Skip to content

Device and interventional therapy

TL;DR — Renal denervation is one of the clearest worked examples in modern medicine of a sham-controlled trial destroying a consensus, and then of the evidence moving again. SYMPLICITY HTN-3 randomised 535 patients 2:1 to denervation or sham and found an office systolic difference of −2.39 mm Hg (95% CI −6.89 to 2.12; p=0.26 for superiority) and a 24-hour ambulatory difference of −1.96 mm Hg (−4.97 to 1.06), against uncontrolled trials that had reported reductions of 25–30 mm Hg (Bhatt 2014, PMID 24678939). The field then rebuilt: standardised medication regimens, verified adherence, ambulatory endpoints, better catheters and patient selection. Second-generation sham-controlled trials are positive but with much smaller effects — SPYRAL HTN-OFF MED Pivotal −3.9 mm Hg 24-hour systolic (Bayesian 95% CrI −6.2 to −1.6) (Böhm 2020, PMID 32234534); pooled ultrasound denervation across three RADIANCE trials −5.9 mm Hg daytime ambulatory systolic (95% CI −8.1 to −3.8) (Kirtane 2023, PMID 36853627). Meta-analysis of 10 sham-controlled trials in 2,478 participants gives −4.4 mm Hg 24-hour ambulatory systolic (95% CI −6.1 to −2.7) and −6.6 mm Hg office systolic (−9.7 to −3.6) (Azizi 2026, PMID 41870448). No device has an outcome trial.

The sequence, in order

Year Study Design Result
2014 SYMPLICITY HTN-3 (PMID 24678939) 535 patients, 2:1 vs sham, resistant hypertension, office SBP ≥160 Office difference −2.39 mm Hg (95% CI −6.89 to 2.12), ns; ambulatory −1.96 (−4.97 to 1.06), ns
2014–15 SYMPLICITY HTN-3 secondary (PMIDs 24858423, 25835443, 25400162) Ambulatory, 12-month and predictor analyses Consistent null; predictors of response identified post hoc
2017 SPYRAL HTN-OFF MED pilot (PMID 28859944) 80 patients off drugs, sham-controlled proof of concept Positive; motivated the pivotal trial
2020 SPYRAL HTN-OFF MED Pivotal (PMID 32234534) 331 patients off drugs, Bayesian design with informative prior 24-h systolic difference −3.9 mm Hg (95% CrI −6.2 to −1.6); office −6.5 (−9.6 to −3.5); posterior probability of superiority >0.999
2018 RADIANCE-HTN SOLO (PMID 29803590) 146 patients, ultrasound, off drugs Daytime ambulatory systolic −8.5 vs −2.2 mm Hg; adjusted difference −6.3 (95% CI −9.4 to −3.1), p=0.0001
2021 RADIANCE-HTN TRIO (PMID 34010611) 136 patients on standardised triple single-pill therapy Median daytime ambulatory systolic −8.0 vs −3.0 mm Hg; between-group −4.5 (95% CI −8.5 to −0.3), adjusted p=0.022
2022 SYMPLICITY HTN-3 final 36-month (PMID 36130612) Unblinded after 6 months, crossover permitted Office systolic change −26.4 vs −5.7 mm Hg (adjusted difference −22.1, 95% CI −27.2 to −17.0); 24-h −15.6 vs −0.3 (−16.5, −20.5 to −12.5); time in therapeutic range 18% vs 9%
2023–24 RADIANCE pooled analyses (PMIDs 36853627, 37883784) 506 patients across three trials 2-month daytime ambulatory difference −5.9 mm Hg (95% CI −8.1 to −3.8); at 6 months after blinded medication escalation, adjusted difference −3.0 mm Hg (−5.7 to −0.2) with fewer drugs in the denervation arm (p=0.004)
2025–26 SPYRAL pooled long-term (PMIDs 40175677, 42137915) 2-year nighttime and 36-month pooled Sustained reductions

The 36-month SYMPLICITY HTN-3 report requires care. After the 6-month primary endpoint, patients were unblinded and sham patients could cross over, so the −22.1 mm Hg adjusted difference at 36 months rests on imputing pre-crossover values for crossovers and comparing against a shrinking, self-selected non-crossover group (Bhatt 2022, PMID 36130612). It is legitimate long-term safety evidence — no late complications emerged, composite safety event rates were 14–15% across groups to 48 months — and suggestive efficacy evidence, but it is not the randomised comparison the 6-month result was.

