Effectiveness of Renal Denervation Therapy for Resistant Hypertension

Accepted Manuscript Effectiveness of Renal Denervation Therapy for Resistant Hypertension: A Systematic Review and Meta-Analysis Mark I. Davis, MD Kristian B. Filion, PhD David Zhang, Mark J. Eisenberg, MD, MPH Jonathan Afilalo, MD, MSc Ernesto L. Schiffrin, MD, PhD Dominique Joyal, MD PII:

S0735-1097(13)01727-0

DOI:

10.1016/j.jacc.2013.04.010

Reference:

JAC 18821

To appear in:

Journal of the American College of Cardiology

Received Date: 10 January 2013 Revised Date:

18 March 2013

Accepted Date: 7 April 2013

Please cite this article as: Davis MI, Filion KB, Zhang D, Eisenberg MJ, Afilalo J, Schiffrin EL, Joyal D, Effectiveness of Renal Denervation Therapy for Resistant Hypertension: A Systematic Review and Meta-Analysis, Journal of the American College of Cardiology (2013), doi: 10.1016/j.jacc.2013.04.010. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Effectiveness of Renal Denervation Therapy for Resistant Hypertension: A Systematic Review and Meta-Analysis Mark I. Davis MD*, Kristian B. Filion PhD†, David Zhang*, Mark J. Eisenberg MD, MPH*‡, Jonathan Afilalo MD, MSc*‡, Ernesto L. Schiffrin MD*§, PhD Dominique Joyal MD*‡ *

SC

RI PT

Division of Internal Medicine, Jewish General Hospital, McGill University, Montreal, Canada † Division of Clinical Epidemiology, Jewish General Hospital, McGill University, Montreal, Canada ‡ Division of Cardiology, Jewish General Hospital, McGill University, Montreal, Canada §Lady Davis Institute for Medical Research, Jewish General Hospital, McGill University, Montreal, Canada

M AN U

Address for correspondence and reprints: Dominique Joyal, MD, FRCPC, FACC, FSCAI 3755 Cote Ste-Catherine Road Montreal, Quebec, Canada H3T 1W3 Tel: 514 340-8222 / Fax: 514 340-7534 [email protected]

Running Head: Renal denervation for resistant hypertension Conflict of Interest: None

AC C

EP

TE D

Word Count: 4443

1

ACCEPTED MANUSCRIPT

M AN U

SC

RI PT

STRUCTURED ABSTRACT Objectives: We sought to determine the current effectiveness and safety of sympathetic renal denervation (RDN) for resistant hypertension (RH). Background: RDN is a novel approach that has been evaluated in multiple small studies. Methods: We performed a systematic review and meta-analysis of published studies evaluating the effect of RDN in patients with RH. Studies were stratified according to controlled vs. uncontrolled design and analyzed using random-effects meta-analysis models. Results: We identified 2 randomized controlled trials (RCT), 1 observational study with a control group and 9 observational studies without a control group. In controlled studies, there was a reduction in mean systolic and diastolic BP at 6 months of -28.9 (95% CI -37.2, -20.6) and -11.0 (95% CI -16.4, -5.7) mmHg, respectively, compared to medically treated patients (for both, p<0.0001). In uncontrolled studies, there was a reduction in mean systolic and diastolic BP at 6 months of -25.0 (95% CI -29.9, -20.1) and -10.0 (95% CI -12.5, -7.5) mmHg, respectively, compared to pre-RDN values (for both, p<0.00001). There was no difference in the effect of RDN according to the five catheters employed. Reported procedural complications included 1 renal artery dissection and 4 femoral pseudoaneurysms. Conclusions: RDN resulted in a substantial reduction in mean BP at 6 months in patients with RH. The decrease in BP was similar irrespective of study design and catheter employed. Large RCTs with long-term follow-up are needed to confirm the sustained efficacy and safety of RDN. Keywords: Renal denervation, Resistant hypertension, Meta-analysis

AC C

EP

TE D

Abbreviations: ACEI = Angiotensin converting enzyme inhibitor ARB = Angiotensin II receptor blocker BMI = Body mass index BP = Blood pressure CAD = Coronary artery disease CCB = Calcium channel blocker DM = Diabetes mellitus GFR = Glomerular filtration rate RDN = Renal sympathetic denervation RH = Resistant hypertension

2

ACCEPTED MANUSCRIPT

INTRODUCTION Resistant hypertension (RH) is defined as uncontrolled systolic blood pressures (BP) despite therapy with ≥3 anti-hypertensive agents from at least 3 different classes including a

RI PT

diuretic. In most studies, 10-15% of hypertensive subjects (1,2), but up to 20% of the

hypertensive population in some publications (3), have RH, particularly those with advanced age, obesity, diabetes mellitus, sleep apnea and chronic kidney disease (4-6). In patients with

SC

RH, pharmacological options are limited. Historically, a surgical option, namely sympathectomy, led to significant reduction in BP but was associated with high surgical morbidity (7-9).

M AN U

Although surgical sympathectomy was largely abandoned in clinical practice, there has been a renewed interest in the concept as animal models (10,11) showed that renal sympathectomy led to significant reduction in BP and improvement in end organ function (12-15). Percutaneous renal sympathectomy has emerged as a safer, although invasive approach

TE D

utilizing radiofrequency probes to ablate the sympathetic fibers along the renal artery. The proof of concept study (16) demonstrated surprisingly good results and was subsequently followed by a series of studies using different catheters. These studies have generated great enthusiasm such

EP

that percutaneous renal denervation therapy (RDN) has been adopted at a rate rarely seen in the hypertension field. RDN for RH is currently approved in Europe and Canada and is pending

AC C

approval in the United States. One RDN catheter system (Medtronic Ardian Inc., Palo Alto, CA) has been used to treat over 4,000 patients worldwide thus far (17). Despite the enthusiasm and rapid uptake, there has yet to be a comprehensive review of the available evidence to support the practice of RDN.

