Does the blood pressure response to renal denervation depend on the degree of sympathetic overactivity?

Hypertension remains one of the leading modifiable causes of cardiovascular (CV) morbidity and mortality and is estimated to affect >1.5 billion adults worldwide 1. Uncontrolled hypertension leads to substantial rises in healthcare expenditure due to the expanding burden of CV and renal complications 1. Given the gap between available therapies and real-world blood pressure (BP) control, there is a need for novel strategies to mitigate end-organ damage and prevent progression to severe hypertensive phenotypes 2. In particular, resistant hypertension (RHT), defined as uncontrolled BP on ≥3 or controlled BP with ≥4 antihypertensive medications at maximally tolerated doses, is among one of the most clinically relevant phenotypes 2. In the United States, RHT affects 8.5–20% of adults with hypertension 3–5 and is more prevalent in older adults, non-Hispanic Black adults, and those with diabetes and chronic kidney disease 6. Achieving BP goals in patients with RHT can be challenging as intensifying pharmacotherapy to five or six agents typically yields diminishing returns and is often limited by adverse side effects 6. In addition, higher pill burden leads to lower adherence, tolerability, and long-term effectiveness 6. Thus, novel strategies to control BP, including nonpharmacologic approaches such as renal denervation (RDN), are needed to reduce end-organ damage, particularly in those with RHT. RDN has emerged as a novel therapeutic alternative or adjunct to pharmacotherapy for the treatment of uncontrolled hypertension 7. RDN is a catheter-based procedure that involves application of either radiofrequency energy, ultrasound energy, or neurotoxic substance (i.e., alcohol) to ablate afferent sensory and efferent sympathetic nerves that innervate the kidneys 7. The rationale for RDN stems from earlier evidence that renal sympathetic nerve activity (RSNA) is elevated in essential hypertension and that both RSNA and BP can be lowered via afferent and efferent denervation of renal nerves in preclinical models 8. In humans, the first proof-of-concept, open-label, cohort study without randomization (SYMPLICITY HTN-1) demonstrated a substantial and sustained reduction in office systolic BP (SBP) (−22 mmHg at 6 months and −27 mmHg at 12 months) following RDN with radiofrequency ablation in patients with RHT 9. The SYMPLICITY HTN-2 trial was a randomized, unblinded, controlled study which documented substantial reductions in office (−32 vs. 1 mmHg) and ambulatory SBP (−11 vs. −3 mmHg) at 6 months following RDN without any changes in BP in the control group (i.e., antihypertensive medications only but no sham) 10. By contrast, the SYMPLICITY HTN-3 trial was a single-blind, randomized, sham-controlled study which reported no group differences in the reduction of BP between RDN or renal angiography (sham control) 11. Both groups displayed similar reductions in office (−14 vs. −12 mmHg) and ambulatory SBP (−7 vs. −5 mmHg) at 6 months 11. However, the study results may have been influenced by selection bias, medication nonadherence, changes in antihypertensive medications, operator experience, incomplete ablation of the renal nerves, and failure to denervate accessory renal arteries 11. Since then, multiple clinical trials have emerged to re-evaluate the efficacy of RDN using improved study methodology and with more homogenous study populations 7. In this issue of the Journal of Hypertension, Nolde et al.12 performed a retrospective analysis of the Global SYMPLICITY Registry study to determine whether the BP-lowering effect of RDN is different between patients concurrently treated with centrally acting sympatholytic agents (CASAs) vs. those that were not treated with CASAs. The rationale behind this important question stemmed from the notion that concurrent use of CASAs may render RDN less effective given that both CASAs and RDN produce a BP-lowering effect by reducing sympathetic nervous system (SNS) activity 12. The primary observation was that while RDN produced a sustained reduction in ambulatory SBP (−7–9 mmHg) in the entire cohort, the BP-lowering effect of RDN was more pronounced in patients who were not taking CASAs compared to those on CASAs (mean absolute difference of 3 mmHg for ambulatory SBP) despite similar levels of ambulatory SBP at baseline 12. Nolde et al.12 concluded that concurrent use of CASAs may have important implications on the effectiveness of RDN on BP in patients with uncontrolled