Personalised medicine for hypertensionBMJ 2011; 343 doi: https://doi.org/10.1136/bmj.d4697 (Published 28 July 2011) Cite this as: BMJ 2011;343:d4697
- Morris J Brown, professor of clinical pharmacology, University of Cambridge
The two broad reasons for tailoring antihypertensive treatment are that susceptibility to complications from treatment varies between people and hypertension has a variety of causes. It is well known that people with asthma should avoid β blockers and that men should avoid higher doses of spironolactone (because of gynaecomastia); the National Institute for Health and Clinical Excellence (NICE) will soon be recommending diuretics over calcium blockade in the very elderly (>80 years) because of superior efficacy in preventing heart failure.1
More interesting but more challenging is whether treatment can be tailored to the pathogenesis of hypertension in individual patients. Hypertension has multiple causes, so treating all patients the same way is as illogical as it would be to treat all cases of anaemia with vitamin B12 or all cases of pneumonia with penicillin. Recent prospective randomised trials and post-hoc analyses of earlier trials and accompanying editorials provide the evidence and impetus needed to implement plasma renin measurement as the “bacteriology” of hypertension.2 3 4 5
Although the early promise of genetic studies in hypertension and the availability of multiple distinct classes of effective drugs first prompted the exploration of individual variation in response, practical pharmacogenetics in hypertension now seems far off. Of 29 common variants so far associated with blood pressure, only one explains even 1 mm Hg of variance in blood pressure.6 The inference is that many hundreds, maybe thousands, of variants are needed to explain the inherited component of hypertension.
In contrast, the definition in physics of pressure as force divided by area suggests that there are only two physiological pathways to increased blood pressure. Either the cross sectional area of small resistance arteries is reduced, or the weight of the column of fluid comprising the circulation—effectively salt and water—is increased, or both. The complex molecular pathogenesis of essential hypertension causes most patients to lie on a spectrum between the two extremes of volume excess and vasoconstriction. The exceptional patients with single causes—so called secondary hypertension—congregate towards one or other extreme. These patients show greater falls of blood pressure than is seen in essential hypertension when they are treated with the appropriate single drug—for example, α blockade for the vasoconstriction of phaeochromocytoma or mineralocorticoid receptor blockade for the Na+ excess of primary hyperaldosteronism (Conn’s syndrome). Such patients therefore offer a proof of principle that tailored treatment works.
The drugs for which we have evidence of long term benefit fall into four main classes—A (angiotensin converting enzyme inhibitors and angiotensin receptor blockers), B (β blockers and α blockers), C (calcium channel blockers), D (diuretics)—serendipitously, the initial letter of the drug class. The first two work by blocking the renin-angiotensin system, the second two by eliminating Na+.7 Crossover studies comparing each class in the same patients show that, as might have been predicted by the physics, there are broadly only two categories of antihypertensive drug. One comprises the “AB” drugs, which target the renin-angiotensin system at one point or another, so reversing the principal cause of vasoconstriction. The other comprises the “CD” drugs, which target Na+ excess. The original AB/CD rule provided an easy mnemonic for determining the right category in an individual patient, and it was based on observations that plasma renin declines with age and is lower in patients of African or Caribbean origin.8 Effectively, AB/CD used age and ethnicity as surrogates for plasma renin. The AB/CD crossover trials were performed in younger people (<55 years) with hypertension, whose blood pressure responded better to AB drugs than to CD ones.8 9 Subsequently, several large trials of people who were a minimum of 55 years old found that CD drugs reduced systolic blood pressure by about 5 mm Hg more than AB drugs.10
The rule proposed switching categories in those who did not respond to the appropriate drug for their age or ethnicity.8 The switch enables about 20% of patients to avoid receiving unnecessary combination treatment and possibly prevents some patients from having a paradoxical pressor response to the “wrong” drug.2 The NICE version of the AB/CD rule dropped this switch and therefore recommends less individualised treatment than the original version. NICE also relegated B drugs (in 2006) and D drugs (in 2011) to third or fourth line treatment of hypertension. Both moves are controversial, relative to other international guidelines, and the relegation of D drugs ignores the importance of Na+ excess in the pathogenesis of hypertension. But reduced prescribing of B drugs after 2006 had the compensation of enabling renin to be measured without changing treatment. A low plasma renin occurs when the kidneys detect an excess of Na+ in the circulation and is an invaluable diagnostic test for Na+ excess. However, B drugs work by blocking renin secretion and so a suppressed plasma renin on β blockade is not a reliable measure of Na+ excess in patients taking these drugs. Conversely, A, C, and D drugs normally increase renin secretion (by a different mechanism in each case), so a plasma renin that is low despite taking these drugs is even stronger evidence than in untreated patients of Na+ retention. Indeed, the detection of a suppressed renin in patients on multiple therapy and the development of a specific scan for the aldosterone secreting adenomas of Conn’s syndrome have greatly helped to identify an endocrine tumour as the most common curable cause of hypertension. This diagnostic application is sufficient reason alone to adopt renin testing in hypertension, and it validates the interpretation of a low plasma renin as a diagnostic measure of Na+ excess. But we now also know that a low renin in patients receiving treatment predicts a better blood pressure response to D drugs than to AB drugs.5 11
Currently, only a few centres in the United Kingdom offer the newer, most accurate measure of renin activity, so a delay in implementation is inevitable. Meanwhile, the British Hypertension Society (BHS) is midway through a pair of BHF funded collaborative studies to establish the predictive value and cost effectiveness of renin measurement, both at initial diagnosis and in resistant hypertension (patients with uncontrolled hypertension despite treatment with A, C, and D drugs, where NICE considers the evidence insufficient to guide choice). Subject to confirmation in the BHF trials, the figure⇓ incorporates available theory and evidence into our α,β,Δ extension of AB/CD as a working mnemonic for resistant hypertension.
The renin assay is now cheaper (about £12; €14; $19) and better evidenced than many routine tests in medicine; there is something perverse about raising the barrier for a condition that is the most common cause of serious morbidity and most frequent indication for long term treatment. Current guidance varies between top-down prescription and trial and error experimentation in the patient. There is now a third way—a 30 minute laboratory test that measures plasma renin.
Cite this as: BMJ 2011;343:d4697
Competing interests: The author has completed the ICMJE uniform disclosure form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declares: no support from any organisation for the submitted work; no financial relationships with any organisations that might have an interest in the submitted work in the previous three years; no other relationships or activities that could appear to have influenced the submitted work.
Provenance and peer review: Commissioned; not externally peer reviewed.