Why the first trial was negative and the later ones were not

Several explanations have been offered, and they are not mutually exclusive:

  1. Procedure quality. SYMPLICITY HTN-3 used a single-electrode catheter, many operators performed few procedures, and circumferential four-quadrant ablation including distal branches was not standard. Later trials used multi-electrode radiofrequency or ultrasound systems with defined ablation protocols (Persu 2014, PMID 30310496; Verdecchia 2023, PMID 37544846).
  2. Medication changes and non-adherence. Drug regimens were not standardised or verified in SYMPLICITY HTN-3; later trials standardised and, in the RADIANCE programme, verified adherence by urine assay (Azizi 2021, PMID 34010611).
  3. Patient selection. Effect is larger in people with higher baseline pressure, higher heart rate and orthostatic hypertension (Kirtane 2023, PMID 36853627), and with particular 24-hour pulsatile haemodynamic profiles (Weber 2022, PMID 35582957).
  4. Endpoint choice. Office systolic pressure in a resistant-hypertension population is noisy and regresses to the mean; ambulatory endpoints are less so.
  5. Regression to the mean and placebo response. The sham arm in SYMPLICITY HTN-3 fell 11.74 mm Hg, which is the size of the entire effect claimed by earlier unblinded studies — the central lesson of the episode (Bhatt 2014, PMID 24678939; Epstein 2015, PMID 25649995).

Whether sham control is indispensable in this field has been examined meta-analytically, with sham-controlled and non-sham-controlled trials giving systematically different answers (Fadl Elmula 2017, PMID 28443356).

What the current effect size means clinically

A 4–6 mm Hg ambulatory systolic reduction is roughly equivalent to one standard-dose antihypertensive drug, and smaller than spironolactone's effect as a fourth agent (Williams 2015, PMID 26414968; Azizi 2026, PMID 41870448). Its distinguishing features are that it is always-on — independent of daily adherence — and that it reduces subsequent drug requirement: in the RADIANCE pooled 6-month analysis, fewer denervation patients were prescribed additional antihypertensives (p=0.004) and fewer drugs were added (p=0.001) (Azizi 2024, PMID 37883784). Durability data extend to 3 years for ultrasound denervation (Rader 2022, PMID 35913759; Bloch 2024, PMID 39333663) and 36 months pooled for radiofrequency (Kandzari 2026, PMID 42137915), with crossover patients showing the expected delayed response (Mahfoud 2021, PMID 34236037). Nocturnal pressure, prognostically the most important index, falls and stays down (Kario 2018, PMID 30354518; Kario 2025, PMID 40175677). Mechanistic markers move as expected — heart rate falls (Böhm 2019, PMID 30608521), renin and aldosterone respond (Fisher 2022, PMID 34433763), and tissue sodium content changes (Ott 2018, PMID 28845508).

Denervation has also been studied in specific subgroups: obstructive sleep apnoea within SYMPLICITY HTN-3 (Kario 2016, PMID 27118620), African American and non-African American participants (Flack 2015, PMID 26362830), and hypertensive urgencies during follow-up (Weber 2022, PMID 35852582). Prespecified 6-month analyses of RADIANCE-HTN TRIO after medication escalation are consistent (Azizi 2022, PMID 36350593).

