Reverse cardiac remodeling after renal denervation: Atrial electrophysiologic and structural changes associated with blood pressure lowering

Author's Accepted Manuscript

Reverse Cardiac Remodelling Following Renal Denervation - Atrial Electrophysiologic and Structural Changes Associated with Blood Pressure Lowering Alex J.A. McLellan MBBS, Markus P. Schlaich MD, Andrew J. Taylor MBBS Ph.D., Sandeep Prabhu MBBS, Dagmara Hering MD Ph.D., Louise Hammond RN, Petra Marusic BAppSci, Jacqueline Duval BBiotech, Yusuke Sata MD, Andris Ellims MBBS, Murray Esler MBBS Ph.D., Karlheinz Peter MD Ph. D., James Shaw MBBS Ph.D., Antony Walton MBBS, Jonathan M. Kalman MBBS Ph.D., Peter M. Kistler MBBS Ph.D.

PII: DOI: Reference:

S1547-5271(15)00138-1 http://dx.doi.org/10.1016/j.hrthm.2015.01.039 HRTHM6110

To appear in:

Heart Rhythm

www.elsevier.com/locate/buildenv

Cite this article as: Alex J.A. McLellan MBBS, Markus P. Schlaich MD, Andrew J. Taylor MBBS Ph.D., Sandeep Prabhu MBBS, Dagmara Hering MD Ph.D., Louise Hammond RN, Petra Marusic BAppSci, Jacqueline Duval BBiotech, Yusuke Sata MD, Andris Ellims MBBS, Murray Esler MBBS Ph.D., Karlheinz Peter MD Ph.D., James Shaw MBBS Ph.D., Antony Walton MBBS, Jonathan M. Kalman MBBS Ph.D., Peter M. Kistler MBBS Ph.D., Reverse Cardiac Remodelling Following Renal Denervation Atrial Electrophysiologic and Structural Changes Associated with Blood Pressure Lowering, Heart Rhythm, http://dx.doi.org/10.1016/j.hrthm.2015.01.039 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 galley proof before it is published in its final citable 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.

1

Reverse Cardiac Remodelling Following Renal Denervation - Atrial Electrophysiologic and Structural Changes Associated with Blood Pressure Lowering Alex JA McLellan MBBS 1-4, Markus P Schlaich MD 1,2,5,6, Andrew J Taylor MBBS PhD 1-2, Sandeep Prabhu MBBS 1-4, Dagmara Hering MD PhD 2, Louise Hammond RN 2, Petra Marusic BAppSci 2, Jacqueline Duval BBiotech 2, Yusuke Sata MD 2, Andris Ellims MBBS 1-2

, Murray Esler MBBS PhD 1-2,5, Karlheinz Peter MD PhD 1-2, James Shaw MBBS PhD 1-2,

Antony Walton MBBS 1-2, Jonathan M Kalman MBBS PhD 3-4, Peter M Kistler MBBS PhD 14

Affiliations 1

Department of Cardiovascular Medicine, Alfred Hospital, Melbourne, Victoria, Australia;

2

Baker IDI Heart and Diabetes Institute, Melbourne, Victoria, Australia;

3

Cardiology

Department, Royal Melbourne Hospital, Victoria, Australia; 4Faculty of Medicine, Dentistry, and Health Sciences, University of Melbourne, Victoria, Australia; 5Faculty of Medicine, Nursing and Health Sciences, Monash University, Melbourne, Australia; 6School of Medicine and Pharmacology-Royal Perth Hospital Campus, University of Western Australia

Corresponding Author A/Prof Peter Kistler Baker IDI Heart and Diabetes Institute 75 Commercial Road, Melbourne, Victoria, Australia 3004 Tel: 03 85321111, Fax: 03 85321111, Email: [email protected]

Word Count: 4972 words Brief Title: Reverse cardiac remodelling post renal denervation

2

ABSTRACT Background Hypertension is the most common modifiable risk factor associated with atrial fibrillation. Objective We sought to determine the effects of blood pressure (BP) lowering following renal denervation on atrial electrophysiological and structural remodelling in humans. Methods Fourteen patients (mean age 64±9years, duration of hypertension 16±11years, on 5±2 antihypertensive-medications) with treatment resistant hypertension underwent baseline 24hour ambulatory BP monitoring, echocardiogram, cardiac magnetic resonance (CMR) imaging and electrophysiology study(EPS). EPS included measurement of P-wave duration, effective refractory periods and conduction times. Electroanatomic mapping of the right atrium was completed using CARTO3 to determine local and regional conduction velocity and tissue voltage. Bilateral renal denervation was then performed and all measurements repeated after 6months. Results Following renal denervation mean 24-hr BP reduced from 152/84mmHg to 141/80mmHg (p<0.01) at 6 months follow-up.

