Search tips
Search criteria 


Logo of heartHeartVisit this articleSubmit a manuscriptReceive email alertsContact usBMJ
Heart. 2007 September; 93(9): 1011–1014.
PMCID: PMC1955006

More about the “ARB MI paradox


“Logic dictates that angiotensin converting enzyme inhibitors should remain the preferred drug across the entire spectrum of cardiometabolic disease”

Keywords: angiotensin, antagonist, infarction, myocardial, receptor
figure ht107185.f2
A Hall
figure ht107185.f3
M Strauss

Introduced into clinical practice in 1995, the angiotensin receptor blockers (ARBs) selectively antagonise the angiotensin II AT1 receptor, producing effects that include reduction of aldosterone release and hence, also, of salt and water retention. ARBs are effective for the treatment of hypertension, symptomatic congestive heart failure, and for the prevention of stroke. Furthermore, ARBs are effective in preventing progression of diabetic nephropathy and may also prevent atrial fibrillation. In all of these clinical situations, ARBs seem to share the same benefits as previously seen from treatment with angiotensin converting enzyme inhibitors (ACEi). ACEi and ARBs both reduce angiotensin AT1 receptor‐mediated events, albeit through distinct pharmacological mechanisms.

Mechanisms of action

The mechanism of action of ACEi differs from that of ARBs in a number of ways, which include differential effects on bradykinin breakdown, nitric oxide, prostacyclin, COX‐2 activity, and each has a diametrically opposite effect on the angiotensin II AT2 receptors. ACEi reduce activation of AT2 by reducing angiotensin II levels, whereas ARBs augment baseline angiotensin II levels and thereby augment AT2 activation. These differences, have been widely used to suggest that the newer ARBs might have important advantages over ACEi—and, in particular, might be more tolerable. Advocates of ARBs have been keen to suggest that “ARBs are ACEi without the cough”—namely, that ARBs have similar (or better) clinical efficacy, while at the same time being better tolerated. Based, at least in part, on this type of thinking, use of ARBs has increased to the extent that a similar amount of money is currently spent on ARBs in the UK each year (£45.9 million per quarter and 1.8 million items) as is spent on ACEi.1,2 This observation correlates with the fact that generic ACEi are cheaper (eg, ramipril capsules 5 mg costs £2.38/28 days; fig 11)) than non‐generic ARBs (eg, losartan 50 mg costs £18.09/28 days). The pattern of increased use of ARBs is incongruous with all the major guidelines, which recommend ACEi as first‐line treatment, and reserve ARBs for patients shown to be intolerant of ACE inhibition.3,4 It is fiscally responsible, as well as clinically important, to determine the relative merits of these drugs.

figure ht107185.f1
Figure 1 Cost comparison charts January 2007. Reprinted with permission from the Regional Drug and Therapeutics Centre (Newcastle).1

In 2004 the VALUE trial reported their findings based on a 4.1‐year comparison of the ARB, valsartan (up to 160 mg/day), and the calcium channel blocker, amlodipine (up to 10 mg/day), in 15 245 patients with high‐risk hypertension. One key observation was that myocardial infarction (MI) was significantly more common in patients given the ARB (hazard ratio = 1.19 (95% CI 1.02 to 1.38), p = 0.02), an effect that some have attributed to the inferior antihypertensive response of the ARB treatment.5,6 Importantly, a debate sparked which has focused on the mechanism, rather than on the reality, of a failure of ARBs to prevent the occurrence of MI.7,8 The fact that valsartan was a less effective antihypertensive drug than amlodipine is an observation that does little to recommend ARBs as first‐line treatment, regardless of the underlying mechanism for the excess MIs observed.

Effective dose

Some have suggested that the dose of valsartan used in VALUE may have been inadequate,9,10 and that 160 mg twice a day should be the recommended target dose, a dose that was used in the VALIANT trial. VALIANT compared the effects of high‐dose valsartan with captopril 50 mg three times a day in patients after MI heart failure.11 The tolerance profile of high‐dose valsartan relative to captopril puts this into a different perspective (table 11).). The occurrence of cough was less common with the ARBs, as also were rash and taste disturbances, both of which are symptoms directly attributable to the sulphydryl molecular subgroup that distinguishes captopril from the more commonly used carboxyl ACEi (eg, ramipril, perindopril). However, high‐dose valsartan was associated with a clear excess of hypotension and renal dysfunction, particularly when used in combination with captopril. Furthermore, the overall treatment intolerance rates were no different in the ACEi and ARB arms. This information challenges the perception that ARBs are intrinsically better tolerated than ACEi—especially when both are given at “effective doses”.

