Experimental Physiology

Transcription

1 1414 Exp Physiol (17) pp Researh Paper Researh Paper Renal denervation dereases suseptibility of the heart to ventriular fibrillation in a anine model of hroni kidney disease Xiaotie Tang, Lang Shi, Xuebin Cui, Yang Yu, Ting Qi, Cheng Chen and Xianglei Tang Department of Nephrology, Puren Hospital of Wuhan, Wuhan University of Siene and Tehnology, Wuhan, China Edited by: Carolyn Barrett Experimental Physiology New Findings What is the entral question of this study? Renal denervation (RDN) has been shown to be effetive and safe, resulting in better ontrol of blood pressure and an improvement in left ventriular hypertrophy in hroni kidney disease () patients. Ventriular arrhythmias and sudden ardia death are ommon auses of death in patients, but previous studies pay almost no attention to the effets of RDN on the risk of ventriular fibrillation assoiated with. What is the main finding and its importane? Renal denervation ould derease suseptibility of the heart to ventriular fibrillation in a anine model. Improvement of left ventriular hypertrophy, sympatheti ativation and inflammation by RDN may be responsible for its benefiial effets. Renal denervation (RDN) has been shown to have therapeuti value in patients with hroni kidney disease (). The aim of this study was to investigate whether RDN ould derease the suseptibility of the heart to ventriular fibrillation in a anine model of. Twenty-one dogs were used. Chroni kidney disease was produed by subtotal nephretomy in 16 dogs with RDN treatment ( + RDN group, n = 8) or sham RDN ( group, n = 8). Another five dogs underwent sham operation and sham RDN to serve as ontrols ( group). Parameters of renal funtion, blood pressure, ehoardiography, ECG, noradrenaline and inflammation were measured at baseline and 6 weeks after the surgial proedure. The ventriular fibrillation threshold (VFT) was determined at the end of the study. Subtotal nephretomy suessfully indued a anine model. When ompared with the group, subtotal nephretomy in the group signifiantly elevated blood pressure; inreased the left ventriular mass, end-diastoli left ventriular internal dimension, left ventriular end-diastoli posterior wall thikness and end-diastoli interventriular septum thikness; prolonged the QT interval, orreted QT interval, the interval from the peak to the end of the T wave (Tp-e) and the orreted Tp-e interval; and inreased the QT dispersion and the Tp-e/QT ratio; dereased the VFT; and inreased the serum onentrations of noradrenaline, C-reative protein and interleukin-6. Renal denervation signifiantly attenuated these hanges indued by. The study demonstrated that RDN ould derease the suseptibility of the heart to ventriular fibrillation in this model. Improvement of left ventriular hypertrophy, sympatheti ativation and inflammation by RDN may be responsible for its benefiial effets. X. Tang and L. Shi ontributed equally to this work. DOI: /EP8637

2 Exp Physiol (17) pp Renal denervation and hroni kidney disease 1415 (Reeived Marh 17; aepted after revision 18 August 17; first published online 21 August 17) Corresponding author X. Tang: Department of Nephrology, Puren Hospital of Wuhan, Wuhan University of Siene and Tehnology, No. 1 Benxi Street, The Fourth Jianshe Road, Qingshan Distrit, Wuhan City, China. slangwh@163.om Introdution Chroni kidney disease () is a hroni, progressive and multifatorial disorder, with an inreased risk of ardiovasular morbidity and mortality (Gansevoort et al. 13; Di Lullo et al. 16). Sudden ardia death (SCD)-related mortality is inreased 14-fold in dialysis patients when ompared with subjets who have normal kidney funtion (Di Lullo et al. 16). Even mild hanges in renal funtion also inrease the risk of SCD, espeially in older populations. It is well reognized that sympatheti nervous system ativation plays a ruial role in disease development, progression and adverse outomes in patients (Ewen et al. 13; Johns, 14). Even in the early stage of, sympatheti overativity is already present and inreases in parallel with the progression of hypertension and impairment of kidney funtion. In reent years, renal denervation (RDN) has emerged as a new approah to derease sympatheti ativity (Krum et al. 9; Shlaih et al. 9; Linz et al. 15). Several linial studies have shown that RDN exerts promising therapeuti effets on left ventriular (LV) hypertrophy (LVH) and blood pressure ontrol in patients with resistant hypertension (Kiuhi et al. 13, 16; Hering et al. 17). However, previous studies have paid almost no attention to the effets of RDN on -assoiated ventriular arrhythmias and SCD. We hypothesized that RDN