NPC Archive Item: A2RAs: no substantially increased risk of MI, but ACE inhibitors still first line

NOTE – This is an archive post from the NPC and has not been updated since first publication. Therefore, some hyperlinks may no longer be working.
MeReC Rapid Review NPC Logo

28 June 2011

A large meta-analysis provides evidence that angiotensin-II receptor antagonists (A2RAs) do not substantially increase the risk of MI, all-cause or CV mortality, or angina, as was once a concern from earlier data. However, there was no significant reduction in risk of these outcomes associated with A2RAs compared with placebo. When compared with active controls, the picture is more complicated, but this meta-analysis provides no compelling evidence to favour A2RAs over ACE inhibitors as first line drug.

Level of evidence:
Level 2 (good quality patient-oriented evidence) according to the SORT criteria.

Prescribers should continue to follow NICE guidance and use an ACE inhibitor as first choice renin-angiotensin system (RAS) drug whenever a drug from this class is indicated. A2RAs (also called angiotensin-receptor blockers [ARBs]) should be used only for patients where an RAS drug is indicated, but an ACE inhibitor has had to be discontinued because of an intolerable ACE inhibitor-induced cough. Combination therapy with an A2RA plus an ACE inhibitor has a limited role and then only in certain indications.

Prescribing managers should review local prescribing trends for RAS drugs as suggested in the document Key therapeutic topics 2010/11 – Medicines management options for local implementation, produced by the NPC as part of the QIPP programme.

What is the background to this?
In all NICE guidance where renin-angiotensin system drugs are indicated, i.e. in hypertension, heart failure, post-MI, type 2 diabetes, type 1 diabetes and chronic kidney disease, ACE inhibitors are recommend as the first-line RAS drugs of choice. They have a more robust evidence base than A2RAs for all indications in terms of evidence for efficacy and safety; and with regard to most factors related to individual patients. This evidence has been summarised in this elearning event and in a MeReC Bulletin from March 2010. NICE guidance on management of hypertension is being updated and will reflect the latest evidence.

In 2004 an editorial in the BMJ raised concerns about increased rates of myocardial infarction (MI) with A2RAs versus other therapies, particularly ACE inhibitors. This was controversial: for example, two companion articles published in Circulation in 2006 elaborated on the arguments for and against this risk.

Since the concerns were originally raised, two meta analyses (MAs), by McDonald et al (19 studies, n=31,569) in 2005 and Volpe et al (20 trials, n=108,909) in 2009, both found no statistically significant difference in risk of MI between A2RAs and placebo and between A2RAs and ACE inhibitors, but could not rule out a small difference. The authors of the MA discussed here included 37 randomised controlled trials (RCTs) of 147,020 participants. They used a technique called sequential trial analysis to try to obtain firm evidence about the risk of A2RAs and MI.

What does this study claim?
The meta-analysis found no statistically significant difference in the relative risk (RR) of MI, all-cause mortality, cardiovascular (CV) mortality or angina (separately) compared with placebo, or active comparators, or overall. Full results are given in the ‘study details’ section, below. The authors state that the sequential trial analysis suggested firm evidence for absence of even an average 7.5% relative risk increase (upper confidence limit 11%) for MI, an average 4% relative risk increase (upper confidence limit 7%) in death, an average 5% relative risk increase (upper confidence limit 5%) in CV death and an average 15% relative risk increase (upper confidence limit 12%) in angina with A2RAs compared with controls.

There was a statistically significant reduction in risk of stroke with A2RAs compared to placebo (RR 0.91, 95% confidence interval [CI] 0.85 to 0.98) and overall (RR 0.90, 95%CI 0.84 to 0.98) but not compared to active treatment (RR 0.91, 95%CI 0.80 to 1.04). There was a statistically significant reduction in risk of heart failure with A2RAs compared to placebo (RR 0.85, 95%CI 0.79 to 0.92) and active treatment (RR 0.88, 95%CI 0.80 to 0.98) and overall; and in the risk of diabetes compared to placebo (RR 0.89, 95%CI 0.85 to 0.94) and active treatment (RR 0.77, 95%CI 0.64 to 0.93) and overall.

So what?
This MA and sequential trial analysis provides reassurance that A2RAs do not substantially increase the risk of MI, all-cause or CV mortality, or angina (see the background to this study, above). The authors acknowledge that they cannot rule out an absolute increase of MI less than 0.3% (NNT >333 over a mean of 3.33 years) compared with controls, but suggest that “this is likely to be clinically less important”.

However, patients taking A2RAs do so in the hope that they will derive a CV benefit, not merely that they will not be harmed by doing so. This MA could not detect a statistically significant benefit from A2RAs regarding the outcomes above when compared with placebo, whereas evidence exists for ACE-inhibitors for a benefit on risks of mortality, MI, coronary heart disease events, stroke, and hospitalisation for heart failure (see the MeReC Bulletin). It is fair to say that, as A2RA studies were generally conducted more recently than ACE inhibitor studies, greater use of other preventative drugs, such as statins, antiplatelets and beta-blockers in A2RA studies is a possibility. The law of cumulative benefits (or diminishing returns) means that statistically significant benefit can become increasingly difficult to demonstrate with the addition of each preventative drug. Nevertheless, prescribers must work with the best evidence available: we have clear evidence of benefit over placebo with ACE inhibitors and an absence of such evidence for A2RAs for these outcomes, and hypothesising about what might have been found had A2RAs been tested in exactly the same patients included in older ACE inhibitor trials must remain speculation.