Other device approaches

Approach Principle Best evidence Status
Baroreflex activation therapy (carotid sinus stimulation) Chronic afferent baroreflex activation reduces sympathetic outflow Rheos Pivotal: 265 implanted, randomised 2:1 to immediate vs delayed activation; met endpoints for sustained efficacy and device/therapy safety but not acute responder rate or procedural safety; 42% vs 24% achieved SBP ≤140 mm Hg at 6 months (p=0.005) (Bisognano 2011, PMID 21816315) Not in routine hypertension practice; second-generation devices and endovascular baroreflex amplification under study (Victor 2015, PMID 26149485; van Kleef 2018, PMID 29744599)
Central iliac arteriovenous anastomosis (ROX coupler) Reduces effective arterial stiffness and systemic vascular resistance ROX CONTROL HTN, open-label randomised: office systolic −25.1±23.3 mm Hg and 24-h ambulatory −12.6±17.4 mm Hg at 12 months; 33% developed ipsilateral venous stenosis requiring stenting (Lobo 2017, PMID 29061728) Not adopted; the venous-stenosis rate and lack of sham control are the obstacles (Kapil 2018, PMID 29084002; Brier 2015, PMID 26403331)
Carotid body ablation Removes chemoreflex-driven sympathetic tone Early-phase human work only (Iturriaga 2018, PMID 29789952; Iturriaga 2018, PMID 29114876) Experimental

General reviews cover the whole device landscape and its physiological rationale (Lauder 2020, PMID 32286512; Mahfoud 2021, PMID 33793330; Lohmeier 2019, PMID 30920919; Ng 2016, PMID 27370788; Gierthmuehlen 2020, PMID 32030509; Tolu-Akinnawo 2024, PMID 39108770; Zhang 2026, PMID 41344842; Lauder 2019, PMID 31278490; Oliva 2014, PMID 25416664).

The methodological lesson

The renal denervation story is the strongest available case for sham control in procedural cardiovascular research. Pre-2014 uncontrolled series reported 25–30 mm Hg reductions; the sham arm of SYMPLICITY HTN-3 alone fell 11.74 mm Hg (Bhatt 2014, PMID 24678939). It is also a case against premature abandonment: the technique was widely written off (Epstein 2015, PMID 25649995) before better-designed trials found a real, if modest, effect (Böhm 2020, PMID 32234534; Kirtane 2023, PMID 36853627). Both errors — over-claiming from unblinded data and over-generalising from a single negative trial — are instructive.

Open questions

  • Does renal denervation reduce cardiovascular events? Every trial to date has a blood-pressure endpoint; no outcome trial has reported (Azizi 2026, PMID 41870448).
  • Can responders be identified prospectively? Higher baseline pressure, higher heart rate and orthostatic hypertension predict response post hoc (Kirtane 2023, PMID 36853627), but no prospective selection algorithm has been validated.
  • Is the 36-month SYMPLICITY HTN-3 difference an efficacy signal or an artefact of unblinding and crossover imputation? (Bhatt 2022, PMID 36130612)
  • Should denervation be positioned before or after a mineralocorticoid receptor antagonist, given that spironolactone's effect is roughly twice as large? (Williams 2015, PMID 26414968; Azizi 2026, PMID 41870448)
  • Why has baroreflex activation therapy, which produced a larger control-rate difference than denervation in Rheos, not progressed, and is that a scientific or a commercial fact? (Bisognano 2011, PMID 21816315)