Table thumbnail
Table 1 Adverse events leading to treatment dose reduction or discontinuation in the VALIANT Trial11

ARB meta‐analyses

In August 2006, the journal Circulation published a detailed analysis of the “ARB‐MI paradox” with a set of papers debating whether ARBs may actively increase the risk of MI across a wide range of disease states.12,13 Both sets of authors agreed14,15 on the following key points:

  • ACEi can prevent both MI and cardiovascular (CV) death.
  • ARBs do not prevent MI or CV death.
  • ACEi should be considered to be better than ARBs in this respect.
  • ACEi should be used as first‐line treatment across the entire cardiometabolic vascular spectrum of disease.

The two papers differed slightly in their results for reported ARB meta‐analyses for the relevant randomised controlled trial data (table 22).). Strauss and Hall reported an 8% (95% CI 1% to 16%, p = 0.03) excess of MI for ARB as compared with non‐ARB control arms; while Tsuyuki and McDonald reported only a 3% (95% CI −7% to 13%, p = 0.59) excess of MI. Although some small differences in the MI rates reported for some trials were apparent, the main distinction between these two meta‐analyses related to (a) the studies included; (b) the statistical test used. Tsuyuki and McDonald claimed to have included all trials irrespective of study size, study duration, concomitant use of ACEi, or indeed the absence of published data. For example, the unpublished MI data from the MOSES Study16 was included based on a “personal communication. Unfortunately, the “communication” did not make it clear whether multiple events for individual patients were included—as reported for stroke in the main published analysis for MOSES.16 If so, then it is conceivable that a single patient randomised to nitrendipine might have had 20 recurrent events as compared with 17 patients given eprosartan who might each have had a fatal MI event. The clinical significance of either of these two possible scenarios is quite different. Although inclusivity is an important attribute of a meta‐analysis, this must be weighed against the use of appropriate, prespecified, systematic inclusion criteria. In this regard, it should be noted that other potentially relevant unpublished data were not included—for example, from the RESOLVD pilot study.17 Perhaps a “personal communication,” would have been relevant, especially considering that the RESOLVD pilot was stopped prematurely owing to excess CV events with candesartan.17

Table thumbnail
Table 2 Comparison of two meta‐analyses of ARB trials and incidence of MI12,13

In addition to the ARB meta‐analyses of Strauss and Hall and Tsuyuki and McDonald, two other ARB meta‐analyses (Verdecchia and Volpe) also confirm that ARBs do not reduce MI or CV death, and both showed a trend for an increase in MI,10 as did Tsuyuki. The lack of reduction in MI and CV death seen in all four ARB meta‐analyses contrasts sharply with an ACEi meta‐analysis by Strauss and Hall of over 150 000 patients. ACEi reduced MI by 14% and CV death by 12% in patients with similar CV risks as the ARB meta‐analysis (risk of MI about 6%, and risk of CV death about 9%). Of note, two recent meta‐regression analyses comparing ACEi with ARBs (BPLTTC)18 or calcium channel blockers (CCB),19 respectively, confirm that ACEi reduce MI and CV death by an additional 9–12% above that predicted by blood pressure lowering alone, confirming a benefit of ACEi that is independent of blood pressure.18

In conclusion, there is no dispute about the major clinical efficacy associated with the use of ACEi across a wide spectrum of CV disease states (table 33).12 In contrast, the efficacy of ARBs seem to be limited mainly to surrogate clinical end points that include blood pressure, renal function, and signs and symptoms of heart failure. Efficacy with regard to major clinical end points (other than stroke) seems to be much more limited with ARBs—and in the case of MI, absent all together. Furthermore, these distinct profiles are based on intention to treat analyses and so take into account any differences in treatment compliance/tolerance. Importantly, direct comparison of higher and presumably more “effective doses” of ARBs with ACEi shows tolerability to be equivalent.

Table thumbnail
Table 3 Meta‐analysis results for ACEi and ARBs assessing a range of “hard” cardiovascular end points12

Given these observations, and also the important concerns about the “ARB‐MI paradox” it seems pertinent to ask why healthcare providers should spend substantially more money (per prescription) on ARBs than ACEi. Logic dictates that ACEi should remain the preferred drug across the entire spectrum of cardiometabolic disease.


