Is renal denervation an alternative or a complement to aldosterone antagonists in treatment of resistant hypertension?

The role of the sympathetic nervous system in the pathophysiology of hypertension, particularly in resistant hypertension, is substantiated by a huge body of experimental and clinical arguments. Despite that, very little evidence is available in favor of a major effect of peripheral blockade of sympathetic receptors in essential hypertension [1–3]. Thus, several new therapeutic strategies have focused on targeting the hyperactivation of the sympathetic nervous system through renal denervation (RDN) or other devices interfering with the baroreflex [4]. Despite encouraging early results, a recent meta-analysis [5] has concluded for an unsatisfactory effect of RDN, and there is still a lot of discussion as to the real long-term extent of the blood pressure (BP)-lowering effect of RDN, especially when compared to the BP-lowering effect of intensified drug treatment, particularly with aldosterone receptor antagonists. In this issue of Journal of Hypertension, Rosa et al.[6] investigated the efficacy at 24 months of RDN versus intensified pharmacological treatment, including spironolactone as an add-on treatment, in patients with true resistant hypertension (24-h ambulatory BP monitoring, secondary causes of hypertension thoroughly excluded and compliance with prescribed antihypertensive medications evaluated at baseline). The Prague-15 trial [7] was a prospective, open-label, randomized evaluation of patients with true resistant hypertension. Patients were randomized in a 1 : 1 ratio either to a catheter-based RDN and optimal antihypertensive treatment group; or to an intensified pharmacological treatment group including spironolactone (25 mg daily) as an add-on treatment. The current study [6] shows that, over a period of 24 months of follow-up, RDN was well tolerated, with no serious side effects; however, it was not superior to intensified pharmacological treatment. Indeed, similar and comparable reductions of 24-h systolic BP after RDN or spironolactone addition after randomization were observed over a period of 24 months (−9.1 mmHg; P = 0.001, and −10.9 mmHg; P = 0.001, respectively). The crossover analysis showed nonsignificantly better efficacy of spironolactone addition on 24-h and office systolic BP reduction than RDN (3.7 mmHg; P = 0.27, and 4.6 mmHg in favor of spironolactone addition; P = 0.28, respectively). This result is particularly striking since patients were true resistant hypertensive, and that drug adherence was evaluated at baseline. The result of this trial cannot be interpreted in mechanistic terms because intervention did not target the same mechanistic pathways. Aldosterone antagonists, including spironolactone, although indirectly modulating the sympathetic nervous system, mainly act by reducing the sodium load (short term) and reversing cardiovascular fibrosis (long term). However, the evidence for benefit of the clinical use of spironolactone in randomized controlled trials in patients with resistant hypertension is scarce. Most of its putative protective effect comes from heart failure studies. The Addition of Spironolactone in Patients with Resistant Arterial Hypertension (ASPIRANT) [8] trial reported that spironolactone, as an add-on treatment, induced decreases in both office systolic BP (−14.6 mmHg) and 24-h systolic BP (−10.6 mmHg) that were significantly higher than corresponding reductions in the control groups. More recently, the results of this trial were extended by the ‘Spironolactone versus placebo, bisoprolol, and doxazosin to determine the optimal treatment for drug-resistant hypertension’ (PATHWAYS 2) trial [9], a 12-month randomized, double-blind, placebo-controlled crossover trial aiming to compare the efficacy of spironolactone versus placebo, bisoprolol, and doxazosin in treatment of resistant hypertension. This trial demonstrated that spironolactone was superior to other antihypertensive agents including bisoprolol and doxazosin in the treatment of resistant hypertension. The results of the present analysis by Rosa et al. further validate the important action of spironolactone in comparison with new invasive therapeutic strategies like RDN, in the setting of true resistant hypertension. The results clearly show similar BP-lowering effects of RDN and spironolactone as an add-on to an intensified pharmacological treatment at a long-term follow-up, thus challenging the belief that resistant hypertension cannot be adequately treated with existing drug therapies. Some aspects of this analysis, however, should be taken into consideration. Although the current analysis highlighted that spironolactone seems to be one of the most effective drugs in the treatment of resistant hypertension, a consistent number of