References

  1. Bhatt DL, et al. A controlled trial of renal denervation for resistant hypertension. N Engl J Med. 2014;370:1393-401. PMID 24678939
  2. Bhatt DL, et al. Long-term outcomes after catheter-based renal artery denervation: final follow-up of SYMPLICITY HTN-3. Lancet. 2022;400:1405-1416. PMID 36130612
  3. Bakris GL, et al. Impact of renal denervation on 24-hour ambulatory blood pressure: SYMPLICITY HTN-3. J Am Coll Cardiol. 2014;64:1071-8. PMID 24858423
  4. Bakris GL, et al. 12-month blood pressure results of catheter-based renal artery denervation: SYMPLICITY HTN-3. J Am Coll Cardiol. 2015;65:1314-1321. PMID 25835443
  5. Kandzari DE, et al. Predictors of blood pressure response in the SYMPLICITY HTN-3 trial. Eur Heart J. 2015;36:219-27. PMID 25400162
  6. Townsend RR, et al. Catheter-based renal denervation in the absence of antihypertensive medications (SPYRAL HTN-OFF MED). Lancet. 2017;390:2160-2170. PMID 28859944
  7. Böhm M, et al. Efficacy of catheter-based renal denervation in the absence of antihypertensive medications (SPYRAL HTN-OFF MED Pivotal). Lancet. 2020;395:1444-1451. PMID 32234534
  8. Azizi M, et al. Endovascular ultrasound renal denervation to treat hypertension (RADIANCE-HTN SOLO). Lancet. 2018;391:2335-2345. PMID 29803590
  9. Azizi M, et al. Ultrasound renal denervation for hypertension resistant to a triple medication pill (RADIANCE-HTN TRIO). Lancet. 2021;397:2476-2486. PMID 34010611
  10. Kirtane AJ, et al. Patient-Level Pooled Analysis of Ultrasound Renal Denervation in the Sham-Controlled RADIANCE Trials. JAMA Cardiol. 2023;8:464-473. PMID 36853627
  11. Azizi M, et al. Patient-Level Pooled Analysis of Endovascular Ultrasound Renal Denervation or a Sham Procedure 6 Months After Medication Escalation. Circulation. 2024;149:747-759. PMID 37883784
  12. Azizi M, et al. Effects of Renal Denervation vs Sham in Resistant Hypertension After Medication Escalation: RADIANCE-HTN TRIO. JAMA Cardiol. 2022;7:1244-1252. PMID 36350593
  13. Rader F, et al. Durability of blood pressure reduction after ultrasound renal denervation: three-year follow-up. EuroIntervention. 2022;18:e677-e685. PMID 35913759
  14. Bloch MJ, et al. 36-month durability of ultrasound renal denervation in RADIANCE-HTN TRIO. Hypertens Res. 2024;47:3467-3472. PMID 39333663
  15. Mahfoud F, et al. Changes in blood pressure after crossover to ultrasound renal denervation in patients initially treated with sham. EuroIntervention. 2021;17:e1024-e1032. PMID 34236037
  16. Kandzari DE, et al. Long-term clinical benefit after radiofrequency renal denervation: pooled 36-month results. EuroIntervention. 2026;22:585-593. PMID 42137915
  17. Kario K, et al. Twenty-Four-Hour Ambulatory Blood Pressure Reduction Patterns After Renal Denervation (SPYRAL HTN-OFF MED). Circulation. 2018;138:1602-1604. PMID 30354518
  18. Kario K, et al. Two-year nighttime blood pressure changes after radiofrequency renal denervation. Hypertens Res. 2025;48:1951-1962. PMID 40175677
  19. Böhm M, et al. Ambulatory heart rate reduction after catheter-based renal denervation. Eur Heart J. 2019;40:743-751. PMID 30608521
  20. Fisher NDL, et al. Plasma renin and aldosterone concentrations related to endovascular ultrasound renal denervation in RADIANCE-HTN SOLO. J Hypertens. 2022;40:221-228. PMID 34433763
  21. Ott C, et al. Impact of renal denervation on tissue Na+ content in treatment-resistant hypertension. Clin Res Cardiol. 2018;107:42-48. PMID 28845508
  22. Kario K, et al. Impact of Renal Denervation on Patients With Obstructive Sleep Apnea and Resistant Hypertension: SYMPLICITY HTN-3. Circ J. 2016;80:1404-12. PMID 27118620
