What is resistant hypertension? This question is important because for several years, it seemed to define a large group of patients who warranted the development of new therapies and defined a basis for selecting patients for trials of interventions, such as renal denervation. It provided the foundation for the entire scientific dialogue in this field. There have been more than 300 publications on renal denervation in resistant hypertension in the last 5 years.1 In short, what are we all talking about?


Although there are formal definitions, clinicians commonly consider resistant hypertension to be a permanent diagnosis of the patient. In routine clinical practice, the diagnosis of resistant hypertension tends to be accepted without formally requiring confirmation that the patient’s body is resistant to the blood pressure-lowering effect of therapy that has definitely reached the bloodstream. Our experience in the cath lab is that significant doses of intravenous nitrate never seem to fail to have a blood pressure-lowering effect, even when this is unwanted. Intravenous adenosine, a similarly familiar drug in the cath lab, also seems very reliable in lowering blood pressure.

As endovascular physicians, we are sure that readers share our curiosity that endovascularly administered medication seems to lower blood pressure in essentially all patients, even when inconvenient to the procedure and unhelpful for the patient. We think that this is because prescription of oral medication is different from direct infusion of intravenous medication in that there are many stages at which the agent may not progress to the bloodstream. First, the prescription may not be filled. Second, even if one obtains the medication, everyone’s lives are filled with many competing demands for attention, and one could easily forget to take the tablet. Third, although it may be surprising to us as physicians, many people do not think it worthwhile to take medications to substantially increase lifespan, even if they are free of cost, side effects, and loss of autonomy.2


Our scientific analysis of the field demonstrated that there is a simple procedure that can be undertaken in out- patients to reduce blood pressure by 35/15 mm Hg. This is slightly larger than that claimed for renal denervation even in its headiest early days.3 It is a molecular approach that has proved effective in controlling an element of the world’s second most common cause of death and should also prove effective if applied systematically for the world’s leading cause of death, which is cardiovascular disease.4

This technology is noninvasive and reversible. Although it requires skilled professional input, the training necessary is not as extensive as for invasive procedures. This would eliminate any possibility that commentators could suggest there were regional differences in the effectiveness with which therapy was delivered.5

Those involved in SYMPLICITY HTN-36 rebutted the suggestion of regional differences, explaining that the procedures were done “with a large degree of care” and close involvement of manufacturers to “make sure the procedures were done in the most effective way.”7 We favor this second view. We suspect that the difference between trials arises from the lack of blinding in the non-North American study measurements. The therapeutic approach we describe would sidestep such an unhelpful innuendo about interventionists of any region.

The therapy that delivers a 35/15-mm Hg reduction should be additive to renal denervation, according to the best available scientific information to date, because it engages multiple therapeutic targets.

There are no regulatory hurdles. This may be important for readers in the United States, where the Food and Drug Administration has come under acerbic criticism8,9 for requiring reliable evidence of nascent10 therapies. Perhaps an enterprising reader could devise a path to profit? There seems ample corporate funding available to support ventures in this arena.3 The key intellectual property is described in the Appendix sidebar.


There may be a chasm between the disease we think we are targeting for clinical study and the patients we are actually recruiting into trials. In recent years, the concept of pseudoresistant hypertension has become established among experts to represent phenomena such as nonadherence. This useful insight may not yet be coming to the forefront of every clinician’s mind.

Moreover, clinical trials of resistant hypertension rarely describe any formal strategy to enroll only patients with biological resistance to the molecule in the blood, and exclude those in whom the resistance occurs before the bloodstream (eg, not taking the medication).


How quickly science moves from nowhere to an accurate answer depends not only on the technology available but also on the evaluators’ freedom from bias. Two thousand years ago, Eratosthenes estimated the circumference of the Earth by peering down a well and using some mathematics.11 His answer was correct to within 0.16%. We believe an important driver for Eratosthenes’ astoundingly accurate answer was the absence of professional pressure upon him to overestimate or underestimate the value. It is for similar reasons that we know the speed of light: no one stands to gain from overstating its value. Indeed, some have lost from doing so.12

Medical research, in contrast, is troubled by our assumption that those who can report large effects from a therapy must have greater clinical skill than those who cannot manage this. In reality, this assumption may fail to reflect other contributors to large reported effects, including unintentional bias in measurements3,13,14 or even research error.15

Clinical therapeutic discoveries have an unenviable track record of early overestimation of effect sizes followed by agonizing years of serial downgrades of expectations. Some leading experts imply that pioneers’ exaggeration of therapeutic effects is well-accepted and even apply a name, the Proteus effect.16 Our alternative view is that science should seek to deliver reliable information without regard to any investment cycle.3

We should compare the efficiency with which the truth is converged upon, between the example of lonely Eratosthenes measuring the 40,000-km circumference of planet Earth and an entire international scientific community counting the millimeters of mercury of blood pressure decrease resulting from renal denervation.