hypertension. A major strength was access to a large cohort of patients who underwent RDN (i.e., n = 2712 for the entire cohort, 38.2% of whom were on CASAs), while an inherent strength included ambulatory BP monitoring 12. Ambulatory BP monitoring also allowed for exploration into the impact of RDN and CASA use on BP profiles given that nighttime BP is superior to daytime BP in predicting future CV morbidity and mortality in patients with RHT 13. The Global SYMPLICITY Registry study is an ongoing prospective, multicenter, single-arm, noninterventional, and open-label registry, with an overall goal of documenting the long-term safety and effectiveness of RDN in a real-world clinical setting 14. Since factors such as physical capacity 15 and sleep quality 16 were not recorded in this cohort, it is unclear if behavioral changes following RDN may have influenced ambulatory BP 17,18. The open-label and observational study design may also introduce potential for bias, including selection bias, performance bias, detection bias, and attrition bias, which likely limits the reliability and generalizability of the observations 19. In addition, although there was no difference between CASA and non-CASA groups in terms of the number of BP medications during all time points, it is unclear whether dosages and/or classes of antihypertensive medications may have changed over time. Importantly, different classes of antihypertensive medications have differential effects on SNS activity. For instance, inhibitors of the renin–angiotensin–aldosterone system (RAAS) including angiotensin converting enzyme inhibitors, angiotensin receptor blockers and mineralocorticoid receptor blockers decrease central SNS outflow, whereas vasodilators and diuretics such as amlodipine and hydrochlorothiazide lead to a reflex increase in SNS activity 20. In addition, approximately 20% of patients with hypertension reported regular use of nonprescription medications (e.g., NSAIDS or nasal decongestants) that may raise BP or compromise antihypertensive drug efficacy 21. Despite some limitations in the original SYMPLICITY trials, the current analysis may inform the potential mechanisms by which RDN lowers BP. Nolde et al.12 found that patients on CASAs had less BP reduction after RDN compared to those not on CASAs. Since patients on CASAs already have lower central SNS output, these findings suggest that one mechanism by which RDN reduces BP is via reduction of SNS activation presumably through afferent denervation. In addition, the observation that CASA patients had a significant reduction in BP (albeit not as great in magnitude as the non-CASA users) may suggest that this group had BP lowering after RDN primarily via efferent denervation. Direct measurements of SNS activity using a microneurography (i.e., central SNS outflow) and isotope dilution (i.e., organ-specific norepinephrine spillover to the heart or kidneys) as well as measurements of RAAS components could further inform the specific mechanisms by which RDN reduces BP in humans. In summary, Nolde et al.12 should be commended for providing clinically relevant insight into the potential impact of CASAs on the BP response to RDN. Given the variable responses to RDN and the heterogeneity of patients with uncontrolled hypertension, a major challenge is identifying the patients to refer for RDN that are most likely to have a robust BP-lowering effect. The current findings raise the question as to whether BP responses to CASAs could potentially serve as a useful predictor to the effectiveness of RDN in the treatment of uncontrolled hypertension. For instance, a greater lowering of BP in response to clonidine may identify patients who have higher baseline SNS activity and sympathetically mediated (i.e., neurogenic) hypertension and might possibly have greater reductions in BP after RDN. Other factors such as volume status and RAAS component levels may also aid in patient selection given that RDN also ablates renal efferent nerves that affect these pathways. Identifying factors that may predict the efficacy of BP-lowering responses to different therapeutic strategies including RDN would be clinically valuable for risk stratification and determining the ideal treatment approach for BP control in individual patients. ACKNOWLEDGEMENTS None. Funding: This work was funded, in part, by the U.S. Department of Veterans Affairs (IK2RX003670, to K.B.) and the National Institutes of Health (R01 HL135183 and R33 AT010457, to J.P). Conflicts of interest There are no conflicts of interest.