Navigating the shoals in hypertension: discovery and guidanceBMJ 2012; 344 doi: https://doi.org/10.1136/bmj.d8218 (Published 13 January 2012) Cite this as: BMJ 2012;344:d8218
All rapid responses
The British Hypertension Society (BHS) strongly supports the views expressed by Morrison et al.1, concerning the lack of availability of chlortalidone in the UK. Based on a careful review of the available evidence, the 2011 NICE guidelines on management of hypertension2, developed in conjunction with the BHS, strongly promoted the use of chlortalidone (or indapamide), rather than commonly used thiazide diuretics such as bendroflumethiazide. Despite the very strong evidence-base for chlortalidone, there was already some difficulty in prescribing it in the UK, because the usual starting dose is 12.5-25mg, and only a 50mg tablet is available via a single supplier. However, it seems that even this will be no longer available. We believe that this will have a significant impact upon the management of hypertension in the UK and should be viewed as a retrograde step. It seems strange that no company feels able to make a profit from the sale of chlortalidone in the UK when hypertension is such a common disorder and thiazides are so widely prescribed. Many other generic drugs are available in the UK despite much lower volume sales.
We strongly urge the Department of Health to investigate the situation urgently and to secure a continuing supply of chlortalidone for UK patients. The alternative of importation on a named-patient basis results in significant delays, and is vastly more expensive.
1. BMJ 2013;346:f3076
2. National Institute for Health and Care Evidence. Hypertension: clinical management of primary hypertension in adults. CG127. 2011. http://publications.nice.org.uk/hypertensioncg127
Competing interests: No competing interests
The National Institute for Health and Clinical Excellence (NICE) Guideline Development Group (GDG), in collaboration with the British Hypertension Society (BHS), published in August 2011 the updated guideline for management of primary hypertension in adults . Among the recommendations, diuretics have been demoted to third-line treatment, mainly as a result of concern about the risk of electrolyte derangement and diabetes mellitus. In addition, the almost universal use of bendroflumethiazide (2.5mg), which remains the most commonly prescribed antihypertensive in England , has been replaced by a recommendation to use ‘thiazide-like’ diuretics; in particular, chlortalidone or indapamide. The GDG make the point that there is no evidence in support of the use of bendroflumethiazide, except in trials using a much larger dose of 10mg.
Morris Brown and colleagues  have identified that chlortalidone (Hygroton®, Alliance Pharmaceuticals UK) is only available as a 50mg tablet and, although this scored tablet could be halved, the GDG recommendation of a 12.5mg starting dose is hard to achieve, and if given on alternate days at 25mg it is likely to raise significant problems with adherence . As guidance, this seems less than ideal. However, things are about to get a whole lot worse. We understand that Alliance Pharmaceuticals have recently informed the Department of Health (DH) that they have temporarily discontinued supply of Hygroton® tablets in the UK and have now exhausted all residual supplies. As they were the sole supplier of this medicine to the UK market, patients and their doctors will soon have major problems implementing the guidance; indeed, some of our patients in Lothian are already unable to source this medication from local pharmacies.
Supply of chlortalidone is a complex issue. NICE recommendations are based on published literature and other evidence, but the existence of a particular medicine, and the range of doses available in the UK, is dependent on there being sufficient demand for a manufacturing company, or supplier, to seek a licence for its use. The Medicines and Healthcare products Regulatory Agency (MHRA), as an independent regulator, cannot solicit applications. At present, the onus rests with the manufacturer to prove that a product is safe, effective and of sufficient quality to meet European legislative standards. One way forward would be to import chlortalidone, which is available as a 12.5mg tablet in the Netherlands and as a scored 25mg preparation in numerous other EU countries.
What are physicians to do? NICE would recommend indapamide as an alternative ‘thiazide-like’ diuretic, but others  has already argued that there are no data from any primary prevention trial in which indapamide met its primary endpoint. Another way forward would be to go back to bendroflumethiazide, but at a higher dose of 5-10mg; an intervention shown to be as good at reducing BP as chlortalidone and perhaps with a similar risk of electrolyte derangement [6,7].