patients experienced several adverse effects, including hyperkalemia and antiandrogen effects (31% of the overall patients in the intensified pharmacotherapy group) – a finding that may limit spironolactone clinical use on the long term. Moreover, the creatinine clearance was significantly reduced after 24 months in the intensified pharmacotherapy group. In addition to this, the BP-lowering effect in the RDN group was less pronounced than in other studies investigating the efficacy of RDN at long term [10]. This may be caused by suboptimal denervation, although the investigators underwent training and reassessment about the procedure quality. The lower baseline systolic BP compared with previous trials might partly explain the smaller BP-lowering effect of RDN. The study by Rosa et al.[6] also has methodological limitations. The crossover character of the study added potential difficulties in precise quantification of BP with the two therapeutic strategies. In fact, crossing over should be considered as an outcome by itself because if RDN had been 100% effective, no crossing over to spironolactone would have been necessary. Moreover, neither RDN nor spironolactone addition was performed when goal BP was reached at 12-month follow-up. Furthermore, RDN as a crossover procedure was stopped at the beginning of 2014 after the negative results of ‘Renal sympathetic denervation in patients with treatment-resistant hypertension’ (SYMPLICITY HTN-3) [11] and an interim analysis of the present study. Thus, the results of this analysis might be biased in terms of BP-lowering effect. Last, although tested at baseline, adherence to the prescribed antihypertensive treatment was not assessed at the time of primary endpoint assessment. Nonadherence tends to increase with follow-up duration, an effect which may be exaggerated in this study because of spironolactone adverse effects. More generally, it is possible that patients included for RDN studies present a selection because they volunteer for an alternative to taking medications and may be less adherent to complex medications. Recently, Azizi et al.[12] reported by using drug screening in urine or plasma by ultrahigh liquid chromatography coupled with tandem mass spectrometry (UPLC-MS/MS) that more than 50% of the patients included in the Renal Denervation in the Renal Denervation for Hypertension (DENERHTN) trial were deemed to be partially or completely nonadherent to their prescribed antihypertensive therapy at 6 months of follow-up. Direct measurement of antihypertensive drugs in plasma or urine seems to be the best way not only for evaluation but also for improvement of drug adherence in resistant hypertension. Indeed, when nonadherent patients are informed about the results of drug screening, BP control improved markedly without intensification of therapy [13]. Thus, whatever its modalities, drug adherence should be measured before interpreting precisely the results of clinical trials, particularly in the setting of resistant hypertension. Finally, one important remaining question is whether RDN can be considered as an alternative for resistant hypertensive patients not tolerating spironolactone or a complement for those not fully responding to spironolactone. The present data cannot answer this question. Taken together, RDN is at best as (in)effective as intensified pharmacological therapy in terms of BP-lowering effect among patients with true resistant hypertension. The present study shows a similar BP-lowering effect of RDN when compared to an intensified pharmacological therapy including spironolactone as add-on treatment. In theory, RDN could be considered as an alternative to spironolactone. However, concerns related to the long-term safety of spironolactone, as discussed above, and the fact that not all patients will respond to this treatment, RDN might also be considered as having a synergistic action with optimized medical treatment and may be considered as complement to aldosterone antagonists. Indeed, the prescription of aldosterone antagonist at baseline was identified as one of predictors for a larger decrease in office systolic BP in a post-hoc analysis of SYMPLICITY HTN-3 [14]. This observation, as reported also by other studies of RDN, highlights the encouraging potential add-on effect of RDN. It is still too early to answer whether those two therapeutic strategies might fit into the overall clinical management of patients with resistant hypertension. Because this task has not been formally and rigorously investigated, clinicians will still remain obliged to prescribe more effective antihypertensive agents, including spironolactone, even in combination with RDN, until a dedicated clinical trial will adequately approach the problem. ACKNOWLEDGEMENTS Conflicts of interest There are no conflicts of interest.