Submit your picks! | @NephMadness | @nephmadness.bsky.social | NephMadness 2025

Swapnil Hiremath

Swapnil Hiremath is a Professor at the University of Ottawa and an Associate Scientist at the Ottawa Hospital Research Institute. His primary clinical and research interests are in improving care of individuals living with high blood pressure, CKD, and those receiving hemodialysis. He is co-founder of the online nephrology journal club, #NephJC.

Competitors for the Resistant Hypertension Region

Team 1: Renal Denervation

vs

Team 2: Novel Drugs for Hypertension

Image generated by Evan Zeitler using DALLE-E 3, accessed via ChatGPT at http://chat.openai.com, February 2025. After using the tool to generate the image, Zeitler and the NephMadness Executive Team reviewed and take full responsibility for the final graphic image.

The journey of renal denervation (RDN) has followed the classic hype cycle: the peak of inflated expectations as a cure for resistant hypertension with the 20-30 mm Hg blood pressure (BP) decrease reported in Symplicity-HTN-1 and 2, the trough of disillusionment with the null results of Symplicity HTN-3, and now in the plateau of productivity of a more modest BP effect after the Spyral and RADIANCE trials. The scouting report from Stephanie Torres Rodrigues does a thorough job of detailing the evidence and the science, so let’s dig up a few nuggets which can help readers understand where, and if, RDN fits in with modern day management of hypertension. 

The Lost Decade

It has been over a decade since the Symplicity HTN-3 results dropped like a lead balloon and seemingly sank the prospects of RDN. But in this intervening decade, Medtronic and the involved hypertension trialists did not give up hope. The issues that sunk HTN-3, related to operator expertise and catheter design, information bias, high variance due to varying adherence, and patient selection were identified and ironed out by White et al. In addition to this, many scientific studies in animal models were conducted (belatedly, one might say) to understand renal sympathetic innervation, ablation techniques, and more. To name a few, 

• Nerve ablation is followed by axonal loss, with reduction in norepinephrine levels, robust till 180 days (Sharp et al).
• Ablation of main renal artery and first order branches lead to a dose dependent decrease in norepinephrine tissue content and axonal density unlike more lesions in main artery alone (Mahfoud et al).
• Temperature, power and impedance considerations for radiofrequency ablation (e.g. nerve damage is optimal with multiple circumferential lesions [Coates et al]). 

In many respects this wasn’t really a lost decade, with the apocryphal slope of enlightenment which led to better trial design, the superior Symplicity Spyral catheter design, and the power delivery control algorithm, among other advancements. 

Trial Data

The trial results are very nicely discussed in the scouting report. With respect to systematic reviews (e.g. Vukadinovic et al) it’s mostly a GIGO issue related to significant clinical and statistical heterogeneity, so I would not trust the pooled estimates very much. The one important outcome for which there was minimal heterogeneity was change in nocturnal BP, which was overall lower (~ 4.5/2.5 mm Hg lower). This is notable as otherwise no intervention (particular bedtime dosing) has been shown to be helpful in fixing nocturnal hypertension.  The full cohort of the SPYRAL HTN-ON MED trial (Kandzari et al) makes for instructive reading as the largest and longest sham-control intervention trial of RDN. The primary outcome of 24 hour ambulatory BP change was a measly -1.9 mm Hg (95% CI: -4.4 to 0.5 mm Hg; P = 0.12). This occurred mostly because more participants in the sham group got more BP lowering drugs, and more people in the intervention/RDN group had their BP drugs lowered or stopped (likely since BP was better and even lower than target). While this could be seen from different perspectives (the authors’ conducted a post hoc win ratio analysis incorporating drug changes in the outcome to overcome this), it remains instructive to what would happen in real life clinical practice. If the BP is not controlled, a clinician provider would simply add a medication, leading to better BP control. Voila, no need for RDN! But what about the pill burden you say? 

Patient Perspectives

Pill burden (or more candidly, pill hatred) is an especially important aspect for asymptomatic conditions like hypertension, and can lead to non-adherence apart from the decreased quality of life that the term itself implies. Two studies (Schmeider et al and Kandzari et al) examine patient preferences around RDN and provide additional insight into how pill burden and RDN might intersect. Interestingly, in the first study the authors did ask clinicians, who almost unanimously preferred RDN use in those on multiple medications and with resistant hypertension. However, no relationship with increasing pill burden and a preference for RDN was seen in both studies from the patient perspective. If anything, there was a slight preference at the low end – where RDN (which would probably result in a reduction of one BP lowering drug on average) might mean going to zero medications. After all (in hindsight), going from 5 to 4 pills is a quantitative reduction, but going from 1 or 2 to zero pills is a qualitative reduction in pill burden. Another win for looking at patient reported outcomes (the NephMadness 2018 Champion). 

The Societal Perspective

The US Food and Drug Administration (and other agencies such as Health Canada) have approved RDN without any rider for its indication being resistant hypertension. So, there is nothing to stop it being performed in the vast hypertensive population. However, it costs a bunch of money (~ US$20,000) and it is important to consider the alternatives. Generic spironolactone is pennies. Can we afford to spend so much money on expensive therapies with a small effect when alternatives are so cheap? There are cost-effectiveness analyses (e.g. Kandzari et al ),but like any model, a look into the inputs tell you this is practically unusable. The authors model a ~10/5 mm Hg BP difference, which is sustained over a lifetime. This makes no practical sense; if BP is not controlled, one would add BP meds. It is not like a 2 month trial, where participants are deliberately left untreated for a short period to demonstrate the efficacy of an intervention. Taking such flawed economic analyses to the bank would be a big mistake. 

Conclusion

In summary, RDN does have a modest effect on BP lowering and some patients may prefer it, especially if it means not taking any BP medication. However, its cost is much higher than the generic BP medications that we mostly use. A deeply unsatisfying aspect of RDN is that it is not linked to any pathophysiology (e.g. renovascular hypertension or primary aldosteronism), coupled with a lack of pre-procedural factors to identify who might benefit. Contrast that with the new drugs that are approved and in the pipeline, such as flozins, GLP1RAs, endothelin antagonism, aldosterone synthase inhibition, and RNA interference injectables: elegant mechanisms, novel delivery systems, and proven end organ benefit data. This choice is a slam dunk for drugs over devices.

– Guest Post written by Swapnil Hiremath

As with all content on the AJKD Blog, the opinions expressed are those of the author of each post and are not necessarily shared or endorsed by the AJKD Blog, AJKD, the National Kidney Foundation, Elsevier, or any other entity unless explicitly stated.

Click to read the Resistant Hypertension Region

Submit your picks! | @NephMadness | @nephmadness.bsky.social | NephMadness 2025