Interventional procedure overview of percutaneous transluminal renal sympathetic denervation for resistant hypertension

Radiofrequency RDN

In a Cochrane review of 15 studies (1,416 patients with resistant hypertension) that compared percutaneous transluminal renal sympathetic denervation (RDN) with antihypertensive therapy or sham procedure (control), there were statistically significantly greater reductions in 24-hour ambulatory systolic BP (mean difference [MD] -5.29 mmHg, 95% CI -10.46 to -0.13, p=0.04; I²=77%; 9 studies, n=1,045; GRADE, moderate quality) and 24-hour ambulatory diastolic BP (MD -3.75 mmHg, 95% CI -7.10 to -0.39, p=0.03; I²=73%; 8 studies, n=1,004; GRADE, moderate quality) after RDN compared with control. The high heterogeneity depended on the type of radiofrequency system (multi-electrode compared with single-electrode catheter). For daytime ambulatory BP monitoring, there were less reductions in systolic and diastolic BP after RDN compared with control, but the effects were not statistically significant (systolic BP, MD 3.87 mmHg, 95% CI -5.02 to 12.76, p=0.39, I²=70%, 5 studies, n=234; diastolic BP, MD 2.93 mmHg, 95% CI -3.22 to 9.08, p=0.35, I²=76%, 5 studies, n=234). For nighttime ambulatory BP monitoring, there were greater reductions in systolic and diastolic BP after RDN compared with control, but the effects were not statistically significant (systolic BP, MD -1.65 mmHg, 95% CI -12.74 to 9.45, p=0.77, I²=75%, 5 studies, n=234; diastolic BP, MD -1.08 mmHg, 95% CI -9.25 to 7.08, p=0.79, I²=87%, 5 studies, n=234; Pisano 2021).

In a case series of 2,237 patients with uncontrolled hypertension (antihypertensive medication classes, mean 4.5) who had RDN (using the single-electrode denervation system), 24-hour ambulatory systolic BP statistically significantly decreased (mean change, -7.2 mmHg, p<0.0001) at 6-month follow up. Statistically significant decrease in 24-hour ambulatory systolic BP was sustained over 3 years (1 year, -7.2 mmHg; 2 years, -8.2 mmHg; 3 years, -8.0 mmHg; all p<0.0001; Mahfoud 2019).

In a case series of 407 patients with resistant hypertension who had RDN (mainly using the single-electrode denervation system), 24-hour systolic and diastolic BP reduced at 3 months (mean change in systolic BP, -8 mmHg [SD 19]; mean change diastolic BP, -4 mmHg [SD 13]), 6 months (-8 mmHg [SD 17] and -5 mmHg [SD 11]) and 12 months (-10 mmHg [SD 18] and -6 mmHg [SD 12]). All reductions were statistically significant (all p<0.001). Both systolic and diastolic BP at daytime and nighttime also statistically significantly reduced at 3 months, 6 months and 12 months after the procedure (all p<0.001). At 6 months, the ambulatory BP response rate (24-hour ambulatory BP reduction of 5 mmHg or more) was 55% (120/220). In total, 22% of patients reached the systolic 24-hour BP goal of less than130 mmHg at every follow up. Subgroup analysis showed that 24-hour BP reduction after the procedure was statistically significantly more apparent in group A (patients with a mean 24-hour BP more than145/90 mmHg, equivalent to an office BP of 160/100 mmHg) compared with group B (patients with a mean 24-h BP of 145/90 mmHg or below; p<0.01 at every follow-up; Zweiker 2016).

In a case series of 253 patients with resistant hypertension who had RDN (mainly using the single-electrode denervation system), daytime ambulatory BP statistically significantly reduced from 170 mmHg (SD 22) at baseline to 158 mmHg (SD 25) at 8.5-month follow up for systolic BP (MD, -12 mmHg; p<0.001), and from 98 mmHg (SD 16) to 91 mmHg (SD 17) for diastolic BP (MD, -7 mmHg; p<0.001). For nighttime ambulatory BP, systolic BP decreased from 154 mmHg (SD 26) to 145 mmHg (SD 26) and diastolic BP from 86 mmHg (SD 18) to 83 mmHg (SD 17). According to baseline daytime ambulatory systolic BP from quartiles 1 to 4 (142 mmHg for quartile 1, 162 mmHg for quartile 2, 176 mmHg for quartile 3 and 199 mmHg for quartile 4), at 8.5-month follow up the mean reduction in daytime ambulatory systolic BP was 0.4 mmHg for quartile 1, 6.5 mmHg for quartile 2, 14.5 mmHg for quartile 3, and 22.1 mmHg for quartile 4 (p value for quartile trend <0.001). Overall, 62% of patients who responded to RDN (daytime ambulatory systolic BP reduction of 5 mmHg or more; Sharp 2016).