might provide protetion against -related ventriular fibrillation (VF) through multiple mehanisms, inluding improvement of LVH, sympatheti ativation and inflammation. In the present study, we reated a anine model by subtotal nephretomy and measured the parameters of renal funtion, blood pressure, ehoardiography, ECG, noradrenaline (NA), inflammation and ventriular fibrillation threshold (VFT) to test the hypotheses. Methods Ethial approval Animal studies were reviewed and approved by the Institutional Animal Care and Use Committee of Wuhan University of Siene and Tehnology (permit no ). All proedures onformed to the Guide for the Care and Use of Loratory Animals published by the US National Institutes of Health (NIH Publiation, revised 11). All experiments omply with the priniples and regulations, as desribed by Grundy (15). All efforts were made to minimize the number of animals used and their suffering. At the end of study, all animals were not allowed to reover, and were killed with an overdose of sodium pentobarbital (15 mg kg 1, I.V., Merk, Darmstadt, Germany). Surgial proedure Twenty-one adult male dogs weighing kg were provided by the animal entre in the medial ollege of Wuhan University of Siene and Tehnology. The dogs were anaesthetized with 3 mg kg 1 (I.V.) sodium pentobarbital via the ephali vein, and general anaesthesia was maintained with an additional dose of 2 mg kg 1 (I.V.) at the end of eah hour. The depth of anaesthesia was monitored throughout the experiment by heking heart rate, breathing rate, the orneal reflex and the toe-pinh response. The dogs were intubated and ventilated with a positive pressure respirator. In 16 dogs, bilateral flank inisions were performed to expose the left and right kidneys. Several dorsal and ventral branhes of the left and right renal arteries were ligated to ahieve a target of 5 6% infartion of both the kidneys (estlished by visual verifiation of topial yanosis; Kingma et al. 14). In eight of 16 dogs, both kidneys were surgially denervated by utting all visible nerves of the adventitia of the renal arteries ( + RDN group), and then the arteries were moistened with a % phenol ethanol solution for 1 15 min. Renal denervation was onsidered suessful when eletrial stimulation of eah renal artery did not produe any hanges of blood pressure. Sham denervation was performed in the remaining eight dogs ( group). Another five sham-operated dogs underwent the same anaestheti and surgial protools as the dogs, but the kidneys were only gently manipulated before losure of the inisions ( group). The inisions were losed in layers. Analgesia was provided with butorphanol (.2 mg kg 1, I.M., twie daily, Hengrui Mediine Co., LTD, Jiangsu, China) for 3 days. Animals were housed in a olony room with ontrolled temperature (22 24 C) and a 12 h 12 h dark light yle (light on at 7. h), with food and water provided ad libitum. After 6 weeks follow-up,all dogs were again anaesthetized (same proedure as desribed in the previous paragraph) for measurements and for the seond operation to determine the VFT (as desribed below, in the subsetion Ventriular fibrillation threshold measurements ). The design and time line of the experiments are shown in Fig. 1.

3 1416 X. Tang and others Exp Physiol (17) pp Serum measurements At baseline and at the 6th week of follow-up, before anaesthesia, blood samples were olleted from the ephali vein into an ie-hilled tube, and immediately entrifuged at 231g for 1 min at 4 C, and then stored at C until further assay. Conentrations of blood urea nitrogen, reatinine, haemoglobin, NA, C-reative protein (CRP) and interleukin-6 (IL-6) were examined using enzyme-linked immunosorbent assay kits (Wuhan Beinglay Bioteh Co., Ltd, Hubei, China). Ehoardiography measurements At baseline and at 6th week of follow-up, transthorai Doppler ehoardiography was performed under general anaesthesia, before the operations, with a Vivid 7 system (GE Vingmed Ultrasound, Horten, Norway). Images of long-axis, short-axis and apial four-hamber views were obtained. The end-diastoli left ventriular internal dimension (LVIDd), the left ventriular end-diastoli posterior wall thikness (PWTd) and the end-diastoli interventriular septum thikness (IVSTd) were measured. The left ventriular (LV) mass was alulated from LV linear dimensions using the Devereux formula (Devereux et al. 1986). All ehoardiograms were measured blind to the investigator in tripliate by the same sonographer, and the averages were taken for analysis. Eletroardiographi measurements At baseline and after 6 weeks follow-up, a 12-lead ECG was reorded using adhesive leads, under general anaesthesia before the