Apparent or pseudo-RHTN, defined as inadequate BP control in a patient who does not have true RTHN but is receiving appropriate treatment, must first be excluded before consigning a an individual to a diagnosis of true RHTN [12]. Pseudo-RHTN most commonly arises from: poor office BP measurement technique, the ‘white coat’ effect, poor patient adherence with prescribed therapy, and/or a suboptimal antihypertensive treatment regimen. Complicated dosing regimens, inadequate patient education, and the rising cost of medication (within some healthcare systems) must also be borne in mind. It is of prime importance to conduct a thorough exploration of these patient and physician barriers to sustained BP control and eliminate them first, before establishing a definitive diagnosis of RHTN. Trials of RDN should ideally stipulate this process has been performed thoroughly before a patient is recruited.

Up to a third of patients, defined as having RHTN according to office BP recordings, are later found to manifest a white coat effect (i.e., a persistently elevated office BP but a normal home or ambulatory daytime average BP of <135/85 mmHg) [16, 21]. This serves to emphasize the importance of using ABPM to confirm a true RHTN diagnosis [3, 4, 5, 22]. The long-term prognostic implications of the white-coat effect appear to be intermediate between sustained hypertension and normotension, and there is evidence to suggest that these individuals do have an increased risk of developing a sustained hypertensive state [23]. A white coat effect should be suspected in any individual with persistently elevated office BP readings but no signs of target organ damage or signs/symptoms of over-treatment such as postural hypotension, dizziness, or syncope [24].

Office BP measurements could be deemed more reliable if taken using an automated device with the patient alone in a quiet room [25]. There is little doubt, however, that out-of-office BP measurements are more reproducible, reflect diurnal variation, reduce observer bias if automated devices are used, and provide multiple recordings from which to base a firm diagnosis or monitor response to therapy [26]. Furthermore two recent meta-analyses have confirmed a more accurate estimation of treatment response with out-of-office (home BP monitoring – HBPM and 24-h ABPM) recordings, ABPM also giving rise to an attenuated reduction in BP, when compared to clinic BP measurements [27, 28].

The landmark PAMELA study drew attention to the significant disparity between clinic and out-of-office BP measurements and established normal values for HBPM and ABPM [29]. It is now widely accepted that CV risk is more tightly correlated to out-of-office BP than to clinic BP [30, 31]. Furthermore, elevated ambulatory BP can predict CV morbidity and mortality in patients with RHTN, whereas clinic BP has less prognostic value [32]. These data provide compelling evidence for the routine use of out-of-office BP monitoring when assessing patients for presumed RHTN and should be a prerequisite inclusion criterion in trials of RDN. Indeed, an International Expert Consensus Statement for the performance of percutaneous RDN recommends confirmation of persistently elevated BP above target by using 24-h ABPM [9]. Similarly, expert consensus documents issued by the European Society of Cardiology (ESC) and the European Society of Hypertension (ESH) on catheter-based RDN also advocate ruling out pseudo-RHTN and thereafter confirming the validity of persistently high office BP recordings with the use of ABPM [6, 8, 10].

Once a diagnosis of RHTN has been reliably established, individuals should be evaluated for potentially remediable lifestyle factors such as obesity, alcohol, and caffeine consumption, and excess dietary sodium, which all contribute to the hypertensive state [11]. Thereafter, a survey for the potential use or abuse of exogenous substances that raise BP should be performed (Table 5.2). Once identified, the offending agents should be discontinued, minimized or substituted as required. Intuitive interventions such as weight loss, regular exercise and a high-fiber, low-salt diet must be attempted.

The next recommended step in the diagnostic pathway is to exclude a potentially remediable secondary cause [6, 8–10]. The prevalence of secondary hypertension is greater in the RHTN cohort than in the general hypertensive population, with studies indicating 5–10 % of RHTN patients with an identifiable cause [33, 34]. The most common causes are hyperaldosteronism, CKD (either the cause or result of chronic, poorly controlled HTN), renal artery stenosis (RAS), and obstructive sleep apnea [11, 35]. Less common causes include renal parenchymal disease, pheochromocytoma, thyroid diseases, Cushing’s syndrome, coarctation of the aorta, and intracranial tumours. It is fundamentally important for an RDN trial to incorporate a selection protocol through which those RHTN patients with a secondary cause are excluded prior to randomization. As such, the trial screening should include a focused history, thorough physical examination, biochemical evaluation, non-invasive imaging, and subsequent onward referral to a specialist clinic if deemed necessary prior to potential enrolment [9, 36].

Adherence to prescribed therapy is of particular relevance in this scenario since RHTN is largely an asymptomatic condition treated with multiple medications, each with their own array of potentially troublesome side effects. Measures to improve drug adherence such as improving patient education, motivation, and ownership of their management program along with setting realistic goals in achieving BP targets warrant careful attention.