Ultrasound RDN

In an RCT of 136 patients with resistant hypertension who had RDN or sham procedure, 24-hour ambulatory systolic BP reduced in both groups (RDN, -6.6 mmHg, 95% CI -10.4 to -2.8; sham, -6.5 mmHg, 95% CI -10.3 to -2.7) at 3-month follow up. There were no statistically significant differences in 24-hour ambulatory systolic and diastolic BP reductions between groups (difference in systolic BP, −0.1 mmHg, p=0.971; difference in diastolic BP, -0.4 mmHg, p=0.806; Kario 2022).

In an RCT of 136 patients with resistant hypertension who had RDN or sham procedure, 24-hour ambulatory systolic BP reduced in the RDN group (median difference, -8.5 mmHg, IQR -15.1 to 0.0) and the sham group (median difference, -2.9 mmHg, IQR -12.6 to 2.5) at 2-month follow up. The median between-group difference was -4.2 mmHg (95% CI -8.3 to -0.3, baseline-adjusted p=0.016). During the same period, 24-hour ambulatory diastolic BP also reduced in both groups (median difference in the RDN group, -5.4 mmHg, IQR -10.4 to 0.0; median difference in the sham group, -2.4 mmHg, IQR -7.8 to 0.5), with a median between-group difference being -2.0 mmHg (95% CI -4.5 to 0.6; baseline-adjusted p=0.12). When considering daytime and nighttime BP separately, there were statistically significantly greater reductions in daytime and nighttime systolic BP after RDN compared with sham (median between-group difference in daytime systolic BP, -4.5 mmHg, 95% CI -8.5 to -0.3, baseline-adjusted p=0.022; median between-group difference in nighttime systolic BP, -3.9 mmHg, 95% CI -8.8 to 1.0, baseline-adjusted p=0.044) but not for diastolic BP (median between-group difference in daytime diastolic BP, -1.8 mmHg; 95% CI -4.5 to 0.8, baseline-adjusted p=0.18; median between-group difference in nighttime diastolic BP, -2.8 mmHg, 95% CI -6.1 to 0.2, baseline-adjusted p=0.053; Azizi 2021).

Radiofrequency and ultrasound RDN

In a 3-arm randomised trial of 117 patients with resistant hypertension who had radiofrequency ablation of the main renal artery (RFM-RDN using the multi-electrode denervation system), radiofrequency ablation of the main renal artery, branches, and accessories (RFB-RDNusing the multi-electrode denervation system) or ultrasound-based ablation of the main renal artery (USM-RDN), there were statistically significant reductions in daytime systolic BP (-9.5 mmHg [SD 12.3]), daytime diastolic BP (-6.3 mmHg [SD 7.8]) and nighttime systolic BP (- 6.1 mmHg [SD 14.2]) in all patients (all p<0.001) at 3 months after treatment. Comparison between groups showed that there was a statistically significantly greater reduction in daytime systolic BP in the USM-RDN group than the RFM-RDN group (MD -6.7 mmHg, 98.3% CI -13.2 to -0.2, adjusted p=0.043) but not between RFM-RDN and RFB-RDN (MD -1.8 mmHg, 98.3% CI –8.5 to 4.9, adjusted p>0.99) and between USM-RDN and RFB-RDN (MD -4.9 mmHg, 98.3% CI -11.5 to 1.7, adjusted p=0.22). The systolic BP response rate (systolic BP reduction of 5 mmHg or more) was observed in 66% of patients who had RFM-RDN compared with 73% in the RFB-RDN group and 67% in the USM-RDN group (p=0.77). Profound BP response was found in 8% of patients who had RFM-RDN, 14% of patients who had RFB-RDN, and 29% of patients who had USM-RDN (p=0.039; Fengler 2019).

In a case series of 296 patients with resistant hypertension who had RDN using radiofrequency or ultrasound, 24-hour ambulatory systolic and diastolic BP statistically significantly reduced by 8.3 mmHg (SD 12.2) and 4.8 mmHg (SD 7.0) at 3 months, by 8.0 mmHg (SD 12.4) and 5.1 mmHg (SD 7.1) at 6 months, and by 8.7 mmHg (SD 14.1) and 5.4 mmHg (SD 7.8) at 12 months (all p<0.001). At 3 months, 61% of patients (180/296) were classified as BP responders (24-hour ambulatory systolic BP reduction of 5 mmHg or more) and 39% (116/296) as non-responders. Systolic BP at 6 months and 12 months remained statistically significantly more reduced in patients who responded to RDN than patients who did not (-12.1 mmHg [SD 2.8] compared with -2.8 mmHg [SD 13.8], and -11.7 mmHg [SD 12.0] compared with -2.0 mmHg [SD 10.7], p<0.001 for both, compared with baseline BP values; Fengler 2021).