Global conduction velocity increased significantly

(0.98±0.13m/s to 1.2±0.16m/s at 6months, p<0.01), conduction time shortened (32±5msec to 27±6msec, p<0.01) and complex fractionated activity (37±14% to 19±12%, p=0.02) reduced. Changes in conduction velocity correlated positively with changes in 24-hr mean systolic BP (R2=0.55, p=0.01). There was a significant reduction in left ventricular mass (139±37g to 120±29g, p<0.01) and diffuse ventricular fibrosis (T1 partition-coefficient 0.39±0.07 to 0.31±0.09; p=0.01) on CMR. Conclusion Blood pressure reduction following RDN is associated with improvements in regional and global atrial conduction and a reduction in ventricular mass and fibrosis. Whether changes in electrical and structural remodelling are solely due to blood pressure lowering, or are in part due to intrinsic effects of renal denervation remains to be determined.

3

KEYWORDS Renal Denervation Remodelling Atrial fibrillation Electrophysiology Study Cardiac MRI

ABBREVIATIONS RDN

Renal Denervation

MRI

Magnetic Resonance Imaging

CMR

Cardiac Magnetic Resonance

TTE

Trans Thoracic Echocardiogram

LV

Left Ventricle

LVH

Left Ventricular Hypertrophy

INTRODUCTION Hypertension affects up to a quarter of the population1 and is the major modifiable risk factor associated with atrial fibrillation2, yet little is known about the effects of blood pressure lowering on human electrophysiology. In animal studies hypertension is associated with adverse electrophysiological remodelling characterized by conduction slowing and an increase in atrial fibrosis on histopathology3. Similar electrophysiological changes were described in human hypertension and associated with an increase in AF inducibility4. Anti-hypertensive medications are successful in achieving blood pressure targets in just onethird of patients5, and renal denervation (RDN) has been tested as an alternative strategy to improve blood pressure control in patients with resistant hypertension6. RDN significantly

4

reduces sympathetic activity, which is known to be elevated both in patients with hypertension and AF7,

8,9

. Early case control studies had demonstrated significant blood

pressure reduction with RDN6, 9. Improved blood pressure control with RDN has also been associated with reductions in LV mass on echocardiogram and cardiac MRI13, 14. Hypertension is associated with an increase in AF following catheter ablation10. A small randomized controlled trial of AF ablation with or without RDN suggested that AF ablation combined with RDN was associated with higher rates of freedom from AF compared to patients who underwent AF ablation alone11. However, the recently published Symplicity HTN-3 study, a single blinded randomised sham controlled trial could not demonstrate a significant difference in blood pressure reduction between the treatment and the sham group neither with office nor ambulatory blood pressure monitoring12. While criticism was raised relating primarily to the inexperience of operators performing RDN in Symplicity HTN-3, significant BP reductions were typically observed, albeit with a large degree of interindividual variability. Furthermore, the exact mechanisms responsible for the blood pressure lowering effects of RDN have been incompletely defined but likely include neuro-hormonal modulation9, 13, 14. Randomized controlled trials have suggested that anti-hypertensive medications may be associated with a reduction in atrial fibrillation15 however the electrophysiologic mechanisms responsible for this clinical observation have not been defined. The aim of the present study was to determine whether blood pressure lowering following RDN is associated with electrophysiologic and electroanatomic changes in the human atrium.

METHODS Study population Between June 2012 and October 2013 we prospectively recruited 14 patients with treatment resistant hypertension (defined as a blood pressure greater than goal target despite the concurrent use of at least 3 anti-hypertensive medications) who were scheduled for renal denervation. Exclusion criteria included age > 75years, secondary causes of hypertension, significant renal disease (eGFR based on the Modification of Diet in Renal Disease criteria of less than 45

5

mL/min per 1·73 m²), ischaemic heart disease, heart failure, valvular heart disease, diabetes, recent stroke, pregnancy, contra-indications to cardiac MRI and the inability to provide informed consent. The study protocol was approved by the Alfred Hospital Ethics review committee and all patients gave informed consent. Study Protocol Patients were intensively studied at two time intervals: (i) at baseline within a week before RDN and (ii) 6 months post RDN. Pre-procedural investigations Prior to renal denervation patients underwent detailed investigations including: 24-hour ambulatory blood pressure monitoring: assessed with an oscillometric Spacelabs 90217 monitor (Spacelabs Healthcare, Issaqua, WA, USA) with daytime readings every 15mins and night-time readings every 30mins6. pathology (full blood examination, Creatinine and electrolytes, BNP and 24-hr urinalysis) 7-day Holter monitoring, Cardiac imaging: Echocardiography and CMR CMR sequences were acquired using a clinical 1.5-T scanner (Signa HD 1.5-T, GE Healthcare, Wisconsin, USA) during breath-holds 10 to 15 s as previously described16. LV volume, function and mass were determined from a short axis stack (8mm slice thickness, no gap) from the mitral annulus to the apex. Assessment of regional ventricular fibrosis was performed with late gadolinium enhancement (LGE) imaging. Diffuse atrial fibrosis and diffuse left ventricular fibrosis were assessed using a myocardial T1 mapping sequence (SMART1Map; Saturation Method using Adaptive Recovery Times for T1 mapping17) in long and short axis.