We have systematically reviewed the current body of evidence for RDN, and quantified its BP-lowering effect in patients with RH using a random effects meta-analysis model.

3

ACCEPTED MANUSCRIPT

METHODS Data sources and search strategy. We performed a systematic review and meta-analysis in accordance with the standards set forth by the Preferred Reporting Items for Systematic

RI PT

Reviews and Meta-Analyses (PRISMA) statement (18,19). We searched PubMed, EMBASE and the Cochrane Collaboration database using the keywords “renal denervation”, “blood pressure” and “hypertension”. The search was limited to English language articles published in the last 5

SC

years (this technology was only developed in that timeframe). In addition, we hand-searched references of retrieved articles and used PubMed’s related articles feature to identify studies not

M AN U

captured by our primary search strategy. The final search was run on December 1, 2012. Study selection. We included randomized and observational studies comparing BP response in patients treated with RDN vs. patients treated with standard medical therapy (controlled studies) and observational studies comparing BP in a single group of patients pre-

TE D

and post-RDN (uncontrolled studies). Inclusion criteria were (1) RDN performed using contemporary percutaneous catheters and radiofrequency probes, (2) patient population with RH (not meeting BP target despite therapy with 3 or more anti-hypertensive agents from at least 3

EP

different classes), (3) at least 10 study participants, and (4) at least 3 months of follow-up for blood pressure response. BP measurements could include manual, automated, or invasive blood

AC C

pressure recordings, as long as the same method was used pre- and post-RDN. Reviews, editorials, letters, animal studies, case reports, and conference abstracts were excluded. Once full articles were retrieved, studies were further excluded if there was an overlap in patients with another study within the same analysis (in which case the larger sample size of the two studies was selected). Thus, while some patients could possibly have been included in both the

4

ACCEPTED MANUSCRIPT

controlled and uncontrolled study analyses, they were only included once in any given analysis. Consequently, there was no overlap in patients included in our meta-analyses. Data Extraction. Data was extracted in duplicate by two independent reviewers (MD,

RI PT

DZ). Disagreements were resolved by consensus. We extracted data pertaining to baseline

characteristics of study subjects (number of subjects, age, sex, comorbidities, antihypertensive agents), trial inclusion and exclusion criteria, method of blood pressure measurement, type of

SC

catheter used, BP response to RDN (including BP pre- and post-RDN), non-responder rate, procedural complications, maximal length of follow-up, and mortality.

M AN U

Outcomes. The primary outcome measure was mean systolic and diastolic BP reduction following RDN between 3 and 6 months of follow-up. Secondary outcome measures included (1) non-responder rate, defined as an achieved decrease in systolic BP of <10 mmHg, (2) mean BP reduction stratified by catheter type, and (3) reported procedural complications and averse

TE D

outcomes including death from any cause.

Methodological Quality. To determine the quality of the included studies, we used the Cochrane Collaboration Risk of Bias Tool (20) for the two randomized control trials and the

EP

Newcastle-Ottawa scale for the observational studies. We set a follow-up rate of greater than 70% at 6 months as a limit to determine high risks of bias at follow-up for studies evaluated with

AC C

the Newcastle-Ottawa scale in the outcome section of this scale (See appendix 2). Data synthesis and statistical analysis. For controlled studies, the difference in BP

change with RDN vs. medical therapy was pooled across studies and analyzed using randomeffects meta-analysis models with inverse variance weighting. Separate models were constructed for 3 and 6 months of follow-up. For uncontrolled studies, the BP change pre vs. post RDN was pooled and analyzed using the same meta-analysis models. The magnitude of heterogeneity

5

ACCEPTED MANUSCRIPT

present was estimated using the I2 statistic, an estimate of the proportion of the total observed variance that is attributed to between-study variance. In order to compare the magnitude of blood pressure reduction based on RDN catheter used we constructed a separate meta-analysis

order to compare heterogeneity using the I2 statistic.

RI PT

stratified by catheter type using a random-effects generic inverse variance weighting model in

Certain studies reported measures of variability other than standard deviation (SD). In

SC

these cases, 95% confidence intervals or standard error of the mean were converted to standard deviation in order to maintain consistency of the reported results. In one study (21), the only

M AN U

measure of variability reported was interquartile range. By including this study in the metaanalysis models, we are assuming a normal distribution of change in BP. In another study (22), no estimate of variance was reported, thus we assumed a SD equal to the mean of other reported SD’s. Sensitivity analyses were performed excluding these two studies. P< 0.05 was considered

TE D

significant. To evaluate the interaction by type of catheter employed, we used the one-way ANOVA and Tukey’s multiple comparisons test. Throughout, values are presented as mean ± SD unless otherwise stated. Analyses were performed using The Cochrane Collaboration Review

RESULTS

EP

Manager 5.1.7 (20) and the GraphPad Prism 6.0 (23) software packages.