Radiofrequency RDN

In the Cochrane review of 15 studies, there was a statistically significantly greater reduction in office diastolic BP after RDN compared with control (MD - 4.61 mmHg, 95% CI -8.23 to -0.99, p=0.01; I²=77%; 8 studies, n=1,049; GRADE, moderate quality) but not for systolic BP (MD -5.92 mmHg, 95% CI -12.94 to 1.10, p=0.10; I²=86%; 9 studies, n=1,090; GRADE, moderate quality). Subgroup analyses showed that benefits on office systolic BP became evident in studies using a multi-electrode radiofrequency catheter (MD -5.10 mmHg, 95% CI -9.14 to -1.06) compared with those using a single-electrode catheter system, also nullifying the heterogeneity among studies (I²=0%; Pisano 2021).

In the case series of 2,237 patients, there was a statistically significant decrease in office systolic BP (mean change, -12.8mmHg, p<0.0001) at 6 months after the procedure. Statistically significant decrease in office systolic BP was sustained over 3 years (1 year, -12.3 mmHg; 2 years, -14.7 mmHg; 3 years, -16.5 mmHg; all p<0.0001; Mahfoud 2019).

In the case series of 407 patients, office systolic and diastolic BP statistically significantly reduced at 3 months (mean change, -16 mmHg [SD 25] and -4 mmHg [SD18]), 6 months (-20 mmHg [SD 26] and -7 mmHg [SD 18]), and 12 months (-20 mmHg [SD 27] and -8 mmHg [SD 18]) after the procedure (all p<0.001). The office BP response rate (office systolic BP decrease of 10 mmHg or more) after 6 months was 69% (128/185). In total, 30% of patients reached the office systolic BP goal of 140 mmHg or less at every follow-up (Zweiker 2016).

In the case series of 253 patients, office BP statistically significantly reduced from 185 mmHg (SD 26) at baseline to 163 mmHg (SD 28) at 11-month follow up for systolic BP (MD -22 mmHg [SD 29]; p<0.001), and from 102 mmHg (SD 19) to 93 mmHg (SD 19) for diastolic BP (MD -9 mmHg [SD 19]; p<0.001). According to baseline daytime ambulatory systolic BP from quartiles 1 to 4 (142 mmHg for quartile 1, 162 mmHg for quartile 2, 176mmHg for quartile 3 and 199 mmHg for quartile 4), at 8.5-month follow up the mean reduction in office systolic BP was 15.2 mmHg for quartile 1, 22.3 mmHg for quartile 2, 22.9 mmHg for quartile 3 and 30.3 mmHg for quartile 4 (p=0.001 for quartile trend). Overall, 65% of patients responded to RDN (office systolic BP reduction of 10 mmHg or more; Sharp 2016).

Ultrasound RDN

In the RCT of 136 patients, office systolic and diastolic BP reduced in the RDN group (-11.0 mmHg and -4.9 mmHg) and the sham group (-9.0 mmHg and -5.0 mmHg) at 3-month follow up. There were no statistically significant differences in office systolic and diastolic BP reductions between groups (difference in systolic BP, −2.0 mmHg, p=0.511; difference in diastolic BP, 0.1 mmHg, p=0.946; Kario 2022).

In the RCT of 136 patients, office systolic BP reduced in the RDN group (median difference, -9.0 mmHg, IQR -19.5 to -1.5) and the sham group (median difference, -4.0 mmHg, IQR -12.0 to 9.0) at 2-month follow up. The median between-group difference was statistically significant (-7.0 mmHg, 95% CI - 13.0 to -0.0, baseline adjusted p=0.037). For office diastolic BP, there were reductions in both groups (RDN, -5.0 mmHg, IQR -13.5 to 2.5; sham, -1.0 mmHg, IQR -7.0 to 6.0) but the median between-group difference was not statistically significant (-4.0 mmHg, 95% CI -9.0 to 0.0, baseline-adjusted p=0.16; Azizi 2021).

