Renal denervation - radiofrequency vs. ultrasound: insights from a mixed treatment comparison meta-analysis of randomized sham controlled trials.

Abstract

Multiple randomized trials have shown that renal denervation (RDN) reduces blood pressure (BP) when compared with sham control but the antihypertensive efficacy of radiofrequency vs. ultrasound-based RDN is uncertain. We aimed to compare the outcomes of radiofrequency RDN (rRDN) and ultrasound RDN (uRDN), when compared with sham in patients with hypertension. PubMed, EMBASE, and clinicaltrials.gov databases were searched for randomized sham-controlled trials (RCTs) of rRDN or uRDN or for trials of rRDN vs. uRDN. Primary efficacy outcome was 24-h ambulatory SBP. A mixed treatment comparison meta-analysis was performed comparing the efficacy and safety against sham and against each other. Among 13 RCTs that enrolled 2285 hypertensive patients, rRDN reduced 24-h ambulatory SBP [(MD = 2.34 mmHg; 95% confidence interval (95% CI): 0.72-3.95], office SBP (MD = 5.04 mmHg; 95% CI: 2.68-7.40)], and office DBP (MD = 2.95 mmHg; 95% CI: 1.68-4.22) when compared with sham. Similarly, uRDN reduced 24-h ambulatory SBP (MD = 4.74 mmHg; 95% CI: 2.80-6.67), day-time ambulatory SBP (MD = 5.40 mmHg; 95% CI: 3.68-7.13), night-time ambulatory SBP (MD = 3.84 mmHg; 95% CI: 0.02-7.67), and office SBP (3.98 mmHg; 95% CI: 0.78-7.19) when compared with sham. There was significantly greater reduction in 24-h ambulatory SBP (MD = 2.40 mmHg; 95% CI: 0.09-4.71), day-time ambulatory SBP (MD = 4.09 mmHg; 95% CI: 1.61-6.56), and night-time ambulatory SBP (MD = 5.76 mmHg; 95% CI: 0.48-11.0) with uRDN when compared with rRDN. For primary efficacy outcome, uRDN ranked #1, followed by rRDN (#2), and sham (#3). In hypertensive patients, rRDN and uRDN significantly reduced 24-h ambulatory and office SBP when compared with sham control with significantly greater reduction in ambulatory BP with uRDN than with rRDN at 4 months (mean) of follow-up. A large-scale randomized head-to-head trial of rRDN or uRDN is warranted to evaluate if there are differences in efficacy.