CrossTalk opposing view: Which technique for controlling resistant hypertension? Renal nerve ablation

Hypertension affects an estimated one billion adults globally, representing a major cardiovascular epidemic. Treatment-resistant hypertension, defined as uncontrolled blood pressure (BP) despite ≥3 antihypertensive medications (Calhoun et al. 2008), is becoming more frequent. Recent reports from the US indicate that ∼13% of adults that are being treated for elevated BP have resistant hypertension (Persell, 2011) with 1 in 50 patients developing resistant hypertension within a median of 1.5 years from the initiation of pharmacotherapy (Daugherty et al. 2012). There is a clear need for alternative therapeutic approaches. Both animal and human studies have clearly confirmed the critical role of sympathetic overactivity in hypertension (DiBona & Kopp, 1997; Schlaich et al. 2004). A variety of other clinically relevant conditions frequently associated with hypertension such as renal disease (Schlaich et al. 2009) have also been found to be characterized by sympathetic overactivity. Post-ganglionic sympathetic nerve fibres form a dense neuronal network within the adventitia of the renal artery (DiBona, 2005) and have been identified as key contributors in the pathophysiology of hypertension (Esler et al. 1989, 1990; Schlaich et al. 2004). Efferent motor fibres innervate all renal structures, including the renal vasculature, tubules and the juxtaglomerular apparatus (Barajas, 1978). Central sympathetic outflow to the kidneys modulates short and long-term BP control by stimulating the release of renin, increasing sodium and water reabsorption, and inducing renal vasoconstriction (DiBona & Kopp, 1997). Communication with integral structures in the central nervous system occurs via afferent renal nerves. Recent studies suggest that renal ischaemia, hypoxia and oxidative stress result in an increase in renal afferent activity (Ye et al. 1998, 2002). Renal sensory afferent nerve activity directly influences sympathetic outflow to the kidneys and other extensively innervated organs involved in cardiovascular control such as the heart and peripheral blood vessels, by modulating posterior hypothalamic activity (Campese & Kogosov, 1995; Campese, 2000). These components of the neural regulation of renal function are considerably stimulated in hypertension and clearly contribute to the rise in BP (Campese, 2000). To identify the contribution of renal sympathetic efferent and sensory afferent nerves to renal and systemic organ function, surgical ligation and re-anastomosis of the renal arteries, or surgical stripping of the renal nerve adventitia with local application of phenol were frequently applied to achieve renal denervation (RDN) in many experimental models including rodents (Katholi, 1985), swine (O'Hagan et al. 1990), canine (Kassab et al. 1995), ovine (Whitworth et al. 1981) and numerous hypertension models (DiBona, 2005). Bilateral RDN prevented or attenuated the development of hypertension in genetic, salt-sensitive, obesity-related, renovascular and other hypertension models (DiBona & Kopp, 1997). The preclinical science surrounding RDN has supported human surgical efforts to modify renal innervation, particularly for the treatment of hypertension. Smithwick and colleagues (Smithwick & Thompson, 1953) investigated the effects of splanchnicectomy in 1266 hypertensive patients and compared their outcomes to those of 467 hypertensive control subjects. Interestingly, while a significant and sustained BP-lowering effect was only seen in ∼45% of surgically treated patients, the 5 year mortality rates were substantially lower than in the control group (19% vs. 54%, respectively), suggesting that inhibition of sympathetic outflow to the splanchnic vascular bed per se contributes to improved cardiovascular outcomes. Documented activation of the renal sympathetic outflow in hypertensive patients, the blood-pressure lowering effect of surgical RDN in experimental models of hypertension, and the anatomical features of renal nerves in their passage to the kidneys provided the framework for the development of interventional RDN approaches using radiofrequency energy delivered by a catheter in the renal artery lumen. The long-term safety and efficacy of RDN to control BP has been investigated in the Symplicity Clinical Trial Program. In 2009, the first proof-of-principle trial (Symplicity HTN-1) (Krum et al. 2009) was undertaken in 45 patients with resistant hypertension. Office BP was significantly reduced at 6 and 12 months after RDN (–22/–11 mmHg and –27/–17 mmHg, respectively). As an important proof of the physiological concept, renal noradrenaline spillover was reduced by 47% as an assessment of reduced renal sympathetic nerve activity, documenting substantial but incomplete denervation. Central sympathetic outflow, as assessed by microneurography, was also reduced in treated patients (Hering et al. 2013). In the extended Symplicity HTN-1 cohort (n = 153) the treatment effect on BP was sustained at 24 and 36 months (Krum et al. 2014) with a favourable safety profile (Figure 1). Figure reproduced from (Krum, 2014) with permission. Symplicity HTN-2 was a multicentre, prospective, randomized controlled trial involving 106 resistant hypertensive patients (Esler et al. 2010). Of the 49 patients who immediately underwent RDN, mean office BP at 6 months significantly decreased by –32/–12 mmHg with no BP change in the control group (n = 51) assigned to standard pharmacological therapy. Renal artery imaging at follow-up (n = 43) confirmed the safety of the procedure with no reported incidence of renal artery stenosis or aneurismal deformation. More recently, 12 months follow-up data from 47 patients in the Symplicity HTN-2 trial was published (Esler et al. 2012). The magnitude of systolic BP reduction at 12 months was consistent with that observed in the first Symplicity HTN-1 trial (−28 vs. –27 mmHg). In both cohorts, RDN did