Catheter-based renal denervation: the next chapter begins

Hypertension remains the most prevalent modifiable risk factor worldwide.1 Both, the most recent European and US guidelines on hypertension management recommend to lower systolic blood pressure (SBP) to >120–130 mmHg in most patients with hypertension.2,3 This is typically achieved by lifestyle interventions and drug treatment. Antihypertensive drugs are widely available, rather cheap and have been shown to not only reduce blood pressure but also to improve cardiovascular outcomes. Non-adherence to anti-hypertensive medication however, is increasingly recognized as a main contributor to insufficient blood pressure control. The rates of non-adherence in hypertension assessed by toxicological analyses are as high as 66%4 and have been associated with younger age, female gender, number of daily doses, and treatment with diuretics.5 The renal sympathetic nerves are involved in the progression and development of hypertension and hypertension-mediated endorgan damage.6 Several experimental models strongly support the concept of modulating renal sympathetic activity by renal denervation and observational data from the 1950 to 1960’s have shown that surgical sympathectomy lowers blood pressure and was associated with substantially improved survival rates in the majority of patients with hypertension.7,8 These experiences set the stages for the development of catheter-based approaches using radio-frequency, ultrasound, or injection of neurotoxic agents such as alcohol to affect the renal afferent and efferent nerves located in the adventitia of the renal arteries.9 The initial clinical trials focused on patients with severe therapy resistant hypertension, a population in which conservative approaches have failed to control blood pressure. Indeed, the results from the early trials (Symplicity HTN-1 and HTN-2) were rather positive, documenting large falls in office blood pressure with a favourable safety profile, which made the procedure quite attractive for patients with resistant hypertension (Figure 1). Rise and fall of enthusiasm for catheter-based renal denervation hypertension treatment. The field suffered a significant setback though in 2014 with the publication of the first randomized, sham-controlled Symplicity HTN-3 trial.10 This trial failed to prove superiority of renal denervation when compared with medication only. Secondary analysis provided critical insights and established the importance of procedural performance, such as the number of ablations and the relevance of achieving a circumferential, four-quadrant ablation pattern, which was associated with larger blood pressure reductions.11 Furthermore, 38% of the patients in the renal denervation group and 40% in the sham group had medication changes in the first 6 months despite the protocol mandating to maintain antihypertensive regimens constant.12 Interestingly, sub-analysis revealed that the neutral findings were predominantly driven by the African-American subgroup, in whom the sham effect was numerically more pronounced (−18 mmHg) than in the renal denervation group (−16 mmHg). These findings have been related to their low renin, volume-dependent form of hypertension, and/or poor medication adherence. Importantly, an extended analysis, including patients from the Symplicity HTN-3 study and the Global Symplicity Registry showed that patients with isolated systolic hypertension and high pulse wave velocity exhibit a blunted effect of renal denervation on blood pressure when compared with patients with systolic-diastolic hypertension.13 Importantly, in 2015 the results of the DENERHTN study, a multicentre, open-label, randomized controlled clinical trial with blinded endpoint evaluation investigated renal denervation plus standardized stepped-care antihypertensive treatment vs. stepped-care antihypertensive treatment alone in 106 subjects and demonstrated superiority in the renal denervation with an additional decrease of 5.9 mmHg in daytime ambulatory SBP at 6 months.14 The prevalence of non-adherence to antihypertensive drugs using toxicological analyses at 6 months was not different in the renal denervation and control group, indicating that regardless of adherence to treatment, renal denervation plus standardized stepped-care antihypertensive treatment was superior to antihypertensive treatment alone.15 A multidisciplinary group of European and US investigators explored the gaps in our knowledge about device-based hypertension treatments and provided recommendations for the design of future randomized controlled trials.16,17 To address open questions and clarify the role of renal denervation, a series of new studies have been designed after carefully considering the shortcomings of previous trials, such as inadequate patient selection, alterations in compliance with antihypertensive medication, insufficient technical performance of the procedure, and several other device features. The first set of new clinical studies, namely the SPYRAL-OFF, SPYRAL-ON, and the RADIANCE-SOLO have been presented at the European Society of Cardiology (ESC) Congress 2017 in Barcelona and at EuroPCR 2018, and subsequently published in The Lancet.18–20 The primary outcomes of these trials are summarized in Figure 2. Primary endpoints of the most recents trials for RDN and sham. *Mean between-group difference adjusted for baseline systolic blood pressure. All values are means (95% confidence interval). The prospective, randomized, double-blind, sham-controlled SPYRAL-OFF medication study included patients with combined hypertension with an office SBP between 150 and 180 mmHg, office diastolic blood pressure (DBP) >90 mmHg, and ambulatory SBP of 140–170 mmHg with no concomitant antihypertensive therapy (either drug-naïve or after wash-out). Urine and toxicological analyses were used to ensure that the patients adhered to the protocol and were not taking antihypertensive medication. A