Diffuse

ventricular fibrosis was quantified by i) the pre-contrast T1 relaxation time, ii) the post-contrast

T1

relaxation

(λ=∆R1(myocardium precontrast).

time,

and

iii)

the

partition

postcontrast-precontrast)/∆R1(blood

pool

co-efficient,

λ

postcontrast-

Post-contrast T1 relaxation times were corrected for GFR and

acquisition time post contrast as previously reported18.

6

Assessment of left ventricular function on echocardiography was performed using the Simpson bi-plane summation of disc method.

Echocardiographic calculation of left

ventricular mass was performed using the Devereux formula.

Electrophysiology study and electroanatomical mapping Renal denervation and electrophysiology study were performed in the fasted state under minimal conscious sedation with electrophysiology study performed immediately prior to RDN.

Patients on beta-blockers or non-dihydropyridine calcium channel blockers were

instructed to withhold such medication for 5 days prior to procedure. SixF and 8F sheaths were inserted in the right femoral vein with a decapolar catheter (St Jude, MN, USA) positioned within the coronary sinus and a Navistar SmartTouch TM catheter (Biosense Webster, CA, USA) utilized for mapping of the right atrium. First, assessment of effective refractory period testing was performed at twice diastolic threshold from two locations (proximal coronary sinus and high right atrium), using an eight beat drive train at two cycle lengths (600ms and 450ms) for each location with an atrial extra advanced in 10ms increments from below refractoriness (150ms) until atrial capture was achieved (ERP was the average of triplicate measurements at each site and drive train). Second, conduction time was assessed during stable CS pacing at 600ms and 450ms measuring the conduction time from pacing stimulus on CS proximal to CS distal (using the average of 10 measurements for conduction time at each drive train). During stable sinus rhythm the P wave duration in lead II was measured (using the average of 10 P-wave measurements). Finally an electroanatomical map (EAM) of the right atrium was created during stable CS pacing at 600ms using CARTO 3 (Biosense Webster). Points were only included if at least 10g of force was achieved with stability of electrogram in time (local activation time <5ms) and catheter position in space (catheter movement < 6mm). An even distribution of points was collected using a fill threshold of 15 with editing and annotation of each point performed off-line post procedure. For the purposes of regional analysis walls were evenly divided into posterior, anterior, septal and lateral right atrial walls. The EAM was assessed for global and regional conduction velocity, complexity of electrograms, and for bipolar and unipolar voltage including areas of low voltage and scar

7

(figure 1).

Conduction velocity (CV) was assessed on the isochronal activation map,

assessing the velocity of wave-front propagation at regions of least isochronal crowding, averaging CV measured from 5 pairs of electrograms at each region, as previously described4.

Regions of low voltage (bipolar voltage of contiguous points <0.5mV) and

scar (bipolar voltage of contiguous points < 0.05mV) were documented.

Assessment of

complex electrograms included assessment of fractionated signals (electrograms with ≥50ms duration with at least 3 deflections) as previously described4. Renal Denervation Renal denervation was performed with femoral arterial access and the endovascular technique as previously reported6. The Symplicity catheter was advanced into the renal artery with 6-10 ablations at maximum of 8W delivered to each renal artery. Visceral pain limiting ablation was managed with intravenous opiate and benzodiazepine as required. Follow-up Changes in anti-hypertensive medication were discouraged during the study period. At six months patients repeated all the investigations performed at baseline, including electrophysiology study as described above. Analysis of ambulatory blood pressure, electrophysiology, and cardiac structure and function (TTE and CMR) was performed by independent investigators with specialization in these fields, blinded to the results of the other investigators and procedural status (pre vs. post procedure).

Statistics Continuous variables are expressed as a mean ± SD with comparisons from baseline to follow-up performed with a paired Student’s t-test. Categorical variables are expressed as numbers and percentages, and were compared with the related samples McNemar test. Correlations between variables were assessed with the Pearson correlation coefficient. A two sided P value <0.05 was considered statistically significant. All statistical analysis was performed using SPSS software version 21.0 (SPSS, Chicago, IL, USA).

RESULTS

8

Baseline Characteristics and Procedural Details The baseline characteristics and procedural details of the study cohort are outlined in table 1. The cohort was predominantly male (67%), middle aged (average age 64±9) with a mean duration of hypertension of 16±11 years (range 1-40), on a mean of 5±2 antihypertensive medications. Renal denervation was performed with an average of 8±1 ablations per renal artery (range 5-12). Changes in Blood Pressure Changes in 24-hour ambulatory BP from baseline to 6 months are presented in figure 2. There was a reduction in mean 24-hour ambulatory BP from 152/84 at baseline to 141/80 at 6 months follow-up, (change in mean systolic BP 11±11mmHg, p<0.01). Maximum 24-hr ambulatory BP reduced from 192/112 to 178/107 at 6 months follow-up (change in maximum systolic BP 13±19mmHg, p=0.03). Structural Remodelling on Cardiac MRI and Echocardiogram (table 2) Cardiac magnetic resonance imaging There was a significant reduction in left ventricular mass from 139±37g at baseline to 120±29g (p<0.01) at 6 month follow up. Left ventricular mass indexed to body surface area (BSA) (66±15g to 57±12g, p<0.01), left atrial (27±6cm2 to 24±6cm2, p=0.047) and right atrial size (24±5cm2 to 20±5cm2, p<0.01) all decreased significantly at follow-up.

