ASCOT-BPLA: Risk factors for new-onset diabetes

Susan Jeffrey

September 06, 2006

Barcelona, Spain - Further analysis of patients who developed diabetes during the Anglo-Scandinavian Cardiac Outcomes Trial-Blood Pressure Lowering Arm (ASCOT-BPLA) has allowed investigators to create a risk score that appears to discriminate those at highest risk for development of new-onset diabetes (NOD) during antihypertensive therapy.

Assignment to amlodipine with or without perindopril therapy was associated with a lower risk of NOD in that trial, a 34% reduction in risk relative to those taking atenolol with or without bendroflumethiazide, they report.

Other factors associated with new diabetes risk that were perhaps "not a surprise" were elevated baseline fasting plasma glucose and triglycerides and decreased HDL cholesterol, ASCOT co-principal investigator Dr Peter S Sever (Imperial College, in London, UK) said.

"Overall, almost 10% of patients in the trial developed diabetes, so if you're hypertensive, you really are at increased risk of getting diabetes, and if you're on a beta blocker, It's even worse," Sever said.

This analysis from ASCOT-BPLA was presented here at the World Congress of Cardiology 2006 by Dr Ajay K Gupta (Imperial College).


The ASCOT trial was designed to provide further data on outcomes with newer agents over standard therapy with beta blockers and diuretics and begin to provide some information on combinations of agents. The trial enrolled 19 257 hypertensive patients with at least three other cardiovascular risk factors from 650 general practices in the UK, Ireland, Sweden, Finland, Denmark, Norway, and Iceland.

Patients were randomized to one of two antihypertensive regimens: amlodipine (5/10 mg) (Norvasc, Pfizer), with perindopril (4/8 mg) (ACEON, CV Therapeutics/Solvay Pharmaceuticals; Coversyl, Servier) added as required; or atenolol (50/100 mg) (Tenormin, AstraZeneca), with or without the thiazide diuretic bendroflumethiazide (1.25-2.5 mg). Further treatment with other agents could be added as required to reach a target BP of 140/90 mm Hg or less.

Of the overall population, 10 305 patients with total cholesterol levels of 6.5 mmol/L or less were further randomized to receive atorvastatin 10 mg/day or placebo in a 2x2 factorial design—the ASCOT-LLA population. The lipid-lowering arm of the trial was stopped for efficacy of atorvastatin over placebo and published in 2003.

A reduction in all-cause mortality seen with the amlodipine/perindopril strategy caused the trial to be stopped in November 2004. A 10% reduction in nonfatal MI and fatal CHD, the primary end point of the trial, did not reach statistical significance, a finding that the researchers attribute to the early stop of the trial.

Recent British recommendations for the treatment of hypertension no longer include beta blockers as first-line or even second- or third-line drugs for patients with uncomplicated hypertension, although they are still recommended for patients who also have CHD.

New-onset diabetes

A variety of secondary analyses of the ASCOT data have subsequently been published or reported. In the present analysis, investigators determined risk predictors for NOD, which had been a prespecified end point in the trial, and hoped to develop a risk score that might help to identify those at higher risk.

In the amlodipine/perindopril treatment group, 7074 patients were not diabetic at the start of the trial, similar to 7046 in the atenolol/diuretic group. As previously reported, more patients on atenolol-based therapy developed NOD than in the amlodipine-based therapy arm.

ASCOT-BPLA: New-onset diabetes by treatment group
  Atenolol-based therapy Amlodipine-based therapy
n (%) 799 (11.4) 567 (8.0)

Fasting plasma glucose (FPG) was the most robust risk factor and antihypertensive therapy with amlodipine with perindopril was the strongest protective factor among baseline predictors of NOD. BMI, HDL-cholesterol, and triglyceride levels were also important baseline predictors.

ASCOT-BPLA: Multivariate Cox proportional hazard model for risk of new-onset diabetes
Factor Hazard ratio 95% CI p
Age >55 y (per 5 y) 0.94 0.90-0.96 .006
FPG >5 mmol/L 5.80 5.24-6.43 <0.001
Male sex 0.98 0.84-1.14 0.750
BMI (per 5 units increase) 1.49 1.38-1.62 <0.001
SBP (per 10 mm Hg) 1.07 1.04-1.10 <0.001
Amlodipine+perindopril 0.66 0.59-0.74 <0.001
HDL-C (per mmol/L) 0.72 0.58-0.89 0.002
Triglyceride (per mmol/L) 1.12 1.07-1.17 <0.001
Total cholesterol (per mmol/L) 0.89 0.84-0.94 <0.001
Use of non-CAD medication (Y/N) 1.25 1.11-1.40 <0.001
Alcohol intake (unit/wk) 0.99 0.99-1.00 .017

They then divided the population into quartiles for risk based on these factors and showed that in the highest quartile, the risk for NOD was almost 20 times higher than in the lowest quartile.

ASCOT-BPLA: New-onset diabetes by risk quartiles
Quartile Hazard ratio (95% CI)
First Referent
Second 2.5 (1.8-3.5)
Third 5.0 (3.7-6.8)
Fourth 19.0 (14.3-25.4)

In each quartile, those in the group taking atenolol with or without thiazide had a higher risk for NOD than those taking amlodipine with or without perindopril, the authors noted.

"Compared with use of atenolol+thiazide, the use of amlodipine+perindopril is associated with 34% reduction in the risk for NOD, and this decrease is irrespective of the baseline risk category," Gupta concluded. "The risk model developed is robust, has an excellent discriminative ability, and could potentially play an important role in clinical practice."

Choosing optimal antihypertensive therapy

Invited discussant for this paper was Dr Jose L Zamorano Gomez (Hospital Clinico San Carlos and the Hospital 12 de Octubre de Madrid, Spain). Zamorano said the end point of NOD has been considered in a number of recent antihypertensive trials such as VALUE and ALLHAT, and although NOD was seen to be increased by certain antihypertensive medications, this did not translate into increased clinical events that make up the primary end points of these trials. However, it is likely that longer follow-up will show that the development of new diabetes does actually increase risk, he said.

"Combining the 11 most important risk factors for NOD" reported in this paper by the ASCOT investigators, he said, "it is possible to calculate a risk score that was extremely good at discriminating between risk groups presented in quartiles."

A variety of factors must be considered when an antihypertensive therapy is chosen for individual patients, Zamorano concluded; whether the patient has heart failure, is post-MI, or has diabetes or renal failure. "We need to add new-onset diabetes to factors considered when selecting the optimal treatment for patients," he said.

The principal sponsor of ASCOT is Pfizer Inc, New York, with support provided by Servier Research Group, Paris, France, and Leo Laboratories, Copenhagen, Denmark. The trial was jointly coordinated by the Imperial College London (International Center for Circulatory Health, National Heart and Lung Institute at St Mary's Hospital, London, United Kingdom) and the Scandinavian Coordinating Center (Scandinavian CRI AB), Göteborg, Sweden.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.