Author + information
- Received April 14, 2014
- Revision received May 10, 2014
- Accepted May 13, 2014
- Published online September 16, 2014.
- George L. Bakris, MD∗∗ (, )
- Raymond R. Townsend, MD†,
- Minglei Liu, PhD‡,
- Sidney A. Cohen, MD, PhD†,‡,
- Ralph D’Agostino, PhD§,
- John M. Flack, MD, MPH‖,
- David E. Kandzari, MD¶,
- Barry T. Katzen, MD#,
- Martin B. Leon, MD∗∗,
- Laura Mauri, MD, MSc††,
- Manuela Negoita, MD‡,
- William W. O’Neill, MD‡‡,
- Suzanne Oparil, MD§§,
- Krishna Rocha-Singh, MD‖‖,
- Deepak L. Bhatt, MD, MPH¶¶,
- SYMPLICITY HTN-3 Investigators
- ∗ASH Comprehensive Hypertension Center, University of Chicago Medicine, Chicago, Illinois
- †School of Medicine, Renal Electrolyte Division, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania
- ‡Medtronic CardioVascular, Santa Rosa, California
- §Department of Medicine, Division of Cardiology, Harvard Clinical Research Institute and Boston University School of Public Health, Boston, Massachusetts
- ‖Department of Medicine, Wayne State University and Detroit Medical Center, Detroit, Michigan
- ¶Department of Medicine, Division of Cardiology, Piedmont Heart Institute, Atlanta, Georgia
- #Department of Medicine, Division of Cardiology, Baptist Cardiac and Vascular Institute, Miami, Florida
- ∗∗Department of Medicine, Division of Cardiology, New York Presbyterian Hospital, Columbia University Medical Center, and Cardiovascular Research Foundation, New York, New York
- ††Department of Medicine, Division of Cardiology, Harvard Clinical Research Institute, Brigham and Women’s Hospital Heart and Vascular Center, and Harvard Medical School, Boston, Massachusetts
- ‡‡Department of Medicine, Division of Cardiology, Henry Ford Hospital, Detroit, Michigan
- §§Department of Medicine, Division of Cardiology, University of Alabama at Birmingham, Birmingham, Alabama
- ‖‖Department of Medicine, Division of Cardiology, Prairie Heart Institute, Springfield, Illinois
- ¶¶Department of Medicine, Division of Cardiology, Brigham and Women’s Hospital Heart and Vascular Center and Harvard Medical School, Boston, Massachusetts
- ↵∗Reprint requests and correspondence:
Dr. George L. Bakris, Director, ASH Comprehensive Hypertension Center, University of Chicago Medicine, 5841 South Maryland Avenue, MC 1027, Chicago, Illinois 60637.
Background Prior studies of catheter-based renal artery denervation have not systematically performed ambulatory blood pressure monitoring (ABPM) to assess the efficacy of the procedure.
Objectives SYMPLICITY HTN-3 (Renal Denervation in Patients With Uncontrolled Hypertension) was a prospective, blinded, randomized, sham-controlled trial. The current analysis details the effect of renal denervation or a sham procedure on ABPM measurements 6 months post-randomization.
Methods Patients with resistant hypertension were randomized 2:1 to renal denervation or sham control. Patients were on a stable antihypertensive regimen including maximally tolerated doses of at least 3 drugs including a diuretic before randomization. The powered secondary efficacy endpoint was a change in mean 24-h ambulatory systolic blood pressure (SBP). Nondipper to dipper (nighttime blood pressure [BP] 10% to 20% lower than daytime BP) conversion was calculated at 6 months.
