not compared for this outcome, but both had significantly fewer cases of new-onset diabetes than the chlorthalidone group.
Similar to the findings of CAPPP and ASCOT-BPLA, the ALLHAT demonstrated an ACE inhibitor’s superiority over thiazides in reducing the rate of new-onset diabetes. However, the trial did note some limitations that could influence outcomes. One such limitation was the use of β-blockers as add-on therapy for patients not achieving goal blood pressure with monotherapy. At five years, researchers reported that 43% of the lisinopril-treated population and 40.7% of the chlorthalidone-treated population were receiving either step 2 (atenolol, clonidine, or reserpine) or step 3 (hydralazine) therapy or both. The trial did not discuss how many persons received the individual agents. Since β-blockers may negatively affect blood glucose levels, it is difficult to interpret if their use may have blunted the diabetes-prevention benefit of the ACE inhibitor or worsened the results of the thiazide. However, the results were still clear that each agent had equal potential to prevent the primary endpoint, suggesting that preventing the progression to diabetes did not affect the frequency of adverse cardiovascular and cerebrovascular outcomes.
A subpopulation of the Studies Of Left Ventricular Dysfunction (SOLVD) trial was evaluated for the effect of enalapril on the incidence of diabetes in patients with left ventricular dysfunction (LVD).36,37 The original SOLVD trial sought to evaluate the effects of enalapril on morbidity and mortality in patients with LVD. Patients with asymptomatic LVD were randomly assigned to receive 2.5–20 mg daily of enalapril (n = 2111) or placebo (n = 2117) for a mean of 3.1 years. Diabetes was defined using the American Diabetes Association’s (ADA’s) diagnostic guidelines.3 For the subpopulation of the patients reevaluated, 40 met the criteria of having new-onset diabetes: 9 (5.9%) of the 153 in the enalapril group and 31 (22.4%) of the 138 in the placebo group (adjusted RR, 16.5%; RR, 0.26; p < 0.0001). For patients with impaired fasting glucose at enrollment, 1 patient (3.3%) of 55 developed diabetes in the enalapril group, compared with 12 (48.0%) of 55 in the placebo group (adjusted RR, 45%; p < 0.0001). The addition of enalapril was the variable most influential in decreasing the rate of new-onset diabetes (HR, 0.22; 95% CI, 0.10–0.46; p < 0.0001).
Several factors must be considered when evaluating the limitations of this trial. First, it was a post hoc analysis of a trial conducted 11 years prior. Second, the subpopulation studied was a small set of the original population and may not have adequately represented the study population. Third, diabetes was not a primary endpoint evaluated by the researchers. Finally, the population studied (heart failure patients with an ejection fraction of <35%) was a relatively exclusive population, thus limiting the applicability of the results to other patient populations.36
The above mentioned trials provide persuasive evidence regarding the decreased rate of diabetes in patients taking ACE inhibitors, though some discrepancies persist. First, new-onset diabetes was not a primary endpoint in any of the studies. Second, no absolute definition of diabetes was used throughout all the trials.1 In order to further evaluate the rate of new-onset type 2 diabetes in patients treated with ACE inhibitors, clinicians and researchers need prospective, randomized trials assessing new-onset diabetes as a de-fined primary endpoint.1,38
The results of the ramipril treatment group of the Diabetes Reduction Approaches with Ramipril and Rosiglitazone Medications (DREAM) trial were published in 2006.1,39,40 In this double-blind, randomized, placebo-controlled trial, 5269 patients without cardiovascular disease but with impaired fasting glucose (fasting plasma glucose [FPG] concentration of >100 mg/dL) received up to 15 mg of ramipril daily (n = 2623) or placebo (n = 2646) and were followed for a median of three years. The patients in this trial were concurrently randomly assigned to receive rosiglitazone, ramipril, placebo, or a combination thereof by a 2 × 2 factorial design.40 The study’s primary outcome was new-onset diabetes or death. The diagnosis of diabetes was made using ADA’s diagnostic guidelines.3 Diabetes or death occurred in 475 patients (18.1%) in the ramipril treatment group compared with 517 patients (19.5%) receiving placebo (HR, 0.91; 95% CI, 0.81–1.03; p = 0.15). There were 31 recorded deaths in the ramipril group compared with 32 in the placebo group; however, these results were not significantly different. Further, the difference in the number of patients who developed diabetes (449 in the ramipril group and 489 in the control group) was not significant (HR, 0.91; 95% CI, 0.80–1.03). Additional secondary outcomes included returning to euglycemia (FPG concentration of <110 mg/dL and two-hour postload glucose concentration of <140 mg/dL). A total of 1116 patients (42.5%) receiving ramipril and 1012 patients (38.2%) receiving placebo returned to euglycemia and normal two-hour postload glucose levels (HR, 1.16; 95% CI, 1.07–1.27; p = 0.001). While ramipril had a significant advantage over placebo in returning glucose levels to euglycemic ranges, it failed to meet its primary outcome of preventing progression to new-onset diabetes or death.