radiofrequency multi-electrode catheter (Spyral, Medtronic, Ireland), designed to enable reliable circumferential four-quadrant ablation, was utilized and the main distal renal artery and the branches and accessory arteries have been treated. Preclinical studies suggested that a revised procedural technique with treatment of the renal artery branches rather than just the main stem, where the renal sympathetic nerves are more closely located around the vessel lumen, should be applied. Indeed, ablations in the distal main renal artery, are associated with a reduced variability and larger treatment effects using this radio-frequency device.21 In total, the interventionalists performed an average of 43.8 ± 13.1 total ablations, and treated an average of 2.2 main arteries and 5.2 branch vessels. More ablations were performed in the branch vessels when compared with the main arteries (25.9 ± 12.8 vs. 17.9 ± 10.5 ablations). Importantly, only centres and operators with advanced experience in renal denervation, were participating and only one dedicated interventionalist per centre performed the intervention. The primary endpoint of the study was 24-h ambulatory blood pressure after 3 months. The first interim analysis at 3 months of 80 patients treated in the Spyral HTN OFF-MED study documented a significant reduction in both office [SBP −7.7 mmHg, 95% confidence interval (CI) −14.0 to −1.5; P = 0.0155 and DBP −4.9 mmHg, 95% CI −8.5 to −1.4; P = 0.0077] and ambulatory blood pressure (−5.0 mmHg, 95% CI −9.9 to −0.2; P = 0.0414 and 24-h DBP −4.4 mmHg, 95% CI −7.2 to −1.6; P = 0.0024), compared with sham treatment at 3 months (Figure 2). There were no major safety events reported in either group. The prospective, randomized, double-blind, sham-controlled SPYRAL-ON medication study used the same blood pressure criteria as the SPYRAL-OFF study, but included moderate, uncontrolled hypertensive patients on 1–3 commonly prescribed antihypertensive drugs (thiazide-type diuretics, dihydropyridine calcium channel blockers, angiotensin-converting enzyme-inhibitor or angiotensin receptor blockers, and β-blockers). Concomitant antihypertensive pharmacotherapy had to be stable for a minimum of 6 weeks prior to the procedure. On average, patients were on 2.2 ± 0.9 antihypertensive drugs. This study also used toxicological analysis to assess adherence to treatment. The Spyral radiofrequency catheter (Medtronic, Ireland) was used and a total of 45.9 ± 13.7 ablations were applied in a comparable manner as in the SPYRAL-OFF with treatment of the branches and the main renal arteries. The first interim analysis documented a progressive fall in both office and ambulatory blood pressure at 3 and 6 months, respectively (Figure 2). The 24-h blood pressure profile indicated a persistent blood pressure suppression at all timepoints during the day and night period for patients treated with renal denervation but not for sham patients (Figure 3). Adherence to antihypertensive drugs was similar between groups at all timepoints and the changes in blood pressure were not explained by alterations in adherence to antihypertensive medication, which was dynamic. Of note, antihypertensive drugs not prescribed by physicians were detected in 10–15% of patients at each timepoint. The safety profile was favourable with no relevant complications. Systolic and diastolic blood pressure profiles for RDN (left panels) and sham (right panels) in the SPYRAL-ON Medication study. Modified from Kandzari et al.19 The RADIANCE-SOLO study is a multicentre, randomized, double-blind, sham-controlled, 2-cohort (off-medication SOLO cohort and on-medication TRIO cohort) superiority trial, powered at 80% to detect a 6 ± 12 mmHg difference in daytime SBP from baseline to 2 months post-procedure between the groups. The RADIANCE-SOLO study included patients at low cardiovascular risk with essential hypertension either controlled on 1–2 antihypertensive medications (average seated office blood pressure <140/90 mmHg) or uncontrolled on 0–2 hypertensive medications (average seated office blood pressure ≥140/90 mmHg but <180/110 mmHg). After a 4-week observation and wash-out period, subjects with daytime SBP ≥135/85 mmHg were enrolled. The study device was a balloon-based catheter (Paradise, Recor, CA, USA) that ablates renal sympathetic nerves circumferentially using ultrasound energy. A minimum of two ultrasound emissions (7 s each) were delivered in the main branch of the right and left renal artery (average total emissions 5.4 ± 1.0; average total ablation time 37.9 ± 6.7 s). A total of 146 patients were randomized to undergo renal denervation (n = 74) or sham procedure (n = 72), respectively. After 2 months, the trial did meet its prespecified primary endpoint as the mean difference between the groups favoured renal denervation for change in daytime ambulatory SBP (baseline-adjusted difference between groups: −6.3 mmHg, 95% CI −9.4 to −3.1, P = 0.0001; Figure 2). At 2 months follow-up, 20% (n = 15) of the patients in the renal denervation and 3% (n = 2) patients in the sham group exhibited daytime ambulatory blood pressure values of <135/85 mmHg in the absence of antihypertensive medications (P = 0.001). There was also a persistent fall in hourly SBP and DBP observed in the renal denervation group but in sham treated patients. There were no major safety events reported in either group. There is a lack of reliable markers of immediate procedural success confirming effective renal denervation.17 As a result, it remains uncertain, if the lack of a blood pressure drop following the procedure is due to sub-optimal application of the technology or inadequate patient selection or both. Further, it is possible that the number of ablations applied in the current trials exceeded in certain patients the extent required to decrease renal sympathetic nerve activity sufficiently. Once such a procedural guidance would become available, individualized treatment plans could