Results The 24-h ambulatory SBP changed −6.8 ± 15.1 mm Hg in the denervation group and −4.8 ± 17.3 mm Hg in the sham group: difference of −2.0 mm Hg (95% confidence interval [CI]: −5.0 to 1.1; p = 0.98 with a 2 mm Hg superiority margin). The daytime ambulatory SBP change difference between groups was −1.1 (95% CI: −4.3 to 2.2; p = 0.52). The nocturnal ambulatory SBP change difference between groups was −3.3 (95 CI: −6.7 to 0.1; p = 0.06). The percent of nondippers converted to dippers was 21.2% in the denervation group and 15.0% in the sham group (95% CI: −3.8% to 16.2%; p = 0.30). Change in 24-h heart rate was −1.4 ± 7.4 in the denervation group and −1.3 ± 7.3 in the sham group; (95% CI: −1.5 to 1.4; p = 0.94).
Conclusions This trial did not demonstrate a benefit of renal artery denervation on reduction in ambulatory BP in either the 24-h or day and night periods compared with sham (Renal Denervation in Patients With Uncontrolled Hypertension [SYMPLICITY HTN-3]; NCT01418261)
Ambulatory blood pressure monitoring (ABPM) is recommended for initial evaluation of all persons with newly diagnosed hypertension and for evaluation of antihypertensive drug efficacy in blood pressure (BP) trials (1,2). Its importance is exemplified by positive outcomes on ABPM in a randomized trial in which office readings were not different (3). Early studies of renal denervation in participants with resistant hypertension did not assess ABPM in all participants studied (4,5). SYMPLICITY HTN-3 (Renal Denervation in Patients With Uncontrolled Hypertension) was a prospective, blinded, randomized, sham-controlled trial. In this trial, ABPM was an inclusion criterion, and all participants were evaluated at baseline and at 6 months; the change in ambulatory systolic blood pressure (SBP) from baseline to 6 months was a pre-specified, powered secondary endpoint (6,7). The primary results of the trial demonstrated the safety of the renal denervation procedure but failed to show a greater reduction in office or ambulatory SBP compared with the sham procedure at 6 months (7). This paper presents the detailed 24-h ABPM results of SYMPLICITY HTN-3.
The design of SYMPLICITY HTN-3 was published previously (6). Briefly, patients 18 to 80 years old with resistant hypertension were randomized 2:1 to either renal artery denervation or a sham procedure and were followed for 6 months. All patients provided signed informed consent.
Patients on a stable antihypertensive drug regimen including maximally tolerated doses of ≥3 antihypertensive medications of complementary classes, including an appropriately dosed diuretic, were required to have seated office SBP ≥160 mm Hg (using the average of 3 measurements) at their first screening visit. For the 2 weeks before the second screening visit, patients recorded their home BP and kept a diary recording their adherence to medical therapy. Antihypertensive medication changes were not allowed during this 2-week period. At the second screening visit, the office SBP ≥160 mm Hg was confirmed, adherence with medications was documented, and 24-h ABPM was performed to ensure an ambulatory SBP ≥135 mm Hg. All ABPMs were performed with the Space Labs 24 Ambulatory Blood Pressure Monitoring System (Space Labs Medical, Issaquah, Washington). Patients were instructed to place the cuff on the same arm as used for office BP measurements, and the importance of leaving the cuff in place was stressed. The ABPM parameters were preset to measure BP every 30 min during the day (7:00 am to 9:59 pm) and for every 30 min at night (10:00 pm to 6:59 am). Patients were instructed to engage in their normal daily activities and to hold their arm still by their side to avoid strenuous exercise during the device readings. All patients kept a diary that documented time to bed and time waking up, medications taken, and any other events of significance. On return of the ABPM machine, the patients’ diaries were collected, and the ABPM data were downloaded to a computer. A 24-h ABPM was considered adequate if the number of successful daytime readings captured was at least 21 and the number of successful nighttime readings captured was at least 12. Only ABPMs with the pre-specified number of readings per protocol were included in the analysis.
Additional clinical exclusion criteria included known secondary causes of hypertension or more than 1 hospitalization for a hypertensive emergency in the past year. Anatomic exclusion criteria included >50% renal artery stenosis, renal artery aneurysm, prior renal artery intervention, multiple renal arteries, renal artery diameter <4 mm, or treatable segment <20 mm in length.