ACEi - angiotensin converting enzyme inhibitors

ARB - angiotensin receptor blocker

CV - cardiovascular

MI - myocardial infarction


Conflict of interest: Professor Hall has received research grants from Astra‐Zeneca, Servier UK, and Sanofi‐Aventis UK; has received honoraria from Astra‐Zeneca and Servier UK; and has been paid consultant fees by Servier UK. Dr Strauss has received honoraria from Sanofi‐Aventis, Pfizer, Abbott, and Tanabe; has served as an expert witness for Sanofi‐Aventis; and has served as a consultant/advisory board member for Sanofi‐Aventis and Pfizer.


1. Regional Drug and Therapeutics Centre (Newcastle) Cost comparison charts. January 2007. Available at (accessed 7 June 2007)
2. NHS Business Service Authority Prescription Pricing Division. PACT Centre Pages ‐ Cardiovascular prescribing Available at‐052004.htm (accessed 7 June 2007)
3. Cooper A, Skinner J, Nherera L. et al Clinical guidelines and evidence review for post myocardial infarction: secondary prevention in primary and secondary care for patients following a myocardial infarction. London: National Collaborating Centre for Primary Care and Royal College of General Practitioners, 2007. Available at (accessed 10 June 2007)
4. National Prescribing Centre The place in therapy of angiotensin II receptor antagonists. MeReC Bulletin. 2003;13. Available at (accessed 7 June 2007)
5. Ruilope L M, Segura J. Hope in life and value of blood pressure control. J Hypertens 2004. 222265–2266.2266 [PubMed]
6. Weber M A, Julius S, Kjeldsen S E. et al Blood pressure dependent and independent effects of antihypertensive treatment on clinical events in the VALUE Trial. Lancet 2004. 3632049–2051.2051 [PubMed]
7. Julius S, Kjeldsen S E, Weber M, for the VALUE trial group Outcomes in hypertensive patients at high cardiovascular risk treated with regimens based on valsartan or amlodipine: the VALUE randomized trial. Lancet 2004. 3632022–2031.2031 [PubMed]
8. Mann J, Julius S. The Valsartan Antihypertensive Long‐term Use Evaluation (VALUE) trial of cardiovascular events in hypertension. Rationale and design. Blood Press 1998. 7176–183.183 [PubMed]
9. Sarzani R, Dessa‐Fulgheri P. Angiotensin receptor blockers and myocardial infarction: the importance of dosage. J Hypertens 2006. 241679–1681.1681 [PubMed]
10. Mann D L, Deswal A. Angiotensin‐receptor blockade in acute myocardial infarction—a matter of dose. N Engl J Med 2003. 3491963–1965.1965 [PubMed]
11. Pfeffer M A, McMurray J J, Velazquez E J. et al Valsartan in Acute Myocardial Infarction Trial Investigators. Valsartan, captopril, or both in myocardial infarction complicated by heart failure, left ventricular dysfunction, or both. N Engl J Med 2003. 3491893–1906.1906 [PubMed]
12. Strauss M H, Hall A S. Angiotensin receptor blockers may increase risk of myocardial infarction: unraveling the ARB‐MI paradox. Circulation 2006. 114838–854.854 [PubMed]
13. Tsuyuki R T, McDonald M A. Angiotensin receptor blockers do not increase risk of myocardial infarction. Circulation 2006. 114855–860.860 [PubMed]
14. Strauss M H, Hall A S. Response to Tsuyuki RT amd McDonald MA. Circulation 2006. 114860
15. Tsuyuki R T, McDonald M A. Response to Strauss MH and Hall AS. Circulation 2006. 114855–860.860 [PubMed]
16. Schrader J, Luders S, Kulschewski A, et al; MOSES Study Group Morbidity and Mortality After Stroke, Eprosartan Compared with Nitrendipine for Secondary Prevention: principal results of a prospective randomized controlled study (MOSES). Stroke 2005. 361218–1226.1226 [PubMed]
17. McKelvie R S, Yusuf S, Pericak D. et al The RESOLVD pilot study investigators. Comparison of candesartan, enalapril, and their combination in congestive heart failure: randomized evaluation of strategies for left ventricular dysfunction (RESOLVD) pilot study. Circulation 1999. 1001056–1064.1064 [PubMed]
18. Blood Pressure Lowering Treatment Trialists' Collaboration, Turnbull F, Neal B, Pfeffer M, et al. Blood pressure-dependent and independent effects of agents that inhibit the renin-angiotensin system. J Hypertens 2007. 25951–958.958 [PubMed]
19. Verdecchia P, Reboldi G, Angeli F. et al Angiotensin‐converting enzyme inhibitors and calcium channel blockers for coronary heart disease and stroke prevention. Hypertension 2005. 46386–392.392 [PubMed]

Articles from Heart are provided here courtesy of BMJ Publishing Group