There

was a significant reduction in diffuse ventricular fibrosis as measured by the T1 partition coefficient (0.39±0.07 to 0.31±0.09; p=0.01), and also a strong trend to reduction in diffuse ventricular fibrosis as measured by the pre and post-contrast T1 relaxation times (table 2). There were no significant changes in left ventricular volumes or function, or diffuse atrial fibrosis. Echocardiography There was a reduction in left ventricular mass (215±60g to 192±50g, p=0.05) and left ventricular mass indexed to BSA (106±27g to 95±24g, p=0.06). There were no significant changes in left ventricular dimensions, systolic function or diastolic function. There was a significant reduction in the right atrial but not in left atrial size. Electrophysiology study

9

At 6 months 11 of 14 patients agreed to undergo repeat invasive electrophysiology mapping post renal denervation. Atrial conduction P-wave duration shortened significantly from 122±8ms to 112±8ms (p=0.02) at 6months. There was a significant increase in global conduction velocity from 0.98±0.13m/s to 1.2±0.16m/s, p<0.01) at 6 months follow up. Conduction velocity was significantly increased at 3 of 4 predefined right atrial regions (figure 3). There was a shortening in conduction time along the coronary sinus (32±5msec to 27±6msec, p<0.01). There was a positive correlation between change in 24-hour mean and maximum systolic BP with change in conduction velocity (CV) from baseline to follow-up (∆24-hr mean SBP vs ∆CV, R2=0.55, p=0.01; ∆24-hr max SBP vs ∆CV, R2=0.66, p<0.01) (figure 4). There were two patients with <10mmHg improvement in mean-24hour systolic BP with improvement in CV at follow-up. There was a significant reduction in fractionated electrograms both globally (37±14% to 19±12% at 6months, p=0.02) and regionally (table 3). There was no change in global tissue voltage or discrete sites of low voltage (table 3). There was no significant change in AF inducibility during EPS. 7-day Holter monitor From baseline to follow-up 7-day Holter monitoring, there was no change in average heart rate, burden of atrial or ventricular premature complexes, non-sustained (or sustained) atrial or ventricular tachy-arrhythmias (table 4), or on heart rate variability (supplementary table 1).

DISCUSSION The main findings of the present study investigating the electrical and structural effects on the human atrium of blood pressure lowering following renal denervation include: 1. A reduction in ambulatory 24-hr mean systolic blood pressure 2. Structural cardiac remodelling characterized by a. A reduction in left ventricular mass,

10

b. Reduction in diffuse ventricular fibrosis, c. Reduction in atrial size. 3. Electrical remodelling characterized by a. An increase in conduction velocity and reduction in conduction time, b. Reduction in complex fractionated electrograms. c. No change in tissue voltage or atrial refractoriness Blood pressure lowering following renal denervation Renal artery denervation is a novel therapy associated with sustained lowering of blood pressure although the mechanisms responsible remain incompletely defined13. In the present study the blood pressure reduction was 11/4mmHg as measured by 24hr ambulatory monitoring, similar to the magnitude reported in a global registry of 346 patients undergoing renal denervation19. In the first randomised sham control trial of renal denervation (Symplicity HTN-3) there was a significant reduction in 24-hr mean systolic BP in both the RDN arm and the sham cohort12 with no significant between group difference. The Symplicity HTN-3 trial highlighted the importance of a sham group in a randomised control study of a new therapy. There are several explanations for the outcomes observed in Symplicity HTN-3 including the Hawthorne effect and that the majority of RDN procedures were performed by inexperienced operators. A recent Symplicity HTN-3 sub-study identified use of aldosterone antagonists, increased number of ablations delivered and non-African-American race to predict BP reduction20. Our cohort was entirely Caucasian, had high use of aldosterone antagonist and had increased number of effective ablations delivered (16±2) compared to Symplicity HTN-3 (9±2). These differences may in part explain the significant improvements in BP our study. There remains no acute procedural endpoint to renal artery ablation during RDN which is the subject of ongoing research. Irrespective of the mechanisms responsible for the observed blood pressure change in the present report our data provide valuable insights into structural and electrical remodelling associated with blood pressure reduction. Change in left ventricular mass on CMR and echocardiogram In the present study there was at least a 14% reduction in left ventricular mass (p<0.01) on CMR at six months. Mafoud et al previously demonstrated a significant reduction in LV mass indexed to BSA on cardiac MRI in 55 patients following RDN21. The time course of

11

response to renal denervation is difficult to ascertain but we would speculate that both reduction in afterload and effects on cardiac sympathetic tone are involved. The time course of structural remodelling post renal denervation has been addressed in limited reports. Brandt et al reported a 15% reduction in indexed LV mass on echocardiogram from baseline to 6 months in patients who underwent RDN, with the majority of LV mass reduction achieved in the first month22.