After undergoing a renal angiogram and an assessment of anatomy, patients were randomized to either renal artery denervation or the sham control group in a 2:1 ratio. Patients were followed for 6 months, at which time the primary and pre-specified powered ambulatory SBP secondary endpoints were assessed. Patients in the control group were then allowed to cross over to receive renal denervation treatment, if they still met the inclusion criteria for the study.
The renal denervation procedure used radiofrequency energy delivered by the Symplicity Renal Denervation System (Medtronic, Santa Rosa, California) to ablate the nerves within the main renal arteries. Patients were blinded to whether they received renal artery denervation or only renal angiography (sham). Assessors of BP and study personnel also were blinded to the treatment received by patients. Details of the blinding procedure and confirmation of adequate blinding have been provided previously (7). Antihypertensive medication changes were not allowed during the 6-month follow-up period unless clinically required.
The primary efficacy endpoint was a comparison of office SBP change from baseline to 6 months in the renal denervation group compared with the SBP change from baseline to 6 months in the sham control group and required a superiority margin of 5 mm Hg for success. The secondary efficacy endpoint was the change in mean 24-h ambulatory SBP at 6 months. In addition to 24-h ABPM, daytime and nighttime ambulatory BP differences from baseline to 6 months, as well as differences in the change between the 2 groups, were assessed. ABPM differences in BP and heart rate variability also were assessed at baseline and at 6 months for both groups.
The proportion of patients who were extreme dippers (decline of >20% in night to day BP), dippers (10% to 20% decline in night to day BP), nondippers (<10% decline in BP at night), and reverse dippers (night BP greater than day BP) were calculated at each follow-up visit.
For the current analysis, patients in each group were further categorized according to tertiles of baseline ambulatory SBP. Baseline and 6-month ABPM measurements were determined for each group.
The primary safety endpoint has been previously described and reported (7). Treated patients will be followed biannually, and non-crossover control patients will be followed annually through 5 years post-randomization.
The analyses were performed based on the intent-to-treat principle. The data were collected and analyzed by the sponsor (Medtronic, Minneapolis, Minnesota) and independently validated by Harvard Clinical Research Institute (Boston, Massachusetts). Means and standard deviations of continuous variables were presented by treatment group. Variability of ambulatory BP in each visit was defined as the standard deviation (SD) or coefficient of variation (%, 100 × SD / mean). Between group differences were compared using confidence intervals (CI) and evaluated using unpaired Student t tests. Within group differences from baseline to follow-up were evaluated using paired Student t tests. For categorical variables, the treatment group presented the counts and percentages. They were tested using the exact test for binary variables and chi-square test for multi-level categorical variables. All subgroups shown were pre-specified.
Differences within and between groups of early morning slope analysis also were calculated. Baseline ambulatory SBP was plotted against early morning time (3 am to 8 am), and the slopes of the regression lines were calculated for each group at baseline and at 6 months. The changes in slopes from baseline to 6 months for each group were compared using analysis of covariance.
All randomized patients (n = 535) had an ABPM placed as part of the inclusion criteria for the study (7). No significant differences in baseline characteristics were noted between the 2 groups with the exception of 24-h and daytime diastolic blood pressure (DBP), which was greater in the sham control group (Table 1). Patients were taking an average of 5 antihypertensive medications, and on average 4 medications were at maximally tolerated doses (7). Antihypertensive medication use remained similar in both groups at 6-month follow-up.
The primary efficacy endpoint was the mean change in office SBP from baseline to 6 months in the denervation group, as compared with the mean change in the sham control group, with a superiority margin of 5 mm Hg. This endpoint was not different from the sham control group (7). The powered secondary efficacy endpoint was the change in mean 24-h ambulatory SBP at 6 months between groups with a superiority margin of 2 mm Hg. This endpoint was also not different between groups (−6.75 ± 15.11 mm Hg in the denervation group and −4.79 ± 17.25 mm Hg in the sham group, for a difference between groups of −1.96 mm Hg [95% CI: −4.97 to 1.06; p value with a 2 mm Hg superiority margin = 0.98]) (Central Illustration). The ambulatory DBPs were consistent with the trends in ambulatory SBPs (Central Illustration). The mean daytime and nighttime BPs in both groups also were not different between groups at 6 months (Figures 1 and 2⇓).
