OUP user menu

Treatment-resistant hypertension: another Cinderella story

Franz H. Messerli, Sripal Bangalore
DOI: http://dx.doi.org/10.1093/eurheartj/eht028 1175-1177 First published online: 5 February 2013

This editorial refers to ‘Resistant hypertension: a frequent and ominous finding among hypertensive patients with atherothrombosis’, by D.J. Kumbhani et al., on page 1204

Once upon a time antihypertensive drugs were universally respected. They were respected not only because of their powerful blood pressure-lowering effect but also because of their equally powerful adverse effects, which not uncommonly made vital functions such as urination, defecation, and fornication exceedingly difficult. In those old days antihypertensive drugs were most often used in monotherapy; polypharmacy and fixed combinations were considered an anathema by teachers and attendings. This state of the art is reflected in the Joint National Committee (JNC) in 1984 reminding practising physicians that ‘a diuretic or a β-blocker alone will control blood pressure (BP) in the majority of patients’.1 Relentless marketing efforts by pharmaceutical companies helped to convince physicians further that patients could be managed on monotherapy. Only rarely were dual or even triple combinations necessary to bring blood pressure to its goal. In those heydays of antihypertensive therapy, resistant hypertension was a nonentity or, if you prefer, a Cinderella that was not to be talked about. Resistant hypertension meant failure and defeat; it meant that despite the patient having been fully ‘reserpinized’ or ‘inderalized’ by the ever so skilful physician, blood pressure levels still left much to be desired. An embarrassing situation indeed!

Over the past few years, this issue has changed drastically. Cinderella made it to the ball and turned into a miraculous princess. As seen in Figure 1, the number of yearly publications on resistant hypertension, in a near exponential pattern, has increased more than four-fold over the past decade. However, once again, the emerging interest in treatment-resistant hypertension seems to be driven predominantly by industry rather than by science. Ever since the pioneering information of Schlaich showing that renal denervation lowers blood pressure,2 numerous companies have started to develop tools for this procedure and investigators are canvassing hypertensive patient populations for so-called ‘treatment-resistant hypertension’.

Figure 1

Yearly publications on treatment-resistant hypertension in the last two decades.

Kumbhani et al.,3 in an evaluation of >50 000 patients from the REACH registry, have observed that 12.7% had treatment-resistant hypertension (TRH). Patients with TRH had a heightened risk for adverse long-term outcomes, especially non-fatal stroke and congestive heart failure. The prevalence of TRH in the study varied between 6.0% and 21.7%, and the effects on outcomes were variable (especially on stroke-related outcomes) based on the definition used (Table 1). However, the prevalence from this registry may not represent the true prevalence of TRH for a number of reasons. First, the prevalence in REACH is probably ‘enriched’ by inclusion of a subgroup of patients who made it into the registry because of systolic pressure ≥150 mmHg. Secondly, for certain classes of medications such as angiotensin-converting enzyme (ACE) inhibitors, beta-blockers, nitrates, and diuretics, it is unknown whether they were prescribed for treatment of hypertension or for other indications such as coronary artery disease or heart failure. For example, inclusion of patients with heart failure (27% of patients) in the TRH group becomes problematic as diuretics, beta-blockers, ACE inhibitors, and nitrates may have been used for heart failure rather than for hypertension—driving the prevalence up artificially. Thirdly, data on drug dosage or on duration of treatment with the medications are lacking. Conceivably the registry included patients on three antihypertensive agents who had true TRH for some length of time as well as patients with uncontrolled hypertension who were only recently switched to three agents. Fourthly, there were no data on secondary causes of resistance or on compliance with the medication regimen. Fifthly, and finally, though a robust multivariable model was used to adjust for baseline differences, a number of variables included in the model suffer from a cause and effect relationship. For example, chronic kidney disease could be both the cause of and an effect of TRH. Despite the potential pitfalls mentioned above, the study of Kumbhani et al. is still the largest series to date and offers important insights into the prognosis of this extremely high-risk group of subjects.

View this table:
Table 1

Definition of treatment resistant hypertension

ParameterJNC 711AHA12ESC13NICE14Eligibility criteria for RDN15
DefinitionFailure to reach goal BP (<140/90 mmHg for the overall population or <130/80 mmHg for those with diabetes mellitus or CKD) in patients who are adhering to full doses of an appropriate three-drug regimen that includes a diureticBP that remains above goal in spite of concurrent use of three antihypertensive agents of different classes. Patients whose BP is controlled with four or more medications should be considered to have resistant hypertensionBP levels above goal in spite of the concurrentuse of three antihypertensive agents in adequate doses from different classes including a diureticBlood pressure that remains higher than 140/90 mmHg with the optimal or best tolerated doses of an ACE inhibitor or an ARB plus a CCB plus a diureticOffice BP at least 160 systolic ( ≥150 mmHg in type 2 diabetes) despite treatment with at least three antihypertensve drugs of different types in adequate doses, including one diuretic
Prevalence12.7%21.6%12.7%Not evaluated6.0%
PrognosisIncrease in death/MI/stroke; non-fatal stroke; heart failure hospitalizationIncrease in death/MI/stroke; death; CV death; non-fatal MI; heart failure hospitalizationIncrease in death/MI/stroke; non-fatal stroke; heart failure hospitalizationNot evaluatedIncrease in death/MI/stroke; non-fatal stroke
  • AHA, American Heart Association; ACE, angiotensin-converting enzyme; ARB, angiotensin receptor blocker; BP, blood pressure; CCB, calcium channel blocker; CKD, chronic kidney disease; CV, cardiovascular; ESC, European Society of Cardiology; JNC, Joint National Committee; MI, myocardial infarction; NICE, National Institute for Health and Clinical Excellence; RDN, renal denervation.

