ARBs Preferable to ACE Inhibitors in Large Observational Hypertension Study

ARBs were shown not only to be as effective but also to carry a lower burden of side effects, including cough and angioedema.

ARBs Preferable to ACE Inhibitors in Large Observational Hypertension Study

In more than 3 million people worldwide treated for hypertension, ARBs were not only just as effective as ACE inhibitors at reducing the risk of major cardiovascular events, but also associated with a lower burden of side effects, including acute angioedema and cough, according to a new observational study.

The results, according to RuiJen Chen, MD (Columbia University Irving Medical Center, New York, NY), and colleagues, challenge the 2017 hypertension guidelines from the American Heart Association/American College of Cardiology (AHA/ACC) and the 2018 guidelines from the European Society of Cardiology/European Society of Hypertension (ESC/ESH), where ACE inhibitors and ARBs are given equal footing as first-line therapy for hypertension.

“Despite being equally guideline-recommended first-line therapies for hypertension, these results support preferentially starting ARBs rather than ACE inhibitors when initiating treatment for hypertension for physicians and patients considering renin-angiotensin system inhibition,” conclude the researchers, who reported their findings this week in Hypertension.

Senior investigator George Hripcsak, MD (Columbia University Irving Medical Center), said there have been no large randomized, head-to-head comparisons of ACE inhibitors versus ARBs and the studies that have been done are limited in size and include high-risk patients, such as those with diabetes or the elderly.

And since the drugs are all available as generics, “it is unlikely that a large-scale randomized trial will be carried out,” Hripcsak told TCTMD. “Yet we have suspected and further confirmed a difference between the drug classes. Therefore, for the sake of our patients, we must begin to trust observational studies that are done well, meaning they are fully open with careful analysis and careful diagnostics on large, diverse populations.”

For the sake of our patients, we must begin to trust observational studies that are done well. George Hripcsak

Nephrologist Jeffrey Turner, MD (Yale Medicine, New Haven, CT), who wasn’t involved in the study, said there has been an evolution in thinking about ACE inhibitors and ARBs in the decades since they were first approved (1981 for captopril, an ACE inhibitor, and 1995 for losartan, an ARB). Last year, Turner, along with Ravi Kodali, MD (Yale Medicine), wrote a provocative editorial asking if ACE inhibitors should ever be used for the treatment of hypertension.

“The short answer as to why ARBs should be the first-line therapy over ACE inhibitors is that though the risk is small, study after study has shown a higher incidence of side effects with ACE inhibitors, in particular angioedema, which can be deadly,” Turner told TCTMD. “It’s not a big risk. It’s a rare event, but the fact that we have a clear alternative with ARBs, which have a much lower incidence [of angioedema], does make the argument as to why we should ever put the patient at that slightly increased risk with an ACE inhibitor when we’re going to get equal efficacy when using an ARB.”    


There has been some debate as to whether ARBs provide the same protection from cardiovascular events as ACE inhibitors. In the earliest ACE inhibitor studies, the drugs were shown to reduce the risk of major cardiovascular events across a range of patients, but the same reduction in clinical events was not seen in the placebo-controlled ARB trials. Some hypertension experts have chalked this up to different eras in which the drugs were studied. Systematic reviews and meta-analyses published subsequently have largely concluded that ARBs have similar efficacy as ACE inhibitors.

The new analysis is part of the LEGEND project, which applies high-level analytics to the Observational Health Data Science and Informatics (OHDSI) network of studies. As part of LEGEND-HTN, the researchers set out to use advanced statistical methods to produce reliable results from available international databases, said Hripcsak. In this way, the project aims to help prioritize many of the available first-line therapies.

The researchers analyzed data from eight observational databases to study the relative risk of 55 different outcomes, including four primary effectiveness outcomes (acute MI, hospitalization for HF, ischemic or hemorrhagic stroke, and a composite endpoint that includes the three former endpoints plus sudden cardiac death). The remaining 51 outcomes represented safety events or adverse effects known or suspected to be associated with antihypertensive medications.

In total, 2,297,881 treatment-naive patients with hypertension started an ACE inhibitor and 673,938 started an ARB. The majority of patients started on an ACE inhibitor received lisinopril (80%) while the most commonly used ARB was losartan (45%).