operations. The QT interval was measured from the beginning of the QRS omplex to the end of the T wave, and orreted for heart rate using the Bazett formula, as QT/(RR).5,whereRRis the R R interval. The QT dispersion (QTd) was defined as the differene between the maximal and minimal QT intervals of the 12 leads. The Tpeak to Tend (Tp-e) interval was defined as the interval from the peak to the end of T wave, and was orreted for heart rate using the Bazett formula, as Tp-e/(RR).5.Theratioof Tp-e/QT was then alulated from these values. The ECG parameters were measured blind to the investigator and by two experts and by averaging three onseutive beats. Ventriular fibrillation threshold measurements At the end of the study, a right thoraotomy was performed under anaesthesia at the fifth interostal spae, followed by a periardiotomy and plaement of one eletrode atheter at the right ventriular apex. The VFT was also measured blind to the investigator and with right ventriular paing using a train of 3 stimuli (3 ms interval), whih was performed at the end of a beat drive train at 3 ms, to ensure that VFT was measured at the same heart rate (Huang et al. 15). The VFT was determined by progressively inreasing the paing voltage in 2 V steps, with a s rest period before the next paing train if no ventriular fibrillation was indued. The VFT was defined as the minimal voltage required to indue sustained VF. One VF ourred, no intervention was adopted to terminate VF. All dogs in whih VF ourred were killed with an overdose of sodium pentobarbital (15 mg kg 1, I.V.). First Operation: plus sham RDN Seond Operation: determination of VFT Group (n = 8) First Operation: plus RDN Seond Operation: determination of VFT Group (n = 8) First Operation: Sham and sham RDN Seond Operation: determination of VFT Group (n = 5) Week End Figure 1. The design and time line of the experiments Abbreviations in this and the following figures:, hroni kidney disease;, ontrol; RDN, renal denervation; UCG, ehoardiography; and VFT, ventriular fibrillation threshold.

4 Exp Physiol (17) pp Renal denervation and hroni kidney disease 1417 Statistial analysis Continuous variles are shown as means ± SD. The differenes between baseline and follow-up measurements were assessed by Student s paired t test. Differenes between groups (exept for VFT) were evaluated with repeated-measures two-way ANOVA followed by Bonferroni test if ANOVA was signifiant. One-way ANOVA followed by Bonferroni test was used to ompare the VFT means between groups. The P values were two-sided and onsidered signifiant when <.5. Statistial analyses were arried out using the SPSS version 18. software pakage (SPSS In., Chiago, IL, USA). Results After 6 weeks follow-up, high-frequeny stimulation of both renal arteries did not indue any hanges in blood pressure in the + RDN group, indiating that renal nerve reinnervation may not our at 6 weeks after RDN in this study. Assessment of The data for body weight, serum reatinine, blood urea nitrogen and haemoglobin are summarized in Tle 1. The baseline values of these indiators were not signifiantly different among the three groups. After 6 weeks follow-up, the body weight and haemoglobin were signifiantly dereased, whereas serum reatinine and blood urea nitrogen were signifiantly inreased in both the group and the + RDN group when ompared with the group, indiating that the anine model was suessfully indued by subtotal nephretomy. Renal denervation attenuates blood pressure elevation Figure 2 shows the hanges in blood pressure in eah group. No signifiant differenes were shown in the baseline blood pressure among the three groups. As a result of, systoli blood pressure, diastoli blood pressure and mean arterial pressure were all signifiantly inreased in the and + RDN groups when ompared with the group. The systoli and diastoli blood pressures and mean arterial pressure were all signifiantly lower in the + RDN group ompared with the group, indiating that RDN attenuated the blood pressure elevation indued by. Renal denervation attenuates LVH As shown in Fig. 3, there was no signifiant differene in the baseline values of LV mass, LVIDd, PWTd and LVSTd A DBP (mmhg) SBP (mmhg) C B MAP (mmhg) Figure 2. Blood pressure analysis Changes of systoli blood pressure (SBP; A), mean arterial pressure (MAP; B) and diastoli blood pressure (DBP; C) in eah group. In this and the following figures, the horizontal blak line and the limits depit mean and SD, respetively. a P <.5 versus baseline; b P <.5 versus group; and P <.5 versus group (a: using Student s paired t test; b and : using repeated-measures two-way ANOVA followed by Bonferroni test).