Even with an exponential increase in the number of publications on TRH, there is no consensus on a uniform definition (Table 1). We are particularly taking issue with the definition of resistant hypertension by JNC 7, which defines it as failure to reach goal in patients who are adhering to full doses of an appropriate three-drug regimen that includes a diuretic. Clearly, this definition has to be amended in that a full dose of a calcium channel blocker (CCB), a renin–angiotensin system (RAS) blocker, and a diuretic should be mandatory, if tolerated. In this context, it is interesting to note that in the REACH registry, CCBs, which are one of the first-line antihypertensive agents according to the National Institute for Health and Clinical Excellence (NICE) guidelines, were prescribed in only one in every two patients with TRH and were only the fourth most commonly prescribed antihypertensive agent (∼50%).

Chlorthalidone is to be preferred over hydrochlorothiazide (HCTZ), which in its usual dose of 12.5–25 mg is not the best antihypertensive agent.4,5 In an analysis of 19 trials with >1400 patients, HCTZ at the dose of 12.5–25 mg was consistently inferior to ACE inhibitors, angiotensin receptor blockers (ARBs), beta-blockers, and CCBs ambulatory blood pressure reduction.5 In addition, diuretics are not exactly the best tolerated class of agents. Two analyses documented adherence to diuretics to be exceedingly low compared with other classes of agents such as ACE inhibitors, ARBs, CCBs, and beta-blockers.6,7 Although diuretics are not well tolerated and HCTZ in monotherapy is a weak antihypertensive drug, the situation changes drastically when they are combined with RAS blockade. As was documented by 24 h monitoring in ACCOMPLISH in combination with a RAS blocker, i.e. benazepril, HCTZ at the dose of 25 mg was as powerful an antihypertensive as was amlodipine 10 mg.8 The enhanced antihypertensive efficacy in combination may be due to a synergistic mechanism in that diuretics by their volume depletion stimulate the activity of the RAS, thereby making it more amenable to RAS blockade.

Of note, for patients with diabetes, JNC 7 uses a lower cut-off goal of 130/80 mmHg in their definition to identify TRH. However, a recent trial and analyses suggest that a goal of <140/90 mmHg may be reasonable even in diabetics.9,10 In fact in the recent update of the NICE guidelines for hypertension, there is no lower blood pressure threshold for subjects with diabetes. Blood pressure control has the largest effect on stroke and heart failure reduction and a modest effect on reduction in myocardial infarction. The variability in the effect on stroke in the study by Kumbhani et al. highlights the problem with various definitions of TRH. Moreover, the American Heart Association (AHA) guidelines consider patients with controlled blood pressure on four antihypertensive agents as also having TRH. Neither this nor any other definition, although helpful in identifying a subgroup of patients who have worse prognosis despite controlled blood pressure, may fully serve the goal of identifying a subgroup of patients who can be selected for more invasive interventions such as renal sympathetic denervation or baroreflex stimulation. We think that a definition of TRH along the lines of ‘Systolic pressure of at least 160 mmHg or higher despite treatment with a full dose of a RAS blocker (either an ACE inhibitor, an ARB, or a renin inhibitor), a CCB (either dihydropyridine or non-dihydropyridine), a diuretic (preferentially chlorthalidone), and, if tolerated, a mineralocorticoid antagonist (spironolactone or eplerenone)’ may perhaps be more clinically useful.

If history is an indication, it appears that as long as therapies such as renal denervation prove to be efficacious and, perhaps more importantly, remain in fashion, there will be powerful motives to get Cinderella to the ball. However, the lack of a solid definition of TRH, particularly one not including a CCB, should alert us that there is blood in the slipper and serve to distinguish Cinderella from her useless and vile stepsisters.

Conflict of interest: S.B. has served on the Advisory board for Daiichii Sankyo and Boehringher Ingelheim. F.M. is a consultant for Novartis, Daiichi Sankyo, Pfizer, Takeda, Abbott, PharmApprove, Gilead, Servier, Bayer, and Medtronic


  • The opinions expressed in this article are not necessarily those of the Editors of the European Heart Journal or of the European Society of Cardiology.


View Abstract