  23. Flack JM, et al. An analysis of blood pressure and safety outcomes to renal denervation in African Americans and Non-African Americans in SYMPLICITY HTN-3. J Am Soc Hypertens. 2015;9:769-779. PMID 26362830
  24. Weber MA, et al. Hypertension urgencies in the SPYRAL HTN-OFF MED Pivotal trial. Clin Res Cardiol. 2022;111:1269-1275. PMID 35852582
  25. Weber T, et al. Twenty-Four-Hour Pulsatile Hemodynamics Predict Brachial Blood Pressure Response to Renal Denervation. Hypertension. 2022;79:1506-1514. PMID 35582957
  26. Persu A, et al. Renal Denervation After Symplicity HTN-3 - Back to Basics. Eur Cardiol. 2014;9:110-114. PMID 30310496
  27. Epstein M, et al. Is the failure of SYMPLICITY HTN-3 to meet its efficacy endpoint the "end of the road" for renal denervation? J Am Soc Hypertens. 2015;9:140-9. PMID 25649995
  28. Verdecchia P, et al. Catheter-based renal artery denervation: facts and expectations. Eur J Intern Med. 2023;117:66-77. PMID 37544846
  29. Fadl Elmula FEM, et al. Sham or no sham control: that is the question in trials of renal denervation. Blood Press. 2017;26:195-203. PMID 28443356
  30. Azizi M, et al. Diagnosis and Management of Resistant Hypertension: A Review. JAMA. 2026;335:1428-1439. PMID 41870448
  31. Williams B, et al. Spironolactone versus placebo, bisoprolol, and doxazosin (PATHWAY-2). Lancet. 2015;386:2059-2068. PMID 26414968
  32. Bisognano JD, et al. Baroreflex activation therapy lowers blood pressure in patients with resistant hypertension: the Rheos Pivotal Trial. J Am Coll Cardiol. 2011;58:765-73. PMID 21816315
  33. Victor RG. Carotid baroreflex activation therapy for resistant hypertension. Nat Rev Cardiol. 2015;12:451-63. PMID 26149485
  34. van Kleef MEAM, et al. Endovascular Baroreflex Amplification for Resistant Hypertension. Curr Hypertens Rep. 2018;20:46. PMID 29744599
  35. Lobo MD, et al. Central Iliac Arteriovenous Anastomosis for Uncontrolled Hypertension: One-Year Results From ROX CONTROL HTN. Hypertension. 2017;70:1099-1105. PMID 29061728
  36. Kapil V, et al. Central arteriovenous anastomosis to treat resistant hypertension. Curr Opin Nephrol Hypertens. 2018;27:8-15. PMID 29084002
  37. Brier TJ, et al. Central arteriovenous anastomosis for hypertension: it is not all about sympathomodulation. Future Cardiol. 2015;11:503-6. PMID 26403331
  38. Iturriaga R. Carotid Body Ablation: a New Target to Address Central Autonomic Dysfunction. Curr Hypertens Rep. 2018;20:53. PMID 29789952
  39. Iturriaga R. Translating carotid body function into clinical medicine. J Physiol. 2018;596:3067-3077. PMID 29114876
  40. Lauder L, et al. Device-based therapies for arterial hypertension. Nat Rev Cardiol. 2020;17:614-628. PMID 32286512
  41. Mahfoud F, et al. Device Therapy of Hypertension. Circ Res. 2021;128:1080-1099. PMID 33793330
  42. Lohmeier TE, et al. Device-Based Neuromodulation for Resistant Hypertension Therapy. Circ Res. 2019;124:1071-1093. PMID 30920919
  43. Ng FL, et al. Device-based Therapy for Hypertension. Curr Hypertens Rep. 2016;18:61. PMID 27370788
  44. Gierthmuehlen M, et al. Implant-Mediated Therapy of Arterial Hypertension. Curr Hypertens Rep. 2020;22:16. PMID 32030509
  45. Tolu-Akinnawo O, et al. Hypertension and Device-Based Therapies for Resistant Hypertension. Cureus. 2024;16:e66304. PMID 39108770
  46. Zhang M, et al. Innovative Device-Based Treatments for Hypertension. Am J Hypertens. 2026;39:844-859. PMID 41344842
  47. Lauder L, et al. Renal Denervation: Is It Ready for Prime Time? Curr Cardiol Rep. 2019;21:80. PMID 31278490
  48. Oliva RV, et al. Sympathetic activation in resistant hypertension: theory and therapy. Semin Nephrol. 2014;34:550-9. PMID 25416664