Four decades ago, Kübler-Ross17 laid out the sequence of events when people face a profound emotional loss. We invite readers to examine Table 1. In the left column, we list the reasons enumerated in the most up-to-date analysis5 by the most eminent authorities of why SYMPLICITY HTN-36 did not match the previous studies. In the right column, we have added our observations. We invite readers to write to us via Endovascular Today to suggest which stage our field has currently reached (evteditorial@bmctoday.com).


We have previously written in Endovascular Today and described in more detail elsewhere18 three little-discussed causes for trials of novel therapy to inadvertently overstate efficacy.

First is what we call big day bias, which is the tendency of patients, recruited for having unusually high values of a variable, to experience a decrease in that value on the next measurement. Imagine measuring the temperatures of 10,000 people and selecting the hundred with the highest values for enrollment in a single-arm trial of an emerging therapy.10 Six months later, those 100 patients will certainly show a statistically significant decline in temperature, even if the therapy was a placebo. Statisticians call this regression to the mean, but we offer the simpler term big day bias to make the mechanism more instantly recognizable. Having a randomized control arm eliminates this bias.

Second is what we call check once more bias, which is our universal habit as clinicians of remeasuring a value, be it blood pressure or ejection fraction, if the first measured value is clinically doubted. The problem is that in routine clinical practice (unblinded to treatment allocation), we generally cannot resist doing this in a way that (clinically) seems sensible but (scientifically) amounts to bias. We have established this publicly by polling numerous audiences totaling thousands of specialists.19 A measure that cannot be easily discarded and remeasured (eg, ambulatory blood pressure monitoring) can eliminate this bias.

Third is what we call I’d better take them now bias, which is the potential for patients to be inspired to greater concordance with their medication regimen once faced with the seriousness of their condition through undergoing an invasive procedure, perhaps sharing a ward with patients suffering the sequelae of unsuccessfully controlled cardiovascular risk factors. Blinding the patients and their clinical team to the randomization arm eliminates this bias.


Just as electorates get the government they deserve, we as clinicians get the science we deserve.20 If we are an inattentive audience, we will be responsible for incorrect information filling lectures and literature. Undoing this will take many years, loss of face, and millions of dollars of unnecessary effort. Fortunately for us, the present liability situation is that when the most carefully bias-resistant research is eventually completed,21 we are safe from ever being handed the bill for the cost of reaching that point, which is the true price of our earlier uncritical applause.22

If we want to help our own field, we should help it recognize good research design and bad research design by pointing out the differences at every opportunity. Hesitancy in pointing out design errors in research should be considered as inexcusable as failure to point out to a colleague that they are about to saw off the wrong leg of a patient. It may not be one’s job, but it is one’s professional responsibility.


We believe that our community could be more explicit about the patients who could benefit from procedural intervention. Instead of defining them by the number of tablets prescribed, their doses, or their nationality,23 we could openly offer to enroll patients who would rather take the potential risk of an invasive procedure than take additional tablets.

There is zero reason for there to be a scientifically solid cutoff in the number of medications to reliably identify for which patients renal denervation is more appropriate than taking additional tablets. Those who think otherwise should ask themselves, “What is the cutoff on the level of nearsightedness that makes a laser eye surgery superior to wearing glasses?” Having done that, they should see if colleagues answering the same question without conferring come to the same answer.

Instead, we suggest considering procedures genuinely complementary to medical therapy.24 We must recognize that individual patients will vary in their preferences: some will ask for procedural approaches as the first choice while, at the other extreme, others will view it as a last resort. To predetermine that the threshold should be the same for all, without rational quantitative reasoning, is not wise.


For the future, we should think carefully about how to manage what we currently describe as resistant hypertension. Even when it occurs due to unwillingness to take medications consistently, simply repeatedly telling patients to take their tablets may not be the only option. By the time they reach the specialist readers of Endovascular Today, they are likely to have had this advice many times; even innocently describing them as resistant risks conveying to patients that they possess a biochemical block to tablet efficacy. It might be preferable to accept that they have simply reached the limit of the medication that they can sustain.

Our clinical scientific community should work together to identify reliable effect size estimate data, which, in our opinion, arises from blinded, randomized, controlled trials.3,6 When we attempt to mix and match effect sizes from different trial designs, there is a risk of introducing confusion at the higher level.

Wrong trial design gives the wrong answer that leads to wrong directions in further research and wrong treatment decisions. It is up to all of us to build the future, and the future begins now.

Graham D. Cole, MA, MRCP, is with the International Centre for Circulatory Health, National Heart and Lung Institute, Imperial College, London, United Kingdom. He has stated that he has no financial interest related to this article.
James P. Howard, MA, MRCP, is with the International Centre for Circulatory Health, National Heart and Lung Institute, Imperial College, London, United Kingdom. He has stated that he has no financial interest related to this article.
Darrel P. Francis, MD, MA, FRCP, is with the International Centre for Circulatory Health, National Heart and Lung Institute, Imperial College, London, United Kingdom. He has disclosed that he is a consultant to Medtronic. He may be reached at +44 207 594 1093; darrel@drfrancis.org.

