Você está na página 1de 4


Angiotensin Receptor Blockers Do Not Reduce

Risk of Myocardial Infarction, Cardiovascular
Death, or Total Mortality
Further Evidence for the ARB-MI Paradox

Opposing Viewpoint, see p 2085 Martin H. Strauss, MD,

Downloaded from http://circ.ahajournals.org/ by guest on June 1, 2017

Individual trials of angiotensin II receptor blockers (ARBs) and systematic meta-anal-

yses have repeatedly demonstrated the ARB class to significantly reduce systemic
blood pressure, stroke, and the subsequent development of heart failure and diabe-
tes mellitus.1 However, no individual trial or meta-analysis has observed an impact
of ARB treatment on the incidence of myocardial infarction (MI), cardiovascular mor-
tality, or all-cause mortality (Table). All-cause mortality is the most comprehensive
summary indicator of cardiovascular benefit of treatment2 and, consequently, it is
surprising that ARBs have no impact given the other clinical benefits we have just
A systematic review/meta-analysis conducted by Bangalore et al3 demonstrated
that in 20 trials of ARB versus placebo (n=66282) and 21 trials of ARB versus active
comparator (n=39738), ARB in no individual trial improved all-cause mortality. Fur-
thermore, when the trials were analyzed in combination, the impact of ARB on mor-
tality was precisely zero compared with placebo (odds ratio, 1.01; 95% confidence
limit, 0.961.06) and virtually zero compared with an active comparator (odds ratio,
0.99; 95% confidence limit, 0.951.03). Placebo-controlled trials are the most rigor-
ous method of assessing treatment efficacy (as well as harm), and, consequently,
it is concerning that other meta-analyses of ARBs versus placebo-controlled trials
have also reported a lack of mortality reduction in patients with high risk (Savarese),4
those with diabetes mellitus (Cheng),5 and patients with hypertension (Thomopoulos
et al),6 as well as no reduction in MI or cardiovascular mortality (Table). The opinions expressed in this
Thomopoulos et al6 also reported the impact of different drug classes compared article are not necessarily those
with placebo in the treatment of systemic hypertension. Indapamide, calcium chan- of the editors or of the American
nel blockers, and angiotensin-converting enzyme inhibitors (ACEIs) all statistically Heart Association.
significantly reduced mortality, whereas 4 other drug classes reduced mortality but Correspondence to: Martin H.
not to a degree that was of statistical significance. ARB was the only drug class Strauss, MD, North York General
that failed to reduce death compared with placebo, despite representing the largest Hospital, 107-4800 Leslie St,
Toronto, Ontario, Canada M2J
group of patients studied (n=65256) and despite an average blood pressure reduc-
2K9. E-mail dr.marty@bellnet.ca
tion of 3.7/2.0 mmHg. Because stroke was reduced by ARBs, cardiovascular and
all-cause mortality must have been driven by an excess risk of fatal MI and sudden Key Words: angiotensin II
receptor blocker angiotensin
death. ACEIs but not ARBs reduced coronary heart disease (including nonfatal MI) converting enzyme inhibitor
and cardiovascular death (including fatal MI6), despite both being inhibitors of the cardiovascular mortality
renin-angiotensin-aldosterone system. hypertension meta-analysis
ACEIs suppress angiotensin II, thereby preventing pathological effects, but they myocardial infarction
also prevent the breakdown of bradykinin, thereby inducing additional cardioprotec- 2017 American Heart
tive effects. This profile is distinct from the one achieved by use of an ARB that Association, Inc.

