Conference Coverage
WASHINGTON – Renal denervation’s not so simple after all.
Highly anticipated data from the large, sham-controlled renal-artery catheter denervation trial, SYMPLICITY HTN-3, gave its investigators reason to believe that inconsistent and inadequate delivery of nerve-ablating radiofrequency energy by inexperienced operators left too many patients undertreated, producing a neutral outcome for the primary endpoint, the study’s top-line result first reported in a company press release in January.
In the SYMPLICITY HTN-3 (Renal Denervation in Patients With Uncontrolled Hypertension) trial, the average systolic blood pressure reduction among the 364 patients randomized to renal denervation was 14 mm Hg, compared with an average 12 mm Hg reduction in 174 control patients who underwent a renal-artery angiogram masked to resemble a denervation procedure, a difference that was not statistically significant, falling substantially short of the prespecified 5 mm Hg between group difference necessary to fulfill superiority, Dr. Deepak L. Bhatt said at the annual meeting of the American College of Cardiology.
Mitchel L. Zoler/Frontline Medical News
Dr. Deepak L. Bhatt
Although several factors likely led to the neutral outcome, including an unexpectedly robust blood pressure drop in the control arm, another major factor, many suspect, was inadequate delivery of the denervation treatment.
"It might be that the way we used the catheter wasn’t ideal," said Dr. Bhatt, coprincipal investigator of the study, professor of medicine at Harvard Medical School and executive director of interventional cardiology at Brigham and Women’s Hospital, both in Boston.
Concurrent with Dr. Bhatt’s report at the meeting a published version appeared online (N. Engl. J. Med. 2014 March 29 [doi:10.1056/NEJMoa1402670]).
"There is the concept that renal denervation is easy. As a catheter procedure it is easy, but to achieve denervation is far from easy," said Dr. Murray D. Esler, a professor of medicine at Monash University and senior director of the Baker IDI Heart and Diabetes Institute, both in Melbourne, and a pioneer of renal denervation who was involved in prior SYMPLICITY studies of the same renal-denervation device but was not a coinvestigator for the new trial.
"The SYMPLICITY 3 study made a mistake thinking it could get away with not giving the proceduralists any hands-on experience" before the trial. "There was none. The operators hadn’t done it before, and the trial’s designers would not have done that unless they thought that denervation was easy. Well, it’s not easy. There is a learning curve, and I think that’s the main explanation of what went wrong. "
Mitchel L. Zoler/Frontline Medical News
Dr. Murray Esler
Exacerbating the operators’ inexperience with denervation (they were all experienced interventional cardiologists) was a second flaw in the trial’s design: No method was used to monitor the efficacy of each denervation procedure, an easy decision because there is still no proven way to measure denervation efficacy during the procedure. Dr. Esler said he believes that he has a solution, one that involves testing for interrupted activity of the parallel, afferent nerves of the renal arteries using adenosine as a surrogate measure for denervation of the target efferent nerves. But while he is optimistic the adenosine approach will work, it is still at least several months from use in a trial.
In the first clinical study of renal denervation, a series of 50 patients treated at five centers in Australia and Europe reported just 5 years ago (Lancet 2009;373:1275-81), Dr. Esler, a lead investigator for that study, and his associates carefully tested the efficacy of denervation in 10 patients by measuring the direct effect of successful denervation, reduction of renal norepinephrine spillover. They reported that, in those 10 patients, renal denervation cut norepinephrine spillover by an average 47%, which correlated with an average reduction in systolic blood pressure of 22 mm Hg after 6 months.
But this effect from denervation was "far from uniform and ranged from zero to 85%," depending on who was doing the denervation. "It’s obviously operator dependent." Dr. Esler said in an interview. This observation of enormous variability in the physiologic impact of denervation "took on a new context with the failed trial. The denervation done by novices explains to some extent the spread [in results] and makes me think the same thing happened in SYMPLICITY 3."
Nick Piegari/Frontline Medical News
Dr. George Bakris
"If you don’t do enough high-frequency burns or if the catheter doesn’t reach enough nerves" because of misplacement within the renal artery, the denervation won’t be very successful, noted Dr. George Bakris, professor of medicine and director of the comprehensive hypertension center at the University of Chicago and coprincipal investigator of SYMPLICITY HTN-3 along with Dr. Bhatt. "It certainly calls into question whether they denervated effectively. One could argue that if denervation were done appropriately" it would add to drug treatment, Dr. Bakris said in an interview. "This issue has come up about the quality of the denervation; the good news is that it can be done correctly," but that will require "a good understanding of what you are doing to the renal nerves and how much energy you need." SYMPLICITY HTN-3 "was a good but expensive learning experience," Dr. Bakris said.