The DREAM trial faced much criticism for its design, duration, and results. The results did not show that ACE inhibitors prevented progression to diabetes as did the CAPPP, HOPE, SOLVD trial, and ALLHAT. The authors discussed various reasons for this difference, including the prospective study design (all previous studies were secondary or post hoc analyses), lack of diligent preenrollment diabetes assessment in other trials, and exclusion of high-risk patients or those with known cardiovascular disease from the DREAM trial. They also commented on the use of placebo versus control medications (β-blockers and thiazides) and the duration of the study (three years versus four or more years for previous trials).40,41 Surprisingly, the authors failed to mention the concurrent use of rosiglitazone in the DREAM trial. Since patients were randomized to receive ramipril, rosiglitazone, placebo, or any combination, ramipril was not solely compared with placebo. In fact, the authors did not report the percentage of patients in the ramipril and placebo groups receiving rosiglitazone, possibly affecting the number of cases of new-onset diabetes.
The secondary analyses of the previously mentioned studies are promising for the use of ACE inhibitors in the prevention of new-onset diabetes. However, with the exception of the DREAM trial, they have limited application because they are secondary analyses and, in some cases, were conducted 10 or more years after the original study. Enrollees were evaluated using different diagnostic guidelines for diabetes or not formally evaluated at all. Even in the most recent of the trials, the validity of the results is questionable because researchers coadministered a disease-modifying drug with the ACE inhibitor, potentially blunting the results.
The results of an ongoing trial may provide additional evidence. The Ongoing Telmisartan Alone and in Combination with Ramipril Global Endpoint Trial (ONTARGET) is a double-blind, parallel-group analysis of three different treatments: telmisartan 80 mg, ramipril 10 mg, and telmisartan 80 mg plus ramipril 10 mg.42 The primary endpoint of the trial is composite cardiovascular death, MI, stroke, or hospitalization. The trial’s secondary endpoints include the physiological effects of angiotensin II, particularly its role in diabetes mellitus, nephropathy, dementia, and atrial fibrillation. A total of 25,620 patients were enrolled and being followed for 3.5–5.5 years.16,43 New-onset diabetes is only a secondary objective, but the trial’s results may be used to further support the findings of the CAPPP, ALLHAT, and SOLVD trial while further questioning the results of the DREAM trial.
The ultimate question is if preventing new-onset diabetes really matters in overall morbidity and mortality. None of the results of the previously mentioned trials indicated improved morbidity and mortality when using an ACE inhibitor versus another antihypertensive. In trials such as the CAPPP, ALLHAT, and ASCOT-BPLA, where researchers’ control groups received active medications, the rates of death and nonfatal cardiovascular events were equal, regardless of the agent used. Even though the rate of new-onset diabetes was greater in the groups not treated with an ACE inhibitor (compared with a thiazide and β-blocker), morbidity and mortality outcomes were not affected. In fact, certain endpoints in the nondiabetic population in the ALLHAT were better with a thiazide rather than an ACE inhibitor, including stroke (RR, 1.23; 95% CI, 1.05–1.44), combined cardiovascular disease (RR, 1.12; 95% CI, 1.05–1.19), and heart failure (RR, 1.20; 95% CI, 1.04–1.38). Again, these results should be interpreted with caution because the study was not powered to detect differences in the subgroups.
As more evidence supports the benefits of ACE inhibitors in preventing diabetes and delaying the progression of impaired fasting glucose to diabetes, ACE inhibitors may become the agents of choice for patients with impaired fasting glucose and concurrent hypertension. However, the lack of overall difference in outcomes when compared with active controls diminishes this potential. The additional benefit of preventing the progression of impaired fasting glucose to diabetes and equivalent morbidity and mortality prevention may be enough to move ACE inhibitors to the front of the line with thiazides for the treatment of hypertension, but the choice of agents should be made on a patient-specific basis.
ACE inhibitors may preserve pancreatic function and prevent new-onset diabetes, especially for patients who are hypertensive with impaired glucose tolerance. Large studies investigating the effect of ACE inhibitors on the prevention of diabetes as a primary outcome are needed to determine the use for this indication.
- Copyright © 2008. American Society of Health-System Pharmacists, Inc. All rights reserved. 1079-2082/04/0602-1242$06.00