NPC Archive Item: ACE-I plus ARB misses the target

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

The ONTARGET investigators.  Telmisartan, ramipril or both in patients at high risk for vascular events.  New Engl J Med 2008; 358; 1547-59

What is the background to this?
ACE-inhibitors (ramipril in the HOPE study and perindopril in the EUROPA study) have been shown to be effective in reducing the risk of cardiovascular (CV) events and death in patients at high risk of these events.  The evidence for angiotensin receptor blockers (ARBs) either alone or added to an ACE-inhibitor is less clear (see below).  This study compared the ARB telmisartan 80 mg/day, ramipril 10 mg/day and the combination of both drugs in patients at high risk of CV events but who did not have heart failure.

What did this study find?
After a median follow-up of 56 months, there was no significant difference in the rates of the primary outcome (a composite of death from cardiovascular causes, MI, stroke or hospitalisation for heart failure). The rate was 16.5% in the ramipril group, 16.7% in the telmisartan group (relative risk [RR] compared to ramipril 1.01, 95% confidence intervals [95%CI] 0.94 to 1.09) and 16.3% in the combination group (RR compared to ramipril 0.99, 95%CI 0.92 to 1.07).

More patients in the ramipril group than in the telmisartan group stopped treatment due to cough (4.2% vs 1.1%, RR 3.8, Number Needed to Harm [NNH] 32) or angioedema (0.3% vs 0.1%, RR 3, NNH 500) but more patients in the telmisartan group stopped treatment due to hypotensive symptoms (2.7% vs 1.7%, RR 1.6, NNH 100).  Patients taking the combination therapy were more likely than those taking ramipril alone to discontinue treatment due to hypotensive symptoms (4.8% vs 1.7%, RR 2.8, NNH 32), diarrhoea (0.5% vs 0.1%, RR 5, NNH 250) and renal impairment (1.1% vs 0.7%, RR 1.58, NNH 250)

How does this fit with the rest of the evidence?
Although ACE-inhibitors have been shown to be effective in patients at high risk of CV events, the evidence for ARBs is less clear. A meta-analysis in 2005 found that ARBs had no significant effect on risk of MI in patients at risk of CV events compared with placebo (Odds Ratio [OR] 0.94, 95%CI 0.75 to 1.16) or ACE-inhibitors (OR 1.01, 95%CI 0.87 to 1.16).  Later meta-analyses produced similar results, although the wide confidence intervals limit the interpretation of the results.

In patients with heart failure, the ELITE II study found no difference in outcomes with losartan compared to captopril.  In the VAL-HeFT study, valsartan improved CV outcomes in patients taking neither an ACE inhibitor nor a beta-blocker, or only one of these, but worse outcomes were seen in patients taking all three drugs.  In VALIANT (which was similar to ONTARGET except that patients also had heart failure and/or left ventricular dysfunction), valsartan was no more effective than captopril, and the combination of both drugs was equally effective but was less well tolerated.  In CHARM-alternative, candesartan reduced all cause mortality and hospital admissions in patients with heart failure unable to take an ACE-inhibitor.  In CHARM-added, adding candesartan to ACE inhibitors improved outcomes, but beta-blocker use was low (so patient were not optimally treated) and the doses of ACE inhibitors used were about half the recommended target doses for heart failure.

So what?
There seems little to be gained from adding telmisartan to an evidence-based dose of an evidence-base ACE-inhibitor in patients at high risk of CV events who do not also have heart failure. Indeed using the combination made the risk of adverse effects serious enough to cause patients to stop treatment (hypotensive episodes, diarrhea and renal failure) more likely.  In patients with heart failure, CHARM-added suggests a possible advantage from adding candesartan to an ACE inhibitor (if the ACE inhibitor cannot be used at maximum dose and also note the questions over concomitant therapy) but VALIANT suggests that adding valsartan to an ACE-inhibitor is not likely to be beneficial.  It is not known if these results are drug-specific or relate to ARBs as a class.

Many health professionals who prescribe ARBs instead of ACE-inhibitors do so because of the risk of cough associated with the ACE-inhibitors.  In this study, only 4% of patients stopped treatment with ramipril due to cough, and 1% of people taking telmisartan did so as well (although note that some patients dropped out for various reasons during a run-in phase).  From these data, 32 people need to be treated with an ARB rather than an ACE inhibitor to prevent one having to stop because of cough, which is similar to the figures seen in other studies. The reduced risk of cough with the ARB must be weighed against the increased risk of hypotensive episodes, which might lead to falls, etc.

Taking into account evidence of efficacy, tolerability and cost, the first choice angiotensin-system drug for people at high risk of CV disease, with or without heart failure, is an ACE-inhibitor such as ramipril.  ARBs are an alternative in patients truly not able to tolerate ACE-inhibitors due to cough, but note that cough can still occur with ARBs and also that ONTARGET suggests that ARBs are more likely to cause hypotensive symptoms.  The combination of ARB + ACE-inhibitor is not useful except in patients with heart failure, although ONTARGET suggests a possible risk of impaired renal function

Study details

Patients: 25,620 patients with cardiovascular disease or diabetes (~74% with coronary artery disease, ~49% with previous MI, ~37% with diabetes, ~12% smokers)
Intervention and comparisons: telmisartan 80 mg/day, ramipril 10 mg/day or both
Outcomes and results: see above

Sponsorship: Supported by a grant from Boehringer Ingelheim, The Heart and Stroke Foundation of Ontario and a Senior Scientist award from the Canadian Institutes of Health Research

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