AC C

Study Selection and Characteristics. Our literature search identified a total of 294

potentially relevant studies as shown in the flow diagram (Figure 1). Of these, 18 studies met the inclusion criteria. Six additional studies were excluded due to overlap of patients (24-29). All six of these studies were controlled studies with overlap of patients with the Symplicity HTN-2 study and could not be included in the controlled study meta-analysis. On the other hand, the Ukena et al. study (30), which is an uncontrolled study, is included in the uncontrolled study

6

ACCEPTED MANUSCRIPT

meta-analysis even though 18 patients (out of 136) overlap with Symplicity HTN-2, since these two studies are included in separate meta-analyses. Thus, a total of 12 studies were included in our systematic review, encompassing 561 patients treated between 2008 and 2012. The 12-

RI PT

month follow-up data for a study was published in a separate article and the follow-up data was extracted (31,32). The follow-up duration varied between 1 and 24 months with a median followup of 6 months.

SC

Table 1 summarizes the design and methods of the included studies. There were 2

randomized controlled trials (N=133) and 1 observational study with a control group (N=50) (i.e.

M AN U

controlled studies), and 9 observational studies without a control group (N=396) (i.e. uncontrolled studies). The inclusion and exclusion criteria were in large part similar, although one study included patients at the lower end of the RH spectrum (33), and another study included only patients with moderate to severe chronic kidney disease (34). Ambulatory BP monitors

TE D

were employed to measure the primary outcome of BP response (and used for the meta-analysis) in 2 studies (22,35), with these measurements being relatively lower than office measurements, which were used as the primary outcome measure (and used for the meta-analysis) in the other

EP

10 studies (16,21,30,32-34,36-39). The risk of bias was low in the majority of studies, with a detailed assessment available in Appendix 1 and 2.

AC C

Tables 2 and 3 summarize the patient characteristics and concomitant therapies,

respectively. Sixty percent of patients were male and the average age was 60 years. Thirty five percent of patients had type 2 diabetes mellitus and 18% had coronary artery disease. Patients with RH were receiving an average of 5 different antihypertensive medications. Effectiveness of RDN. In controlled studies, there was a reduction in mean systolic and diastolic BP at 6 months of -28.9 (95% CI -37.2, -20.6) and -11.0 (95% CI -16.4, -5.7) mmHg,

7

ACCEPTED MANUSCRIPT

respectively, with RDN compared to medical therapy (for both, p<0.0001; Figure 2). At 3 months, there were reductions of -20.8 mmHg systolic (95% CI -26.4, -15.2) and -7.6 mmHg diastolic BP (95% CI -11.0, -4.2). At 12 months, there were reductions of -25.4 mmHg systolic

RI PT

(95% CI -27.8, -23.0) and -10.0 mmHg diastolic (95% CI -11.0, -9.0). There was a modest

amount of heterogeneity (I2 = 0 at 3 months and 50-60% at 6 months) between these studies. In uncontrolled studies, there was a reduction in mean systolic and diastolic BP at 6

SC

months of -25.0 (95% CI -29.9, -20.1) and -10.0 (95% CI -12.5, -7.5) mmHg, respectively, precompared to post-RDN (for both, p<0.00001; Figure 3). At 3 months, there were reductions of -

M AN U

22.8 mmHg systolic (95% CI -26.3, -18.5) and -9.1 mmHg diastolic BP (95% CI -12.1, -6.1). At 12 months, there were reductions of -22.8 mmHg systolic (95% CI -29.6, -16.0) and -10.6 mmHg diastolic BP (95% CI -15.0, -6.0). There was a significant amount of heterogeneity at the 3 month (I2 = 66% and 80% for sBP and dBP respectively) and 6 month (I2 = 70% and 50% for

TE D

sBP and dBP respectively) time points. However, sensitivity analysis excluding the Kaltenbach study (which included patients with lower baseline BP), the heterogeneity was no longer apparent (I2 = 0) whereas the BP change was maintained (-24/-10 and -27/-11 mmHg at 3 and 6

EP

months respectively). Only six studies (22,32-35,39) had adequate 6-month follow-up of ambulatory BP following RDN. The overall BP response was smaller than that seen in studies

AC C

solely evaluating office BP. For these six studies, there was a reduction in mean systolic and diastolic BP at 6 months of -13.2 (95% CI -19.4, -7.0) and -7.3 (95% CI -10.2, -4.5) mmHg, respectively, pre- compared to post-RDN (p<0.0001 for sBP and p<0.00001 for dBP). Statistical heterogeneity was present for the systolic BP response only (I2 = 76% for sBP and 0% for dBP). Catheter Comparison. A total of 5 different catheters were employed among the 12 studies: the SymplicityTM/Flex catheter (Medtronic) is an RF ablation catheter designed for

8

ACCEPTED MANUSCRIPT

RDN, the Celcius® ThermoCool® and Navistar® ThermoCool® (Biosense Webster) are irrigated RF ablation catheters, the Marinr® (Medtronic) is a standard steerable RF ablation catheter, and the PARADISETM (ReCor Medical) is an ultrasound ablation catheter. Given the

RI PT

overlap of 18 patients between two studies with separate study designs, we excluded the

Symplicity HTN-2 from the comparison of catheter meta-analysis and included the Ukena et al. study given the larger amount of included patients treated with RDN. With all 95% CI

SC

overlapping, there is no evidence of difference in the achieved BP response post RDN among the different catheters used (Figure 4).