Availability of a wider and appropriate range of doses of chlortalidone would make it much easier to implement the existing guidance. However, NICE, BHS, MHRA and DH are unable to mandate the availability of this evidence-based and NICE-recommended medicine for one of the major risk factors contributing to ill health in the UK . While this situation continues, patients and their physicians are in major difficulty.
1. National Institute for Health and Clinical Evidence. Hypertension: clinical management of primary hypertension in adults. CG127. 2011 (online). Available at: http://www.nice.org.uk/nicemedia/live/13561/56007/56007.pdf (Accessed 1st May 2013)
2. Health and Social Care Information Centre: Prescriptive cost analysis – England, 2012 (published online 4th April 2013). Available at: https://catalogue.ic.nhs.uk/publications/prescribing/primary/pres-cost-a... (assessed 1st May 2013)
3. Medical Research Council Working Party. MRC trial of treatment of mild hypertension: principal results. BMJ. 1985; 291: 97-104
4. Brown MJ, Cruikshank JK, MacDonald TM. Navigating the shoals in hypertension: discovery and guidance. BMJ. Published online First. 2012; 344: d8218
5. National Institute for Health and Clinical Evidence. Medicine Adherence. CH76. 2009 (online). Available at: http://www.nice.org.uk/nicemedia/live/11766/42971/42971.pdf (accessed 1st May 2013).
6. Peterzan MA, Hardy R, Chaturvedi N, Hughes AD. Meta-analysis of dose response relationships for hydrochlorthiazide, chlorthalidone and bendroflumethiazide on blood pressure, serum potassium and urate. Hypertension. 2012; 59: 1104-1109
7. Hood SJ, Taylor KP, Ashby MJ, Brown MJ. The spironolactone, amiloride, losartan, and thiazide (SALT) double-blind crossover trial in patients with low-renin hypertension and elevated aldosterone-renin ratio. Circulation. 2007; 116:268-275
8. Lim SS, Vos T, Flaxman AD, et al. A comparative risk assessment of burden of disease and injury attributable to 67 risk factors and risk factor clusters in 21 regions, 1990-2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet. 2012; 380: 2224-2260
Competing interests: No competing interests
The revision of Hypertension guidance by NICE has been criticised on many fronts. To try to make sense of the controversies and adjust my practice in ways I could feel justified according to available evidence, I went round the main sources of evidence and looked at the discussions on the BMJ.
Three areas seemed to attract most attention - the changes to the process of diagnosis, the treatment of older adults and the medication choices proposed. This is an appraisal of the issues pertaining to the diagnosis of Hypertension.
The process of diagnosis of Hypertension has been under fire for a long time. When we reported the results from a Trent RDSU grant financed project (Hutchinson 2007), the background read: "Ambulatory blood pressure monitor (ABPM) results correlate well to 'research standard' measurement of blood pressure, while day-to-day 'clinic standard' measurements do not (Campbell 2005, Fagard 1995 (metaanalysis), Mancia 1997). According to the most influential hypertension management guidelines, ABPM correlates better with target organ injury (JNC-7 (Chobanian 2003), BHS (Williams 2004), ESH/ESC (Guidelines Committee 2003)) and CVD risk (BHS, ESH/ESC) and is better in assessing treatment effect (BHS, ESH/ESC) than clinic measurements."
However, Brown, Cruikshank and MacDonald (2012) worry about what they claim amounts to a redefinition of hypertension, doubting that "the condition diagnosed by measuring blood pressure in the clinic is the same as that diagnosed by ambulatory measurement." This doubt is grounded on the fact that "clinic measured hypertension has been studied in about 500,000 patients in outcome trials", a claim ABPM diagnosed hypertension cannot compete with.
The comparison seems thus to be between clinic blood pressure measurement (CBPM) and ABPM (with home blood pressure HBPM as an alternative). McManus, Caulfield and Williams (2012) replied with evidence that ABPM seems better than CBPM (clinic) at predicting risk and outcomes, further claiming that White Coat effect does not carry additional risk when compared to normotension. This is a bit of a stalemate, as Brown and McManus use different arguments and it is difficult to dismiss either.