  1. 1. http://www.ncbi.nlm.nih.gov/pubmed/?term=%22resistant+hypertension%22+AND+%22renal+denervation%22 Accessed: August 12, 2014.
  2. Fontana M, Asaria P, Moraldo M, et al. Patient-accessible tool for shared decision making in cardiovascular primary prevention: balancing longevity benefits against medication disutility. Circulation. 2014;129:2539-2546.
  3. Shun-Shin MJ, Howard JP, Francis DP. Removing the hype from hypertension. BMJ. 2014;348:g 1937.
  4. Nowbar AN, Howard JP, Finegold JA, et al. 2014 global geographic analysis of mortality from ischaemic heart disease by country, age and income: statistics from World Health Organisation and United Nations. Int J Cardiol. 2014;174:293-298.
  5. Lüscher TF, Mahfoud F. Renal nerve ablation after SYMPLICITY HTN-3: confused at the higher level? Eur Heart J. 2014;35:1706-1711.
  6. Bhatt DL, Kandzari DE, O’Neil lWW, et al; SYMPLICITY HTN-3 Investigators. A controlled trial of renal denervation for resistant hypertension. N Engl J Med. 2014;370:1393-1401.
  7. Kirtane AJ. Ajay Kirtane on treatment of difficult to control hypertension in the US. RHC Online. Available at: http://www. resistanthypertensioncourse.com/videos/ajay-kirtane-treatment-difficult-control-hypertension-us.
  8. Gottlieb S. The FDA wants you for sham surgery. The Wall Street Journal. Feb 18 2014. Available at: http://online.wsj.com/ news/article_email/SB10001424052702304680904579365414108916816-lMyQjAxMTA0MDEwODExNDgyWj. Accessed August 12, 2014.
  9. Gottlieb S. The FDA wants you for sham surgery. American Enterprise Institute. Feb 19 2014. Available at: http://tinyurl.com/fda-demands-placebo-control. Accessed August 12, 2014.
  10. Rosen MR, Myerberg RJ, Francis DP, et al. Translating stem cell research to cardiac disease therapies: pitfalls and prospects for improvement. J Am Coll Cardiol. In press.
  11. Roller, Duane W. Eratosthenes’ Geography. New Jersey: Princeton University Press, 2010.
  12. Reich ES. Embattled neutrino project leaders step down. Nature. April 2, 2012. Available at: http://www.nature.com/news/embattled-neutrino-project-leaders-step-down-1.10371. Accessed June 17, 2012.
  13. Shun-Shin MJ, Francis DP. Why even more clinical research studies may be false: effect of asymmetrical handling of clinically unexpected values. PLoS One. 2013;8:e65323.
  14. Francis DP. How easily can omission of patients, or selection amongst poorly-reproducible measurements, create artificial correlations? Methods for detection and implications for observational research design in cardiology. Int J Cardiol. 2012;167:102-113. 15. Nowbar AN, Mielewczik M, Karavassilis M, et al. Discrepancies in autologous bone marrow stem cell trials and enhancement of ejection fraction (DAMASCENE): weighted regression and meta-analysis. BMJ. 2014;348:g2688.
  15. Moyé L. DAMASCENE and meta-ecological research: a bridge too far. Circ Res. 2014; Jul 8: pii: CIRCRESAHA.114.304767 [Epub ahead of print]
  16. Kübler-Ross E.On death and dying. Bull Am Coll Surg.1975;60:15-17.
  17. Howard JP, Cole GD, Sievert H, et al. Unintentional overestimation of an expected antihypertensive effect in drug and device trials: mechanisms and solutions. Int J Cardiol. 2014;172:29-35.
  18. British HTN Society BHS. Debate: sympathetic denervation—the case against—Darrel Francis. November 24, 2013. Available at: https://www.youtube.com/watch?v=kTtK0FJJ1W4. Accessed June 6, 2014.
  19. Cole GD, Francis DP. The challenge of delivering reliable science and guidelines: opportunities for all to participate. Eur Heart J. 2014. In press.
  20. Ruschitzka F, Abraham WT, Singh JP, et al. Cardiac-resynchronization therapy in heart failure with an arrow QRS complex. N Engl J Med. 2013;369:1395-1405.
  21. Nijjer SS, Pabari PA, Stegemann B, et al. The limit of plausibility for predictors of response: application to biventricular pacing. JACC Cardiovasc Imaging. 2012;5:1046-1065.
  22. Wood S. It ain’t over: renal denervation data dominate Euro PCR as research presses on. Medscape. May 26, 2014. Available at: http://www.medscape.com/viewarticle/825695#2. Accessed August 12, 2014.
  23. Fadl Elmula FE1, Hoffmann P, Larstorp AC, et al. Adjusted drug treatment is superior to renal sympathetic denervation in patients with true treatment-resistant hypertension. Hypertension. 2014;63:991-999. doi: 10.1161/HYPERTENSIONAHA.114.03246. Epub 2014 Mar 3.
  24. Bunker J, Callister W, Chang CL, Sever PS. How common is true resistant hypertension? J Hum Hypertens. 2011;25:137-140.