Hypertension-induced left ventricular hypertrophy is associated with

increased sympathetic activity as measured by cardiac noradrenaline spillover7 and one may speculate that some of the effects on LV mass may be due to RDN induced reduction in noradrenaline levels. In concert with a reduction in left ventricular mass we have also identified the novel finding of a reduction in diffuse left ventricular fibrosis, as indicated by a reduction in the left ventricular partition coefficient, as well as strong trends for changes in both pre- and post contrast myocardial T1 times, both of which have previously been shown to reliably correlate with histological fibrosis23, 24. To our knowledge this has not been previously reported. A reduction in diffuse ventricular fibrosis and mass were associated with a reduction in atrial size which may in part explain the reverse atrial remodelling observed in the present study. Electrophysiological Remodelling Blood pressure lowering was associated with a significant change in a range of measures of atrial conduction including global and regional conduction velocity, conduction time, and complex atrial activity. Previous animal and human studies have shown that elevated blood pressure is associated with a reduction in conduction velocity and increase in complex electrograms3, 4. Sanders and colleagues have elegantly shown in an animal model that hypertension is associated with progressive electrical remodelling characterized by prolongation of atrial refractoriness, conduction slowing and heterogeneity of conduction25. Human studies similarly demonstrated global and regional conduction slowing without changes in tissue voltage in hypertensive patients compared with age matched controls 4. There have been limited animal studies assessing the effects of blood pressure lowering by RDN on atrial electrophysiology. Linz et al26 performed an acute study of RDN in a porcine model with EPS immediately prior and 1.5 hours post RDN. The investigators showed no change in ERP or AF inducibility however AF episodes were significantly shorter in the RDN group. In contrast Hou et al27 used a hyper-sympathetic tone canine model of AF followed by RDN and included a sham control group. Renal denervation acutely attenuated

12

the shortening of atrial refractoriness and propensity to AF, and reduced sympathetic activity compared with the sham control. Effect of pharmacological treatment of hypertension on atrial fibrillation The impact of pharmacologic treatment for hypertension on the occurrence of atrial fibrillation has generated conflicting results. A meta-analysis of angiotensin-convertingenzyme inhibitors (ACE-I) or angiotensin receptor blockers (ARB) versus controls did not demonstrate a significant difference in the development of AF with ACE-I/ARB15. Interestingly the two studies within the meta-analysis where ARB therapy was associated with a reduction in incident AF was related more to regression in LVH and LA size rather than impact on blood pressure. In contrast, angiotensin receptor blockade is associated with a reduction in AF recurrence in hypertensive patients with documented AF. A meta-analysis demonstrated a 63% reduction in recurrent paroxysmal AF and a 45% reduction in recurrent AF post cardioversion15. The addition of spironolactone results in further reduction in AF burden compared to an ACE-I or beta-blocker, despite no additional impact on blood pressure28. These findings highlight the importance of neuro-hormonal modulation and tissue fibrosis in addition to direct blood pressure lowering in the prevention of AF.

In the present study two patients without

significant BP reduction had improvement in conduction velocity, raising the possibility of electrophysiological remodelling induced by autonomic modulation. Clinical Implications Hypertension is the main modifiable risk factor for the development of atrial fibrillation 2, and is heralded by changes in atrial electrophysiology4.

The present study demonstrates

improvements in global and regional conduction and it seems possible that the favorable electrical remodelling associated with blood pressure reduction in the current report may represent the physiological mechanism underlying a reduction in AF burden associated with anti-hypertensive medications or renal denervation. Pokushalov et al performed a small randomized controlled trial in 27 patients with AF and resistant hypertension, and identified greater freedom from AF at one year in patients who were randomized to RDN plus AF ablation compared to AF ablation alone11. The role of RDN as an adjunct to AF ablation in patients with co-morbid resistant hypertension is the subject of ongoing randomized controlled trials.

13

Limitations In the present study global and regional conduction changes were associated with a significant reduction in ambulatory blood pressure, left ventricular mass and diffuse fibrosis as measured by ventricular T1 mapping following renal denervation. In the absence of a sham control group the mechanisms responsible for the electrophysiologic and structural changes are beyond the scope of this small detailed study. In particular there were two patients with a significant improvement in conduction velocity without a significant change in BP suggesting that other factors such as autonomic modification may play a role. However, the purpose of the present study was to assess the atrial electrophysiologic and structural changes in response to blood pressure reduction rather than elucidating the mechanism of action of renal denervation. The incidence of AF may be higher than identified on Holter monitoring due to asymptomatic arrhythmia, noting a low incidence of AF on Holter monitoring and clinical symptoms despite significant electrophysiological remodelling and AF inducibility in 36% of the cohort.

Electrophysiologic study was limited to the right atrium given the ethical

concerns of non-clinically indicated transseptal puncture to obtain left atrial data.

Conclusion Blood pressure reduction post renal denervation is associated with improved atrial electrophysiology, reduction in LV mass and reduction in LV diffuse fibrosis.

These

physiologic changes might explain the reduction in AF burden associated with improved blood pressure control. Whether RDN may have intrinsic effects beyond that of blood pressure lowering on atrial electrophysiologic and structural parameters remains to be determined.