Radiofrequency RDN

In the Cochrane review of 15 studies, there were greater reductions in home systolic and diastolic BP in the RDN group than the control group, but the effects were not statistically significant (between-group difference in systolic BP, -5.6 mmHg, 95% CI -14.5 to 3.2, p=0.205; between-group difference in diastolic BP, -4.8 mmHg, 95% CI -9.8 to 0.3, p=0.065) in the HTN-JAPAN 2015 study. In the DENER-HTN 2015 study, home systolic and diastolic BP reduced in the RDN group (MD in systolic BP, -15.4 mmHg, 95% CI -20.4 to -10.4; MD in diastolic BP, -8.7 mmHg, 95% CI -12.1 to -5.4) and the control group (MD in systolic BP, -11.8 mmHg, 95% CI -16.5 to -7.1; MD in diastolic BP, -6.7 mmHg, 95% CI -9.8 to -3.5), and the between-group differences were not statistically significant for both systolic and diastolic BP (Pisano 2021).

Ultrasound RDN

In the RCT of 136 patients, home systolic BP reduced by 8.7 mmHg and diastolic BP by 3.6 mmHg in patients who had RDN, and home systolic BP reduced by 6.9 mmHg and diastolic BP by 3.7 mmHg in patients who had sham procedures at 3-month follow up. There were no statistically significant between-group differences in both home systolic and diastolic BP reductions (between-group difference in systolic BP, −1.8 mmHg, p=0.488; between-group difference in diastolic BP, 0.1 mmHg, p=0.949; Kario 2022).

In the RCT of 136 patients, home systolic and diastolic BP reduced in the RDN group (median difference in systolic BP, -6.0 mmHg, IQR -17.0 to 1.5; median difference in diastolic BP, -4.0 mmHg, IQR -9.0 to 2.0) and in the sham group (median difference in systolic BP, -2.0 mmHg, IQR -9.5 to 2.0; median difference in diastolic BP, -1.0 mmHg, IQR -5.0 to 4.0) at 2-month follow up. The median between-group differences were not statistically significant for both systolic BP (- 4.0 mmHg, 95% CI -8.0 to 0.0, baseline-adjusted p=0.052) and diastolic BP reductions (-3.0 mmHg, 95% CI -6.0 to 0.0, baseline-adjusted p=0.053; Azizi 2021).

Radiofrequency RDN

In the case series of 2,237 patients, patients were prescribed 4.5 (SD 1.4) antihypertensive medication classes at baseline and 4.4 (SD 1.5) at 3 years (p<0.001), reflecting a statistically significant decrease in angiotensin-converting enzyme inhibitor and centrally acting sympatholytic use with a concomitant increase in aldosterone antagonist use (Mahfoud 2019).

In the case series of 407 patients, the median number of antihypertensive drugs used was 4 (IQR 4 to 5) at baseline and 4 (IQR 3 to 5) at 12 months after the procedure (difference, 0; IQR -1 to 0, p<0.05; Zweiker 2016).

In the case series of 253 patients, the median number of antihypertensive drugs was 5.0 before the procedure. After the procedure, the mean number of drugs added per patient was 0.36 and the mean number of drugs withdrawn was 0.91. The mean dose changes were 0.21 doses up-titrated per patient and 0.17 doses decreased per patient (Sharp 2016).

Ultrasound RDN

In the RCT of 136 patients, there was no statistically significant difference in the reduction in antihypertensive medication load between the RDN and sham groups (RDN, 4.2 [SD 1.7] at baseline, 4.3 [SD 1.7] at 3 months; sham, 3.9 [SD 1.2] at baseline, 3.9 [SD 1.1] at 3 months; Kario 2022).

In the RCT of 136 patients, 93% (64/69) of patients in the RDN group and 85% (57/67) of patients in the sham group had no change in their antihypertensive treatment (p=0.15) at 2-month follow up. The proportion of patients who had additional antihypertensive medication was 4% (3/69) in the RDN group and 12% (8/67) in the sham group (p=0.10). The proportion of patients who reduced their antihypertensive medications was 3% for each group (p=1.0). The proportion of patients who fully adhered to antihypertensive medications with urine samples was 83% (49/59) at baseline and 82% (42/51) at 2 months in the RDN group, and 76% (44/58) at baseline and 82% (47/57) at 2 months in the sham group (Azizi 2021).

Radiofrequency and ultrasound RDN

In the 3-arm randomised trial of 117 patients, 9% (11/117) of patients changed their medications at 3-month follow up. Of these patients, 3 patients who increased medication dose or number were in the RFM-RDN group, 7 patients (2 decreased, 5 increased) in the RFB-RDN group, and 1 patient who decreased the number of drugs in the USM-RDN group (Fengler 2019).