M AN U

Procedural Safety. The pooled non-responder rate was 13.3%. The Kaltenbach study had the highest non-responder rate and excluding this study from the pooled estimate decreased the non-responder rate to 11.7%. No deaths were reported during the stipulated follow-up periods. A total of 5 procedural complications were reported (<1%). These included one renal

TE D

artery dissection (remote from where RDN was performed) (16) and 4 pseudoaneurysms at the site of arterial puncture (16,32,36). DISCUSSION

EP

We have conducted the first systematic review and meta-analysis of the published body of literature pertaining to RDN in patients with RH. Our study was designed to evaluate the

AC C

effect of RDN on BP reduction in a RH population. We found that there was a substantial reduction in BP after RDN at 6 months, which was apparent as early as 3 months and sustained up to 12 months. Importantly, the rate of procedural complications was quite low. We chose to pool the results of the studies based on the study design. Hence the two

randomized controlled trials were pooled with the only controlled cohort study, and the uncontrolled observational studies were pooled together. Observational studies tend to

9

ACCEPTED MANUSCRIPT

overestimate treatment effects by confounding by indication. In our analysis however, all studies have demonstrated a consistent BP reduction regardless of study design. The current study population is composed of only 561 patients, which is a small number

RI PT

for a meta-analysis. It represents, however, most of the published experience of RDN in patients with RH. In a meta-analysis, more important than the number of patients is the number of

included studies, especially when the outcome is continuous. As RDN is now used for clinical

SC

care in Europe, Canada, and parts of Asia, knowledge of the current extent of effectiveness of the procedure, and the source of the evidence, is of outmost important. Before large studies are

must be based on current data.

M AN U

completed and reported, the quality of the evidence and effectiveness and safety of the procedure

The short-to-intermediate term data suggests that RDN is safe and well tolerated, with the most common peri-procedural complaint being abdominal pain that responds to sedatives and/or

TE D

narcotics (16,40). Serial vascular imaging follow-up up to 6 months in Symplicity HTN-1 did not reveal renal artery abnormalities secondary to RDN (based on renal duplex, MRI or CT angiography) (36). Serial biochemical follow-up did not demonstrate deterioration in renal

EP

function (32,36). The only complications identified in our review were pseudoaneurysms at the vascular access site in 4 subjects, and one case of renal artery dissection on initial placement of

AC C

the catheter prior to delivery of the radiofrequency signal. Hence, the renal artery dissection was the only non-access-related complication reported. Among the 12 studies included in the metaanalysis, 8 studies presented data on follow-up imaging of the renal arteries via ultrasound duplex, CT angiography or MRI angiography (16,21,22,32,33,35,36,38). Of the 191 patients with follow-up imaging, there were no documented renal artery stenosis, and only 2 patients had progression of previously visualized renal artery atherosclerosis. To date, two case reports

10

ACCEPTED MANUSCRIPT

(41,42) have been published of individual patients whose blood pressure initially responded to RDN but subsequently had increase in blood pressure on follow-up visits. In both instances, renal doppler and angiography demonstrated renal artery stenosis, which were treated by renal

RI PT

artery stenting. It is still unclear what proportion of patients develops reno-vascular

abnormalities post RDN. Further studies are needed to evaluate long term changes in renal artery anatomy post RDN as well as determine the appropriate imaging follow-up.

SC

The average non-responder rate, defined as a reduction in BP of <10 mmHg after RDN, was 13%. Careful examination of the studies shows that patients who had a lower baseline

M AN U

systolic blood pressure (<150mmHg) were less likely to respond to RDN. Kaltenbach’s study included patients with a systolic blood pressure between 140-160 mmHg and had the highest non-responder rate of nearly 50%. Therefore, it appears that patients with severe elevations of BP may derive the greatest relative benefit from RDN. It remains unclear if the response persists

TE D

in this patient population at long term follow-up or if a rebound phenomenon may occur. In addition to the primary BP lowering effect, secondary effects of RDN have been documented. Brant et al. showed that RDN led to a decrease in left ventricular hypertrophy,

EP

decrease in end systolic volume and increase in left ventricular ejection fraction (24). Mahfoud et al. showed that RDN led to a significant reduction in fasting blood glucose as well as a reduction

AC C

in insulin and c-peptide levels (26). Additionally, Mahfoud et al. showed that RDN reduced albuminuria (25). A recent double blind study showed that RDN prevented recurrence of atrial fibrillation post pulmonary vein isolation (38). Another study showed RDN led to a reduction in heart rate and prolongation of the PR interval (30). We did not identify any studies directly comparing the effectiveness of different catheters. Based on the studies included in this meta-analysis, the magnitude of the achieved BP

11

ACCEPTED MANUSCRIPT

reduction was consistent with all of the different catheters employed. Although RDN-specific catheters with multiple electrodes are being evaluated, it remains to be seen whether they are more effective at lowering BP.

RI PT

Current Studies. A number of RDN trials are currently underway. Symplicity-HTN 3 (clinicaltrials.gov NCT01418261) is a single-blind clinical trial that is randomizing 530 patients in a 2:1 fashion to RDN or control with a follow-up period of 6 months (43). The INSPiRED

SC

study (clinicaltrials.gov NCT01505010) is a randomized clinical trial with a longer follow-up period of 36 months. The DEPART study (clinicaltrials.gov NCT01522430) is a double-blind

M AN U

clinical trial that is randomizing patients to RDN or sham procedure with a follow-up period of 6 months. The SymplicityHF study (clinicaltrials.gov NCT01392196) is evaluating the effect of RDN in patients with heart failure, and contrary to the aforementioned studies that will be focussing on BP response, this study will focus on procedural safety and renal and cardiac

TE D

function. The DREAMS study (clinicaltrials.gov NCT01465724) is evaluating the effect of RDN in patients with metabolic syndrome, with the primary outcome being change in insulin resistance parameters after 12 months of follow-up.

