Drug specific benefits and cardiovascular and stroke outcomes

The treatment of hypertension is one of the most studied areas of clinical medicine. It is therefore perhaps surprising that debate still rages as to whether the benefits of treating high blood pressure are simply determined by the quality of blood pressure control, or whether the choice of drug therapy adds or detracts from the expected benefits of blood pressure lowering. The latter component of this debate has been fuelled by claims of specific, ‘beyond blood pressure’ benefits due to blockade of the renin–angiotensin system (RAS). This concept is further addressed by a paper by Sleight and Yusuf in this issue of the journal [1]. The experimental basis for advocating the benefits of RAS blockade beyond the haemodynamic actions of angiotensin II are compelling. Angiotensin II has been implicated in many biological processes relevant to the pathophysiology of structural and functional disturbances in the cardiovascular, cerebrovascular and renal systems. Indeed, the experimental evidence is so compelling, it is surprising that there is still a debate at all about the merits of RAS blockade as the primary treatment for hypertension. The debate exists because meta-analyses of blood pressure lowering trials have failed to reveal the expected ‘added value’ of angiotensin-converting enzyme (ACE)-inhibition compared to alternative blood pressure lowering therapies in terms of reduced cardio-cerebrovascular morbidity and mortality [2,3]. This is also true of more recent (yet to be published) authorative meta-analyses that include the most recent studies. Some potentially important reasons for this discrepancy between expectation and the weight of evidence are explored by Sleight and Yusuf, including: (i) the probability that blood pressure reduction per se is such a powerful determinant of benefit that any potential drug specific effects are ‘swamped', especially when the data are ‘lumped together’ for a meta-analyses; (ii) trials being too short to reveal drug-specific benefits because of the powerful effects of blood pressure reduction on outcome; and (iii) the benefits of ACE-inhibition being particularly relevant to protection against ischaemic heart disease, although few trials have randomized people at sufficiently high risk of cardiac disease to reveal drug-specific benefits over the short duration of a trial (the exception being the HOPE study, where individuals were randomized on the basis of the cardiac risk rather than their blood pressure). The ALLHAT study added to the perception that drugs were less important than blood pressure reduction [4]. Specifically with regard to ACE-inhibition, ALLHAT failed to demonstrate any advantage of ACE-inhibition over diuretic therapy. By contrast, stroke rates were significantly higher in patients receiving ACE-inhibition. Clearly, this inferior outcome with regard to stroke for ACE-inhibition could well be attributed to inferior blood pressure control compared to diuretic therapy. However, this acknowledges, rightly in my view, the importance of active comparators and equivalence in blood pressure control before any claim can be made about the merits, or otherwise, of a drug used to lower blood pressure. Such discrepancies between blood pressures in different treatment arms in clinical trials are often countered by statisticians’ claims that they can correct for the ‘effects of differences in blood pressure'. We must remain wary and be more critical of such claims. This is mathematical chicanery not biology. Herein lies the problem in interpreting the findings of the HOPE study, upon which much of the claim for superiority of ACE-inhibition over other drugs is founded [5]. The HOPE study was a landmark study and clearly demonstrated the benefits of ramipril therapy over placebo in individuals at high cardiovascular risk. The controversy lies in trying to interpret the data. Was this a drug-specific effect (i.e. blockade of the RAS), or the benefit of lowering blood pressure? That blood pressure was lowered more by ramipril than placebo is not in dispute. The question that can never be resolved is by how much? Nevertheless, the fact that two potentially important determinants of cardiovascular and stroke outcome did change means that it is impossible to define the mechanism for benefit. Moreover, from a scientific perspective, it is impossible to dismiss the potential for either mechanism to have been dominant. Suffice to say that the evidence for the power of even small blood pressure reductions in high-risk patients to produce benefit is compelling [6], whereas the evidence in favour of ACE-inhibition in head-to-head trials with other blood pressure lowering agents is at best scant. There is also a trend to use ACE-inhibition and angiotensin II receptor blockade (ARBs) interchangeably in support of the RAS hypothesis. In my view, this is not appropriate. These are two very different pharmacological tools for inhibiting the RAS system. Notably, ACE-inhibition will suppress angiotensin II production and thus stimulation of all angiotensin II receptor subtypes, whereas ARBs will inhibit only the AT-1 receptor, leaving the others exposed for stimulation by angiotensin II. The biological significance of this difference in mechanism of action in man is unclear. However, its clinical significance may be real. In the LIFE trial, ARB-based therapy (losartan) was clearly superior to β-blocker-based therapy (atenolol) at reducing the incidence of stroke [7]. This contrasts with data for ACE-inhibition, which has failed to show a reduction in the incidence of stroke in head-to-head trials with active comparators. Whether an ACE-inhibitor would have fared as well as the ARB in the patients studied in the LIFE trial is unknown. Until this is the case, the discrepancy in outcomes between ACE- inhibitor treated patients versus active comparators in general, and the ARB versus conventional therapy in LIFE, raises the possibility that the interchangeability of ACE-inhibitors and ARBs may not be appropriate. Further head-to-head studies, such as VALUE, will shed light on the stroke-preventive properties of ARBs. Moreover, head-to-head trials comparing ARBs and ACE-inhibition, such as ON-TARGET, will be important in defining whether there are real differences between these two approaches to RAS blockade. Clinical trials have their limitations, and these are further compounded by meta-analyses. Presently, there is inadequate data to support many of the claims for drug specific benefits for cardiovascular protection beyond the powerful effects of blood pressure reduction. This does not mean that the basic experimental data supporting the rationale for RAS blockade is wrong, By contrast, it is compelling. It simply means that clinical trials cannot answer all of the questions. Moreover, for the most part, their short duration with respect to the duration of disease processes will favour the conclusion that mechanisms acting rapidly to impart benefit (i.e. blood pressure reduction) are more important determinants of outcome than alternative, drug specific mechanisms. The reality in practice is that, to achieve modern blood pressure goals, most patients will require more than one drug therapy. It is reassuring that the major classes of antihypertensive therapy have all been shown to be safe and effective, and the most effective strategy over the longer term is likely to be one that includes RAS blockade, a diuretic and a calcium channel blocker, not least because it will be effective at lowering the blood pressure of most people [8,9].