be derived from such information and unnecessary ablations and prolonged procedures avoided. Renal nerve stimulation (which acutely raises blood pressure) has been suggested as a test to establish that ablation has been achieved, and to predict a favourable blood pressure response during follow-up.22 Another indicative parameter may be change in renal blood flow or renal vascular resistance, which should be altered following renal denervation, given the vasoconstrictive effect of renal sympathetic nerves.23 However, none of these tools has yet been investigated in larger, randomized, controlled trials, which is urgently needed to validate their usefulness. Although several potential confounders have been taken into account in the design of the most recent trials, a common feature of all studies in renal denervation is the variability of the treatment effects (Figure 4). The response to the procedure has commonly been defined as a fall of 5 mmHg or more in ambulatory SBP. Thus, the identification of patients with a high likelihood of a relevant blood pressure lowering is of great importance for various reasons. Firstly, while renal denervation procedures have been shown to be safe, patients should not be exposed to an invasive procedure when alterations in renal sympathetic nerve activity are unlikely to be a relevant contributor to the increase in blood pressure. Secondly, patients with minor blood pressure effects should be excluded in order to increase the magnitude of the mean blood pressure reduction. For instance, patients with isolated systolic hypertension and high pulse wave velocity respond less to renal denervation.13,24 Whether this is related to structural alterations of the vascular tree, a different pathophysiology with negligible involvement of the sympathetic nervous system, or ineffectiveness of currently used devices to achieve successful nerve damage through calcified renal arteries remains to be investigated. It would be a wasted opportunity to exclude this large cohort of hypertensive individuals a priori from future clinical trial efforts. Individual blood pressure response in the RADIANCE-SOLO study. Modified from Azizi et al.20 Renal sympathetic nerves do have the potential to regrow and re-innervate the kidneys after being injured. Indeed, nerve regeneration after months to years has been observed in animal models with devices used in the clinical setting25 and in some organ transplant recipients. However, so far late rises of blood pressure have not been observed largely in patients undergoing renal nerve ablation. The long-term durability of the blood pressure lowering effect is quite difficult to investigate, as the majority of patients in the SPYRAL-OFF and RADIANCE-SOLO trials were back on antihypertensive medication using a standardized titration regimen after primary endpoint collection at 2 and 3 months, respectively. These time points were considered to be the longest period hypertensive patients can be safely left without drugs. Whether patients following renal denervation will require less intense antihypertensive drug regimens to obtain blood pressure control remains to be shown. A so far unanswered question is whether trials in renal denervation should not only demonstrate the antihypertensive efficacy of the procedure, but also its ability to improve clinical outcomes. Indeed, such a trial was planned some years ago, but then postponed after the publication of Symplicity HTN-3. Of note, blood pressure has a very strong and almost linear relation to hard cardiovascular outcomes, such as myocardial infarction, stroke, and death. As such, the U.S. Federal Drug Administration does accept blood pressure as a surrogate endpoint for approval. However, as surges in sympathetic drive are known to trigger myocardial infarction, arrhythmias, and sudden cardiac death, and given the fact that renal denervation led to consistent reductions in blood pressure over the entire 24-h period in the most recent proof-of-concept trials, it might well be possible that interventional antihypertensive strategies provide better protection than some of the currently used pharmacological means to lower blood pressure, especially given the high rates of non-adherence in clinical care. Thus, clinical outcomes trials are worth to be reconsidered after the publication of the above-mentioned proof-of-concept trials. With the publication of the most recent well-designed and rigorously executed proof-of-concept trials, convincing evidence is accumulating to indicate that catheter-based renal denervation is indeed an effective antihypertensive strategy with or without concomitant antihypertensive medication. Further studies with designs relevant to clinical reality also taking into account patient preference will be needed to ultimately establish the position of new device-based treatment approaches in the armamentarium of antihypertensive therapy. The next chapter begins—stay tuned. Conflict of interest: This work was supported by research grants of Medtronic, Recor Medical and St. Jude Medical (now Abbott). F.M. and M.B. are supported by Deutsche Gesellschaft für Kardiologie, Deutsche Hochdruckliga, and Deutsche Forschungsgemeinschaft (SFB TRR 219) and have received grant support and personal fees from Medtronic and Recor Medical. M.P.S. is supported by an NHMRC Research Fellowship and has received consulting fees, and/or travel and research support from Medtronic, Abbott, Novartis, Servier, Pfizer and Boehringer-Ingelheim. M.B. has received honoraria for scientific advisory activities and study committee membership from Astra-Zeneca, Boehringer-Ingelheim, Servier, and Vifor. M.E. receives salary support in the form of a Senior Principal Research Fellowship from the National Health and Medical Research Council of Australia. He has received personal fees from Medtronic. Outside this work TFL has received educational grants of AstraZeneca, Daichi-Sankyo and the Menarini Foundation.