London, UK - Authors of a new analysis of the Anglo-Scandinavian Cardiac Outcomes Trial—Blood Pressure Lowering Arm (ASCOT-BPLA) say that treating hypertensive patients with a calcium-channel blocker/ACE-inhibitor combination instead of a beta blocker/diuretic may help prevent new-onset diabetes [1]. But at least one expert not involved in the study says that the analysis offers "nothing new" and should not dissuade physicians from using diuretics as first-line therapy.

Authors of the analysis, led by Dr Ajay K Gupta (Imperial College London, UK), calculate that physicians could prevent one case of new-onset diabetes by treating just 30 patients for five years with amlodipine and perindopril, instead of atenolol and a thiazide diuretic. "Pending further definitive evidence related to cardiovascular morbidity and mortality with antihypertensive-associated incident diabetes, it seems at best unwise, except where compelling indications apply, to use beta blockers and diuretics in preference to other combinations, such as a calcium-channel blocker plus an ACE inhibitor," they write in the May 2008 issue of Diabetes Care [1].

Gupta and colleagues first presented details from this analysis during the World Congress of Cardiology 2006, as reported by heartwire. Then, as now, Gupta et al reported that assignment to amlodipine with or without perindopril therapy in the ASCOT-BPLA trial was associated with a 34% reduction in risk of developing diabetes, relative to those taking atenolol with or without bendroflumethiazide. In all, more than 14 000 patients of the more than 19 000 in the study were at risk of developing diabetes at baseline; 9.7% (1366 patients) went on to develop diabetes during the 5.5 years of follow-up.

In an interview with heartwire, senior study author Dr Neil Poulter (Imperial College London) explained that the aim of the study was to look for any factors that are predictive of new-onset diabetes or factors that appear to protect against developing diabetes. As expected, the authors found that higher body-mass index (BMI) and higher blood-pressure, fasting plasma glucose, and triglyceride levels at baseline were all associated with increased risk of developing new-onset diabetes. By contrast, higher HDL levels and taking a combination of a calcium-channel blocker and ACE inhibitor were associated with reduced risk of developing diabetes over follow-up.

"The biggest single predictor of new-onset diabetes was blood glucose, and that began from 5 mmol onward," Poulter said. "For each mmol you get an almost sixfold increase in risk of getting diabetes, and starting at 5 mmol, not at six, but at five. And other obvious things like BMI, increased BP, high triglycerides, and low HDL cholesterol were all predictors. And the biggest, strongest preventive predictor of new-onset diabetes was what drugs you were on: the amlodipine/perindopril combination was the biggest single protective factor, compared with being on atenolol and a diuretic."

That protective effect, Poulter added, is likely a combination of the fact that beta blockers and diuretics have a known adverse effect on glucose and the fact that some studies suggest ACE inhibitors and angiotensin-receptor blockers are protective, while calcium-channel blockers are likely "neutral."

But commenting on the study for heartwire, Dr Jackson T Wright (Case Western Reserve University, Cleveland, OH) points out that the ASCOT-BPLA trial randomized patients only to amlodipine vs atenolol, with the ACE inhibitor and diuretics being add-on therapy. "As such, you cannot extrapolate from this study about the implications of ACE inhibitors or, more important, about whether or not diuretics should be first-line therapy, as stipulated in the current JNC 7 guidelines," he notes. What's more, he says, "We've known for more than 40 years that diuretics and beta blockers will elevate glucose, usually on the order of around 5 mg/dL, so that's nothing new."

He points to other studies that have looked at the impact of drug choice on new-onset diabetes, noting that DREAM showed no influence of ramipril on diabetes onset and that in the ALLHAT trial, the increase in new-onset diabetes associated with a diuretic was not linked to an increase in CV events.

Pouter, however, argues that in DREAM, the effect of ramipril on new-onset diabetes, while not as pronounced as the comparator—rosiglitazone—was in the same direction. "It was commensurate with all the other trials that renin-angiotensin-system blockade does protect, I believe, against new-onset diabetes."

Poulter acknowledges that there is "nothing new" in the paper, "it's merely looking at this issue in a hypertensive population and providing a pecking order of all the relatively important things. This is the first time in a large population of hypertensives, a priori, that the real determinants of new-onset diabetes have been listed and their relative importance outlined. And the most protective factor was the kind of drugs given. The other factors are not new news, but this does give you a sense of what's important. And Dr Gupta has produced this algorithm that I think we will be able to turn into a gizmo, some kind of tool, to be able to use to risk-stratify patients."

Poulter and his coinvestigators conclude that their analysis, on top of other information, supports a strategy of using a calcium-channel blocker and an ACE inhibitor, instead of a diuretic and beta blocker, in a patient with no coronary disease and no heart failure, particularly if he or she had any of the baseline characteristics predictive of new-onset diabetes. "If you've got coronary disease you should be on a beta blocker, if you've got heart failure, you should be on a diuretic and probably a beta blocker. But for straight up-and-down hypertension, I cannot think of a reason why you would give a beta blocker and diuretic," he concluded.

But Wright, by contrast, says that until the guidelines are changed, he will start with a diuretic in "most patients."

"There's nothing in this paper to persuade me that doing so would increase their risk of new-onset diabetes," he said. "There's nothing in this paper that is, in fact, new."

If anything, Wright adds, the trial that provides the most insight into the effect of antihypertensive drug choice on outcomes is ACCOMPLISH, reported at this year's ACC Scientific Sessions. "If anything, ACCOMPLISH would be the study that has the greater relevance to this discussion, more so than this ASCOT analysis, but even there, it needs to be looked at in the context of other studies that contradict the ACCOMPLISH results."