ASCOT: a tale of two treatment regimens

Each year in the United Kingdom alone there are 20 000 preventable deaths from cardiovascular disease attributable to hypertension. Much of the excess mortality and associated morbidity arises from poor control of blood pressure among people known to have hypertension. For the past two years in the United Kingdom, general practitioners have had the prime responsibility for tackling this problem, along with financial incentives to meet targets for detecting and controlling high blood pressure. Yet, despite many clinical trials and guidelines, they may be unsure about which antihypertensive drug to use first and how to combine treatments.

In 2004 the National Institute for Health and Clinical Excellence (NICE) recommended thiazide or thiazide-like diuretics as the first line treatment for most patients, with the addition of β blockers as the next step.^w1 This echoed the advice given in the US Joint National Committee's guidelines the previous year.^w2 Near simultaneous guidance from the British Hypertension Society, however, recommended for the first time drugs acting on the renin-angiotensin system—angiotensin converting enzyme (ACE) inhibitors or angiotensin receptor blockers—as first line treatment for “younger, non-black” patients.^w3 In effect, the resulting confusion endorsed earlier European guidelines which advocated leaving the choice of drug to individual practitioners.^w4

An eclectic approach to prescribing would make sense if all antihypertensive agents were equal. But they are not, according to results published last month from the largest antihypertensive trial ever conducted in Europe, the Anglo-Scandinavian cardiac outcomes trial (ASCOT).^w5 The trial compared strategies combining more expensive newer drugs with cheaper older ones. Nearly 20 000 patients with hypertension and at high risk of cardiovascular disease, aged 40-79, were randomised to two treatment regimens (rather than drug classes, as in most preceding trials).

The newer regimen comprised a calcium channel blocker (amlodipine) with an angiotensin converting enzyme inhibitor (perindopril) if required, while the older regimen comprised a β blocker (atenolol) with or without a diuretic (bendroflumethiazide). An extended release formulation of the β blocker doxazosin (4-8 mg) was added for all patients whose blood pressure did not reach the target value of < 140/90 mm Hg, or < 130/80 mm Hg for patients with diabetes. The trial was well conducted over a median follow-up of 5.4 years, with fewer than 2% of patients lost to follow-up and a high proportion continuing the treatment regimen to which they had been randomised.

The study was designed to stop after 1150 fatal or non-fatal coronary heart disease (primary endpoint) events in participants, but it was halted early after 903 such events, when the data and safety monitoring board reported that patients on the atenolol-based regimen were being increasingly disadvantaged as the trial progressed. In the final analysis, rates of fatal and non-fatal coronary heart disease were lower in patients on the amlodipine-perindopril regimen (by 10%), although this was not statistically significant (P = 0.1052). However, clinically and statistically significant reductions in rates of the main secondary end points occurred, including all cause mortality (11%) and cardiovascular mortality (24%), as well as all coronary events (13%) and fatal and non-fatal stroke (23%).

Owing to changes in cardiological practice over the duration of the trial the ASCOT investigators added an end point in the final analysis, combining the primary end point with coronary revascularisation. This modified primary end point was reduced significantly by 14%. Moreover, a combined end point of cardiovascular death plus non-fatal myocardial infarction and stroke similar to that used in many previous hypertension trials was 16% lower in patients taking the newer rather than the older drugs (P < 0.0003).

Were the better outcomes among patients allocated to the amlodipine-perindopril regimen attributable only to differences in achieved blood pressure or were they due to other effects of the drug classes used? The initial fall in blood pressure with the newer agents was brisker, with a difference of 5.9/2.4 mm Hg at three months and an overall mean difference at the end of the trial of 2.7/1.9 mm Hg. An analysis reported in an accompanying paper matched patients who had achieved similar blood pressures on the two regimens in ASCOT and compared outcomes, adjusting for baseline differences.^w6 This analysis found it was unlikely that all the benefits seen in the patients allocated to the amlodipine-perindopril regimen were attributable solely to more effective systolic blood pressure lowering.

Other potential benefits of the newer drugs in ASCOT included a statistically significant 30% reduction in the incidence of type 2 diabetes mellitus. Conversely, β blockers may have promoted the development of diabetes among participants in ASCOT, or this may have represented an interaction between β blockers and thiazides.^w1 Insulin resistance is closely associated with atherogenesis and is potentially modifiable by agents which block the renin-angiotensin system, such as perindopril. Reduced incidence of type 2 diabetes has been reported in several recent trials of ACE inhibitors and of angiotensin receptor blockers. Although concerns were raised in an editorial last year regarding the safety of angiotensin receptor blockers as a class,^w7 a systematic review in this issue (p 873)^w8 provides solid support for a previous rebuttal.^w9

Recent meta-analyses have shown that four or five cases of type 2 diabetes will be prevented when 100 patients are treated for 10 years with newer rather than older antihypertensive drugs.^{w1 10} In a similar population over 10 years, four coronary heart disease events and two strokes will be prevented by lowering blood pressure.^w1 A cost effectiveness analysis of ASCOT, which is clearly needed, should take into account the prevention of type 2 diabetes and its associated long term morbidity and mortality as well as the falling costs of generic calcium channel blockers and ACE inhibitors.

It makes sense for the British Hypertension Society and NICE to develop joint UK guidelines to help practitioners control their patients' blood pressure, with the aim of reducing unnecessary deaths from hypertension.^w11 Both bodies are set to collaborate, probably recommending more frequent prescription of ACE inhibitors or angiotensin receptor blockers plus calcium channel blockers as first line treatments and in combinations. Thiazides will probably be used mainly for adjunctive treatment to achieve targets for lowering blood pressure (particularly in useful combination, fixed dose preparations), while initial treatment with atenolol will be questionable.^w12