Think Before You Make a RAS Decision: Evaluating the Use of RAS Blockers in Patients with Diabetes


Written By

Reviewed By


Bangalore S, Fakheri R, Toklu B et al. Diabetes mellitus as a compelling indication for use of renin angiotensin system blockers: systematic review and meta-analysis of randomized trials. BMJ. 2016; 352: i438.

Hypertension affects more than 70% of patients with type 2 diabetes mellitus and further increases the risk of cardiovascular disease in this high risk population.1 While renin angiotensin system (RAS) blockers are clearly indicated in patients with heart failure, chronic kidney disease with proteinuria, and coronary artery disease (CAD), experts have come to different conclusions regarding their role as initial antihypertensive therapy for patients with diabetes (See Table 1).2,3,4 


Table 1. Selected Guideline Recommendations


Guideline or Clinical Practice Recommendation

Recommended Initial Hypertension Therapy

Non-Black Patients with Diabetes

Black Patients with Diabetes

Eighth Joint National Committee (JNC8) 20142

ACEi, ARB, CCB or Thiazide diuretic

CCB or Thiazide diuretic

American Society for Hypertension (ASH) and the International Society of Hypertension (ISH) 20133

 ACEi or ARB


“acceptable to start with a CCB or thiazide”

American Diabetes Association Standards of Care 20164†


ACEi = angiotensin converting enzyme inhibitor; ARB = angiotensin receptor blocker; CCB = calcium channel blocker
† ADA 2017 has since been released since original submission of commentary


Race may also impact the decision to use a RAS blocker for patients with diabetes.  African-Americans on average  have lower plasma renin levels, which may make RAS blockers less effective when used alone in typical doses.5


RAS inhibition not only has the capacity to reduce systemic blood pressure, but may also prevent the progression of target end organ damage.  Mechanistically, RAS blockers cause selective vasodilation to vital organs and increase concentrations of the endogenous vasodilator, bradykinin.1,6


Several landmark clinical trials, including placebo-controlled (HOPE)6 as well as active-comparator studies (FACET and ABCD)4, have documented improved cardiovascular outcomes from RAS inhibition.4 In addition, the BENEDICT and IDNT trials showed improved renal outcomes.7  However, RAS inhibition was not superior to calcium channel blockers or thiazide-type diuretics in the ALLHAT study – one of the largest clinical trial every conducted with more than 33,000 subjects.1  Given the size and duration of ALLHAT, even small differences could have been detected. Given disparate recommendations in clinical practice guidelines and the inconsistent findings from clinical trials, clinicians are left wondering:


Are RAS blockers superior to other anti-hypertensive agents for the prevention of cardiovascular and renal events in people with diabetes?


A recently published meta-analysis attempted to answer this question.8  The investigators included trials meeting two criteria: 1) randomized, controlled trials comparing RAS blockers to active comparator(s) in patients with diabetes or impaired fasting blood glucose, and 2) a sample size > 100 participants with a follow-up of at least one year.  Studies that enrolled patients with heart failure and those comparing ACE inhibitors (ACEi) to angiotensin receptor blockers (ARB) or combined an ACEi with an ARB were excluded. Three authors independently determined study eligibility, assessed bias risk, and performed data abstraction. Outcomes analyzed were death, end stage renal disease, major cardiovascular events including cardiovascular death, myocardial infarction, angina, stroke, heart failure, and  revascularization, as well as drug withdrawal owing to adverse events. The authors utilized an intention to treat approach comparing RAS blockers with other antihypertensive agents as well as individual comparisons to calcium channel blockers, diuretics, and beta-blockers.8


More than 11,120 trials published from 1998-2012 were screened. After eligibility assessment, 19 studies were included in the meta-analysis.  These studies enrolled 25,414 participants with diabetes with a mean follow-up time of 3.8 years resulting in >95,000 patient-years of follow-up. Mean age of patients ranged from upper 50 years to middle 70 years of age. Black patients represented 7-39% of the study populations in trials reporting race; however race was not reported in 13 trials.   Seventeen trials enrolled patients with diabetes and hypertension while three trials enrolled patients with diabetes and albuminuria. Patients with diabetes and confirmed CAD were only represented in two trials. Calcium channel blockers were compared to RAS blockers in 15 trials, while thiazide-type diuretics or beta-blockers were the comparitor in 5 of the identified studies.8  The primary endpoints were diverse among the included trials. (See Table 2) Eleven trials had a primary endpoint related to at least one major cardiovascular outcome.  Seven trials reported a renal outcome as a primary endpoint.9 


Table 2. Primary Endpoint Variability Among Included Trials9

Cardiovascular Outcomes



CAMELOT, diabetes subgroup; CASE-J, diabetes subgroup; JMIC-B; diabetes subgroup

Cardiovascular events

ANBP2, diabetes subgroup

Cardiovascular event or death


Combined fatal coronary heart disease or non-fatal MI

LIFE, diabetes subgroup

Composite of cardiovascular death, MI and stroke


Composite of MI, stroke, coronary revascularization, admission for CHF, or sudden cardiac death

MOSES, diabetes subgroup

Composite of total mortality and all cardiovascular events

STOP-Hypertension-2, diabetes subgroup

Fatal cardiovascular disease


First clinical endpoint related to diabetes, death related to diabetes and death from all causes

Renal Outcomes



ABCD [hypertensive]

Change in 24 hour CrCl

ABCD [normotensive]

