Perindopril+indapamide improves outcomes in high-risk pts with DM (ADVANCE)

Clinical Question

Is perindopril plus indapamide effective in decreasing bad outcomes in patients with type 2 diabetes and existing macrovascular disease?

Bottom Line

Perindopril (Aceon) plus indapamide (Lozol) is better than placebo in decreasing clinically relevant events in patients with type 2 diabetes who are at high risk of cardiovascular complications. Whether the combination is better than other medications -- like aspirin -- isn't addressed by this study. (LOE = 1b)


Patel A, et al, and the ADVANCE Collaborative Group. Effects of a fixed combination of perindopril and indapamide on macrovascular and microvascular outcomes in patients with type 2 diabetes mellitus (the ADVANCE trial): a randomised controlled trial. Lancet 2007;370(9590):829-840.  [PMID:17765963]

Study Design

Randomized controlled trial (double-blinded)


Industry + govt




Outpatient (any)


This study included more than 11,000 patients older than 55 years who had type 2 diabetes and either a prior cardiovascular event (eg, myocardial infarction, stroke, amputation) or at least one other risk factor for cardiovascular disease, including microvascular complications, tobacco use, an elevated cholesterol level, a low HDL cholesterol level, microalbuminuria, more than 10 years of diabetes, or more than 65 years of age. After completing a 6-week run-in period the "surviving" patients were randomly assigned to perindopril 2 mg plus indapamide 0.625 mg daily or matching placebo. Since the authors weeded out patients who didn't comply or who were unable to tolerate the side effects, any data about drop-outs after randomization are unlikely to reflect tolerability in the real world. After 3 months of treatment, the doses were doubled. The authors were interested in a composite outcome that included cause-specific mortality, cardiovascular events, new or worsening nephropathy, retinopathy, and so forth. I dislike composite end points like this; some outcomes don't make sense to combine. When there is no difference in the occurrence of a more frequent and devastating outcome (like death), but an improvement in some less important outcomes (like the number of photocoagulations as opposed to loss of vision), and the results all get lumped together, the devastating outcome suddenly looks good. For a more cogent example, all one has to do is look at all the bastardizations of the UKPDS trial (BMJ 2000;320:1720-23.). After an average of 4.3 years, the researchers found that 15.5% of the treatment group had a bad outcome compared with 16.8% of the placebo group (number needed to treat [NNT] = 77; 95% CI, 37 - 10,257). All-cause mortality was also lower in the treatment group (7.3% vs 8.5%; NNT = 84; 95% CI, 47 - 790) as was the total number of coronary events (8.4% vs 9.6%; NNT = 84; 95% CI, 45 - 734). Only 15 patients were lost to follow-up after randomization. During the follow-up sessions, the randomized treatment was continued in 83% of the treatment group and 87% of the placebo group.