Funding Sources/ Disclosures Drs. McLellan, Prahbu and Ellims are supported by co-funded NHMRC/NHF Postgraduate Scholarships and BakerIDI Bright Sparks scholarships. Professors Schlaich and Peter are supported by NHMRC project grants and NHMRC research fellowships. A/Professor Taylor is supported by an NHMRC project grant. A/Professor Kistler is supported by a practitioner fellowship from the NHMRC. This research is supported in part by the Victorian

14

Government’s Operational Infrastructure Funding. Professor Schlaich has received research support from Medtronic.

REFERENCES 1.

2.

3.

4.

5.

6.

7.

8.

9.

10. 11.

12. 13.

14.

15.

Burt VL, Whelton P, Roccella EJ, et al. Prevalence of hypertension in the US adult population. Results from the Third National Health and Nutrition Examination Survey, 19881991. Hypertension Mar 1995;25:305-313. Benjamin EJ, Levy D, Vaziri SM, D'Agostino RB, Belanger AJ, Wolf PA. Independent risk factors for atrial fibrillation in a population-based cohort. The Framingham Heart Study. JAMA Mar 16 1994;271:840-844. Kistler PM, Sanders P, Dodic M, et al. Atrial electrical and structural abnormalities in an ovine model of chronic blood pressure elevation after prenatal corticosteroid exposure: implications for development of atrial fibrillation. Eur Heart J Dec 2006;27:3045-3056. Medi C, Kalman JM, Spence SJ, et al. Atrial electrical and structural changes associated with longstanding hypertension in humans: implications for the substrate for atrial fibrillation. J Cardiovasc Electrophysiol Dec 2011;22:1317-1324. Chobanian AV, Bakris GL, Black HR, et al. Seventh report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure. Hypertension Dec 2003;42:1206-1252. Esler MD, Krum H, Sobotka PA, Schlaich MP, Schmieder RE, Bohm M. Renal sympathetic denervation in patients with treatment-resistant hypertension (The Symplicity HTN-2 Trial): a randomised controlled trial. Lancet Dec 4 2010;376:1903-1909. Schlaich MP, Kaye DM, Lambert E, Sommerville M, Socratous F, Esler MD. Relation between cardiac sympathetic activity and hypertensive left ventricular hypertrophy. Circulation Aug 5 2003;108:560-565. Arimoto T, Tada H, Igarashi M, et al. High washout rate of iodine-123metaiodobenzylguanidine imaging predicts the outcome of catheter ablation of atrial fibrillation. J Cardiovasc Electrophysiol Dec 2011;22:1297-1304. Krum H, Schlaich M, Whitbourn R, et al. Catheter-based renal sympathetic denervation for resistant hypertension: a multicentre safety and proof-of-principle cohort study. Lancet Apr 11 2009;373:1275-1281. Berruezo A, Tamborero D, Mont L, et al. Pre-procedural predictors of atrial fibrillation recurrence after circumferential pulmonary vein ablation. Eur Heart J Apr 2007;28:836-841. Pokushalov E, Romanov A, Corbucci G, et al. A randomized comparison of pulmonary vein isolation with versus without concomitant renal artery denervation in patients with refractory symptomatic atrial fibrillation and resistant hypertension. J Am Coll Cardiol Sep 25 2012;60:1163-1170. Bhatt DL, Kandzari DE, O'Neill WW, et al. A controlled trial of renal denervation for resistant hypertension. The New England journal of medicine Apr 10 2014;370:1393-1401. Hering D, Lambert EA, Marusic P, et al. Substantial reduction in single sympathetic nerve firing after renal denervation in patients with resistant hypertension. Hypertension Feb 2013;61:457-464. Schlaich MP, Sobotka PA, Krum H, Lambert E, Esler MD. Renal sympathetic-nerve ablation for uncontrolled hypertension. The New England journal of medicine Aug 27 2009;361:932934. Schneider MP, Hua TA, Bohm M, Wachtell K, Kjeldsen SE, Schmieder RE. Prevention of atrial fibrillation by Renin-Angiotensin system inhibition a meta-analysis. J Am Coll Cardiol May 25 2010;55:2299-2307.

15 16.

17.

18.

19.

20. 21.

22.

23. 24.

25.

26.

27.

28.

