Comparative Effectiveness of ACEIs and ARBs

Authors


To the Editor:

We thank McMurray and colleagues1 for their letter. They interpret a statistically significant protective association for valasartan in one trial2 and no significant association for ramipril in a separate trial3 as evidence that angiotensin receptor blockers are superior to angiotensin-converting enzyme inhibitors in preventing diabetes. However, the hazard ratios from these trials were actually quite similar (0.86 for valasartan vs 0.91 for ramipril) and cannot be directly compared because of the differences in the designs and populations of the trials. In our paper,4 we point out that Ongoing Telmisartan Alone and in Combination with Ramipril Global Endpoint Trial (ONTARGET), a prior head-to-head trial,5 found a slightly (but not statistically significantly) higher incidence of diabetes for telmisartan vs ramipril (hazard ratio, 1.12, 95% confidence interval, 0.97–1.29). Even though ONTARGET studied patients with vascular disease and our study was in patients with hypertension, the lack of consistency between ONTARGET and our results contributed to our cautious interpretation of the diabetes result (“warrants further investigation”).4

Randomized trials provide invaluable evidence for comparative effectiveness but are not available for every important study question in every population of interest. Therefore, well-controlled and cautiously interpreted nonrandomized studies will continue to provide valuable information to patients, clinicians, and policymakers.6

Ancillary