Existing antihypertensive drug treatment requires optimization. Recent medication changes or dose adjustments should be given time to work – this can take up to a month to take effect [12]. If the standard A+C+D treatment algorithm is pursued, as recommended by NICE [4], the optimal fourth line agent for the pharmacologic treatment of RHTN remains open to debate. Although mineralocorticoid receptor antagonists appear to be the obvious contender, their long-term safety, particularly in patients with impaired renal function, is still unknown. As such, an individualized approach to tailor the best antihypertensive regimen for the patient is currently advocated with the addition of aldosterone antagonists if they are considered safe in the circumstances [6, 8–10]. At present there is currently no robust evidence available to define with absolute confidence the most clinically effective 4th or 5th line agent to attain target BP in RHTN [12]. It is little wonder, therefore, that RDN has been seen as a potential game-changer. It is imperative that trials of RDN should ensure potential recruits have had their current medications optimized and that their treatment regimen is in alignment with international consensus standards.

This was the original proof-of-concept non-randomized cohort study designed to show that RDN was feasible, safe and effective [38]. Krum and colleagues treated 45 patients from five centers in Australia and Europe, including the patient described in the case report above [37], using the Symplicity™ Renal Denervation system (Medtronic, Santa Rosa, CA, USA). An office systolic BP ≥160 mmHg (or ≥150 mmHg in Type 2 diabetics) based on an average of three readings was required for trial entry. Mean baseline office systolic and diastolic blood pressures were 177 ± 20 and 101 ± 15 mmHg respectively. Patients were taking an average of 4.7 medications, although diuretic therapy was not taken by all trial participants (95 %). Given the predominance of volume and sodium overload in these patients, attempts at achieving sustained BP control with diuretic medication is now regarded as a prerequisite therapeutic intervention before a RHTN diagnosis can be made [4, 22]. Secondary causes of RHTN had to be excluded prior to inclusion into the study. The trial required preserved renal function (estimated glomerular filtration rate – eGFR >45 mL/min/1.73 m²) to be eligible for RDN.

Office systolic and diastolic blood pressures after the procedure (while maintaining patients on their usual antihypertensive medication therapy) were reduced by 14/10, 21/10, 22/11, 24/11, and 27/17 mmHg at 1, 3, 6, 9, and 12 months, respectively. At the 12-month follow-up stage RDN appeared to be safe with successful completion of the procedure in 43 out of 45 patients. A single intraprocedural renal artery dissection was recorded. This had occurred prior to the application of RF energy without further sequelae. There were no other renovascular complications (such as vessel spasm or RAS) [38]. One patient developed a pseudoaneurysm at the femoral access site.

By 3 years, HTN-1 had recruited 153 patients, 88 of whom had complete data for the entire follow-up period [39]. The premise was to determine the durability of the BP lowering effect seen after RDN in light of concerns that renal afferent and efferent re-innervation may occur in the medium term post procedure. Again, the intervention was shown to be safe with a total of four complications out of the entire patient cohort (one renal artery dissection previously noted in the first 12-month report of HTN-1 [38] and three access-related groin complications). A single patient developed a significant right RAS 24 months after RDN which required angioplasty [39]. At 36 months, significant reductions in office systolic (−32.0 mmHg, 95 % confidence interval [CI] −35.7 to −28.2) and diastolic BP (−14.4 mmHg, CI−16.9 to −11.9) were noted in the 88 patients in whom data were available. In terms of response rate, 55/80 (69 %) had reductions in systolic BP ≥10 mmHg at 1 month. This rose to 82/88 (93 %) at 36 months, which some interpreted as a delayed response phenomenon, although the pathophysiological basis of this remains unclear. Of note, the need for antihypertensive medications actually rose over the 36-month period from a mean of 5.1 at baseline to 5.2 at study termination despite the purported gains in BP control [39]. Further analysis of medication dose or drug changes is precluded, however, by lack of medication data after 12 months – a major limitation of the study.