Ling LH, McLellan AJ, Taylor AJ, et al. Magnetic resonance post-contrast T1 mapping in the human atrium: Validation and impact on clinical outcome after catheter ablation for atrial fibrillation. Heart Rhythm Sep 2014;11:1551-1559. Menini A, Slavin GS, Stainsby JA, Ferry P, Felblinger J, Odille F. Motion correction of multi-contrast images applied to T1 and T2 quantification in cardiac MRI. MAGMA Mar 22 2014. Gai N, Turkbey EB, Nazarian S, et al. T1 mapping of the gadolinium-enhanced myocardium: adjustment for factors affecting interpatient comparison. Magn Reson Med May 2011;65:1407-1415. Mahfoud F, Ukena C, Schmieder RE, et al. Ambulatory blood pressure changes after renal sympathetic denervation in patients with resistant hypertension. Circulation Jul 9 2013;128:132-140. Kandzari DE, Bhatt DL, Brar S, et al. Predictors of blood pressure response in the SYMPLICITY HTN-3 trial. Eur Heart J Nov 16 2014. Mahfoud F, Urban D, Teller D, et al. Effect of renal denervation on left ventricular mass and function in patients with resistant hypertension: data from a multi-centre cardiovascular magnetic resonance imaging trial. Eur Heart J Mar 6 2014. Brandt MC, Mahfoud F, Reda S, et al. Renal sympathetic denervation reduces left ventricular hypertrophy and improves cardiac function in patients with resistant hypertension. J Am Coll Cardiol Mar 6 2012;59:901-909. Bull S, White SK, Piechnik SK, et al. Human non-contrast T1 values and correlation with histology in diffuse fibrosis. Heart Jul 2013;99:932-937. Iles L, Pfluger H, Phrommintikul A, et al. Evaluation of diffuse myocardial fibrosis in heart failure with cardiac magnetic resonance contrast-enhanced T1 mapping. J Am Coll Cardiol Nov 4 2008;52:1574-1580. Lau DH, Mackenzie L, Kelly DJ, et al. Hypertension and atrial fibrillation: evidence of progressive atrial remodeling with electrostructural correlate in a conscious chronically instrumented ovine model. Heart Rhythm Sep 2010;7:1282-1290. Linz D, Mahfoud F, Schotten U, et al. Renal sympathetic denervation provides ventricular rate control but does not prevent atrial electrical remodeling during atrial fibrillation. Hypertension Jan 2013;61:225-231. Hou Y, Hu J, Po SS, et al. Catheter-based renal sympathetic denervation significantly inhibits atrial fibrillation induced by electrical stimulation of the left stellate ganglion and rapid atrial pacing. PLoS One 2013;8:e78218. Dabrowski R, Borowiec A, Smolis-Bak E, et al. Effect of combined spironolactone-betablocker +/- enalapril treatment on occurrence of symptomatic atrial fibrillation episodes in patients with a history of paroxysmal atrial fibrillation (SPIR-AF study). The American journal of cardiology Dec 1 2010;106:1609-1614.

Clinical Perspectives This study identified blood pressure reduction 6 months post renal denervation to be associated with favourable electrical and structural remodelling. These results contribute to the understanding of the role of hypertension as the main modifiable risk factor in the genesis of AF.

Specifically our results suggest that reducing blood pressure in patients with

treatment resistant hypertension may improve the electrophysiologic substrate that promotes AF. Future studies will be required to establish whether electrophysiologic remodelling post

16

renal denervation is solely due to blood pressure reduction or whether alteration of autonomic tone plays a significant role. In patients with AF and treatment resistant hypertension, a randomized controlled trial of renal denervation or sham procedure (potentially in addition to pulmonary vein isolation) will be required prior to recommending renal denervation to this population for the treatment of arrhythmia.

Figure Legend Figure 1: Baseline and follow-up electroanatomical map of the right atrium viewed from a posterior projection. Isochronal activation map (A and C) and voltage map (B and D). The arrow is aligned with the conduction wave-front demonstrating a shortening in the distance between isochrones at follow-up (versus baseline). Brown dots represent fragmented electrograms, light blue dots represent double potentials. IVC: inferior vena cava, SVC: superior vena cava, CS: coronary sinus.

Figure 2: 24-hour ambulatory blood pressure changes from baseline to 6 months. SBP systolic blood pressure, DBP diastolic blood pressure

Figure 3: Global and regional conduction velocity from baseline to follow-up

Figure 4: Correlations for change in mean (A) and maximum (B) ambulatory 24-hour systolic blood pressure versus change in conduction velocity from baseline to follow-up. The arrow indicates a physiological improvement from baseline to follow-up.

17

Table 1: Baseline characteristics and procedural details Baseline characteristics (n=14) Age, years Gender male, n (%) BMI, (kg/m2)

64±9 10 (67) 31±3

Duration of HT, (years)

16±11

Serum creatinine, (mmol/L)

84±22

eGFR, (mL/min)

75±15

DM, n (%)

2 (14)

Dyslipidaemia, n (%)

10 (71)

IHD, n (%)

2 (14)

CVA, n (%)

3 (21)

AF, n (%)

2 (14)

Antihypertensive medications No of antihypertensive medications baseline ACE-I, n (%) ARB

4.9±1.8 5 (36) 10 (71)

Beta-blocker

8 (57)

Calcium channel blocker

8 (57)

Diuretic Aldosterone antagonists Moxonidine

14 (100) 7 (50) 10 (71)

α blocker

4 (29)

Other (minoxidil, nitrate, or methyldopa)

4 (29)

Procedural Characteristics Procedure duration (EPS plus RDN), min Fluoroscopy duration, min Average number of ablations per renal artery

Table 2: Structural Remodelling on Cardiac MRI and Echocardiogram

188±56 31.8±6.9 8±1

18

Baseline

6months

P Value

LVEDV, mL

166±33

174±46

0.39

LVESV, mL

57±18

65±24

0.12

LVEF, %

66±5

63±5

0.18

LV mass, g

139±37

120±29

<0.01

LV mass indexed, g/BSA

66±15

57±12

<0.01

LA area, cm2

27±6

24±6

0.05

RA area, cm2

24±5

20±5

<0.01

LV T1 Partition co-efficient (λ)

0.39±0.07

0.31±0.09

0.01

LV pre-contrast T1 time, ms

1136±215

946±192

0.06

LV post-contrast T1 time, ms

405±42

437±62

0.13

Atrial T1 relaxation time, ms

209±32

214±49

0.80

LVEDD, mm

48±5

47±4

0.50

LVESD, mm

29±6

32±6

0.12

LVEF, %

60±7

60±6

1.00

LV mass, g

215±60

192±50

0.05

LV mass indexed, g/BSA

106±27

95±24

0.06

IVS, mm

13±2

12±2

0.36

PW, mm

11±2

10±2

0.05

LA area, cm2

22±4

23±5

0.28

RA area, cm2

18±5

15±5

<0.01

E/e’

11±4

11±3

0.68

Deceleration time, ms

253±31

249±61

0.84

CMR

Echocardiogram

Table 3: Electrical remodelling post renal denervation at electrophysiology study

19

Baseline

6 months

P Value

CS 600ms

262±28

276±25

0.33

CS 450ms

245±21

254±20

0.31

HRA 600ms

275±24

255±14

0.06

HRA 450ms

263±24

255±14

0.48

AF induction, %

36

36

1.0

CS 600ms

32±5

27±6

<0.01

CS 450ms

31±6

26±5

<0.01

P-wave duration, ms

122±8

112±8

0.02

Global, m/s

0.98±0.13

1.20±0.16

<0.01

Posterior

0.91±0.20

1.13±0.17

<0.01

Anterior

0.98±0.24

1.28±0.38

0.08

Septal

0.94±0.28

1.21±0.22

0.01

Lateral

0.95±0.14

1.26±0.32

0.04

Global, %

37±14

19±12

0.02

Posterior

39±14

22±15

0.01

Anterior

27±11

9±6

<0.01

Septal

41±18

21±13

<0.01

Lateral

35±13

19±15

<0.01

Global Unipolar, mV

2.8±0.9

2.3±0.6

0.14

Global Bipolar, mV

2.1±0.5

2.0±0.5

0.63

Low voltage, %

10±8

11±8

0.80

Atrial ERP

Conduction time

Conduction velocity

Fractionated Potentials

Voltage

20

Table 4: Changes on Holter monitoring and procedural hemodynamics Baseline

Follow-up

P Value

Heart Rate

68±11

69±8

0.62

APC/ 24 hours

165±168

198±244

0.79

VPC/ 24 hours

41±64

18±28

0.22

NSVT, %

1

1

1.00

AF > 30sec, %

1

0

1.00

Systolic BP, mmHg

163±12

152±21

0.13

Diastolic BP, mmHg

87±13

79±12

0.05

Heart rate, bpm

73±12

66±11

0.13

Trends

Procedural hemodynamics

21

Figure Legend Figure 1: Baseline and follow-up electroanatomical map of the right atrium viewed from a posterior projection. Isochronal activation map (A and C) and voltage map (B and D). The arrow is aligned with the conduction wave-front demonstrating a shortening in the distance between isochrones at follow-up (versus baseline). Brown dots represent fragmented electrograms, light blue dots represent double potentials. IVC: inferior vena cava, SVC: superior vena cava, CS: coronary sinus.

Figure 1: Right atrial electroanatomic maps

A

Conduction

B

Voltage and complex electrograms

SVC

SVC

CS Baseline

IVC

IVC SVC

C

D

6 months

CS CS

SVC

22

Figure 2: 24-hour ambulatory blood pressure changes from baseline to 6 months. SBP systolic blood pressure, DBP diastolic blood pressure

Figure2:

Ambulatory BP p=0.03

230

p<0.01

Ambulatory blood pressure (mmHg)

210 192

190

178

p=0.36

170 152

150

Baseline

p=0.04

141

6 months

130 112

110

107

90

84

80

70 50 Mean SBP

Mean DBP

Max SBP

Max DBP

Figure 3: Global and regional conduction velocity from baseline to follow-up

23

Figure3: 1.8

Global and regional conduction velocity p<0.01

p<0.01

p=0.08

p=0.01

p=0.04

1.6 1.4

m/s

1.2 Baseline Follow up

1 0.8 0.6 0.4 Global CV

Posterior

Anterior

Septal

Lateral

24

Figure 4: Correlations for change in mean (A) and maximum (B) ambulatory 24-hour systolic blood pressure versus change in conduction velocity from baseline to follow-up. The arrow indicates a physiological improvement from baseline to follow-up.

Figure 4: Correlation of delta systolic BP vs delta CV A

B

60 R² = 0.5512

60

40

40

30

30

20

10

20 10 0

0 -0.4

-0.2

R² = 0.6601

50

Delta BP (mmHg)

Dellta BP (mmHg)

50

Max systolic BP

improvement

improvement

Mean systolic BP

0.0

0.2

0.4

0.6

-0.4

-0.2

0

-10

-10

-20

-20 Delta CV (m/sec)

improvement

0.2

0.4

0.6