The MA did find a benefit from A2RAs versus placebo in regard to stroke, heart failure, and new onset diabetes. However, even if the absence of evidence for benefit in other outcomes is ignored, this does not provide an argument for choosing A2RAs first line, since the choice is not between A2RA and placebo but between A2RA and an alternative active treatment, primarily an ACE inhibitor.

The MA is less helpful in this regard, since the comparisons with active treatment are complicated by the different patient populations examined (with their different baseline risks), and the variety of active comparators used. In eight trials the active comparator was an ACE inhibitor, but in the 14 others the comparator agents included diuretics, calcium channel blockers, and atenolol. The MA found a benefit in regard to stroke from A2RA compared to placebo, but not compared to active controls. The largest single A2RA vs. ACE inhibitor RCT was ONTARGET, which recruited 25,620 patients with cardiovascular disease or diabetes, and compared A2RA, ACE inhibitor and combination therapy. In this study, telmisartan 80mg/day did not produce significantly different reduction in the primary outcome (a composite of death from cardiovascular causes, MI, stroke or hospitalisation for heart failure), or any constituents of this composite, compared to ramipril 10mg/day. This study also found no significant difference in risk of new-onset diabetes (RR [A2RA vs ACE inhibitor] 1.10, 95%CI 0.95 to 1.26). None of the individual RCTs in the MA which found a reduced risk of diabetes or heart failure with A2RAs versus active controls compared A2RAs with ACE inhibitors.

Many health professionals who prescribe A2RAs instead of ACE-inhibitors do so because of the risk of cough associated with the ACE-inhibitors. In ONTARGET, only 4% of patients stopped treatment with ramipril due to cough compared with 1% of people taking telmisartan (although some patients from both groups dropped out for various reasons during a run-in phase). The NNT was 32; in other words, an average of 1 in 32 people treated with telmisartan for 56 months avoided developing an intolerable cough, who would have done had all 32 taken ramipril. The other 31 developed, or did not develop a cough, just as they would have done had they taken ramipril. On the other hand, more patients in the telmisartan group developed intolerable hypotensive symptoms and had to stop treatment than in the ramipril group (2.7% vs 1.7%, RR 1.6, NNH 100). The reduced risk of cough with the A2RA must be weighed against the increased risk of hypotensive episodes, which might lead to, for example, falls.

Taking all this into account it seems sensible to continue to use ACE inhibitors as first choice RAS drugs, reserving A2RAs only for patients where an RAS drug is indicated, but an ACE inhibitor has to be discontinued because of an intolerable ACE inhibitor-induced cough (and other possible causes of the cough have been excluded). Combination therapy with an A2RA plus an ACE inhibitor has a limited role and then only in certain indications, as discussed in the MeReC Bulletin.

Study details
Bangalore S, et al. Angiotensin receptor blockers and risk of myocardial infarction: meta-analyses and trial sequential analyses of 147 020 patients from randomised trials. BMJ 2011;342:d2234

Design: Systematic review of RCTs with meta-analysis and trial sequential analysis

Patients: 37 RCTs included 147,020 participants and had a total follow-up of 485,166 patient years (mean follow up 3.3 years per trial). Trials included patients with a variety of conditions (in different combinations) including acute MI, atrial fibrillation, coronary artery disease, diabetes (type 1 and type 2), heart failure, hypertension, nephropathy, renal impairment, and stroke

Intervention and comparison: A2RAs versus placebo or active controls (various)

Outcomes and results
Results of the random effects MA are shown in the table.


Relative risk (95%CI)

Vs. placebo

Vs. active control



0.93 (0.81 to 1.07) NS

1.04 (0.98 to 1.11) NS

0.99 (0.92 to 1.07) NS

All-cause mortality

0.99 (0.95 to 1.03) NS

0.99 (0.94 to 1.06) NS

1.00 (0.97 to 1.02) NS

CV mortality

0.97 (0.91 to 1.02) NS

1.01 (0.94 to 1.10) NS

0.99 (0.94 to 1.04) NS


0.97 (0.90 to 1.04) NS

0.87 (0.72 to 1.06) NS

0.95 (0.85 to 1.06) NS


0.91 (0.85 to 0.98)

0.91 (0.80 to 1.04) NS

0.90 (0.84 to 0.98)

Heart failure

0.85 (0.79 to 0.92)

0.88 (0.80 to 0.98)

0.87 (0.81 to 0.93

New onset diabetes

0.90 (0.85 to 0.94)

0.77 (0.64 to 0.93)

0.85 (0.78 to 0.93)

NS = not statistically significant. Statistically significant outcomes highlighted in bold

Sponsorship: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors

More information on A2RAs can be found on the Renin-angiotensin system drugs section of the NPC website

Please comment on this rapid review in the NPC discussion rooms, or using our feedback form.

Make sure you are signed up to NPC Email updates — the free email alerting system that keeps you up to date with the NPC news and outputs relevant to you