Blood pressure drugs have no mortality benefit in diabetic patients

To the Editor:

We question the following statement in the article by Campbell et al (1): “Treatment of high blood pressure in people with diabetes results in large reductions in death...within a short period of time...” This statement was based on references 2 to 10. However, the first two references (2,3) found no mortality benefit from these angiotensin receptor blockers.

The next reference (4) reported a nonsignificant mortality difference after nine years from “tight blood pressure control”. Reference 5 is a subgroup of the Appropriate Blood Pressure Control in Diabetes (ABCD) trial. This trial showed no mortality benefit (P=0.8) from aggressive blood pressure control after 5.3 years, while reference 6 represented the balance of ABCD participants – those with higher baseline blood pressure – and demonstrated borderline significant mortality benefit (the given P=0.037 was erroneous).

The Action in Diabetes and Vascular Disease: Preterax and Diamicron MR Controlled Evaluation (ADVANCE) trial (7) reported P=0.03 for mortality, but 79 diabetic patients would have to take perindopril plus indapamide for five years to postpone the death of one patient (78 representing the number needlessly treated).

The Micro Heart Outcomes Prevention Evaluation (MICRO-HOPE) substudy (8) found a nonsignificant mortality difference emerging after two years on ramipril, and 31 diabetic patients would have to take this drug for 4.5 years to postpone one death. Interestingly, the authors propose that much of the action of this angiotensin-converting enzyme inhibitor may be from mechanisms other than those that lower blood pressure (11).

The Systolic Hypertension in Europe (Syst-Eur) trial (9) reported no significant mortality benefit (P=0.09) from calcium channel blockade after two years. Ominously, the study mentioned potential harm from calcium channel blockers in diabetic patients.

Reference 10 was a meta-analysis of diabetic and nondiabetic patients suffering a total of 17,000 major cardiovascular events. In the diabetic patients, the above angiotensin receptor blocker and angiotensin-converting enzyme inhibitor effects were reflected, but because treatment durations were not given, numbers needed or needlessly treated cannot be calculated.

The authors, therefore, are not supported by evidence when suggesting short-term “major reductions in death”, and diabetic patients must be told. What is urgently needed are numbers needed or needlessly treated for individual end points – ie, death, ischemic (nontransient ischemic attack) and hemorrhagic stroke, myocardial infarction, heart and kidney failure and microangiopathies – for each of the available blood pressure drugs, for one to five years of treatment, rather than ‘relative combined end point risk reductions’ without defined treatment periods.

With such numbers needed or needlessly treated (and it is likely that none of them are less than 100 patient treatment-years per end point, which in itself, would make clinical relevance doubtful), we can clearly inform patients and consider other avenues to tackle diabetes, which, after all, is mostly a preventable metabolic disease (insulin resistance secondary to excess visceral fat due to junk food addiction) and of which blood pressure can be a symptom but not the cause.