EP

Study Limitations. Our study has several potential limitations. First, the majority of included studies were observational in nature and thus may be affected by confounding by

AC C

indication and/or selection bias. While the BP reduction was modestly greater in the observational controlled study than in either of the randomized controlled studies, large clinically important treatment effects were reported in all controlled studies. Second, with the use of published aggregate data, we were unable to examine the effect of RDN in patient subgroups. Third, inclusion was restricted to published studies and may therefore be affected by publication bias. Fourth, the follow-up rate was quite limited in many of the included studies, resulting in

12

ACCEPTED MANUSCRIPT

less than 70% 6-month follow-up. Fifth, while we conducted secondary analyses that were stratified by catheter type, there were 5 different catheters used in the 12 included studies and studies of catheters other than the Symplicity catheter were of modest size. Consequently, there

RI PT

were insufficient data to draw meaningful conclusions regarding the comparative efficacy of the different catheters. Lastly, given the recent evolution of RDN, the pooled person-time remains modest.

SC

In summary, the current available data suggest that RDN results in a substantial BP reduction at 6 months follow-up in patients with RH. With few adverse events reported,

M AN U

available data also suggest that RDN has a favourable safety profile. Nonetheless, large randomized controlled trials with long-term follow-up are needed to confirm the sustained

AC C

EP

TE D

efficacy and safety of RDN in this patient population.

13

ACCEPTED MANUSCRIPT

REFERENCES 1.

Persell SD. Prevalence of Resistant Hypertension in the United States, 2003–2008. Hypertension 2011;57:1076-1080. de la Sierra A, Segura J, Banegas JR et al. Clinical Features of 8295 Patients With

RI PT

2.

Resistant Hypertension Classified on the Basis of Ambulatory Blood Pressure Monitoring. Hypertension 2011;57:898-902.

Roberie DR, Elliott WJ. What is the prevalence of resistant hypertension in the United

SC

3.

States? Current Opinion in Cardiology 2012;27:386-391

Sarafidis PA, Bakris GL. Resistant Hypertension: An Overview of Evaluation and

M AN U

4.

Treatment. J Am Coll Cardiol 2008;52:1749-1757. 5.

Sander GE, Giles TD. Resistant hypertension: concepts and approach to management. Curr Hypertens Rep 2011;13:347-55.

Calhoun DA, Jones D, Textor S et al. Resistant Hypertension: Diagnosis, Evaluation, and

TE D

6.

Treatment. Hypertension 2008;51:1403-19. 7.

Evelyn KA, Singh MM, Chapman WP, Perera GA, Thaler H. Effect of thoracolumbar

EP

sympathectomy on the clinical course of primary (essential) hypertension: A ten-year study of 100 sympathectomized patients compared with individually matched,

AC C

symptomatically treated control subjects. The American journal of medicine 1960;28:188-221.

8.

Hoobler SW, Manning JT, Paine WG et al. The Effects of Splanchnicectomy on the Blood Pressure in Hypertension. Circulation 1951;4:173-183.

9.

Smithwick Rh TJE. Splanchnicectomy for essential hypertension: Results in 1,266 cases. JAMA: The Journal of the American Medical Association 1953;152:1501-1504.

14

ACCEPTED MANUSCRIPT

10.

DiBona GF. Sympathetic nervous system and the kidney in hypertension. Curr Opin Nephrol Hypertens 2002;11:197-200.

11.

DiBona GF, Kopp UC. Neural control of renal function. Physiological Reviews

12.

RI PT

1997;77:75-197.

Joles JA, Koomans HA. Causes and Consequences of Increased Sympathetic Activity in Renal Disease. Hypertension 2004;43:699-706.

Recordati G, Zorzoli F, Zanchetti A. Effects of a reversible 'nephrectomy' on renal

SC

13.

sympathetic activity and blood pressure in the rat: evidence for an acute angiotensin-

14.

M AN U

mediated hypertension. J Hypertens 2000;18:1277-1287.

ID D, Bertog SC, Wunderlich N, Sievert H. Catheter-based renal sympathectomy. J Cardiovasc Surg 2010;51:721-739.

15.

Rippy MK, Zarins D, Barman NC, Wu A, Duncan KL, Zarins CK. Catheter-based renal

TE D

sympathetic denervation: chronic preclinical evidence for renal artery safety. Clin Res Cardiol 2011;100:1095-101. 16.

Krum H, Schlaich M, Whitbourn R et al. Catheter-based renal sympathetic denervation

EP

for resistant hypertension: a multicentre safety and proof-of-principle cohort study. The Lancet 2009;373:1275-1281.

Medtronic I. Symplicity™ RDN System Clinical Trial Data. Mountain View, California:

AC C

17.

Medtronic, Inc, 2012.

18.

Liberati A, Altman DG, Tetzlaff J et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. PLoS Med 2009;6:e1000100.

15

ACCEPTED MANUSCRIPT

19.

Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med 2009;6:e1000097. Review Manager (RevMan). 5.1 ed. Copenhagen: The Cochrane Collaboration, 2011.

21.

Witkowski A, Prejbisz A, Florczak E et al. Effects of renal sympathetic denervation on

RI PT

20.

blood pressure, sleep apnea course, and glycemic control in patients with resistant hypertension and sleep apnea. Hypertension 2011;58:559-65.

Prochnau D, Figulla HR, Surber R. Efficacy of renal denervation with a standard EP

SC

22.

Cardiol 2012;157:447-8.

M AN U

catheter in the 24-h ambulatory blood pressure monitoring-long-term follow-up. Int J

23.

GraphPad Prism. 6.01 ed. La Jolla, California: GraphPad Software, 2012.

24.

Brandt MC, Mahfoud F, Reda S et al. Renal sympathetic denervation reduces left ventricular hypertrophy and improves cardiac function in patients with resistant

25.