not adversely affect kidney function. This observation is supported by a recent study of 88 patients with resistant hypertension who had a preserved estimated glomerular filtration rate 6 months post RDN (Mahfoud et al. 2012). Similarly, a recently published subject-level meta-analysis including 109 patients with resistant hypertension from 10 European centres demonstrated favourable safety and a 72% response rate, defined as a systolic BP reduction ≥10 mmHg, in patients with baseline office systolic BP ≥160 mmHg (Persu et al. 2014). Data from studies using alternative RDN devices have confirmed significant BP lowering efficacy without major safety concerns (Worthley et al. 2013). Several criticisms raised in relation to the Symplicity HTN-2 trial protocol were addressed in the very recently published Symplicity HTN-3 trial (Bhatt et al. 2014). This US study randomized a total of 535 patients with resistant hypertension on maximally tolerated doses of at least three antihypertensive agents, including a diuretic, in a 2:1 ratio to RDN or sham procedure with mandatory 24 h ambulatory BP measurements. While the primary safety endpoint was met and a reduction in systolic office BP (−14.1 ± 23.9 mmHg; P < 0.001) from baseline was seen with RDN, the sham control group had a similar BP reduction (−11.7 ± 25.9 mmHg; P < 0.001) resulting in a non-significant difference of −2.39 mmHg (95% confidence interval, –6.89 to 2.12). Although rigorously designed, in-depth analysis of the Symplicity HTN-3 data suggests that the trial execution was less than ideal: concerns raised by Symplicity HTN-3 investigators (Kandzari, 2014) relate to the inexperience of the 111 participating interventionalists, many performing a renal denervation procedure for the very first time within the trial, possibly resulting in insufficient renal denervation. Indeed, a circumferential ablation pattern considered essential to effectively denervate the kidneys with the single-electrode device used in the trial was achieved in only 26% of treated patients. Furthermore, while the protocol mandated a stable drug regimen, approximately 40% of patients in each group had medication changes within the 6 months period between the RDN/sham procedure and assessment of the primary endpoint. The ethnic background of study participants may also have influenced the study outcome with the BP fall in the sham control arm in African Americans being double that of non-African Americans. Further detailed analysis of the Symplicity HTN-3 data is ongoing and will help to clarify whether these factors may have contributed to the neutral findings of the study. Irrespective of these findings, the conceptual basis for renal denervation as a treatment approach is very strong, as exemplified very recently with catheter-based renal denervation reducing renal noradrenaline content and abolishing hypertension in an obese dog model (Henegar et al. 2014) to an extent very similar in magnitude to that seen in hypertensive patients in Symplicity HTN-1. To bring the field forward, further adequately designed clinical trials with the various RDN devices will have to include some form of testing for renal denervation (e.g. renal noradrenaline spillover, electrical stimulation of renal nerves, intra-renal adenosine infusion), select appropriate patient cohorts characterized by sympathetic overactivity, and ensure adequate training and proctoring of interventionalists to ultimately define the future role of renal denervation in clinical medicine. Longer term follow-up of patients and repeated testing for renal denervation will help to address the unresolved question of possible reinnervation. Readers are invited to give their views on this and the accompanying CrossTalk articles in this issue by submitting a brief (250 word) comment. Comments may be submitted up to 6 weeks after publication of the article, at which point the discussion will close and the CrossTalk authors will be invited to submit a ‘Last Word’. Please email your comment to journals@physoc.org. Markus Schlaich currently heads the Neurovascular Hypertension & Kidney Disease Laboratory at Baker Heart Research. He is a renal physician with specific interest in hypertension, its causes and consequences. He has a strong clinical research background in hypertension, nephrology and cardiovascular disease with a specific interest in treatment modalities targeting the sympathetic nervous system and has contributed to the development of renal denervation as a novel therapeutic approach to hypertension. Yusuke Sata is Research Officer at Baker IDI Heart and Diabetes Institute. He obtained his medical degree from Fukuoka University, Japan. He subsequently trained in internal medicine and cardiology at the National Cerebral and Cardiovascular Center, Osaka, Japan. His research areas include baroreflex control of renal and cardiovascular dynamics, and central vs. peripheral control of cardiac and vascular reflexes and stress responses. Murray Esler is a cardiologist and medical scientist, based at the Baker IDI Heart and Diabetes Institute, where he is Senior Director, and at the Alfred Hospital. He is also Adjunct Professor of Medicine at Monash University. His research interests are the human sympathetic nervous system, stress and its effects on the heart and blood pressure, causes and treatment of high blood pressure and heart failure, and neurotransmitters of the human brain. Disclaimer: Supplementary materials have been peer-reviewed but not copyedited. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article. M. P. Schlaich has received consulting fees, and/or travel and research support from Medtronic, Abbott, Novartis, Servier, Pfizer and Boehringer-Ingelheim. Y. Sata has nothing to disclose. M. D. Esler has received consulting fees, and travel and research support from Medtronic and serves on the scientific advisory boards of Abbott Pharmaceuticals and Medtronic. M. P. Schlaich is supported by an NHMRC Research Fellowship.