In fact, Brown and McManus talk about different things when then talk about CBPM. The issue is what I described above as the 'research standard' against the 'clinic standard' of blood pressure measurement. As reported by the authors cited, differences in setting, time available and other stressors, as well as the set of patient worries concurring with the blood pressure being measured, introduce significant differences between the research standard (cited by Brown) and the clinic standard (referred to by McManus) of measuring blood pressure 'in the office'. The issue is that we know very well we have been treating a condition different from the hypertension studied in outcome trials, we have been treating a sort of hypertension which diagnosis correlates badly with the diagnosis within the trials (Campbell 2005).
The issue of concern here is heterogeneity. And it comes in several fashions. The global assumption is the broad equivalence (give or take a certain correction factor) of research standard blood pressure measurement (RBPM) where outcomes are studied, CBPM (clinic) which is the usual process for diagnosis, ABPM and HBPM. To confuse the issue further, some of the studies used to determine this equivalence have HBPM measured by patient while in other studies it is measured by an investigator; some studies take RBPM as comparator (with a specific setting and staff to collect BP measurements) while others rely on 'usual' CBPM standards. Off note, RBPM and HBPM both try to establish a standardised, relaxed setting for the measurements, while CBPM is variable to this respect and ABPM positively ignores the setting (in what respects day time measurements). It must be realised that the very reason why ABPM and HBPM are interesting methods for diagnosis of hypertension is precisely because they tell us different things from CBPM, otherwise we could simply apply a correction factor to the CBPM and all would be sorted; it is not, because CBPM is different from RBPM, or ABPM, or HBPM, and because heterogeneity is a significant concern and cannot be corrected that simply.
Part of the heterogeneity may resolve as a consequence of the population studied/treated. If there is a predominance of sedentary or retired patients, only a few will have the day time average influenced by more strenuous physical activity or more stressful jobs. But it remains that we are measuring different things, at different times, in different settings:
RBPM - less time constraints, less associated to stressful worries, standardised setting, standard used for current knowledge
CBPM - stringent time constraints, associated to stressful setting and other coexisting health worries, current diagnostic standard
ABPM - no time constraints, heterogeneity of stressors, exertion, setting regarding day time, less so for night time
HBPM - less time constraints, less associated to stressful worries, standardised setting
Therefore, if heterogeneity is a significant conceptual concern, and we want to know what we are measuring, we cannot compare CBPM or ABPM with any of the others without a feeling that we are comparing apples and oranges. While we may be able to establish all sorts of statistical correlations between them, and it may be the case that the productivity of the apple tree may be correlated to the yield of oranges, there is no way we can get an orange out a certain quantity of apples. It would be tempting to conclude that the only blood pressure measurements that can be compared are RBPM (from where our knowledge originated, as Brown points out) and HBPM, which can be equally standardised to offer a useful baseline; possibly a night time ABPM could also fit the bill.
In conclusion, I will be very keen to have any diagnosis of hypertension validated by HBPM, or ABPM overnight, and will look carefully at the night time readings for the latter. This is so because I know my clinic BP measurements are not up to research standard, and because I have no idea what my ABPM patients were up to during waking hours. So I will look carefully at their HBPM diary and at the night readings of the ABPM and, while I may well be wrong, I do not think that either Brown or McManus would have a problem with that.
