Diabetogenic effects of “old” antihypertensive drugs
Data obtained from observational studies have shown that patients with untreated hypertension display an increased incidence of type II diabetes mellitus compared to the normotensive state (1).
They have also shown that this phenomenon is made worse when thiazide diuretics or beta-blockers are administered chronically, i.e. that new onset diabetes is more frequent in hypertensive patients treated with these drugs than in untreated individuals, particularly when their administration is combined (2-4).
This issue has become hotly debated, because of its implications in terms of the choice of drugs to employ for initial antihypertensive treatment in seeking to ensure maximal cardiovascular protection for the patient. The question is whether physicians should
1) disregard the above metabolic inconvenience, given the excellent blood pressure lowering effect of thiazide diuretics and beta-blockers as well as their ability to protect hypertensive patients from cardiovascular disease,
2) give great importance to this inconvenience and refrain from any large use of these drugs or
3) privilege a “flexible” attitude, i.e. impose no limits as to the use of diuretics and beta-blockers when patients are at high cardiovascular risk - and thus needing an effective and timely blood pressure reduction to less than 140 mmHg systolic and 90 mmHg diastolic-, but use them with caution when, as in younger or mild hypertensive patients, the more limited benefit to be expected from a blood pressure reduction may be neutralised or superseded by the increased cardiovascular risk associated with the occurrence of diabetes.
An additional problem related to this last opinion is that of identifying those patients at high risk of becoming diabetics in whom diuretics and beta-blockers should not commonly represent first choice drugs.
New onset diabetes and “new” antihypertensive drugs
One of the most striking findings from recent clinical trials has been the one concerning the class-differential effects on new-onset diabetes. Indeed, calcium channel blockers, ACE-inhibitors and angiotensin II receptor blockers all have a remarkable impact on the risk of new-onset diabetes, whereas any impact is less frequent when compared with older drug classes over relatively short trial durations of 3-6 years. There has been some debate over whether such data reflect a prodiabetic effect of older drugs, or an antidiabetic effect of the new ones. Evidence from trials, however, also shows a reduced incidence of new-onset diabetes in patients treated with ACE inhibitors or angiotensin II receptor blockers as compared with placebo (3,5,6), which suggests a “true” antidiabetic effect of new drugs. In this regard, some other trials are important.
This is particularly the case when it comes to the Valsartan Long-term Use Evaluation (VALUE), which showed a significant benefit for the angiotensin II receptor blocker valsartan over amlodipine, a metabolically neutral calcium channel blocker (7).
Clinical value of drug-induced diabetes
A possibility which has been advanced is that the apparent diabetogenic effect of diuretics and beta-blockers simply reflects a “cosmetic” increase in blood glucose, with none of the adverse cardiovascular effects of true type 2 diabetes.
Evidence that this is not the case has been provided by the CAPtopril Prevention Project (CAPPP) study performed on 6886 hypertensive patients enrolled into a treatment programme between 1973 and 1993 for an average of 6.3 years (8).
For most of the treatment period, diuretics or beta-blockers were the first-line drugs. As expected, the incidence of cardiovascular disease during the study was significantly associated with raised blood glucose, whether measured at baseline or during treatment. Furthermore, patients who did not have raised blood glucose at baseline, but did have significantly raised levels (> 7.75 mmol/l) at some point in the study, had a 50% increased incidence of cardiovascular disease, a rate very similar to that of patients with raised blood glucose at baseline.
Furthermore, in another study, a treatment-induced increase in blood glucose in patients aged 50 years was the major predicting factor for the occurrence of myocardial infarction at the age of 60 years (9). Based on these data, it seems that increases in blood glucose associated with antihypertensive treatment are just as serious as those that are a result of conventional risk factors.
Many more questions need to be answered to understand what a diabetic condition induced by antihypertensive drugs mean for cardiovascular preventive strategies. We need to know, for example,
1) whether the increased risk of developing diabetes with diuretics and beta-blockers becomes progressively greater with the increased duration of exposure to these drugs or 2) whether their effect just anticipates the inevitable, i.e. bringing on diabetic hypertensive individuals who will develop this condition a few years later anyway.
We also need to know whether and how accurately patients that are more at risk of developing diabetes during antihypertensive treatment can be identified based on their demographic, clinical and/or genetic characteristics.
Finally, although there is no question that the dysmetabolic effects of diuretics and beta-blockers are dose-dependent, it would be extremely important to know
1) whether diuretics have any diabetogenic effect when employed at low doses and
2) whether this effect can be entirely or partially counteracted by the antihypertensive agents with which low-dose diuretics are usually combined.
The content of this article reflects the personal opinion of the author/s and is not necessarily the official position of the European Society of Cardiology.