The greatest reduction from baseline in mean 24-h SBP occurred in patients who were in the highest baseline ambulatory SBP tertile (Figure 3). There was no significant change from baseline for the lowest tertile group and no difference in any of the ABPM changes between the denervation and sham groups. Mean 24-h heart rates were similar at baseline and at 6 months in the 2 lower tertiles. Details of 24-h, daytime, nighttime, and heart rate measures are reported in Online Table 1.
The proportion of patients who converted from nondipper to dipper at 6 months post-randomization was 21.2% in the denervation group and 15.0% in the sham control group (difference, 6.2%; 95% CI: −3.8 to 16.2%). Early morning (3 am to 8 am) SBP increases were not significant for either group, with a slope change of −0.05 at 6 months for the denervation group (p = 0.90) and −0.52 in the sham group (p = 0.35).
The differences in ambulatory SBP from baseline to 6 months in various subgroups of participants are shown in the Central Illustration. The absolute magnitudes of difference were small (<6 mm Hg) and not significant in any of the pre-specified subgroups. No significant differences were noted between groups as a function of baseline ambulatory SBP. There was no significant difference in 24-h heart rate between baseline and 6 months −1.36 ± 7.41 for denervation and −1.30 ± 7.25 for sham control (95% CI: −1.45 to 1.35; p = 0.94).
The primary safety endpoint and other safety events have been described previously (7). Few major adverse events occurred in the trial: 1 (0.6%) in the sham control arm and 5 (1.4%) in the treatment arm, for a difference of 0.8% (95% CI: −0.9%, 2.5%; p = 0.67).
This randomized, sham-controlled, blinded trial failed to show a benefit of renal artery denervation on the powered secondary endpoint of 24-h ambulatory SBP. Daytime and nighttime ABPM and heart rate change were also not different between the denervation and sham control groups. Earlier unblinded trials of renal denervation demonstrated significant reductions in ABPM measurements 6 months post-denervation that were similar in magnitude to the current trial, but without the benefit of a blinded control (8,9). Previous renal denervation studies reported smaller decreases in 24-h and daytime ABPM than office BP; this finding is consistent with our observations (8,9). This BP difference between office and ambulatory settings was predicted by a meta-analysis of antihypertensive drug and renal denervation trials that noted that differences in office and ambulatory BP reductions disappear in double-blind placebo controlled drug trials (10,11).
ABPM has the advantage of less bias compared with office readings and provides a complete picture of BP throughout the day. This concept is further supported by this trial because disappearance of the difference between office and ABPM effect in a double-blind, placebo-controlled trial suggests that the placebo effect is stronger for office than ABPM visits. ABPM is generally believed to be less vulnerable to placebo-like effects because it is assessed in a patient’s daily life (12). Given that area under the curve of the 24-h ABPM is more accurate than office readings taken at fixed times, differences in BP may be detected by ABPM when office readings fail to show differences (3). In the current trial, no such differences were noted, although a trend was noted for a greater nighttime BP reduction in the denervation group compared with the sham control group.
High BP variability correlates with higher cardiovascular event rates (13,14). In this trial, BP variability was not significantly different between groups at 6 months. Additionally, a reduction in early morning BP surge or restoration of dipping status was not affected by renal denervation (15,16).
Small reductions in office-measured heart rate as well as glycemic control were observed in previous trials of renal denervation. In the current trial, 24-h heart rate and glycemic control did not significantly change from baseline in either group. There are a few possible reasons for this observation. First, effective denervation may not have occurred in all patients randomized to denervation because there was no definitive way to assess denervation. A second possibility, however, is that the drug doses affecting heart rate, which also would adversely affect glucose control, may have been maximized in both groups, hence blunting the small previous effect seen on heart reduction and glycemic control. Regardless of the possible causes, an effect on heart rate was not seen in this trial.