5 1418 X. Tang and others Exp Physiol (17) pp Tle 1. Assessment of hroni kidney disease group (n = 5) group (n = 8) + RDN group (n = 8) Parameter BS 6W BS 6W BS 6W Body weight (kg) 11. ± ± ± ± ± ±.3 Serum reatinine (μm) 64 ± 14 7 ± 16 6 ± ± ± ± 25 Blood urea nitrogen (mm) 4.2 ± ± ± ± ± ± 2. Haemoglobin (g dl 1 ) 11.5 ± ± ± ± ± ± 1.2 Abbreviations: 6W, after 6 weeks follow-up; BS, baseline;, hroni kidney disease;, ontrol; and RDN, renal denervation. P <.5 versus BS (using Student s paired t test); P <.5 versus group (using repeated-measures two-way ANOVA followed by Bonferroni test). among the three groups. However, 6 weeks after, the LV mass (92.7 ± 25.9 versus 56.7 ± 1.9 g, P <.5), LVIDd (3.1 ± 1.7 versus 26.8 ±.8 mm, P <.5), PWTd (1.6 ± 1.6 versus 8.6 ± 1.2 mm, P <.5) and LVSTd (1.5 ± 1.6 versus 8.6 ± 1.1 mm, P <.5) were all signifiantly inreased in the group when ompared with the group. However, -indued hanges in LV mass (61.4 ± 11.4 versus 92.7 ± 25.9 g, P <.5), LVIDd (28.1 ± 2.1 versus 3.1 ± 1.7 mm, P <.5), PWTd (8.8 ±.7 versus 1.6 ± 1.6 mm, P <.5) and LVSTd (8.6 ±.8 versus 1.5 ± 1.6, P <.5) were signifiantly attenuated by RDN. The LVEF was not signifiantly hanged after 6 weeks follow-up in all three groups. Renal denervation attenuates the hanges of ECG parameters Figure 4A shows representative ECG traes in eah group before and 6 weeks after operations. The values of ECG parameters at baseline showed no signifiant differene among the three groups. After 6 weeks follow-up, the dogs in the group exhibited shorter RR interval, longer QT interval, orreted QT (QT) interval, Tp-e interval A LV mass (g) B LVIDd (mm) C PWTd (mm) 4 D E 16 1 IVSTd (mm) 12 8 LVEF (%) Figure 3. Ehoardiogram parameters Changes of left ventriular (LV) mass (A), the end-diastoli left ventriular internal dimension (LVIDd; B), the left ventriular end-diastoli posterior wall thikness (PWTd; C), the end-diastoli interventriular septum thikness (IVSTd; D) and the left ventriular ejetion fration (LVEF; E) in eah group. a P <.5 versus baseline; b P <.5 versus group; and P <.5 versus group (a: using Student s paired t test; b and : using repeated-measures two-way ANOVA followed by Bonferroni test).

6 Exp Physiol (17) pp Renal denervation and hroni kidney disease 1419 and orreted Tp-e (Tp-e) interval, and larger QTd and Tp-e/QT when ompared with the group. However, ombined with RDN signifiantly attenuated the -indued ECG hanges (Fig. 4B H). Renal denervation attenuates the hanges of VFT The VFT was measured at the end of study (Fig. 5). When ompared with the group, the VFT was signifiantly lower in the group (1.1 ± 4.2 versus.1 ± 5.2 V, P <.5) and the + RDN group (16.2 ± 4. versus.1 ± 5.2 V, P <.5). However, the VFT was even lower in the group when ompared with the + RDN group (1.1 ± 4.2 versus 16.2 ± 4. V, P <.5). Renal denervation attenuates the hanges of serum NA, CRP and IL-6 As shown in Fig. 6, the serum onentrations of NA, CRP and IL-6 at baseline were not signifiantly different among the three groups. After 6 weeks follow-up, serum NA, CRP and IL-6 onentrations were markedly inreased in the and + RDN groups. However, when ompared with the group, RDN signifiantly dereased the serum NA, CRP and IL-6. Disussion The main findings of this study were as follows: (i) subtotal nephretomy-indued was assoiated with an inreased risk of ventriular fibrillation (longer A B R-R interval (ms) D C QT interval (ms) E RDN QT interval (ms) 4 35 QTd 4 F Tp-e interval (ms) 6 4 G Tp-e interval (ms) H Tp-e/QT Figure 4. Eletroardiogram parameters A shows representative ECG traes in eah group before and 6 weeks after operations. B H shows the hanges of R R interval (B), QT interval (C), orreted QT (QT) interval (D), QT dispersion (QTd; E), Tp-e interval (F), orreted Tp-e (Tp-e) interval (G) and Tp-e/QT ratio (H). a P <.5 versus baseline; b P <.5 versus group; and P <.5 versus group (a: using Student s paired t test; b and : using repeated-measures two-way ANOVA followed by Bonferroni test).

7 14 X. Tang and others Exp Physiol (17) pp QT interval, QT interval, Tp-e interval and Tp-e interval, larger QTd and Tp-e/QT, and lower VFT) whih, however, was attenuated by RDN; and (ii) improvement of LVH, sympatheti ativation and inflammation may be responsible for the protetive effets of RDN against ventriular fibrillation. The findings suggest that RDN might be a promising approah for prevention of malignant ventriular arrhythmias and SCD in patients. Chroni kidney disease-related ventriular fibrillation and SCD and the therapeuti role of RDN It is well aepted that ardiovasular ompliations are the leading auses of deaths in patients with (Gansevoort et al. 13). Ventriular tahyarrhythmias are ommon in patients, with a progressive inrease in the inidene of SCD with loss of renal funtion (Green et al. 11). Data from the US Renal Data System Annual Data Report suggested that SCD aounts for 25% of all-ause mortality. In the present study, after 6 weeks follow-up, subtotal nephretomy indued signifiant inreases in serum reatinine and blood urea nitrogen, and dereases in body weight and serum haemoglobin, in both the group and the + RDN group, indiating that the model was suessfully reated. To evaluate the risk of ventriular fibrillation, we measured the ECG parameters and VFT. It has been shown that prolonged repolarization or inreased dispersion of repolarization might predispose patients to malignant ventriular arrhythmias (Watane et al. 4). Previously, QT and QT intervals and QT dispersion have been used to evaluate myoardial repolarization, and thus estimate the arrhythmogeni risk to the heart (Topilski et al. 7). Besides, Tp-e and Tp-e intervals and the Tp-e/QT ratio have been widely used to evaluate the risk of malignant ventriular arrhythmias (Castro Hevia et al. 6; Alizade et al. 17). Dobre et al. (12) examined the ECG parameters in 1192 VFT (V) 3 1 a Figure 5. Ventriular fibrillation threshold (VFT) in eah group a P <.5 versus group; and b P <.5 versus group (using one-way ANOVA followed by Bonferroni test). partiipants with, and their results showed that 68.5% of these patients had major (38.8%) or minor (29.7%) ECG normalities, inluding prolongation of the QT and QT intervals, whih are independent risk markers for ardiovasular death. In the present study, we also found that the dogs in the group exhibited prolonged QT, QT, Tp-e and Tp-e intervals and inreased QTd and Tp-e/QT ratio, suggesting an inreased risk of VF in dogs. In addition, the VFT was signifiantly dereased in the group when ompared with the group, further indiating an inreased suseptibility of the heart to VF in the dogs. Renal denervation for the treatment of resistant hypertension has been met with great enthusiasm by ardiologists during reent years (Krum et al. 9; Linz et al. 17). Although the Sympliity HTN-3 trial failed to demonstrate a benefiial effet of RDN in patients with resistant hypertension (Bhatt et al. 14), several fators, suh as inreased plaebo effet, inadequate denervation or an inility to test for suessful denervation, might have ontributed to the negative results. In addition to resistant hypertension, RDN might have great potential in the treatment of other diseases assoiated with exessive sympatheti ativation (Bohm et al. 14). Reently, several linial studies have suggested that RDN also exerts benefiial effets in patients. Kiuhi et al. (13) performed RDN in 24 patients. After 6 months follow-up, the results showed that RDN was effetive and safe, resulting in a better ontrol of blood pressure, a short-term improvement in renal funtion, and redued albuminuria. In a study with a longer follow-up period (24 months), Hering et al. (17) also showed that RDN ould halt the progression of and improve blood pressure ontrol in patients. However, in these studies almost no attention has been paid to the effets of RDN on -assoiated ventriular fibrillation and SCD. In normal healthy dogs, Huang et al. (14) showed that RDN signifiantly inreased the ventriular eletrophysiologial stility. Basi and linial studies have also shown that RDN an derease ventriular tahyarrhythmias in different linial settings (Linz et al. 13; Remo et al. 14; Ukena et al. 16). In the present study, we showed that RDN signifiantly attenuated -indued VFT redution and ECG hanges, inluding QT, QT, Tp-e and Tp-e interval prolongation and QTd and Tp-e/QT ratio elevations. The present study indiates that RDN is aple of proteting against VF in this model. Possible mehanisms underlying the protetive effets of RDN against VF There is a progressive inrease in the prevalene of LVH, and LV mass inreases when the estimated glomerular filtration rate dereases (Cerasola et al. 1). At the start

8 Exp Physiol (17) pp Renal denervation and hroni kidney disease 1421 of dialysis, LVH is present in >75% of patients with advaned (Foley et al. 1) and has been linked to an inreased risk of sustained ventriular arrhythmias and predisposition to SCD (Haider et al. 1998). The ehoardiogram results in the present study showed that LV mass, LVIDd, PWTd and LVSTd were all signifiantly inreased in the group when ompared with the group, suggesting a progression of LVH, whih might ontribute to the inreased risk of VF. However, + RDN signifiantly dereased LV mass, LVIDd, PWTd and LVSTd when ompared with the group, indiating that RDN attenuated the progression of LVH in dogs, onsistent with a reent linial study by Kiuhi et al. (16), who showed that RDN was effetively and safely le to improve the LVH in refratory hypertensive patients. Therefore, improvement of -indued LVH by RDN might ontribute to its anti-vf ation. Inreased sympatheti ativation is found in the initial linial stages of and greatly ontributes to the inreased ardiovasular risk (Ewen et al. 13). Afferent signals from renal sensory nerves ould ativate the sympatheti nervous system (Hering et al. 13), resulting in NA release, whih is involved in the pathogenesis of LVH and reates a substrate for SCD. In the present study, we found that blood pressure and serum NA were signifiantly inreased in the group, indiating a signifiant inrease in sympatheti ativity. Previous studies have also found that patients have a withdrawal in parasympatheti modulation of the heart rate in onjuntion with an inrease in sympatheti input to the sinoatrial node (Chan, 8). In the present study, the R R interval was markedly dereased in dogs, onsistent with previous studies (Liang et al. 14). However, + RDN signifiantly attenuated the elevations of blood pressure and NA and inreased the R R interval when ompared with the group, indiating that inhibition of sympatheti ativation might be another mehanism underlying the protetive effets of RDN against VF. In addition to LVH and sympatheti ativation, inflammation also has been found to be assoiated with SCD independently of traditional ardiovasular risk fators (Di Lullo et al. 16). Serum pro-inflammatory ytokine onentrations inrease with a deline in renal funtion. Both CRP and IL-6 have been assoiated with ventriular arrhythmias (Parekh et al. 8). The inflammatory proess ould promote ventriular remodelling, affeting ventriular ondution, ausing a delay in repolarization and thus leading to ventriular arrhythmias and SCD (Parekh et al. 8). In the present study, dogs had higher serum onentrations of A Serum NA (pg/ml) B CRP (pg/ml) 1 C IL-6 (pg/ml) Figure 6. Serum onentrations of noradrenaline (NA; A), C-reative protein (CRP; B) and interleukin-6 (IL-6; C) a P <.5 versus baseline; b P <.5 versus group; and P <.5 versus group (a: using Student s paired t test; b and : using repeated-measures two-way ANOVA followed by Bonferroni test).

9 1422 X. Tang and others Exp Physiol (17) pp CRP and IL-6, whih were attenuated by RDN. Therefore, suppression of the inflammatory response by RDN might also ontribute to its anti-vf ation. Study limitations There are several limitations in this study. First, the pathophysiologial proess and severity of renal impairment in our model are not ompletely in aord with linial situations. Further, linial studies might be designed to evaluate whether atheter-based RDN ould attenuate the ECG parameters and prevent ventriular arrhythmias and SCD in patients. Seond, this study has a small sample size, whih might redue power to detet small but linially signifiant effets. Third, we performed ombined surgial and hemial RDN, whih might be more omplete ompared with the atheter-based RDN. However, it has been shown that NA redution in the kidney was similar between surgial RDN and atheter-based RDN (Mahfoud et al. 13). Fourth, there were no ourrenes of spontaneous VF in this study, beause the dogs were followed up for only 6 weeks. Besides, we measured only the VFT but did not determine other invasive ventriular eletrophysiologial parameters, suh as ventriular effetive refratory period and ation potential duration. However, the non-invasive ECG parameters were analysed in the present study to reflet the risk of ventriular arrhythmias. Fifth, further studies might be performed to assess the additional values of RDN in ombination with angiotensin-onverting enzyme inhibitors, β-adrenoeptor blokers, or both, in the management of patients. Conlusions The present study demonstrated that RDN ould derease suseptibility of the heart to VF in a anine model of indued by subtotal nephretomy. Improvement of LVH, sympatheti ativation and inflammation by RDN may be responsible for its benefiial effets. Renal denervation may be a promising approah for prevention of malignant ventriular arrhythmias and SCD in patients. Referenes Alizade E, Yesin M, YaziioğluMV, KaraayvazEB, Atii A, Arslan S, Avi A, Aar G, Taki M, Izi S & Pala S (17). Evaluation of Tp-e interval, Tp-e/QT ratio, and Tp-e/QT ratio in patients with asymptomati arrhythmogeni right ventriular ardiomyopathy. Ann Noninvasive Eletroardiol 22, e Bhatt DL, Kandzari DE, O Neill WW, D Agostino R, Flak JM, Katzen BT, Leon MB, Liu M, Mauri L, Negoita M, Cohen SA, Oparil S, Roha-Singh K, Townsend RR & Bakris GL (14). A ontrolled trial of renal denervation for resistant hypertension. NEnglJMed37, Bohm M, Linz D, Ukena C, Esler M & Mahfoud F (14). Renal denervation for the treatment of ardiovasular high risk-hypertension or beyond? Cir Res 115,4 49. Castro Havia HJ, Antzelevith C, Tornés Bárzaga F, Dorantes Sánhez M, Dortiós Balea F, Zayas Molina R, Quiñones Pérez MA & Fayad Rodríguez Y (6). Tpeak-Tend and Tpeak-Tend dispersion as risk fators for ventriular tahyardia/ventriular fibrillation in patients with the Brugada syndrome. J Am Coll Cardiol 47, Cerasola G, Nardi E, Mule G, Palermo A, Cusimano P, Guarneri M, Arsena R, Giammarresi G, Carola FA & Cottone S (1). Left ventriular mass in hypertensive patients with mild-to-moderate redution of renal funtion. Nephrology (Carlton) 15, Chan CT (8). Heart rate variility in patients with endstage renal disease: an emerging preditive tool for sudden ardia death? Nephrol Dial Transplant 23, Devereux RB, Alonso DR, Lutas EM, Gottlieb GJ, Campo E, Sahs I & Reihek N (1986). Ehoardiographi assessment of left ventriular hypertrophy: omparison to neropsy findings. Am J Cardiol 57, Di Lullo L, Rivera R, Barbera V, Bellasi A, Cozzolino M, Russo D, De Pasalis A, Banerjee D, Floari F & Rono C (16). Sudden ardia death and hroni kidney disease: from pathophysiology to treatment strategies. Int J Cardiol 217, Dobre M, Brateanu A, Rashidi A & Rahman M (12). Eletroardiogram normalities and ardiovasular mortality in elderly patients with. Clin J Am So Nephrol 7, Ewen S, Ukena C, Linz D, Shmieder RE, Böhm M & Mahfoud F (13). The sympatheti nervous system in hroni kidney disease. Curr Hypertens Rep 15, Foley RN, Curtis BM, Randell EW & Parfrey PS (1). Left ventriular hypertrophy in new hemodialysis patients without symptomati ardia disease. Clin J Am So Nephrol 5, Gansevoort RT, Correa-Rotter R, Hemmelgarn BR, Jafar TH, Heerspink HJ, Mann JF, Matsushita K & Wen CP (13). Chroni kidney disease and ardiovasular risk: epidemiology, mehanisms, and prevention. Lanet 382, Green D, Roberts PR, New DI & Kalra PA (11). Sudden ardia death in hemodialysis patients: an in-depth review. Am J Kidney Dis 57, Grundy D (15). Priniples and standards for reporting animal experiments in The Journal of Physiology and Experimental Physiology. Exp Physiol 1, Haider AW, Larson MG, Benjamin EJ & Levy D (1998). Inreased left ventriular mass and hypertrophy are assoiated with inreased risk for sudden death. JAmColl Cardiol 32, Hering D, Esler MD & Shlaih MP (13). Chroni kidney disease: role of sympatheti nervous system ativation and potential benefits of renal denervation. EuroIntervention 9(Suppl R), R127 R135. Hering D, Marusi P, Duval J, Sata Y, Head GA, Denton KM, Burrows S, Walton AS, Esler MD & Shlaih MP (17). Effet of renal denervation on kidney funtion in patients with hroni kidney disease. Int J Cardiol 232,

10 Exp Physiol (17) pp Renal denervation and hroni kidney disease 1423 HuangB,YuL,HeB,LuZ,WangS,HeW,YangK,LiaoK, Zhang L & Jiang H (14). Renal sympatheti denervation modulates ventriular eletrophysiology and has a protetive effet on ishaemia-indued ventriular arrhythmia. Exp Physiol 99, Huang J, Qian J, Yao W, Wang N, Zhang Z, Cao C, Song B & Zhang Z (15). Vagus nerve stimulation reverses ventriular eletrophysiologial hanges indued by hypersympatheti nerve ativity. Exp Physiol 1, Johns EJ (14). The neural regulation of the kidney in hypertension and renal failure. Exp Physiol 99, Kingma JJ, Simard D, Voisine P & Rouleau JR (14). Impat of hroni kidney disease on myoardial blood flow regulation in dogs. Nephron Exp Nephrol 126, Kiuhi MG, Maia GL, de Queiroz Carreira MA, Kiuhi T, Chen S, Andrea BR, Graiano ML & Lugon JR (13). Effets of renal denervation with a standard irrigated ardia lation atheter on blood pressure and renal funtion in patients with hroni kidney disease and resistant hypertension. Eur Heart J 34, Kiuhi MG, Mion DJ, Graiano ML, de Queiroz Carreira MA, Kiuhi T, Chen S & Lugon JR (16). Proof of onept study: improvement of ehoardiographi parameters after renal sympatheti denervation in refratory hypertensive patients. Int J Cardiol 7, Krum H, Shlaih M, Whitbourn R, Sobotka PA, Sadowski J, Bartus K, Kapelak B, Walton A, Sievert H, Thambar S, Abraham WT & Esler M (9). Catheter-based renal sympatheti denervation for resistant hypertension: a multientre safety and proof-of-priniple ohort study. Lanet 373, Liang Z, Liu LF, Chen XP, Shi XM, Guo HY, Lin K, Guo JP, Shan ZL & Wang YT (14). Estlishment of a model of renal impairment with mild renal insuffiieny assoiated with atrial fibrillation in anines. PLoS One 9, e Linz D, Hohl M, Shütze J, Mahfoud F, Speer T, Linz B, Hübshle T, Juretshke HP, Dehend R, Geisel J, Rütten H & Böhm M (15). Progression of kidney injury and ardia remodeling in obese spontaneously hypertensive rats: the role of renal sympatheti innervation. Am J Hypertens 28, Linz D, Mania G, Mahfoud F, Narkiewiz K, Ruilope L, Shlaih M, Kindermann I, Shmieder RE, Ewen S, Williams B&Böhm; Global SYMPLICITY Registry Investigators (17). Renal artery denervation for treatment of patients with self-reported obstrutive sleep apnea and resistant hypertension: results from the Global SYMPLICITY Registry. JHypertens35, Linz D, Wirth K, Ukena C, Mahfoud F, Poss J, Linz B, Böhm M & Neuberger HR (13). Renal denervation suppresses ventriular arrhythmias during aute ventriular ishemia in pigs. Heart Rhythm 1, Mahfoud F, Lüsher TF, Andersson B, Baumgartner I, Cifkova R, Dimario C, Doevendans P, Fagard R, Fajadet J, Komajda M, Lefèvre T, Lotan C, Sievert H, Volpe M, Widimsky P, Wijns W, Williams B, Windeker S, Witkowski A, Zeller T & Böhm M; European Soiety of Cardiology (13). Expert onsensus doument from the European Soiety of Cardiology on atheter-based renal denervation. Eur Heart J 34, Parekh RS, Plantinga LC, Kao WH, Meoni LA, Jaar BG, Fink NE, Powe NR, Coresh J & Klag MJ (8). The assoiation of sudden ardia death with inflammation and other traditional risk fators. Kidney Int 74, Remo BF, Preminger M, Bradfield J, Mittal S, Boyle N, Gupta A, Shivkumar K, Steinberg JS & Dikfeld T (14). Safety and effiay of renal denervation as a novel treatment of ventriular tahyardia storm in patients with ardiomyopathy. Heart Rhythm 11, Shlaih MP, Sobotka PA, Krum H, Lambert E & Esler MD (9). Renal sympatheti-nerve lation for unontrolled hypertension. NEnglJMed361, Topilski I, Rogowski O, Rosso R, Justo D, Copperman Y, Glikson M, Belhassen B, Hohenberg M & Viskin S (7). The morphology of the QT interval predits torsade de pointes during aquired bradyarrhythmias. J Am Coll Cardiol 49, Ukena C, Mahfoud F, Ewen S, Bollmann A, Hindriks G, Hoffmann BA, Linz D, Musat D, Pavliek V, Sholz E, Thomas D, Willems S, Böhm M & Steinberg JS (16). Renal denervation for treatment of ventriular arrhythmias: data from an International Multienter Registry. Clin Res Cardiol 15, Watane N, Kobayashi Y, Tanno K, Miyoshi F, Asano T, Kawamura M, Mikami Y, Adahi T, Ryu S, Miyata A & Katagiri T (4). Transmural dispersion of repolarization and ventriular tahyarrhythmias. JEletroardiol37, 191. Additional information Competing interests None delared. Author ontributions Coneption or design of the work: Xiaotie Tang and L.S. Aquisition, analysis or interpretation of data for the word: X.C., Y.Y., T.Q., C.C. and Xianglei Tang. Drafting the work or revising it ritially for important intelletual ontent: Xiaotie Tang, L.S., X.C., Y.Y., T.Q., C.C. and Xianglei Tang. All authors approved the final version of the manusript and agree to be aountle for all aspets of the work in ensuring that questions related to the auray or integrity of any part of the work are appropriately investigated and resolved. All persons designated as authors qualify for authorship, and all those who qualify for authorship are listed. Funding This study was supported by Health Bureau Clinial Mediine Researh Projet of Wuhan (WX13C32).