Disappointing Results from SYMPLICITY-HTN 3 Trial of Renal Denervation in Resistant Hypertension

Renal artery denervation’s effectiveness in treating resistant hypertension was called into question during the Late-Breaking Clinical Trials session on opening day of the American College of Cardiology 2014 Scientific Session.

Deepak Bhat, MD, MPH, (Harvard Medical School and Brigham and Women’s Hospital), presented the results of SYMPLICITY-HTN 3 and gave the packed main hall early news of results of the first prospective, multi-center, randomized, blinded, sham controlled study of percutaneous renal artery denervation.

The trial, which included 535 patients enrolled in 88 participating centers in the US, tested Medtronic’s renal artery denervation device against a sham procedure. In a tightly constructed study design, patients were included if their office systolic blood pressure (SBP) readings averaged >/= 160 mm Hg in three readings over two visits, and they were on a stable regime of three or more hypertensive medications of different classes, including a diuretic.

Patients were excluded if 24-hour blood pressure monitoring yielded average SBP of <135 mm Hg (to capture “white coat hypertension”), or if their history revealed significant renal artery stenosis, other renal artery pathophysiology, or history of prior renal artery intervention; also excluded were subjects with an eGFR of less than 45 mL/min/1.73 m².

After a baseline renal angiogram, subjects were then randomized to a sham procedure, essentially renal angiogram alone, or renal denervation. Subjects were followed for six months, with office visits at one month and three months, two weeks of home blood pressure and hypertension medication confirmation conducted before the six-month point, and the opportunity for crossover to the treatment arm for sham recipients at six months (assuming study criteria were still met).

No medication changes were made during the six-month interval, and patient assessors and study personnel were all blinded to treatment status during this period as well. Peri-procedural blinding precautions, including the use of conscious sedation, blindfolds, and headphones with music for study subjects, were utilized as well.

Major adverse events in the treatment group were benchmarked against an objective performance criterion, derived from historic data. Change over time of office systolic blood pressure from baseline to six months was compared in the renal denervation group compared to the sham group. A superiority margin of 5 mm Hg was established. A secondary effectiveness endpoint comparing mean 24-hr ambulatory (ABPM) SBP change over time from baseline to six months in both the denervation arm and the control arm was also identified, with a superiority margin of 2 mm Hg.

Of the 535 subjects randomized, 364 were randomly allocated to renal denervation, while 171 were assigned to the sham control procedure. Mean age was57.9 years for the intervention group and 56.2 years for the sham procedure. Not quite two-thirds of the subjects were male in both groups. Study groups did not differ significantly in race, comorbidities, or other aspects of medical history. Antihypertensive medication profiles were also similar between the two groups. 96.2% of the intervention group and 98.8% of the sham group followed through to the six-month end point.

The primary safety endpoint was met, with a 1.4% major adverse event rate, well below the performance goal of 9.8% (p <0.001).

Both treatment arms showed a significant reduction in office systolic blood pressure at the six-month end point (14.1 points intervention vs. 11.7 sham, p < 0.001 for both arms); similarly, ABPM SBP was significantly reduced in both groups as well, with the change from baseline reaching statistical significance in both arms. However, the difference in reduction of office SBP between the two groups was 2.4 mm Hg (p=0.26), so statistical significance was not reached for the primary efficacy endpoint. For the secondary efficacy endpoint of ABPM SBP, the difference in reduction of 1.96 mm Hg also failed to reach significance (p=0.98).

A number of limitations were identified by the study’s authors, and also reviewed by the discussion panel after the presentation. Blood levels of antihypertensive medications were not obtained, so drug adherence was not directly measured. The large and significant change over time in both groups may have been related to the placebo effect, or it may be that study participants in both arms achieved improved medication compliance after enrollment in the study. Further, there is no commonly accepted and easily obtained measure of denervation of the renal sympathetic system, so it is possible that renal denervation did not occur, although the study authors could confirm that the appropriate energy had been delivered by the device.

The results of this study, failing to confirm efficacy of renal artery denervation for resistant hypertension, stand in contrast to several other unblinded single-center studies, as well as to animal models and evidence showing that left ventricular hypertrophy is ameliorated after renal artery denervation. At the morning’s wrap-up press conference, UC San Diego’s Anthony DeMaria, MD, one of the discussion panelists, commented, “There was a general assumption that this would be a very effective therapy… So the SYMPLICITY 3 trial, which was conducted extremely well… surprised almost everyone… that the results were negative.”

When asked what advice he would give to clinicians regarding whether renal denervation should be pursued for individuals with resistant hypertension, especially in light of the fact this procedure with this device is clinically approved in Europe, DeMaria urged taking the middle ground. Bhatt noted that more study is indicated, and advocated more rigorous investigation of surgical procedures utilizing a sham arm. However, he stopped short of advising clinicians to stop using the procedure, especially for patients whose hypertension has been stubborn and resistant to maximal multidrug therapy.