Change in 24 hour CrCl


Composite endpoint of doubling of serum creatinine, development of ESRD, or death






Persistent microalbuminuria

Fogari et al

Urinary albumin excretion

Miscellaneous Outcomes




Common carotid artery intima-media thickness


Lipid profile and glucose metabolism

ABCD=Appropriate Blood Pressure Control in Diabetes; ALLHAT=Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial; ANBP2=Second Australian National Blood Pressure Study; BENEDICT=Bergamo Nephrologic Diabetes Complications Trial; CAMELOT=Comparison of Amlodipine vs Enalapril to Limit Occurrences of Thrombosis; CASE-J= Candesartan Antihypertensive Survival Evaluation in Japan; CHF = chronic heart failure; CrCl = creatinine clearance; ESRD= end stage renal disease; FACET=Fosinopril Versus Amlodipine Cardiovascular Events Randomized Trial; IDNT=Irbesartan Type II Diabetes Nephropathy Trial; JMIC-B=Japan Multicenter Investigation for Cardiovascular Disease-B;  J-MIND= Japan Multicenter Investigation of Antihypertensive Treatment for Nephropathy in Diabetes; LIFE=Losartan Intervention For Endpoint reduction; MI= myocardial infarction; MITEC=Media Intima Thickness Evaluation with Candesartan cilexetil; MOSES=Morbidity and Mortality After Stroke, Eprosartan Compared With Nitrendipine for Secondary Prevention; NAGOYA HEART= Comparison between valsartan and amlodipine regarding morbidity and mortality in patients with hypertension and glucose intolerance; NESTOR=Natrilix SR versus Enalapril Study in Type 2 diabetic hypertensives with microalbuminuria;  STOP-Hypertension-2= Swedish Trial in Old Patients with Hypertension; UKPDS 39=UK Prospective Diabetes Study Group.


No statistical difference was demonstrated when comparing RAS blockers to other antihypertensive agents in patients with diabetes for all outcomes assessed (See Table 3). Not surprisingly, the ALLHAT trial contributed most heavily — representing 50% or more of the outcome events for death, stroke, and heart failure. Heterogeneity — the amount of variability in the observed outcomes across the included studies — was measured using the I2 statistic.  A value less than 25% is considered low heterogeneity.  In the majority of comparisons, heterogeneity was low except for myocardial infarction (I2= 48.1%, P=0.032), heart failure (I2= 47.7%, P=0.039), end-stage renal disease (I2= 47.7%, P=0.089), and drug withdrawal owing to adverse effects (I2= 48.4%, P=0.060).8


Table 3. Outcomes with RAS Blockers Compared to Other Antihypertensive Agents in Patients with Diabetes8



Relative Risk (95% Confidence Interval)



 0.99 (0.93 to 1.05)

Cardiovascular Death


1.02 (0.83 to 1.24)

Myocardial Infarction


0.87 (0.64 to 1.18)

Angina Pectoris


0.80 (0.58 to 1.11)



1.04 (0.92 to 1.17)

Heart Failure


0.90 (0.76 to 1.07)



0.97 (0.77 to 1.22)

End Stage Renal Disease


0.99 (0.78 to 1.28)

Drug Withdrawal Owing to Adverse Effects


0.80 (0.61 to 1.05)


Comparing RAS blockers to individual antihypertensive classes did not reveal any statistically significant differences for the majority of outcomes assessed.  RAS blockers were superior to calcium channel blockers in preventing heart failure (RR = 0.78; 95% CI 0.70-0.88; I2 = 0.0%).  This result was largely driven by the ALLHAT trial. Compared to diuretics or beta-blockers, RAS blockers did not demonstrate superiority in any outcome assessed.9


This study has several limitations. Ideally, a meta-analysis combines and statistically analyzes studies with similar patient populations and outcomes to answer a clinical question. Although a comprehensive search was used to identify the trials included in this meta-analysis, the baseline characteristics and makeup of the study populations included in these studies were quite different.  This limits direct comparisons.8 One noteworthy element missing in the inclusion criteria was a diagnosis of hypertension, and the study population did include normotensive patients as well as those with hypertension. Other differences within the study populations were the presence or absence of cardiovascular or renal disease, the degree of renal dysfunction (if present), and race.


In addition to differing patient populations, primary endpoints were considerably different. Renal outcomes were assessed in only 37% of trials. Cardiovascular events were a secondary outcome in > 40% of the trials.8 Some trials included cardiovascular and renal outcomes as endpoints, while others used surrogate markers (Table 2).  Thus, the diversity of the studies included in the meta-analysis make it difficult to determine the true effect of the intervention.11  


The evidence to support the use of RAS blockers over other major antihypertensive agents is underwhelming. Placebo controlled trials, such as HOPE, likely were the primary factor making RAS blockers the preferred first-line agents for decades. This meta-analysis in a heterogeneous population failed to show superiority of RAS blockers over other antihypertensive agents for the prevention of hard cardiovascular and renal outcomes. Interestingly, the 2017 ADA Standards of Care revised their recommendations.12  Rather than favoring an ACEi or ARB for patients with diabetes and hypertension, the ADA now considers all four main classes of agents to be appropriate first line agents, citing this meta-analysis as supporting evidence. Is this meta-analysis strong enough to end the debate? Does the choice of initial antihypertensive agent really matter given that most patients require two or more agents to adequately control blood pressure?7 What do you think?