Whereas HTN-1 was a single-arm proof-of-principle study, and therefore exposed to confounding issues such as regression to the mean, placebo, and Hawthorne effects, HTN-2 was a prospective, multicenter trial in which 106 patients were randomly allocated to RDN (n = 52) or control (n = 54) [40]. A 2-week ‘Baseline Evaluation Period’ was used to analyze BP patterns in potential recruits with twice-daily home BP monitoring and a daily medication log to monitor compliance prior to formal randomization. Despite the opportunity to use these out-of-office BP recordings, the investigators restricted their RHTN diagnosis to the mean of office BP measurements. This potentially limits the generalizability of the trial’s results [12]. Much like HTN-1, therefore, patients with a systolic BP ≥160 mmHg (≥150 mmHg in Type 2 diabetics) despite adherence to ≥3 antihypertensive agents were eligible for recruitment. Furthermore, inclusion criteria stipulated a ‘stable’ treatment regimen of ≥3 antihypertensives, which prevented any change in drug or dose 2 weeks prior to randomization and maintenance of the same baseline combination for at least 6 months post RDN to avoid adjustments confounding the results [12, 40]. In HTN-1, 22 % of patients were taking an aldosterone antagonist at baseline [38] and only 17 % in HTN-2 [40], perhaps reflecting the relative lack of conclusive data on what constitutes best practice in terms of pharmacotherapy in RHTN patients. Exclusion of patients with a known secondary cause of HTN is routine at present and was actively performed in HTN-1 [38, 41]. The protocol for HTN-2 excluded Type I diabetics but did not explicitly exclude those with a known secondary cause – the reasoning behind the shift in recruitment protocols is unclear [40].

At the 6-month time point, office BP fell by 32/12 ± 23/11 mmHg from 178/96 mmHg ± 18/16 mmHg at baseline in those receiving RDN compared to a change of only 1/0 ± 21/10 mmHg from 178/97 ± 17/16 mmHg at baseline in the control arm. Note, however, the wide standard deviation in the latter cohort. When ABPM measurements were analysed, the reductions were less impressive. In the RDN group, there was a mean reduction of 11/7 ± 15/11 mmHg (n = 20) at 6 months using out-of-office recordings. Much like in HTN-1, RDN was shown to be safe with no serious complications related to the device or the procedure [40]. Renal function remained the same overall in both groups at 6 months.

Results from HTN-2 at 1 year included the original RDN group at baseline (n = 47) and control subjects who crossed over to the RDN arm and had the procedure performed per protocol at 6 months (n = 35) [42]. An overall fall in BP in the original RDN group of 28/10 mmHg remained durable at 12 months. The crossover group also demonstrated a fall in BP of 24/8 mmHg at 6 months. A crossover patient suffered renal artery dissection during guide catheter insertion for angiography. Aside from this, there were no other renovascular complications reported thereby reaffirming the safety of this procedure. No significant changes in estimated glomerular filtration rate (eGFR) were noted in either group at this time point. At 36-month follow up, the data had been locked and was only available in the RDN group (n = 40). The BP lowering effect remained durable with a reduction of 33/14 mmHg [43].

Both HTN-1 and HTN-2 were trials that introduced a seemingly safe and effective procedure to the interventional world at large. There were, however, a number of limitations to the studies, which prevent the widespread generalizability of the data:

The Global SYMPLICITY Registry (GSR) (ClinicalTrials.gov Identifier: NCT01534299) is consecutively enrolling up to 5,000 patients from over 200 sites worldwide to determine the real-world durability of effect and safety of the Symplicity RDN catheter. It also incorporates the GREAT SYMPLICITY registry initiated in Germany [51]. The registry aims to establish procedural benchmarking and practice patterns, assess the effect of geographical variation and procedural characteristics on outcome, and to collect quality of life data post procedure and in relation to patient comorbidity [51].

Data from the first 1,000 patients consecutively enrolled to the registry and followed up for 6 months were presented at the American College of Cardiology Scientific Sessions in March 2014 [52]. Safety solely related to the procedure appeared to be maintained with vascular complications occurring in only 0.4 % (n = 4) of the cohort (n = 913) at 6 months. There were no new cases of RAS. Both new onset end stage renal disease and a doubling of the serum creatinine were also rare (0.2 % each).

Perhaps most striking were those patients with a baseline office systolic BP ≥160 mmHg (i.e., the RHTN cut-off used in HTN-1 and HTN-2) or an ambulatory systolic BP ≥135 mmHg on ≥3 antihypertensive medications, which represented only a third (n = 327) of the 1,000-patient cohort studied for this preliminary analysis. This cohort achieved the greatest reduction in office BP post procedure, whereas those patients with a relatively “normal” baseline BP actually saw an increase over time (Table 5.3). It is not clear whether this latter cohort corresponded to those patients in whom the primary indication for RDN therapy was to treat a disease state characterized by SNS hyperactivation outside of the standard uncontrolled HTN parameter (i.e., heart failure, sleep apnea, insulin resistance, chronic kidney disease, or atrial fibrillation). Overall, the average reduction in office systolic BP from baseline to 6 months was a modest 11.9 mmHg for all patients. Among the subset of patients who met the BP criteria used in HTN-1 and HTN-2 and who were taking maximally tolerated doses of ≥3 antihypertensive agents the fall in mean BP was 17.3 mmHg.