Changes in the usage of antihypertensive drugs: implications and prospects

In their paper [1], Cheung and colleagues describe changes in antihypertensive drug usage over 8 years at a university hospital in Hong Kong. The quantitative analysis of usage frequency is supplemented by estimates of the percentage of patients with blood pressure at target levels. Their findings prompt a number of questions, about patterns of drug usage in different populations worldwide, the possible influences on these patterns and how these might change over the next few years. Of course, such detailed information is not available everywhere, so some comparisons are likely to remain speculative. In this particular case there are apparently no specific Hong Kong or Chinese national guidelines which are applicable, unlike the USA or Europe [2–4]: it would indeed be very interesting to establish how far such guidelines do actually modify prescribing practice in countries where they are supposed to be applicable. Regardless of that, Cheung's paper can serve as a focus for several current issues of interest and debate.

The high usage of calcium channel blockers

Hypertension in Chinese populations is predominantly of the low renin phenotype [5], so it is understandable that they are widely used and appear to be effective. This is especially true in older patients and the authors comment on the higher usage in this age group. The controversy regarding the safety of the dihydropyridine calcium channel blockers has largely abated [6], partly because achievement of lower blood pressures is seen as a priority which outweighs small increases in risk due to sympathetic activation, but also because such activation is much more likely with short-acting drugs. These are now little used, if at all.

The increasing use of β-blockers

The efficacy of β-blockers in hypertension is currently highly controversial. Studies in elderly patients, many with isolated systolic hypertension, suggested that β-blockers were less effective in terms of clinical outcomes than thiazide diuretics or the long-acting dihydropyridine calcium channel blockers [7, 8]. Very recently, Lindholm and his colleagues have concluded that atenolol is associated with increased mortality across a much wider range of clinical trials, although its blood pressure-lowering effect was comparable [9]. Atenolol is a very commonly used reference drug in clinical trials of new antihypertensive drugs, but fewer data are available for the many other β-blockers. It is unclear therefore whether this apparent disadvantage is a possible class effect or peculiar to atenolol: it has been suggested that its hydrophilicity may be a factor, though it is difficult to fit this into a comprehensive explanation. It is also too early to say whether this report has influenced prescribing anywhere. A striking observation is the excellent tolerability of β-blockers in this population, contrary to the perception of many prescribers and patients. It is possible that there are important ethnic differences in this context, but there is limited available information [10].

Increasing numbers of drugs per patient

All current guidelines and a large body of other opinion, such as the enormous meta-analysis of Law and his colleagues [11], support the idea of combination therapy in hypertension. In fact, the reality is that it will be impossible to achieve anything resembling target blood pressures using monotherapy, certainly not without unacceptable side-effects. These may well cause patients simply to abandon medication altogether. However, the increase in medications per patients is rather modest (from 1.6 to 2) and this seems low in comparison with the HOT and UKPDS trials [12, 13] and some surveys [14]. The main issues are whether combination therapy should be the initial approach to treatment and whether fixed-dose combinations should be more generally accepted. In fact, they are in most countries other than the UK (the influence of clinical pharmacologists, perhaps?). Many combinations of ACE inhibitors or angiotensin II receptor blockers with thiazide diuretics are available and these are pharmacologically rational and clinically effective, at least in terms of blood pressure reduction. They also, in most cases, have the advantage of providing genuinely low doses of the diuretics and therefore minimize side-effects: often the lowest doses of thiazides available alone are higher than they need to be. In many countries other combinations are also marketed, particularly ACE inhibitors and calcium channel blockers. In any case, it is clear that multiple therapies, in whatever format, will be increasingly the norm in the management of hypertension. A practical approach to choosing combinations (the ABCD system) has been proposed in the recent guidelines of the British Hypertension Society [4, 15], based on the interaction of the major classes of antihypertensive drugs with the renin–angiotensin system. This has been criticised on the grounds that it is not based on evidence of outcomes, but it is difficult to envisage any large-scale trial which will provide such evidence: meanwhile the clinical challenges remain, as discussed below. Drug adverse effects are rightly regarded as a significant obstacle to the successful treatment of hypertension. Most of these are dose-related, so that the problem could be minimized by the use of low-dose combinations: the meta-analysis mentioned above [11] clearly shows that halving the dose of drugs used in combination will greatly reduce adverse effects, but not efficacy.