Brown M, Cruikshank K, MacDonald T. Navigating the shoals in hypertension: discovery and guidance. BMJ, 2012;344:d8218
Campbell N, Culleton B and McKay D. Misclassification of blood pressure measurement in ambulatory physician practices. AmJHypert, Dec 2005; 18(12):1522-1527
Chobanian A et al. The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure: the JNC 7 report. JAMA. 2003 May 21;289(19):2560-72
Guidelines Committee. 2003 European Society of Hypertension--European Society of Cardiology guidelines for the management of arterial hypertension. J Hypertens 2003;21: 1011-53
Fagard R, Staessen J, Thijs L and Amery A. Multiple standardized clinic blood pressures may predict left ventricular mass as well as ambulatory monitoring: A metaanalysis of comparative studies. AmJHypert, May 1995; 8(5): 533-540
Hutchinson L, Calinas-Correia J. Preliminary results of the evaluation of the positive predictive value of current standard methodology for the diagnosis of hypertension. Report. Trent Research Development Support Unit, 2007. (Research capacity development grant.) (unpublished)
McManus R, Caulfield M, Williams B. Hypertension Guidelines 2011: NICE authors respond. BMJ 2012;344:e181
Mancia G et al. Ambulatory blood pressure is superior to clinic blood pressure in predicting treatment-induced regression of left ventricular hypertrophy. SAMPLE Study Group. Study on Ambulatory Monitoring of Blood Pressure and Lisinopril Evaluation. Circulation, March 1997; 95(6): 1464-1470
Williams B et al. Guidelines for management of hypertension: report of the fourth working party of the British Hypertension Society, 2004—BHS IV. J Hum Hypertens 2004;18: 139-85
Competing interests: No competing interests
I sympathise with Dr Rogers’ frustration, but unfortunately there are two answers – depending whether absolute or relative risk is plotted against blood pressure (Figure 1). The former is what Dr Rogers calls the raw data, but the latter is, in the Oxford view of risk, the more valid way of disentangling the influence of blood pressure from confounders like age (1). The log-linear relation between cardiovascular risk and blood pressure is proven, for all ages, down to 115/75 mmHg. It does not automatically follow that the threshold for treatment should be reduced from 140/90 mmHg. But it would be comic, if not so tragic, that NICE can without evidence recommend raising the threshold, with a near-certain increase in strokes and myocardial infarction; whilst simultaneous proposals (from the Stroke Research Network and British Hypertension Society) to test a lower threshold were rejected by the Health Technology Assessment arm of the NHS as dangerous and impractical. Our article’s objection to NICE’s guidance was partly to their unjustified conclusions. But our greater concern was the increasing temptation for guideline and funding committees to skip the obstacles facing high-impact, high-risk research seeking novel evidence, in favour of derivative meta-analyses and cost-effectiveness reviews of existing evidence (2).
The threshold question raised by Dr Rogers bears also on the issue of whether hypertension is ‘just’ the skewed end of normal blood pressure distribution – the old Platt-Pickering debate. The failure of genome-wide association studies, of 100,000 people with largely ‘normal’ blood pressure, to account for more than a few mmHg of blood pressure contrasts with the excitement of finding single mutations which explain a whole syndrome, e.g. hypokalaemic hypertension, in 5% of hypertensive patients (3-5). It is unlikely that the only monogenic syndromes have phenotypes which trip off the autoanalyser. It will be splitting clinicians, not lumping epidemiologists, who recognise new phenotypes.
1. Lewington S, Clarke R, Qizilbash N, Peto R, Collins R. Age-specific relevance of usual blood pressure to vascular mortality: a meta-analysis of individual data for one million adults in 61 prospective studies. Lancet 2002;360(9349):1903-13.
2. Rawlins M. A new pathway for the regulation and governance of health research. http://www.acmedsci.ac.uk/index.php?pid=47&prid=88 2011.
3. Ehret GB, Munroe PB, Rice KM, Bochud M, Johnson AD, Chasman DI, et al. Genetic variants in novel pathways influence blood pressure and cardiovascular disease risk. Nature 2011;478(7367):103-9.
4. Choi M, Scholl UI, Yue P, Bjorklund P, Zhao B, Nelson-Williams C, et al. K+ channel mutations in adrenal aldosterone-producing adenomas and hereditary hypertension. Science 2011;331(6018):768-72.
5. Azizan EA, Lam B, Newhouse S, Zhou J, Clarke J, Happerfield L, et al. Microarray, qPCR and KCNJ5 sequencing of aldosterone-producing adenomas reveal differences in genotype and phenotype between zona glomerulosa- and zona fasciculata-like tumors. J Clin Endocrinol Metab 2012;97:(in press).
Competing interests: No competing interests
Should treatment be extended to all those whose blood pressure increases their cardiovascular risk?
As a GP without an extensive knowledge of statistical methods I cannot fathom the workings of Lewington et al in reference 4.
I am sure that I am not alone.
The problem is simply: is the relationship between BP and mortality a threshold one, as exhibited when raw data is analysed , or is the relationship a linear one as exhibited when log surrogates are substituted. (European Heart Journal (2000) 21, 1635–1638
doi:10.1053/euhj.2000.2227,There is a non-linear relationship between mortality and
It's either one or the other.
Competing interests: No competing interests