March 29th, 2014

Negative Findings on Renal Denervation for Hypertension: Sweet SYMPLICITY?

CardioExchange’s Rick Lange and David Hillis interview Deepak L. Bhatt, co-principal investigator of the SYMPLICITY HTN-3 trial, which was presented at the 2014 American College of Cardiology conference in Washington, DC, and published in The New England Journal of Medicine. For CardioExchange’s news coverage of the trial, click here.


Catheter-based radiofrequency denervation of the renal arteries is currently used to treat hypertension in more than 80 countries, largely on the basis of evidence from nonrandomized studies and unblinded randomized trials showing large reductions in blood pressure (BP) after renal denervation.

In the industry-funded SYMPLICITY HTN-3 trial, patients with resistant hypertension (≥160 mm Hg) were randomly assigned to undergo renal-artery denervation or only renal angiography (sham control), but the participants were unaware of their randomized assignment. At 6 months, the two groups did not differ significantly in reduction of office or 24-hour ambulatory systolic BP.


Lange and Hillis: What do you consider the most likely explanation for the discrepancy between the findings from SYMPLICITY HTN-3 and those from previous studies of renal denervation?

Bhatt: At least part of the difference is due to randomization, blinding, and use of a sham control. These are particularly difficult (and expensive) strategies in trials of procedures. However, they do remain important basic elements of rigorous trial design. The control arm had a significant reduction in blood pressure, and I think at a minimum, that amount should be subtracted from prior reports of the magnitude of blood pressure reduction with renal artery denervation. Why our reduction in blood pressure in the treatment arm was so much lower than previously reported remains unknown. It could be specific to the device we studied or it could be the way that the proctors instructed physicians to use the device in this trial. We are examining all these possibilities and hope in the coming months to provide some insight into whether there is a way forward – I remain cautiously optimistic.

Lange and Hillis: What does SYMPLICITY HTN-3 teach us about conducting interventional trials and the importance of blinding?
Unintentional, often well-meaning, bias tends to creep into non-blinded studies. This is less of an issue for a trial powered for “hard” endpoints, such as mortality. For a surrogate endpoint such as blood pressure, even though it is a powerful surrogate, a variety of different measurement biases can occur without blinding.

Lange and Hillis: Would you say that renal denervation for treating hypertension has now been blown out of the water?

Bhatt: I think it is premature to say that. Just as there has been exuberance about the procedure by understandably enthusiastic physicians hoping for a new treatment for a difficult patient subset, it would be extreme to pronounce the field dead on the basis of one trial with one device used in a certain way. The technology may work in other conditions, such as heart failure. Other devices, or even the device we studied used differently, might still have value in renal denervation. I would not say that the renal denervation story is ending, but really is just rebooting, hopefully in a more rigorous way. I think one key is to be certain that we really are effectively denervating with any catheter system before further clinical trials are performed. The procedure may not be as “simple” as we had assumed.

Lange and Hillis: Would you advise any patients with hypertension to undergo renal denervation? If so, whom?

Bhatt: No, I probably would not – at least not at the present time. It is conceivable that the procedure may still have value in patients dissimilar from the ones we enrolled. For example, nonadherent patients with out of control blood pressure were not (knowingly) included in our study, but these patients exist in real life and are a challenge to treat. Our negative study does not mean that the procedure would not work in that type of patient. On the other hand, we have raised the bar for evidence, and I don’t think it is sufficient for physicians to rely on anecdotal successes even in the patient type I described. So, bottom line, I think that further study should continue in renal denervation for resistant hypertension and other conditions, such as heart failure. Clinical use, however, should really wait until there are solid preclinical data followed by proper randomized data supporting performance of the procedure.

Share your thoughts on Dr. Bhatt’s analysis of the implications of the SYMPLICITY HTN-3 trial.

To view all of our coverage from the ACC meeting, go to our ACC.14 Headquarters page.

One Response to “Negative Findings on Renal Denervation for Hypertension: Sweet SYMPLICITY?”

  1. Leon Hyman, Ms M.D. says:

    I wonder why they don’t go back to 2 operations used in the 1940’s and 50’s which cut the splancnic nerves to the legs. The Peet operation cut 40%0f the nerves and worked about 40% of the time and the Smithwick cut 60% percent of the nerves and worked 605 of the time albeit with mire side effects especially orthostatic Hypotension. As a student in Michigan’s hypertension clinic in the summer of 1958 I worked up a man with a blood pressure of 320 over 190 according to his family Dr. who chad an older cuff that went up to 400. When i took it, The systolic was still beeping at 300 and the diastolic cut off sharply at 190. The next day they did the Peet procedure on him, and one month later his BP was 140/90. on 0.25 mg of reserpine a drug no longer on the market. This was before clorthiazide(Diurel) came out. All evidence of malignant hypertension in him was gone including fundal pappiledema