Failure to achieve target blood pressures

In this paper it is reported that about 40% of hypertensive patients attained blood pressures of 140/90 mmHg or less. This is very similar to the results in the EUROASPIRE II survey [16] and is representative of apparently improving blood pressure control worldwide, or at least in some centres; clearly though, much more needs to be achieved. Other studies are less optimistic, however, [17]. In the current Hong Kong study only 20% of diabetic patients achieved the more stringent blood pressure target of 130/80 mmHg. There are certainly a variety of reasons why achievements fall short of targets everywhere and it is difficult to generalize in any useful way. One issue is likely to be patients’ adherence to treatment. In this paper 96% of patients claimed to comply without fail to the prescribed regimes. This would be a remarkably high figure by any international standards and perhaps should be regarded with slight scepticism. It is well recognized that long-term treatment of hypertension may have particularly marked problems with adherence for several reasons: the asymptomatic nature of most hypertension; the often life-long nature of medication; and the relatively high incidence of adverse effects with many classes of antihypertensive drugs. As many clinicians will know, it can be difficult to persuade patients with moderate hypertension of the advantages of treatments which may make them tired, cause impotence and precipitate gout, though we can truthfully say that we can usually avoid these problems.

Differential effects on systolic and diastolic blood pressure

The authors of this paper note that diastolic blood pressure is better controlled than systolic pressure in their patient population; two-thirds of patients achieved diastolic blood pressure <90 mmHg but less than one-third had systolic pressure <140 mmHg. It is noted that this is particularly the case in the elderly and the authors suggest that this because older patients are more likely to be on multiple drugs. This is not likely to be the whole explanation, however, and raises a number of interesting and interconnected issues:

• Above the ages of 45–50 years, which is the mean age of this population, systolic pressure becomes progressively more important as a predictor of cardiovascular events [18].

• In most populations systolic blood pressure rises steadily with age, while diastolic pressure reaches a plateau or declines [19].

• None of the currently used classes of antihypertensive drugs is truly selective for systolic (or that matter diastolic) pressure and they tend to lower both.

It is therefore not surprising that achieved diastolic pressures are generally lower: they may have been lower to begin with and have been subjected to multiple drugs in an effort to reduce systolic pressures. The lowering of diastolic pressure to ‘unnecessarily’ low levels may not be wholly innocuous, let alone beneficial. Very low as well as high diastolic pressure may be associated with ischaemic hemispheric white matter lesions [20].

Future prospects

We are not on the threshold of finding the magic bullet for hypertension, but at the same time should acknowledge how much better equipped today's prescriber is in comparison with his or her predecessor 20 years ago. We still have an incomplete understanding of the aetiology of ‘essential’ or primary hypertension, which is the basis of raised blood pressure in the overwhelming majority of patients, and therefore are hampered in finding new treatments for this extremely common condition. Progress in genetic research is not likely to have any major impact on management in the near future, though we may be able to reinforce the rational basis for choosing particular existing drugs [21]. It is also inevitable that combination therapy, whether with fixed-dose or other formulations, will be accepted generally as the norm in the management of hypertension. The one major new class of antihypertensive likely to appear in the next year or so is that of renin inhibitors. After many years of trying it seems that a drug, aliskiren, has been developed which has good efficacy and usable pharmacokinetics, notably in terms of bioavailability [22]. Its introduction will mean that it will be possible to block therapeutically almost every step in the renin–angiotensin–aldosterone system, with the exception of non-ACE-mediated angiotensin II synthesis. However, its role in therapy has still to be determined. The vasopeptidase inhibitors, dual inhibitors of ACE and neutral endopeptidase, appeared to have promising efficacy but the lead drug in this class, omapatrilat, has been associated with increased incidence of angioedema compared with ACE inhibitors [23]. This may outweigh any presumptive advantages over these and other existing antihypertensive drugs and seems likely to end or at least greatly restrict their further development.

So it does not seem likely that we will see dramatic advances in antihypertensive therapy over the next few years. The emphasis needs to be on better use of what we have already, and this may include new uses for existing drugs. For instance, spironolactone appears to be effective in many patients with resistant hypertension even when they do not have detectable hyperaldosteronism, though the reason for this is not understood [24]. A different change may relate to changes in definitions: the JNCVII guidelines introduced the concept of ‘prehypertension’, with systolic blood pressure levels between 120 and 139 mmHg [1]. It is debatable whether the increased risk associated with these levels of blood pressures warrants drug therapy in terms of absolute risk reduction [25] but life-style modifications may be helpful. Indeed, this points to the danger of overmedicalization as well as overmedication: would we really want to give antihypertensive medication to over half the adult population? Writing as a physician who works in a hypertension clinic, my answer would be an unequivocal ‘no’!