2088 May 30, 2017 Circulation. 2017;135:20882090. DOI: 10.1161/CIRCULATIONAHA.117.026112

Angiotensin Receptor Blockers: Mortality Is Not Reduced

Table. Risk of Myocardial Infarction, Cardiovascular Mortality, and All-Cause Mortality in Parallel Meta-
Analyses of Placebo-Controlled Trials of Angiotensin-Converting Enzyme Inhibitors and Angiotensin Receptor
ACEI vs Placebo ARB vs Placebo
Myocardial Cardiovascular All-Cause Myocardial Cardiovascular All-Cause
Infarction Death Death N Infarction Death Death N
High risk, Bangalore 0.83 0.83 0.89 0.93 1.02 1.01
62398 66282
et al3 (0.780.9) (0.70.99) (0.801.0) (0.851.03) (0.921.14) (0.961.06)
High risk, Savarese4 0.81 0.9 0.91 0.9 1.03 1.01
53791 54421
(0.750.88) (0.781.03) (0.850.98) (0.81.02) (0.851.26) (0.941.08)
Diabetes mellitus, 0.83 0.89 1.21 1.03
NA 21997 NA 13304
Cheng5 (0.700.99) (0.790.99) (0.811.8) (0.891.18)
Hypertension, 0.87 0.91 1.03 1.01
NA 49440 NA 65256
Thomopoulos et al6 (0.780.98) (0.850.98) (0.941.13) (0.971.06)
Values indicate hazard ratio (95% confidence interval). ACEI indicates angiotensin-converting enzyme inhibitor; ARB, angiotensin II receptor blocker.

induces selective antagonism of the AT1 angiotensin II (n=146838) by the BPLTTC (Blood Pressure Lowering
Downloaded from http://circ.ahajournals.org/ by guest on June 1, 2017

receptor, inhibiting a negative feedback loop and con- Treatment Trialists Collaboration), as well as being in-
sequently increasing angiotensin II levels by 200% to dependent of blood pressure lowering.1 For any given
300%. This may lead to MI as a result of stimulation of blood pressure reduction, ACEIs also reduce the risk of
the AT2 receptors, which are markedly upregulated and coronary heart disease by an additional 9% (P=0.002),
active in atheromatous plaques.1 ACEIs in meta-analyses independently of the effects of blood pressure lowering.
of placebo-controlled trials conducted in parallel to the In contrast, ARBs had no blood pressureindependent
ARB meta-analyses discussed above (Thomopoulos et effects on coronary heart disease; rather, there was a
al,6 Savarese,4 Cheng,5 Bangalore et al3) demonstrated small, nonsignificant increase in the risk of harm of 7%
a robust 9% to 11% risk reduction in all-cause mortality, (7%: 95% confidence interval, 7 to 24; P=NS), a phe-
10% to 17% risk reduction in cardiovascular mortality, nomenon we have called the ARB MI paradox. Specifi-
and 17% to 19% risk reduction in MI (all P<0.05; Table). cally, it is paradoxical that no net benefits are observed
The divergent effects of ACEI and ARB on both all-cause with ARBs. Furthermore, and most important, for any
and cardiovascular mortality are despite their compara- given blood pressure reduction, ACEIs reduced the risk
tor arms having similar risks of both (7.8% versus 9%, of MI and death an additional 15% (P=0.002) above that
4.7% versus 5.2%, respectively).3 of an ARB.
These observations are consistent with an earlier Head-to-head trials of ARB versus ACEI are of inter-
meta-analysis1 that reported that ARBs versus all com- est, but there is a paucity of data. In the ONTARGET
parators (11 trials, n=55050) did not reduce mortality trial (Ongoing Telmisartan Alone and in Combination With
(odds ratio, 1.01; 95% confidence interval, 0.961.06, Ramipril Global Endpoint Trial),2 the largest ACEI versus
P=0.8) and the risk of MI actually increased 8% (odds ra- ARB controlled trial reported to date (n=17118), the
tio, 1.08; 95% confidence interval, 1.011.16; P=0.03), ARB telmisartan was not statistically equivalent to the
with that increase qualitatively apparent in 9 of the trials ACEI ramipril for a combined cardiovascular end point.
and statistically significant in 2. This article also included The US Food and Drug Administration approved telmisar-
a parallel meta-analysis of ACEI trials. ACEIs versus all tan as a second-line therapy for those high-risk patients
comparators (39 trials, n=150943) reduced the relative who are ACEI intolerant (New Drug Application 20850).
risk of total mortality 9% (P<0.0001) and MI by 14% This decision was influenced by a lack of superiority of
(P<0.0001). The divergent effects in the parallel ACEI telmisartan compared with placebo in the parallel TRAN-
and ARB meta-analyses on MI and death were marked, SCEND trial (Telmisartan Randomized Assessment Study
despite their comparator arms having similar cardiovas- in ACE Intolerant Subjects With Cardiovascular Disease),
cular risks; rates of global death, MI, and stroke (cere- consistent also with the HOPE-3 trial (Heart Outcomes
brovascular accident) were essentially no different (13% Prevention Evaluation-3) conducted in patients at inter-
versus 14%, 5.8% versus 6.3%, and 4.2% versus 4.4%, mediate risk (n=12705; follow-up, 5.6 years)4 in whom
respectively). In addition, blood pressure lowering com- candesartan combined with hydrochlorothiazide failed to
parable to the risk reduction in stroke, a blood pressure reduce a combination of cardiovascular events despite
dependent phenomenon, was also similar. a robust reduction in blood pressure of 6.0/3.0 mmHg.
Divergent cardiovascular effects of ACEIs and ARBs Logic dictates that practice guidelines should recognize
were also confirmed in a meta-regression analysis the unique cardiovascular benefits of ACEIs and their prefer-