Overall, there were no statistically significant differences in any of the major CVD events, including the composite endpoint. In contrast, use of an ACE inhibitor was associated with a significantly higher risk of acute pancreatitis (HR 1.32; 95% CI 1.04-1.70), angioedema (HR 3.31; 95% CI 2.55-4.51), cough (HR 1.32; 95% CI 1.11-1.59), and GI bleeding (HR 1.18; 95% CI 1.01-1.41). ACE inhibitors were also associated with a higher risk of abnormal weight loss.   

As part of the observational analysis, investigators used multiple approaches to identify and adjust for residual bias, including propensity-score matching and calibration of the risk model using positive and negative controls.

It challenges us to really believe what we experience, which is the sense that you get more complications from ACE inhibitors than with ARBs. Willie E. Lawrence

“We believe that this analysis is the most advanced observational study on this topic, both because of the confounding adjustment methods we used and because of the extensive diagnostics that we employ to detect residual bias,” said Hripcsak. “The 76 negative controls aim to uncover confounding and other bias that might be missed by our analysis and allow us to calibrate the result to account for systematic error.”

Overall, the results matched up to their expectations, said Hripcsak, although he cautioned there still may be some differences in effectiveness across different patient populations, or between specific drugs within a class. On the whole, though, there is no difference in effectiveness between ACE inhibitors and ARBs, he said. As to why ACE inhibitors remain more commonly prescribed despite their known side effects, Hripcsak believes it might simply come down to comfort given that they’ve been around longer. Also, prior to ARBs coming off patent, there had been a large difference in price, he noted.

Cough, Yes, but Pancreatitis and GI Bleeds?   

To TCTMD, Turner said the MI/ARB paradox has been discounted by the majority of physicians. In the ON-TARGET trial, which compared telmisartan against ramipril in patients at high risk of vascular events, there was no difference in the primary endpoint of death from cardiovascular causes, MI, stroke, or hospitalization, but more angioedema with the ACE inhibitor ramipril. Like others, Turner said the theory suggesting less protection with ACE inhibitors emerged from the placebo-controlled trials testing the drug classes in different eras.

“The event rates in the control group were vastly different because a generation had passed,” said Turner. “Our ability to manage cardiovascular events greatly improved. The signal that arose was really more because the event rates were so much lower with the ARBs than they were with the ACE inhibitors.”

Willie E. Lawrence, MD (Midwest Heart and Vascular Specialists, Kansas City, MO), who wasn’t involved in the study, said that physicians have historically leaned toward an ACE inhibitor over an ARB because they have been on the market much longer. The new study, however, challenges current thinking about going with ACE inhibitors first, noting that the results line up with his clinical experience.

“It challenges us to really believe what we experience, which is the sense that you get more complications from ACE inhibitors than with ARBs, mainly as it relates to cough that people develop,” said Lawrence. “It does give some support that we should choose an ARB over and ACE inhibitor in most instances.”

Lawrence, head of the AHA’s National Hypertension Control Initiative oversight committee, pointed out that the present study, while thought-provoking and well done, is observational.

“The study isn’t designed to answer the question—it’s designed to ask the question,” said Lawrence. “When you do these kinds of studies you can often get some bizarre results. In this one, they’re reporting a higher incidence of pancreatitis and GI bleeding. That speaks to the fact that this wasn’t a double-blind, controlled trial. It’s not a rigorous trial that we’d do to answer these types of questions. For me, I’ve never seen pancreatitis with an ACE inhibitor or an ARB. I’ve never seen GI bleeding with an ACE inhibitor or ARB. In these types of studies, you can sometimes see things that you really can’t explain.”

Similarly, Turner said pancreatitis and GI bleeding are not side effects encountered in his clinical practice. “I’m a little bit suspicious about this being a true side effect or if this a chance finding,” he commented.

Michael O’Riordan is the Associate Managing Editor for TCTMD and a Senior Journalist. He completed his undergraduate degrees at Queen’s…

Read Full Bio
  • Hripcsak reports grants from the National Library of Medicine and Janssen Research.
  • Lawrence and Turner report no relevant conflicts of interest.