Several possible explanations exist for the discordant findings between the prior renal denervation data and the present results (5,8,9). Previous renal denervation studies described the change in ABPM from baseline without a control arm, and thus treatment efficacy was assumed but could not be verified. This trial also observed significant ambulatory BP reductions from baseline to 6 months following renal denervation, but the similar significant 24-h ambulatory BP drop in the sham control group resulted in a lack of significant difference between the 24-h ambulatory BP reductions between the 2 groups. It is intriguing that the nocturnal SBP and DBP dropped significantly from baseline, but the BP reductions in the sham control group were less and were not significantly different from baseline (Figure 2). The similar overall ambulatory BP reduction seen in the sham control arm may be partly attributed to participation in a trial that provided a high degree of patient support and oversight that may have led to improved adherence with medications and diet, an observation known as the Hawthorne effect (17,18). These data suggest that the prior renal denervation studies overestimated the treatment effect of the procedure.
The potential contribution of a placebo effect to the BP reductions attributed to renal denervation in prior studies cannot be ascertained (19,20). Our analysis suggests that a placebo effect, perhaps enhanced by the interventional procedure in the control group, did affect BP change. Alternatively, however, one could hypothesize that the placebo effect could have worsened BP in the intervention group because the patient may have been concerned that he or she received sham treatment. Both these statements are speculative at best; however, this observation has important implications for future trial designs of medications and devices for hypertension and other diseases.
There are limitations in interpreting the data from this trial. There was no validation of medication adherence by assessing urine levels for metabolites of antihypertensive drugs. It is well documented that more than 50% of patients with resistant hypertension are nonadherent with medications (21,22). In this trial, however, assessment of patient diaries completed before treatment and again before unblinding at 6 months fails to suggest a difference in medication adherence between the groups. Moreover, only 5.8% of subjects had medication changes during the 2 weeks before the second screening visit. Although that may have affected baseline BP, no significant differences in office BP or ABPM were noted between groups at screening visits. Medication adherence may have improved as a result of the intervention, as suggested by changes in ABPM at 6 months in the sham control group. This finding further supports the concept of an objective measure of medication adherence in patients enrolled in trials to treat drug-resistant hypertension. Finally, there was no formal prospective assessment of generator values of impedance, and this may account for a variable response between centers.
The trial was not powered to detect small differences in ambulatory BP or any potential differences in pre-specified subgroups. The trend for improved nighttime BP control following renal denervation that was not detected during the day suggests the need for further research on the physiologic effects of renal denervation. A relative lack of operator experience could have affected the outcomes in the renal denervation group, although all procedures were proctored. However, we found no evidence of an operator learning curve when results of early procedures were compared with later procedures. The catheter generator system provides confirmation of energy delivery, but no biomarker or easily applied test is available to confirm adequate denervation at the time of the treatment. Finally, these trial results are specific to the Symplicity catheter and may not be generalizable to other renal denervation systems.
The current trial confirms the safety of renal denervation with the Symplicity catheter; however, a significant BP-lowering effect on 24-h ambulatory BP was not observed. Further clinical research using rigorous trial design will be required to understand whether renal denervation has any role in the treatment of resistant hypertension.
COMPETENCY IN MEDICAL KNOWLEDGE: Compared with a sham procedure, catheter-based renal artery sympathetic denervation did not reduce BP either during daytime or nighttime periods, as assessed by systematic ABPM.
COMPETENCY IN PATIENT CARE: Renal artery denervation is an investigational procedure not currently approved for clinical use in the United States, and it should not be recommended for patients with resistant hypertension outside the context of properly designed and regulated clinical trials.
TRANSLATIONAL OUTLOOK: Additional studies are needed to assess whether selection of hypertensive patients according to different criteria, alternative methods, or more complete renal artery sympathetic denervation could provide more effective BP control.
The authors would like to thank Xiaohua Chen, MS, and Lanyu Lei, MS, from Harvard Clinical Research Institute for statistical analyses funded by Medtronic and Colleen Gilbert, PharmD, from Medtronic for editorial support. The authors also would like to thank Sandeep Brar, MD, and Juan Wu, MS, for data analysis support and Vanessa DeBruin, MS, Denise Jones, RN, BSN, Dan Jolivette, MD, and the entire SYMPLICITY HTN-3 study team for dedicated research support.
This study was funded by Medtronichttp://dx.doi.org/10.13039/100004374. Dr. Bakris has received consultant fees from Takeda Pharmaceutical Company, AbbVie, Novartis Pharmaceuticals Corporation, Janssen Pharmaceuticals, Bristol-Myers Squibb (BMS), Bayer Healthcare, Medtronic, Relypsa, Inc., Orexigen Therapeutics, Merck & Co., Inc., and GlaxoSmithKline; has grant funding of an investigator-initiated grant from Takeda; is the Editor of the American Journal of Nephrology and the hypertension section of UpToDate; and is Associate Editor of Diabetes Care and Section Editor of Nephrology Dialysis Transplantation. Dr. Bhatt is on advisory boards for Elsevier Practice Update Cardiology, Medscape Cardiology, and Regado Biosciences; is on the board of directors for the Boston Veterans Affairs Research Institute, Society of Cardiovascular Patient Care; is Chair of the American Heart Association’s Get With The Guidelines Steering Committee; is on Data Monitoring Committees for Duke Clinical Research Institute, Harvard Clinical Research Institute, Mayo Clinic, and Population Health Research Institute (including for EnligHTNment); and receives honoraria from the American College of Cardiology (Editor, Clinical Trials, Cardiosource), Belvoir Publications (Editor-in-Chief, Harvard Heart Letter), Duke Clinical Research Institute (clinical trial steering committees), Harvard Clinical Research Institute (clinical trial steering committee), HMP Communications (Editor-in-Chief, Journal of Invasive Cardiology), Population Health Research Institute (clinical trial steering committee), Slack Publications (Chief Medical Editor, Cardiology Today’s Intervention), WebMD (CME steering committees); and is an editor of Clinical Cardiology (Associate Editor), and Journal of the American College of Cardiology (Section Editor, Pharmacology). Research grants were received from the following companies: Amarin Corporation, AstraZenecahttp://dx.doi.org/10.13039/100004325, BMS, Eisai, Inc., Ethicon Endo-Surgery, Inc., Medtronichttp://dx.doi.org/10.13039/100004374, Hoffman-La Roche, Inc., Sanofi Aventis, The Medicines Company; unfunded research: FlowCo, Inc., PLx Pharma, and Takeda Pharmaceutical Company. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- ambulatory blood pressure monitoring
- diastolic blood pressure
- systolic blood pressure
- Received April 14, 2014.
- Revision received May 10, 2014.
- Accepted May 13, 2014.
- American College of Cardiology Foundation
- ↵U.S. Food and Drug Administration. Guidance for Industry Hypertension Indication: Drug Labeling for Cardiovascular Outcome Claims. 2011. Available at: http://www.fda.gov/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/default.htm. Accessed May 23, 2014.
- National Clinical Guideline Centre. Hypertension: The Clinical Management of Primary Hypertension in Adults. Clinical Guideline 127: Methods, Evidence and Recommendations: National Institute for Health and Clinical Excellence. 2011:328.
- Esler M.D.,
- Krum H.,
- Schlaich M.,
- et al.
- Mahfoud F.,
- Ukena C.,
- Schmieder R.E.,
- et al.
- Howard J.P.,
- Nowbar A.N.,
- Francis D.P.
- Parati G.,
- Ochoa J.E.,
- Lombardi C.,
- et al.
- Kario K.,
- Pickering T.G.,
- Umeda Y.,
- et al.
- Gale E.A.,
- Beattie S.D.,
- Hu J.,
- et al.
- Tomaszewski M.,
- White C.,
- Patel P.,
- et al.