Circulation. 2017;135:20882090. DOI: 10.1161/CIRCULATIONAHA.117.026112 May 30, 2017 2089

Strauss and Hall

ential use compared with ARBs. Were such advice to be giv- dox. Circulation. 2006;114:838854. doi: 10.1161/CIRCULA-
en, the predicted impact on lives saved would be profound. TIONAHA.105.594986.
2. Ferrari R, Boersma E. The impact of ACE inhibition on all-cause
and cardiovascular mortality in contemporary hypertension trials:
a review. Expert Rev Cardiovasc Ther. 2013;11:705717. doi:
DISCLOSURES 10.1586/erc.13.42.
Dr Strauss has received speaker honorarium from and par- 3. Bangalore S, Fakheri R, Toklu B, Ogedegbe G, Weintraub H, Mes-
ticipated in advisory boards for Servier. Dr Hall has received serli FH. Angiotensin-converting enzyme inhibitors or angioten-
sin receptor blockers in patients without heart failure? Insights
speaker honorarium from Servier.
from 254,301 patients from randomized trials. Mayo Clin Proc.
2016;91:5160. doi: 10.1016/j.mayocp.2015.10.019.
4. Savarese G, Costanzo P, Cleland JG, Vassallo E, Ruggiero D,
AFFILIATIONS Rosano G, Perrone-Filardi P. A meta-analysis reporting effects of
From North York General Hospital, Toronto, Ontario, Canada angiotensin-converting enzyme inhibitors and angiotensin recep-
tor blockers in patients without heart failure. J Am Coll Cardiol.
(M.H.S.); and Leeds MRC Medical Bioinformatics Centre,
2013;61:131142. doi: 10.1016/j.jacc.2012.10.011.
Leeds, West Yorkshire, United Kingdom (A.S.H.).
5. Cheng J, Zhang W, Zhang X, Han F, Li X, He X, Chen J. Effect of
angiotensin-converting enzyme inhibitors and angiotensin II recep-
tor blockers on all-cause mortality, cardiovascular deaths, and
FOOTNOTES cardiovascular events in patients with diabetes mellitus: a meta-
Circulation is available at http://circ.ahajournals.org. analysis. JAMA Intern Med. 2014;174:773785. doi: 10.1001/
Downloaded from http://circ.ahajournals.org/ by guest on June 1, 2017

6. Thomopoulos C, Parati G, Zanchetti A. Effects of blood pres-

sure lowering on outcome incidence in hypertension, 4: effects
REFERENCES of various classes of antihypertensive drugs: overview and me-
1. Strauss MH, Hall AS. Angiotensin receptor blockers may in- ta-analyses. J Hypertens. 2015;33:195211. doi: 10.1097/
crease risk of myocardial infarction: unraveling the ARB-MI para- HJH.0000000000000447.

2090 May 30, 2017 Circulation. 2017;135:20882090. DOI: 10.1161/CIRCULATIONAHA.117.026112

Angiotensin Receptor Blockers Do Not Reduce Risk of Myocardial Infarction,
Cardiovascular Death, or Total Mortality: Further Evidence for the ARB-MI Paradox
Martin H. Strauss and Alistair S. Hall

Circulation. 2017;135:2088-2090
doi: 10.1161/CIRCULATIONAHA.117.026112
Downloaded from http://circ.ahajournals.org/ by guest on June 1, 2017

Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright 2017 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7322. Online ISSN: 1524-4539

The online version of this article, along with updated information and services, is located on the
World Wide Web at:

Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Circulation can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial
Office. Once the online version of the published article for which permission is being requested is located,
click Request Permissions in the middle column of the Web page under Services. Further information about
this process is available in the Permissions and Rights Question and Answer document.

Reprints: Information about reprints can be found online at:


Subscriptions: Information about subscribing to Circulation is online at: