Author + information
- Received January 22, 2013
- Revision received June 5, 2013
- Accepted June 11, 2013
- Published online November 12, 2013.
- Christian Ott, MD∗,
- Felix Mahfoud, MD†,
- Axel Schmid, MD‡,
- Tilmann Ditting, MD∗,
- Paul A. Sobotka, MD§,‖,
- Roland Veelken, MD∗,
- Aline Spies, MD†,
- Christian Ukena, MD†,
- Ulrich Laufs, MD†,
- Michael Uder, MD‡,
- Michael Böhm, MD† and
- Roland E. Schmieder, MD∗∗ ()
- ∗Department of Nephrology and Hypertension, University Hospital at the University of Erlangen-Nuremberg, Erlangen, Germany
- †Klinik für Innere Medizin III, Universitätsklinikum des Saarlandes, Homburg/Saar, Germany
- ‡Department of Radiology, University of Erlangen-Nuremberg, Erlangen, Germany
- §The Ohio State University, Columbus, Ohio
- ‖Coridea-NC1, New York, New York
- ↵∗Reprint requests and correspondence:
Dr. Roland E. Schmieder, Department of Nephrology and Hypertension, University of Erlangen-Nuremberg, Ulmenweg 18, 91054 Erlangen, Germany.
Objectives This study sought to investigate the effect of renal denervation (RDN) in patients with treatment-resistant hypertension according to the established definition (Joint National Committee VII and European Society of Hypertension/European Society of Cardiology guidelines), that is, office blood pressure (BP) ≥140/90 mm Hg (with at least three antihypertensive drugs, including a diuretic, in adequate doses) and confirmed by 24-h ambulatory BP monitoring (ABPM).
Background RDN emerged as an innovative interventional antihypertensive therapy. However, so far, only patients with severe hypertension (systolic BP ≥160 mm Hg or ≥150 mm Hg for patients with type 2 diabetes) have been investigated.
Methods In this study, there were 54 patients with moderate treatment-resistant hypertension (office BP ≥140/90 mm Hg and <160/100 mm Hg and diagnosis confirmed by 24-h ABPM of ≥130/80 mm Hg) who underwent catheter-based RDN using the Symplicity catheter (Medtronic Inc., Mountain View, California).
Results Patients were treated with 5.1 ± 1.4 antihypertensive drugs on average. Office BP was significantly reduced by 13/7 mm Hg 6 months after RDN (systolic: 151 ± 6 mm Hg vs. 138 ± 21 mm Hg, p < 0.001; diastolic: 83 ± 11 mm Hg vs. 75 ± 11 mm Hg, p < 0.001). In patients (n = 36) who underwent ABPM 6 months after treatment, there was a reduction in average 24-h ABPM by 14/7 mm Hg (systolic: 150 ± 16 mm Hg vs. 136 ± 16 mm Hg, p < 0.001; diastolic: 83 ± 10 mm Hg vs. 76 ± 10 mm Hg, p < 0.001). In 51% of patients, office BP was controlled below 140/90 mm Hg after RDN. In addition, heart rate decreased from 67 ± 11 to 63 ± 10 beats/min (p = 0.006).
Conclusions Our data indicate that RDN may reduce office and 24-h ambulatory BP substantially in patients with moderate treatment-resistant hypertension. (Renal Denervation in Treatment Resistant Hypertension; NCT01687725)
Arterial hypertension remains a global public health burden because it is the most prevalent risk factor for cardiovascular (CV) morbidity and mortality (1). Treatment-resistant hypertension (TRH) is defined as blood pressure (BP) above treatment goal despite use of at least three antihypertensive drugs in adequate doses from different classes including a diuretic (according to Joint National Committee [JNC]-7  and European Society of Hypertension/European Society of Cardiology [ESH/ESC] guidelines criteria [3,4]). Increased sympathetic activity plays a crucial pathogenetic role in the development, maintenance, and aggravation of arterial hypertension (5), especially in TRH. Renal denervation (RDN) has emerged as an interventional approach to treat severe TRH by altering sympathetic afferent and efferent nerve activity (6). The Symplicity Hypertension (HTN)-1 (including extended long-term follow-up) and HTN-2 studies have shown that catheter-based RDN is an apparently safe approach for achieving a substantial and sustained BP reduction in patients with severe TRH (7–9). Furthermore, after RDN, BP during cardiopulmonary exercise was also reduced without alteration in cardiorespiratory response (10). Moreover, regression of left ventricular mass and improvement in diastolic function were observed after RDN (11). One of the main inclusion criteria of these studies was an office systolic blood pressure (SBP) ≥160 mm Hg (≥150 mm Hg for patients with type 2 diabetes), despite the patient being treated with 3 or more antihypertensive drugs (including 1 diuretic). The BP inclusion criterion (SBP ≥160 mm Hg) of the Symplicity studies is well specified, but it is obvious that most patients with TRH have office SBP readings between 140 and 160 mm Hg (12).
In the face of the evidence from antihypertensive trials, there is little doubt that these patients have a substantially elevated CV risk. It has been repeatedly shown that irrespective of the starting BP level, even modest BP reduction is accompanied by significant attenuation of CV morbidity and mortality (3,4,13,14). Therefore, new antihypertensive strategies to achieve BP control, and therefore reduce CV risk, in less severe forms of TRH are needed and should be carefully evaluated; so far, only 1 small retrospective analysis focused on the influence of RDN in moderate hypertensive cases (15).
We, therefore, initiated a prospective multicenter open-label pilot study aimed at assessing the efficacy of RDN in cases of true moderate TRH. The antihypertensive efficacy of RDN in patients demonstrated by ambulatory blood pressure monitoring (ABPM) to have true TRH was assessed by office BP and ABPM (16) change. This study was undertaken in order to extend the evidence base of RDN beyond the findings of previous trials that were limited to severe TRH.
Study cohort and design
In this investigator-initiated, prospective, multicenter pilot study, 54 adult patients with TRH defined by JNC-7 (2) and ESH/ESC (3,4) guidelines were consecutively included if the patient's office BP was ≥140/90 mm Hg and <160/100 mm Hg. In addition, in every patient, true resistant hypertension was confirmed by initial 24-h ABPM (≥130/80 mm Hg), thereby excluding “white coat” (elevated office BP, but normal values out of the office, either on ABPM or home blood pressure monitoring [HBPM]) or pseudoresistant hypertension; but in contrast to Symplicity HTN-2 trial, no home BP measurement for 2 weeks was required (8). Patients were required to be on an unchanged antihypertensive drug regimen for at least 2 months, without any allowance for dose or regimen adjustments. Thus, office BP measurements and 24-h ABPM were taken under stable conditions. Main exclusion criteria were renal artery anatomy (main renal arteries <4 mm in diameter or <20 mm in length, hemodynamically or anatomically significant renal artery abnormality or stenosis in either renal artery, history of renal artery intervention including balloon angioplasty or stenting, multiple main renal arteries in either kidney) and any secondary cause of hypertension, except for treated obstructive sleep apnea syndrome and chronic kidney disease. Altogether, 19% of patients referred to our tertiary universities' outpatient clinics were eligible for this pilot study and were treated between February 2011 and March 2012. The study protocol was approved by the local ethics committees from the 2 participating centers, and the study was performed according to Declaration of Helsinki and Good Clinical Practice guidelines. Written informed consent was obtained from all patients before study entry. The study was registered at U.S. National Institutes of Health Clinical Trials website (NCT01687725).
Office and 24-h ambulatory BP
Office BP was measured in both arms after 5 min of rest in a sitting position, with an oscillometric device (study center Erlangen: Dinamap Pro100V2 [Criticon, Norderstedt, Germany]; study center Homburg/Saar: Omron HEM-705 monitor [Omron Healthcare, Vernon Hills, Illinois], with a printer for documentation). Subsequent BP measurements were made in the arm with the higher BP reading, and the average of the last 3 measurements was taken. Ambulatory 24-h BP measurements were taken with an automatic portable device that was validated according the ESH International Protocol (17) (e.g., Spacelab no. 90207, Redmont, California) prior to RDN. Twelve patients refused ABPM, and in 6 patients, data readings were insufficient (in 5 cases there were no nighttime recordings; and in 1 case, recordings <80% successful were obtained) 6 months after RDN. Hence, in a subgroup (n = 36), successful ABPM was measured 6 months after RDN. In every patient, the same device was used before and 6 months after RDN. Patients were divided according their dipping pattern into “dippers” (night-time BP fall >10%) and “non-dippers” (night-time BP fall <10%).
Heart rate was obtained using 12-lead electrocardiography, performed 10 min after supine rest at standard sensitivity (10 mm = 1 mV) and a paper speed of 50 mm/s.
Catheter-based renal denervation
For RDN, the femoral artery was accessed with standard endovascular technique. Radiofrequency catheter (Symplicity RDN system, Medtronic Inc.) was advanced in each renal artery by angiography. As described previously in detail (7), at least 4 radiofrequency ablations (energy delivery for 120 s each), controlled and regulated by a radiofrequency generator, were applied within the lengths of each renal artery. Patients received 5,000 IU of heparin to achieve an activated clotting time of >250 s. Diffuse visceral pain during the procedure was managed with anxiolytics and narcotics.
All analyses were performed using SPSS version 19.0 software (SPSS Inc., Chicago, Illinois). Normal distribution of data was confirmed by Kolmogorov-Smirnov test before further analyses. Data were compared by paired and unpaired Student t-tests and by Wilcoxon and McNemar tests where appropriate. Data are mean ± SD. Univariate correlation was performed using the Pearson correlation coefficient. A two-sided p value of <0.05 was considered statistically significant.
Clinical characteristics of the study population are shown in Table 1. Most patients were middle-aged, male, and overweight. The mean office BP was 151/83 mm Hg, and mean 24-h ambulatory BP was ≥130/80 mm Hg (systolic: 149 ± 13 mm Hg, and diastolic: 81 ± 11 mm Hg) in all patients, thereby confirming true resistant hypertension and excluding “white coat” hypertension (18). Patients were treated with a mean of 5.1 ± 1.4 antihypertensive drugs, and 3 patients were not receiving diuretic therapy because of acknowledged contraindications such as secondary polycythemia or documented intolerance.
No severe adverse events were observed, including renal vascular and renal damage, associated with the procedure.
Three months after RDN, office BP was reduced (systolic: 151 ± 6 mm Hg vs. 146 ± 23 mm Hg, p = 0.164; diastolic: 83 ± 11 mm Hg vs. 79 ± 12 mm Hg, p = 0.011) (Figs. 1A, 1B, and 2), and 6 months after RDN, a further BP reduction was documented (systolic: 151 ± 6 mm Hg vs. 138 ± 21 mm Hg, p < 0.001; diastolic: 83 ± 11 mm Hg vs. 75 ± 11 mm Hg, p < 0.001) (Figs. 1A, 1B, and 2). In 51% of patients, office BP was controlled after RDN, defined as office BP <140/90 mm Hg. Additionally, 61.1% of patients had an office SBP reduction of ≥10 mm Hg 6 months after RDN (defined as treatment response).
In patients with available ABPM before and after RDN (n = 36), office BP was reduced (systolic: 151 ± 5 mm Hg vs. 142 ± 20 mm Hg, p = 0.012; diastolic: 84 ± 10 mm Hg vs. 79 ± 11 mm Hg, p = 0.003) 3 months after RDN. There was a further reduction of SBP (151 ± 5 mm Hg vs. 133 ± 19 mm Hg, p < 0.001) and diastolic BP (84 ± 10 mm Hg vs. 75 ± 11 mm Hg, p < 0.001) 6 months after RDN (Fig. 3). In addition, we observed a reduction in mean 24-h ABPM (systolic: 150 ± 16 mm Hg vs. 136 ± 16 mm Hg, p < 0.001; diastolic: 83 ± 10 mm Hg vs. 76 ± 10 mm Hg, p < 0.001) 6 months after RDN (Fig. 3). RDN reduced both daytime ABP, by 14/6 mm Hg (systolic: 154 ± 17 mm Hg vs. 140 ± 16 mm Hg, p < 0.001; diastolic: 85 ± 11 mm Hg vs. 79 ± 11 mm Hg, p = 0.001) and nighttime ABP, by 13/5 mm Hg (systolic: 141 ± 17 mm Hg vs. 128 ± 18 mm Hg, p < 0.001; diastolic: 75 ± 11 mm Hg vs. 70 ± 11 mm Hg, p = 0.005). Dipping pattern was absent in 60% before RDN and in 51.4% after RDN (p = 0.607), respectively.
Six months after RDN, HR was also significantly reduced (67 ± 11 mm Hg vs. 63 ± 10 beats/min, p = 0.006) (Fig. 4). The change of HR did not correlate with the reduction of SBP at 3 months (r = −0.042, p = 0.768) and at 6 months (r = −0.095, p = 0.538), indicating that RDN resulted in a reduction of HR independently of its BP-lowering effects. In accordance, HR reduction did not differ between nonresponder and responder of SBP, defined as a decrease of SBP ≥10 mm Hg at 6 months after RDN. Furthermore, SBP reduction did not differ between nonresponder and responder of HR, defined as HR decrease of ≥2 beats/min (i.e., median HR change) at 6 months after RDN (Fig. 5).
In 37% of patients, antihypertensive medication was reduced during the 6-month follow-up period despite the guidance of the study protocol not to do so. According to an exploratory analysis, BP dropped by 13/6 mm Hg (p = 0.024/0.015) in those whose medication was maintained and by 13/9 mm Hg (p = 0.024/0.007) in those whose medication was reduced 6 months after RDN. Antihypertensive medication was not increased in any patient.
Recent studies have shown that RDN is an effective antihypertensive approach for severe TRH, but its effect in less severe forms of hypertension is unknown. In a retrospective, observational pilot study of an electronic medical database, consisting of approximately 30,000 patients, the prevalence of TRH in an ambulatory care setting was assessed. It was reported that 9.1% of patients had TRH and that their baseline SBP was 148 (± 20.8) mm Hg (12). Ensuring a normal distribution, it is estimated that approximately 70% of patients had a SBP <160 mm Hg. In contrast to previous studies (7,8), we have included patients with moderate TRH, whose office BP was between ≥140/90 mm Hg and <160/100 mm Hg, despite use of 3 antihypertensive drugs in full doses of different classes including a diuretic. Of note, in every patient, true resistant hypertension was confirmed by 24-h ABPM, thereby excluding “white coat” or pseudoresistant hypertension. According to the Spanish registry study, 37.5% of patients with presumed TRH based on office BP values were found to have pseudoresistant hypertension (i.e., 24-h ABPM <130/80 mm Hg) with a substantial white coat effect (18). The major finding of our prospective, multicenter study is that RDN resulted in a significant reduction of office systolic and diastolic BP as well as 24-h ABP in moderate true TRH.
Not surprisingly, the magnitude of office BP reduction after RDN and the responder rate (SBP reduction of at least 10 mm Hg after 6 months) was less pronounced than in previous reports (7,8) because the entry BP criteria were lower and pre-treatment BP was the major determinant of the magnitude of BP fall (9,19). However, it has been repeatedly shown that irrespective of starting BP level, even modest BP reductions are accompanied by significant attenuation of CV morbidity and mortality (3,4,13,14). Although the prognosis of TRH has not been robustly evaluated, recent evidence indicates that TRH has a severe prognosis (20,21). Recently, a single-center prospective study consisting of approximately 2,300 patients with uncomplicated hypertension with a mean follow-up of 3.5 years showed that TRH leads to a higher incidence of coronary artery damage (p = 0.004), stroke (p = 0.027), and atrial fibrillation (p = 0.001) than non-TRH (22). Moreover, TRH was an independent predictor of composite CV endpoints. Post-hoc analysis of randomized trials analyzing the CV outcome in relation to achieved BP levels during treatment have shown that, regardless of the type of treatment, patients in whom SBP was reduced to <140 mm Hg have a reduced incidence of CV events compared with patients with SBP values >140 mm Hg (23–26). Recently, a meta-analysis of major trials with antihypertensive agents revealed that in patients with high CV risk, the “residual risk,” meaning the risk level attained by intensive therapy, can rarely be decreased below the cut-off value defining a high-risk condition (i.e., 20% risk of CV events in 10 years). This indicates that delaying therapeutic correction of CV risk factors until a high level is achieved blunts the full benefits of interventions (27). Herein, in our study, over 50% of the patients were newly controlled according to target office BP values 6 months after RDN.
RDN is associated with reductions of additional markers of risk for CV morbidity and mortality. Amongst these are reductions in markers of arterial stiffness (28), insulin resistance (29,30) and left ventricular hypertrophy (11). These findings suggest earlier consideration of therapeutic RDN may result in further patient benefits.
National and international guidelines provide thresholds for office BP values; however, threshold values of ABPM for various severity stages of hypertension are not established. A meta-analysis by Mancia and Parati (16) indicated that while patients achieved target office BP values (<140/90 mm Hg), BP values measured by 24-h ABPM remained above the corresponding value, 125 to 130/80 mm Hg, generally used as a threshold for hypertension. Moreover, no prospective study has been conducted in which patients were stratified and treated based on ambulatory BP criteria, and subsequent assessment of incidence of CV events during follow-up has been performed. The value and magnitude of 24-h ABPM reduction is comparable to that of office BP reduction in our study. This is in contrast to the findings in the Symplicity studies but is related to the lower BP values before RDN. In an Australian study consisting of 8,575 patients with a wide range of BP values, it was found that the lower the BP before the intervention, the closer the agreement between ABPM and office BP reductions were (31). No clear improvement in dipping status was found 6 months after RDN, which is not surprising as it has been shown that classification of patients into dippers and non-dippers is poorly reproducible over time (32,33).
In approximately 40% of the patients, antihypertensive therapy was reduced in the follow-up, which, however, is not interchangeable with BP control. This unintended reduction of antihypertensive medication implies that the reported effect of BP reduction and control underestimates the real BP reduction following RDN in our cohort of patients with moderate TRH. It was the decision of the individual primary care physician to reduce the antihypertensive medication despite the strict study protocol to keep medication stable unless SBP decreased <120 mm Hg or hypotensive side effects became clinically relevant; no severe hypotensive episode or clinical symptoms were observed. It seems this is rather the result that physicians are willing to accept higher levels of BP in TRH, a finding that has been reported in the Supporting Hypertension Awareness and Research Europe-wide (SHARE) survey (34) and other studies (35).
Our data are encouraging because they potentially indicate an innovative interventional approach to treating patients with moderate TRH, with the potential consequence of lower CV morbidity and mortality. Furthermore, the socioeconomic consequences need to be carefully analyzed because both a lower incidence of CV complications and reduced costs of antihypertensive medication need to be balanced with the costs of RDN intervention.
A major limitation of our study is the lack of a control group, and we feel our sample size is rather small. The study population we examined represents a large population of uncontrolled treated hypertensive patients. We thus need a large-scale prospective, randomized, multicenter, controlled clinical trial in this group of TRH patients to precisely define the therapeutic role of RDN in moderate TRH patients. Indeed, on March 7, 2013, Medtronic submitted an Investigational Device Exemption to the U.S. Food and Drug Administration to assess in a randomized, controlled clinical trial the safety and efficacy for RDN in patients with moderate TRH (Symplicity HTN-4).
Our first prospective pilot study indicates that RDN reduces office BP and 24-h ABPM substantially and to a similar extent in patients with moderate true TRH.
Dr. Mahfoud has a financial relationship with Medtronic, Vessix, Recor, and St. Jude. Dr. Sobotka is an employee of Cibiem Inc.; receives royalties from sales related to Medtronic Ardian renal denervation system; and is a consultant with Medtronic Inc., Ardelyx Inc., and Rox, Inc. Dr. Ukena has received speaker's honoraria from Medtronic. Dr. Böhm is on advisory boards of AstraZeneca, Bayer AG, Boehringer-Ingelheim, Daiichi-Sankyo, Merck Sharp & Dohme, Novartis, Pfizer, Sanofi-Aventis, and Servier; is a member of the speaker's bureaus of AstraZeneca, AWD Dresden, Bayer, Boehringer-Ingelheim, Berlin-Chemie, Daiichi-Sankyo, Merck Sharp & Dohme, Novartis, Pfizer, Sanofi-Aventis, and Servier; and provides study support to AstraZeneca, Bayer AG, Boehringer-Ingelheim, Novartis, Pfizer, Sanofi-Aventis, Servier, and Adrian-Medtronic. Prof. Schmieder has served as advisor and speaker and the University Hospital received grants from Medtronic. All other authors have reported they have no relationships relevant to the contents of this paper to disclose. Drs. Ott and Mahfoud contributed equally to this work.
- Abbreviations and Acronyms
- ambulatory blood pressure monitoring
- blood pressure
- European Society of Cardiology
- European Society of Hypertension
- heart rate
- Joint National Committee
- renal denervation
- systolic blood pressure
- treatment-resistant hypertension
- Received January 22, 2013.
- Revision received June 5, 2013.
- Accepted June 11, 2013.
- American College of Cardiology Foundation
- Mancia G.,
- De Backer G.,
- Dominiczak A.,
- et al.
- Ukena C.,
- Mahfoud F.,
- Kindermann I.,
- et al.
- Brandt M.C.,
- Mahfoud F.,
- Reda S.,
- et al.
- McAdam-Marx C.,
- Ye X.,
- Sung J.C.,
- Brixner D.I.,
- Kahler K.H.
- Zhang Y.,
- Zhang X.,
- Liu L.,
- Zanchetti A.
- Calhoun D.A.,
- Jones D.,
- Textor S.,
- et al.
- Daugherty S.L.,
- Powers J.D.,
- Magid D.J.,
- et al.
- Tsiachris D.,
- Tsioufis C.,
- Kasiakogias A.,
- et al.
- Mancia G.,
- Grassi G.,
- Zanchetti A.
- Brandt M.C.,
- Reda S.,
- Mahfoud F.,
- Lenski M.,
- Bohm M.,
- Hoppe U.C.
- Mahfoud F.,
- Schlaich M.,
- Kindermann I.,
- et al.
- Head G.A.,
- Mihailidou A.S.,
- Duggan K.A.,
- et al.