TE D

hypertension. J Am Coll Cardiol 2012;59:901-9.

Mahfoud F, Cremers B, Janker J et al. Renal hemodynamics and renal function after catheter-based renal sympathetic denervation in patients with resistant hypertension.

26.

EP

Hypertension 2012;60:419-24.

Mahfoud F, Schlaich M, Kindermann I et al. Effect of renal sympathetic denervation on

AC C

glucose metabolism in patients with resistant hypertension: a pilot study. Circulation 2011;123:1940-6.

27.

Ukena C, Mahfoud F, Kindermann I et al. Cardiorespiratory Response to Exercise After Renal Sympathetic Denervation in Patients With Resistant Hypertension. J Am Coll Cardiol 2011;58:1176-1182.

16

ACCEPTED MANUSCRIPT

28.

Prochnau D, Lucas N, Kuehnert H, Figulla HR, Surber R. Catheter-based renal denervation for drug-resistant hypertension by using a standard electrophysiology catheter. EuroIntervention : journal of EuroPCR in collaboration with the Working Group

29.

RI PT

on Interventional Cardiology of the European Society of Cardiology 2012;7:1077-80. Brandt MC, Reda S, Mahfoud F, Lenski M, Bohm M, Hoppe UC. Effects of renal

sympathetic denervation on arterial stiffness and central hemodynamics in patients with

30.

SC

resistant hypertension. J Am Coll Cardiol 2012;60:1956-65.

Ukena C, Mahfoud F, Spies A et al. Effects of renal sympathetic denervation on heart

M AN U

rate and atrioventricular conduction in patients with resistant hypertension. Int J Cardiol 2012. 31.

Esler MD, Krum H, Schlaich M, Schmieder RE, Bohm M, Sobotka PA. Renal Sympathetic Denervation for Treatment of Drug-Resistant Hypertension: One-Year

TE D

Results From the Symplicity HTN-2 Randomized, Controlled Trial. Circulation 2012;126:2976-2982. 32.

Symplicity HTN-2 Investigators. Renal sympathetic denervation in patients with

EP

treatment-resistant hypertension (The Symplicity HTN-2 Trial): a randomised controlled trial. The Lancet 2010;376:1903-1909. Kaltenbach B, Franke J, Bertog SC, Steinberg DH, Hofmann I, Sievert H. Renal

AC C

33.

sympathetic denervation as second-line therapy in mild resistant hypertension: A pilot study. Catheter Cardiovasc Interv 2013;81:335-339.

34.

Hering D, Mahfoud F, Walton AS et al. Renal Denervation in Moderate to Severe CKD. J Am Soc Nephrol 2012;23:1250-7.

17

ACCEPTED MANUSCRIPT

35.

Ahmed H, Neuzil P, Skoda J et al. Renal Sympathetic Denervation Using an Irrigated Radiofrequency Ablation Catheter for the Management of Drug-Resistant Hypertension. JACC: Cardiovascular Interventions 2012;5:758-765. Symplicity HTN-1 Investigators. Catheter-Based Renal Sympathetic Denervation for Resistant Hypertension. Hypertension 2011;57:911-917.

37.

RI PT

36.

Mabin T, Sapoval M, Cabane V, Stemmett J, Iyer M. First experience with endovascular

SC

ultrasound renal denervation for the treatment of resistant hypertension. EuroIntervention 2012;8:57-61.

Pokushalov E, Romanov A, Corbucci G et al. A Randomized Comparison of Pulmonary

M AN U

38.

Vein Isolation With Versus Without Concomitant Renal Artery Denervation in Patients With Refractory Symptomatic Atrial Fibrillation and Resistant Hypertension. J Am Coll Cardiol 2012;60:1163-70.

Zuern CS, Rizas KD, Eick C et al. Effects of Renal Sympathetic Denervation on 24-hour

TE D

39.

Blood Pressure Variability. Front Physiol 2012;3:134. 40.

Sapoval M, Azizi M, Bobrie G, Cholley B, Pagny JY, Plouin PF. Endovascular renal

41.

EP

artery denervation: why, when, and how? Cardiovasc Intervent Radiol 2012;35:463-71. Kaltenbach B, Id D, Franke JC et al. Renal Artery Stenosis After Renal Sympathetic

42.

AC C

Denervation. J Am Coll Cardiol 2012;60:2694-2695.

Vonend O, Antoch G, Rump LC, Blondin D. Secondary rise in blood pressure after renal denervation. The Lancet;380:778.

43.

Kandzari DE, Bhatt DL, Sobotka PA et al. Catheter-Based Renal Denervation for Resistant Hypertension: Rationale and Design of the SYMPLICITY HTN-3 Trial. Clinical Cardiology 2012;35:528-35.

18

ACCEPTED MANUSCRIPT

FIGURE LEGENDS Figure 1: Flow diagram for study selection – Studies were selected by a three stage exclusion process. Studies were first screened based on abstract. Following selection of appropriate studies,

were excluded due to an overlap in patient populations.

RI PT

studies were excluded based on amount of participants and follow-up. Finally 6 further studies

Figure 2: Meta-analysis of controlled studies - Forest plot demonstrating the changes in

SC

systolic and diastolic blood pressures stratified by follow-up time amongst controlled studies post-RDN.

M AN U

Figure 3: Meta-analysis of uncontrolled studies - Forest plot demonstrating the changes in systolic and diastolic blood pressures stratified by follow-up time amongst uncontrolled studies post-RDN.

Figure 4: Catheters used for RDN - Forest plot stratified by catheter type demonstrating the

TE D

change in systolic blood pressure at 3 months follow-up post RDN. Testing for heterogeneity between catheter and blood pressure lowering effect is represented at the bottom of the figure using the I2 statistic. Three month follow-up time was used for comparison of catheters as all

EP

catheters employed had this minimal length of follow-up. Note, Symplicity HTN-2 is not included in this analysis as 18 patients overlapped with the Ukena et al. study, and this study had

AC C

a greater number of patients treated with RDN.

19

ACCEPTED MANUSCRIPT

Table 1: Characteristics and inclusion criteria of all included studies Year

Type of

Method of

Inclusion / Exclusion

Type of

Study

BP

Criteria

Catheter

measurement

M AN U

Randomized Control Trials Pokushalov 2012

Randomized Office BP in

Inclusion: Symptomatic

et al.(38)

Control

drug refractory AF on ≥ 2 ThermoCool®

Navistar®

antiarrhythmics;

catheter

Paroxysmal AF;

(Biosense

Standard criteria* but no

Webster)

EP

TE D

triplicate

patients control

Length

Non-

of

Responder

patients follow-

Rate (%)

up (months)

13

14

3, 6, 9,

0

12

exception to valvular

AC C

Trial

#

SC

for analysis

# RDN

RI PT

Study

disease, ICD/pacemaker, or medication changes; Exclusion: no NYHA

20

ACCEPTED MANUSCRIPT

class III or IV CHF; no

AF ablation; no treatment

Symplicity

2010, Randomized Office BP in

HTN-

2012

Trial

Prospective

Office BP in

Cohort

triplicate

Observational Studies

al.(35)

2012

Prospective Cohort

24 hour

AC C

Ahmed et

Standard criteria*

TE D

al.(16)

54

1, 3, 6,

16

12

Medtronic)

EP

2009

52

(Ardian,

triplicate

Observational Control Study Krum et

Symplicity™

M AN U

2(31,32) †,‡

Control

Standard criteria*

SC

with amiodarone

RI PT

LVEF <35%; no previous

AMBP

Symplicity™

45

5

(Ardian,

1, 3, 6, 9, 13 12

Medtronic)

Standard criteria* except

Celcius®

sBP ≥140 and no

ThermoCool®

exception to valvular

catheter

disease, ICD/pacemaker

(Biosense

10

N/A

1, 3, 6

0

21

ACCEPTED MANUSCRIPT

2012

al.(34)

Webster)

Prospective

Office BP in

Inclusion: patients with

Symplicity™

Cohort

triplicate

stage 3-4 chronic kidney

(Ardian,

disease and resistant

Medtronic)

2012

et al.(33)

N/A

Symplicity™

Prospective

Office BP in

Inclusion: SBP 140-160;

Cohort

triplicate

≥ 3 antihypertensives;

(Ardian,

Exclusion: no secondary

Medtronic)

1, 3, 6,

NR

12

20

N/A

3, 6

45

11

N/A

0.5, 1, 2,

NR

M AN U

Kaltenbach

SC

HTN

15

RI PT

Hering et

or medication changes;

causes of HTN 2012

Office BP in

Cohort

triplicate

Standard criteria* but BP

of valvular disease,

catheter

ICD/pacemaker, or

(ReCor

medication changes

medical)

et al.(22)

2012

Prospective

24-hour

Inclusion: mean 24 hour

Marinr

Cohort

ABPM

ambulatory sBP ≥150, ≥

®standard

AC C

Prochnau

PARADISE

≥140/90 and no exclusion ™ Ultrasound

EP

al.(37)

Prospective

TE D

Mabin et

3

30

N/A

1, 3, 6,

NR (25 in

12

initial

22

ACCEPTED MANUSCRIPT

HTN-1(36)

Ukena et

2012

al.(30) ‡

ablation

causes of HTN; no

catheter

known RVA

(Medtronic)

study)

RI PT

Exclusion: no secondary

Prospective

Office BP in

Standard criteria* except

Symplicity™

Cohort

triplicate

no exception to

(Ardian,

SC

2011

steerable RF

M AN U

Symplicity

4 antihypertensives;

medication changes

Medtronic)

Prospective

Office BP in

Standard criteria* except

Flex

Cohort

triplicate

no medication changes

(Medtronic)

153

N/A

1, 3, 6,

8

12, 18, 24 136

N/A

3, 6

17

10

N/A

3, 6

0

TE D

within 2 weeks of study enrollment

Prospective

Office BP in

Cohort

triplicate

Standard criteria* except

Symplicity™

no medication changes

(Ardian,

within 2 weeks of study

Medtronic)

EP

et al.(21)

2011

AC C

Witkowski

enrollment; diagnosed sleep apnea

23

ACCEPTED MANUSCRIPT

al.(39)

2012

Prospective

Office BP in

Standard criteria* except

Flex (Ardian,

Cohort

triplicate

no medication changes

Medtronic)

within 2 weeks of study

N/A

6

18

SC

enrollment

11

RI PT

Zuern et

M AN U

* Standard criteria: these are the initial inclusion and exclusion criteria of the original RDN trial by Krum = age ≥ 18; sBP ≥160 (≥150 if type 2 diabetes mellitus); ≥ 3 antihypertensives (including a diuretic); eGFR> 45 ml/min/1.73 m2; no secondary causes of HTN; no type 1 diabetes mellitus; no known renovascular abnormalities; no valvular disease; no ICD/pacemaker; not pregnant; no medication changes within 3 months of study enrollment

TE D

† Includes the 12 month follow-up data of the same clinical cohort

EP

‡ There exists an overlap of 18 patients between these studies

AC C

ABPM = ambulatory blood pressure monitoring; BP = blood pressure; HTN = hypertension; eGFR = estimated glomerular filtration rate; sBP = systolic blood pressure; NYHA = New York heart association; CHF = congestive heart failure; AF = atrial fibrillation, ICD = implantable cardiac defibrillator

24

ACCEPTED MANUSCRIPT

Table 2: Baseline characteristics of the included study subjects Treatment

Age

Group

Male Heart

GFR

(%)

(ml/min (kg/m2) T2DM CAD (%)

Rate

BMI

(BPM) per 1.73

(%)

79

NR

78±6

et al.(38)

56±9

71

NR

80.2±5

Medical

RDN

58±12 65

75±15

HTN-2(32)

Medical

58±12 50

71±15

*

Therapy

Krum et

RDN

58±9

15

23

181±7 /

28±5

14

14

21

97±6 178±8 / 96±4

31±5

40

19

52

178±18 /

86±20

31±5

28

7

52

97±16 178±16 /

AC C

Observational Control Study

8

77±19

EP

Symplicity

56

72±11

(mmHg)

28±6

TE D

Therapy

M AN U

Randomized Control Trials 57±8

(%)

BP

SC

m2)

Pokushalov RDN

Dyslipidemia Baseline

RI PT

Study

98±17

81±23

NR

31

22

64

177±20 /

25

ACCEPTED MANUSCRIPT

al.(16)

Medical

51±8

80

79±9

95±15

NR

40

20

100

Therapy

61±12 80

NR

NR

33±5

Hering et

RDN

61±9

60

64±9

31±9

al.(34) RDN

61±11 55

NR

et al.(33) RDN

55±14 36

al.(37) RDN

et al.(22) Symplicity HTN-1(36)

RDN

62±13 67

NR

AC C

Prochnau

NR

57±11 61

NR

EP

Mabin et

77±25

73±13

NR

NR

TE D

Kaltenbach

M AN U

al.(35)

NR

30

NR

NR

20

SC

RDN

RI PT

173±8 /

Observational Studies Ahmed et

101±15

73

50

NR

60

98±9

158±16 / 88±15

NR

174±22 / 91±16

50

NR

148±7 / 83±11

27

36

64

180±20 / 109±13

50

20

NR

166±22 / 88±13

83±20

NR

31

22

68

176±17 / 98±15

26

ACCEPTED MANUSCRIPT

Ukena et

RDN

62±9

58

66±12

NR

31±5

39

10

NR

al.(30) *

93±15 RDN

50±12 70

NR

81±11

32±7

40

et al.(21) RDN

69±7

73

NR

75±18

29±3

36

NR

36

174±9 / 103±12 189±23 / 92±15

M AN U

al.(39)

36

NR

SC

Zuern et

RI PT

Witkowski

177±21 /

Data is presented as mean ±SD. NR = Not Reported; RDN = renal sympathetic denervation; BPM = beats per minute; GFR = glomerular filtration rate; BMI = body mass index; T2DM = type 2 diabetes mellitus; CAD = coronary artery disease.

AC C

EP

TE D

* There exists an overlap of 18 patients between these studies

27

ACCEPTED MANUSCRIPT

Table 3: Baseline antihypertensive usage of the included studies Treatment

# of Anti-

ACEI

Direct

Beta

CCB

Group

hypertensives / ARB

Renin

Blocker

(%)

Inhibitor

(%)

(%)

Pokushalov RDN

3.8±0.4

92

NR

et al.(38)

3.6±0.6

100

NR

M AN U

Randomized Control Trials

Medical

96

HTN-

Medical

5.3±1.8

94

2(32)*

Therapy

Blocker

(%)

(%)

76

100

NR

NR

NR

78

71

92

NR

NR

NR

15

83

79

89

52

15

33

19

69

83

91

52

17

19

TE D

5.2±1.5

Sympatholytic (%)

EP

RDN

(%)

77

Therapy Symplicity

Vasodilator Alpha 1

SC

(%)

Diuretic Central

RI PT

Study

AC C

Observational Control Study Krum et

RDN

4.7±1.5

96

NR

76

69

96

NR

18

NR

al.(16)

Medical

4.6±0.5

80

NR

100

100

60

NR

0

NR

Therapy

28

ACCEPTED MANUSCRIPT

RDN

6.7±1

100

NR

80

90

RDN

5.6±1.3

100

29

73

80

RDN

5.4±1.5

100

45

90

RDN

4.5

82-100

NR

RDN

6

RDN

5.1±1.4

RDN

5.5±1.2

RDN

5.0±0.9

al.(35) Hering et

Kaltenbach

Ukena et al.(30)* Witkowski

47

27

27

80

85

60

15

40

NR

9

27

87

77

97

73

33

43

14

82

75

95

33

19

19

86

NR

88

72

100

46

NR

NR

100

NR

100

80

100

20

NR

20

EP

91

TE D

97

AC C

HTN-1(36)

100

100

et al.(22) Symplicity

NR

91

al.(37) Prochnau

NR

36

et al.(33) Mabin et

100

M AN U

al.(34)

100

SC

Ahmed et

RI PT

Observational Studies

29

ACCEPTED MANUSCRIPT

et al.(21) RDN

5.6±2.1

100

55

73

82

100

73

18

36

RI PT

Zuern et al.(3939)

Data is presented as mean ±SD. NR = Not Reported. For the Prochnau et al. study, they grouped patients with ACEI, ARB and direct

AC C

EP

TE D

M AN U

SC

renin inhibitor together in their study. * There exists an overlap of 18 patients between these studies.

30

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT