Listen to this manuscript’s

audio summary by

JACC Editor-in-Chief

Dr. Valentin Fuster.

J O U R N A L O F T H E AM E R I C A N C O L L E G E O F C A R D I O L O G Y V O L . 6 9 , N O . 8 , 2 0 1 7

ª 2 0 1 7 B Y T H E AM E R I C A N C O L L E G E O F C A R D I O L O G Y F O U N D A T I O N

P U B L I S H E D B Y E L S E V I E R

I S S N 0 7 3 5 - 1 0 9 7 / $ 3 6 . 0 0

h t t p : / / d x . d o i . o r g / 1 0 . 1 0 1 6 / j . j a c c . 2 0 1 6 . 1 2 . 0 1 4

Catheter-Based Renal DenervationExacerbates Blood Pressure FallDuring Hemorrhage

Reetu R. Singh, PHD,a Varsha Sajeesh, BSCHONS,a Lindsea C. Booth, PHD,b Zoe McArdle, BSCHONS,a Clive N. May, PHD,b

Geoffrey A. Head, PHD,c Karen M. Moritz, PHD,d Markus P. Schlaich, MD,c,e Kate M. Denton, PHDa

ABSTRACT

Fro

Me

Vic

Un

Un

(#1

(#1

Me

for

ho

ho

rel

Ma

BACKGROUND Clinical trials applying catheter-based radiofrequency renal denervation (RDN) demonstrated a

favorable safety profile with minimal acute or procedural adverse events. Whether ablation of renal nerves adversely

affects compensatory responses to hemodynamic challenge has not been extensively investigated.

OBJECTIVES The aim of this study was to examine the effect of RDN on mean arterial pressure, renal function, and the

reflex response to hemorrhage in sheep with normotension (control) or with hypertensive chronic kidney disease (CKD).

METHODS Sheep underwent RDN (control-RDN, n ¼ 8; CKD-RDN, n ¼ 7) or sham procedures (control-intact, n ¼ 6;

CKD-intact, n ¼ 7). Response to hemorrhage (20% loss of blood volume), including plasma renin activity, was assessed at

2 and 5 months post-procedure.

RESULTS RDN caused a complete reversal of hypertension and improved renal function in CKD-RDN sheep (p < 0.0001

for 2 and 5 months vs. pre-RDN). In response to hemorrhage, mean arterial pressure fell in all groups, with the fall being

greater in the RDN than the intact group (2-month fall in mean arterial pressure: control-intact, �10 � 1 mm Hg; control-

RDN, �15 � 1 mm Hg; p < 0.05; CKD-intact, �11 � 3 mm Hg; CKD-RDN, �19 � 9 mm Hg; p < 0.001). Hemorrhage

increased heart rate and plasma renin activity in intact sheep, but these responses were significantly attenuated in

control-RDN and CKD-RDN animals. Responses to hemorrhage were remarkably similar at 2 and 5 months post-RDN,

which suggests that nerve function had not returned within this time frame.

CONCLUSIONS In hypertensive CKD sheep, RDN reduced blood pressure and improved basal renal function but

markedly compromised compensatory hemodynamic responses to hemorrhage. Therefore, the capacity to

respond to a physiological challenge to body fluid homeostasis may be compromised following RDN.

(J Am Coll Cardiol 2017;69:951–64) © 2017 by the American College of Cardiology Foundation.

T reating human hypertension using radiofre-quency catheter-based renal denervation(RDN) was met with great enthusiasm

following the promising results of Symplicity HTN-1

m the aCardiovascular Program, Monash Biomedicine Discovery Institute

lbourne, Victoria, Australia; bThe Florey Institute of Neuroscience and Me

toria, Australia; cBaker IDI Heart and Diabetes Institute, Melbourne, Vict

iversity of Queensland, Brisbane, Queensland, Australia; and the eSchool

it, University of Western Australia, Perth, Western Australia, Australia

046594) and fellowships to Prof. Denton (#1041844), Prof. Moritz (#10

054619) from the National Health Research Council of Australia. Prof.

dtronic, and his laboratories have received research funding fromMedtron

Abbott (formerly Solvay) Pharmaceuticals, Boehringer Ingelheim, Novart

noraria and travel support from Abbott, Boehringer Ingelheim, Servier

noraria and travel support for presentations from Medtronic. All other au

evant to the contents of this paper to disclose.

nuscript received October 16, 2016; revised manuscript received Decemb

(1) and HTN-2 (2) trials, but this was curtailed byHTN-3 (3), which failed to demonstrate a blood pres-sure (BP)–lowering effect beyond that of a sham con-trol group. However, in the more recent DENERHTN

and Department of Physiology, Monash University,

ntal Health, The University of Melbourne, Parkville,

oria, Australia; dSchool of Biomedical Sciences, The

of Medicine and Pharmacology–Royal Perth Hospital

. This research was supported by a project grant

78164), Prof. Schlaich (#1080404), and Dr. Booth

Schlaich is an investigator in studies sponsored by

ic. Prof. Schlaich serves on scientific advisory boards

is Pharmaceuticals, and Medtronic; and has received

, Novartis, and Medtronic. Prof. May has received

thors have reported that they have no relationships

er 6, 2016, accepted December 13, 2016.

ABBR EV I A T I ON S

AND ACRONYMS

BP = blood pressure

CKD = chronic kidney disease

DBP = diastolic blood pressure

GFR = glomerular filtration

rate

MAP = mean arterial pressure

PRA = plasma renin activity

RBF = renal blood flow

RDN = renal denervation

RVR = renal vascular resistance

SBP = systolic blood pressure

TPR = total peripheral

resistance

UFR = urine flow rate

UNaV = urinary sodium

excretion

Singh et al. J A C C V O L . 6 9 , N O . 8 , 2 0 1 7

Renal Denervation Impairs Responses to Hemorrhage F E B R U A R Y 2 8 , 2 0 1 7 : 9 5 1 – 6 4

952

(Renal Denervation for Hypertension) trial,though without sham control, RDN plus anti-hypertensive medication led to a greaterdecrease in BP compared with antihyperten-sive medication use alone (4). Sympatheticoveractivity is implicated in the pathogenesisof hypertension and chronic kidney disease(CKD), and these conditions commonlycoexist (5). Achieving BP control is the mostimportant avenue for minimizing adverserenal outcomes in patients with CKD (6),providing a basis for RDN as an interventionfor hypertensive CKD (5). In patients withconcomitant resistant hypertension and mod-erate to severe CKD, RDN was demonstratedto be safe and associated with reduced BP(7,8) and improved renal function (8).

On the basis of sound underlying patho-physiology (5), and the acknowledged limita-

tions of HTN-3 (9), RDN research continues tounequivocally determine whether it is an effectivetreatment for hypertension (10). Clinical studies arenow outpacing experimental studies, which may be ofconcern because of the limited information availableas to the functional impact of RDN for homeostasis.Although the procedure clearly causes incompleteRDN (11), it still may be associated with adverse con-sequences, specifically with regard to a limited abilityto respond to physiological challenges. For example,hemorrhage causes reflex activation of the sympa-thetic nerves (12), and lack of renal sympathetic acti-vation may affect restoration of BP in the short termfollowing blood loss. This is not a hypothesis that canbe systematically examined in humans.

SEE PAGE 965

In this study, we used our established model ofhypertension and CKD in sheep (13) with demon-strated evidence of renal sympathetic overactivityand hypothesized that RDN would reduce resting BPin hypertensive CKD without major effects on renalfunction. Additionally, we hypothesized that cardio-vascular and renal responses to hypotensive hemor-rhage (20% loss of blood volume) would be impairedin animals subjected to RDN. Thus, we examined: 1)resting BP and renal function; and 2) cardiovascularand renal changes in response to hemorrhage at 2 and5 months post-RDN in animals undergoing RDN or asham procedure.

METHODS

Experiments were approved by the animal ethicscommittee of Monash University and performed in

accordance with the guidelines of the National Healthand Medical Research Council of Australia. The OnlineAppendix contains additional procedural details. Hy-pertension with CKD was induced by performing uni-lateral nephrectomy in the sheep fetus (day 100 of 150-day gestation) in which the left renal artery, vein, andureter were ligated and the left kidney excised (CKDgroup, n ¼ 14). To generate a normotensive controlgroup, a sham surgical procedure was performed(control, n ¼ 14) as previously described (13). The fullsurgical procedure produces low-renin hypertension,with about a 25% reduction in glomerularfiltration rate(GFR) at 6 months of age (14).

At 10 months of age, animals underwent either adenervation procedure (a total of 6 2-min radio-frequency ablations per artery) or the sham RDNprocedure, as previously described (15), resulting in4 groups: control-RDN (n ¼ 8), CKD-RDN (n ¼ 7),control-intact (n ¼ 6), and CKD-intact (n ¼ 7).

At 6 (pre-RDN), 12 (2 months post-RDN), and 15(5 months post-RDN) months of age, BP (systolicblood pressure [SBP], diastolic blood pressure [DBP],and mean BP) and heart rate were measured via anindwelling carotid artery catheter (14). These mea-surements were acquired over a 72-h period and theaverage is reported as basal mean arterial pressure(MAP) and heart rate.

At these same ages, basal GFR and renal plasmaflow were examined via clearance of chromium-51ethylenediaminetetraacetic acid (dose: 15 mCi/h) andpara-aminohippuric acid (dose: 750 mg/h), respec-tively, over a 4-h period. During this period, urinewas collected at 20-min intervals, with an arterialblood sample (3 ml) taken at the midpoint. Additional3-ml blood samples were collected at hourly intervalsfor analysis of plasma renin activity (PRA) via radio-immunoassay as described previously (14). Renalblood flow (RBF), renal vascular resistance (RVR),filtration fraction, and urinary sodium excretion(UNaV) were calculated as detailed previously (16).This basal experiment served as the time control forthe hemorrhage experiment.

Response to hemorrhage was examined at 2 and5 months post-RDN or sham procedure. Prior tohemorrhage experiments, blood volume was deter-mined via the clearance of 250-kDa fluoresceinisothiocyanate-dextran as previously described (14).

Following a 1-h baseline period, 20% blood volumewas withdrawn over a 15-min period, with variablesmeasured for a further 180 min to assess recovery.MAP, heart rate, urine flow rate (UFR), and UNaVwere measured continuously. GFR and RBF weremeasured during the baseline period and between 40and 180 min post-hemorrhage, once steady state of

TABLE 1 Cardiovascular and Renal Variables

Control (n ¼ 14) CKD (n ¼ 14) p Value

MAP, mm Hg 80 � 2 87 � 1 <0.0001

Heart rate, beats/min 86 � 3 87 � 4 NS

GFR, ml/min/bw* 2.4 � 0.1 1.7 � 0.1 <0.0001

RBF, ml/min/bw* 15.9 � 1.2 12.1 � 0.6 <0.0001

RVR, mm Hg/ml/min/bw* 4.9 � 0.3 7.2 � 0.4 <0.0001

Filtration fraction 0.23 � 0.20 0.21 � 0.40 0.001

UFR, ml/min/bw* 0.024 � 0.007 0.027 � 0.007 NS

UNaV, mmol/min/bw* 1.37 � 0.30 0.80 � 0.10 <0.001

Plasma sodium, mmol/l 139 � 3 138 � 4 NS

Hematocrit 0.36 � 0.07 0.35 � 0.07 NS

PRA, ng/ml/h 1.5 � 0.5 1.0 � 0.4 0.007

Values are mean � SD. *All renal variables are corrected for body weight (bw).

GFR ¼ glomerular filtration rate; MAP ¼ mean arterial pressure; PRA ¼ plasmarenin activity; RBF ¼ renal blood flow; RVR ¼ renal vascular resistance;UFR ¼ urine flow rate; UNaV ¼ urinary sodium excretion.

J A C C V O L . 6 9 , N O . 8 , 2 0 1 7 Singh et al.F E B R U A R Y 2 8 , 2 0 1 7 : 9 5 1 – 6 4 Renal Denervation Impairs Responses to Hemorrhage

953

the clearance markers was reestablished. PRA wasmeasured at baseline, at the end of the 15-min hem-orrhage, and each hour thereafter.

STATISTICAL ANALYSIS. Values are presented asmean � SD. Statistical analysis was performed usingPrism 6 for Windows (GraphPad Software, La Jolla,California), with the level of significance set atp # 0.05. Baseline effects of RDN or the sham pro-cedure were analyzed using an unpaired Student ttest. In response to hemorrhage, renal and cardio-vascular variables in the control or CKD groups wereanalyzed by repeated-measures analysis of variancewith factors group (intact or RDN) and time and theirinteraction, with data analyzed separately at 2 and5 months post-procedure, followed, where appro-priate, by Sidak or Bonferroni post hoc analysis.

RESULTS

All lambs were born at 150 � 1 day of gestation. Bodyweight was not different between the groups at anyage of study (Online Appendix).

At baseline, MAP was significantly higher in CKDsheep compared with controls, but heart rate wassimilar between the groups (Table 1). CKD sheep hadsignificantly lower GFR, RBF, filtration fraction,UNaV, and PRA but higher RVR compared with con-trol counterparts (Table 1). UFR, plasma sodium, andhematocrit were not different between the groups(Table 1).

MAP increased with age in control-intact and CKD-intact sheep but not in control-RDN sheep (Figure 1A).In CKD sheep, when compared with pre-RDN, RDNcaused a decrease in MAP at 2 months (DMAP �6.2 �0.6 mm Hg) and 5 months (DMAP �6.3 � 0.6 mm Hg)post-RDN (p < 0.0001 for both) (Figure 1B). This wasassociated with a similar decrease in SBP and DBP(not shown). Following RDN, heart rate declinedsignificantly in the control and CKD groups at 2 and5 months, respectively (Figures 1C and 1D). RDNreduced basal PRA in the control and CKD groups(Figures 1E and 1F).RENAL FUNCTION. GFR and RBF were not affectedby RDN in control sheep, but both increased in theCKD-RDN sheep (p < 0.001 for both) (Figures 2A to 2D)at 5 months post-RDN. RVR was unaffected in controlsheep (Figure 2E) and significantly increased in theCKD-intact sheep with age (Figure 2F), but in the CKD-RDN sheep, significant decreases in RVR wereobserved at 2 and 5 months post-procedure(Figure 2F). Filtration fraction did not change withage or RDN in any group (Figures 2G and 2H). UFR andUNaV increased following RDN in both groups (OnlineAppendix).

HEMORRHAGE. There was no difference in bloodvolume or the percentage and rate of blood with-drawal between the groups at any age (OnlineAppendix). The response to hemorrhage wasremarkably similar at 2 and 5 months post-procedure.Therefore, the following text does not differentiatebetween the 2 time points; the responses at both timepoints post-procedure are presented in Figures 3 to 5.Data for plasma sodium and hematocrit are providedin the Online Appendix.

Changes in BP and heart rate during hemorrhagewere not different between the intact-control andintact-CKD groups. During the blood-loss period, MAPfell in all groups, but the fall was greater in the RDNgroups compared with their intact counterparts(Figures 3A and 3B). The fall inMAPwas associatedwithdecreases in both SBP and DBP. The fall in SBP wassimilar between RDN and intact groups (Figures 3C and3D), whereas the DBP drop was greater in the RDNversus intact counterparts (Figures 3E and 3F). Heartrate increased in response to hemorrhage in the intactgroups, but this response was significantly attenuatedin the RDN groups (Figures 3G and 3H).

Following hemorrhage, the maximal decreases inMAP and DBP from baseline were greater in both RDNgroups compared with intact counterparts, whereasthe fall in SBP was similar in all groups (Table 2). Inthe control-intact sheep at both ages, MAP wasrestored to baseline levels during the recovery periodafter hemorrhage, but not in the control-RDN group(Figure 4A). In both CKD groups, MAP was not fullyrestored to baseline levels during the post-hemorrhage period, though the recovery was greaterin the CKD-intact than CKD-RDN sheep (Figure 4B).

The maximal increase in heart rate in response tohemorrhage was significantly less in the RDN groups

Singh et al. J A C C V O L . 6 9 , N O . 8 , 2 0 1 7

Renal Denervation Impairs Responses to Hemorrhage F E B R U A R Y 2 8 , 2 0 1 7 : 9 5 1 – 6 4

954

compared with the respective intact groups (Table 2,Figures 4C and 4D). Heart rate was not restored tobaseline in any group during the post-hemorrhageperiod (Figures 4C and 4D).

At both 2 and 5 months, in response to hemor-rhage, PRA increased significantly in both intactgroups, but this response was significantly attenuatedby RDN in both control and CKD groups (Table 2,Figures 4E and 4F). PRA did not return to baselineduring the recovery period in any group.

In response to hemorrhage, there was a lesser fallin GFR in the RDN groups than in their respectiveintact group (Table 2, Figures 5A and 5B). GFRreturned to baseline levels in the control-intactsheep, but not in the control-RDN sheep during thepost-hemorrhage recovery period (Figure 5A). In theCKD-intact sheep, although recovery was observed,GFR was not fully restored to baseline levels and inthe CKD-RDN group, no recovery of GFR wasobserved (Figure 5B).

RBF fell in response to hemorrhage in all groups(Figures 5C and 5D), but the response was attenuatedin the RDN sheep (Figures 5C and 5D, Table 2). Duringthe post-hemorrhage recovery period, RBF returnedto baseline levels only in the control-intact group, notin the control-RDN (Figure 5C), CKD-intact, or CKD-RDN group (Figure 5D).

In response to hemorrhage, filtration fractiondecreased significantly in the control-intact and CKD-intact groups by 60 min post-hemorrhage and thenrecovered to baseline levels thereafter (Figures 5E and5F). This decrease in filtration fraction was absent inboth the control-RDN and CKD-RDN groups comparedwith intact counterparts (Table 2, Figures 5E and 5F).

In the intact sheep, RVR increased significantly(Figures 5G and 5H) in response to hemorrhage.RVR returned to baseline levels by 120 min post-hemorrhage in the control-intact sheep (Figure 5G),whereas RVR remained elevated in the CKD-intactgroup throughout the recovery period (Figure 5H). Inboth RDN groups, the increase in RVR in response tohemorrhage was markedly attenuated (Table 2,Figures 5G and 5H).

UFR and UNaV fell in response to hemorrhage in allgroups and did not recover to baseline levels in anygroup (Figure 3, Online Appendix). The maximal fallin UNaV was significantly attenuated in the RDNgroups, but the fall in UFR was not different betweenthe intact and RDN groups (Table 2).

DISCUSSION

At 2 and 5 months following RDN, BP wasreduced to normotensive levels in sheep with

hypertensive CKD, confirming RDN’s BP-lowering ef-ficacy (1,2) (Central Illustration). These observations areclinically relevant given the similarity of approach andalgorithm used for energy application in several clin-ical trials and this animal study. Additionally, thisstudy showed that although GFR was not changed at2 months post-procedure, increases in both GFR andRBF were observed at 5 months post-procedure inCKD-RDN sheep, demonstrating improved renalfunction. This was in line with available data fromstudies in humans indicative of the potential reno-protective effects of RDN irrespective of its BP-lowering efficacy (5).

Despite these benefits, in the present studycompensatory cardiovascular and renal response tohypotensive hemorrhage (20% blood volume loss)was markedly impaired in sheep that underwentRDN. Responses to hemorrhage at 2 and 5 monthsfollowing RDN were remarkably similar, suggestingthat renal nerve function had not recovered withinthis time frame. Thus, our findings confirmed thatRDN is effective in reducing BP in a hypertensive CKDmodel but also highlighted that compensatory re-sponses to hemodynamic challenges, such as criticalblood loss, might be impaired and, potentially, resultin adverse consequences. Whether compensatory re-sponses to more common and less severe forms ofalterations in fluid homeostasis, such as dehydrationor gastroenteritis, are affected by RDN should beaddressed in future studies.

BASAL CARDIOVASCULAR AND RENAL FUNCTION

FOLLOWING RDN. A major strength of our study,unlike some of the human (1,2) and experimental(17,18) studies, was that it included all importantcontrols, including normotensive controls and a shamprocedure. Furthermore, the ablation procedure fol-lowed the same protocol (6 circumferential ablationsof the renal artery) using the same catheter system asused in the Symplicity HTN studies, and these criteriawere met in all animals.

The sustained reduction in BP in the CKD-RDNsheep, although contrary to the reports of HTN-3(3), corroborated those in animal models (17,18) andthe majority of the clinical trials (Global SymplicityRegistry, DENERHTN) involving the catheter evalu-ated in resistant hypertensive (1,2,4,19) and hyper-tensive CKD (7,8) populations. Consistent withprevious observations (20), BP was not reduced inthe control animals following RDN, but interestingly,the age-related increase in BP observed in the intactanimals was not observed in the RDN sheep.Although speculative and not tested in humans, thenotion of targeting renal sympathetic overactivity at

FIGURE 1 Mean Arterial Pressure, Heart Rate, and Plasma Renin Activity

MAP

(mm

Hg)

CONTROL95

90

85

80

75

70

65

pre-RDN/sham post-RDN/sham

6

RDN

10 12

** **

15

95

90

85

80

75MAP

(mm

Hg)

CKD

70

65

pre-RDN/sham post-RDN/sham

6

RDN

10 12

****

*****

****

15HR

(Bea

ts/m

in)

100

95

90

85

80

75

70

pre-RDN/sham post-RDN/sham

6

RDN

10 12

*******

15

HR (B

eats

/min

)

100

95

90

85

80

75

70

pre-RDN/sham post-RDN/sham

6

RDN

10 12

***

15

PRA

(ng/

ml/h

)

Age (Months)

2.5

2.0

1.5

1.0

0.5

0.0

pre-RDN/sham

control-intact control-RDN CKD-intact CKD-RDN

post-RDN/sham

6

RDN

10 12

**

**

15

PRA

(ng/

ml/h

)

Age (Months)

2.5

2.0

1.5

1.0

0.5

0.0

pre-RDN/sham post-RDN/sham

6

RDN

10 12

***

15

A B

C D

E F

Mean arterial pressure (MAP) (A, B), heart rate (HR) (C, D), and plasma renin activity (PRA) (E, F) were evaluated in normotensive and hy-

pertensive sheep before and after renal denervation (RDN) or sham procedure. *p # 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001

comparing before and after RDN or sham procedure within groups. CKD ¼ chronic kidney disease.

J A C C V O L . 6 9 , N O . 8 , 2 0 1 7 Singh et al.F E B R U A R Y 2 8 , 2 0 1 7 : 9 5 1 – 6 4 Renal Denervation Impairs Responses to Hemorrhage

955

FIGURE 2 Renal Function

3

2

1GF

R (m

l/min

/bw

)

CONTROL

0

pre-RDN/sham post-RDN/sham

6

RDN

10 12 15

3

2

1

GFR

(ml/m

in/b

w)

CKD

0

pre-RDN/sham post-RDN/sham

6

RDN

10 12 15

***

***

Age (Months)

control-intact control-RDN CKD-intact CKD-RDN

A B

C D

E F

G H

20

15

10

5RBF

(ml/m

in/b

w)

0

pre-RDN/sham post-RDN/sham

6

RDN

10 12 15

20

15

10

5RBF

(ml/m

in/b

w)

0

pre-RDN/sham post-RDN/sham

6

RDN

10 12 15

****

****

****

8

6

4

2

RVR

(mm

Hg/

ml/m

in/b

w)

0

pre-RDN/sham post-RDN/sham

6

RDN

10 12 15

8

6

4

2

RVR

(mm

Hg/

ml/m

in/b

w)

0

pre-RDN/sham post-RDN/sham

6

RDN

10 12 15

0.30

0.25

0.20

0.15

Filtr

atio

n Fr

actio

n

0.00

0.05

0.10

pre-RDN/sham post-RDN/sham

6

RDN

10 12 15

Age (Months)

0.30

0.25

0.20

0.15

Filtr

atio

n Fr

actio

n

0.00

0.05

0.10

pre-RDN/sham post-RDN/sham

6

RDN

10 12 15

Glomerular filtration rate (GFR) (A, B), renal blood flow (RBF) (C, D), renal vascular resistance (RVR) (E, F), and filtration fraction (G, H) were

assessed pre- and post-RDN or sham procedure. ***p < 0.001 and ****p < 0.0001 comparing before and after RDN or sham procedure

within groups. Abbreviations as in Figure 1.

Singh et al. J A C C V O L . 6 9 , N O . 8 , 2 0 1 7

Renal Denervation Impairs Responses to Hemorrhage F E B R U A R Y 2 8 , 2 0 1 7 : 9 5 1 – 6 4

956

FIGURE 3 Blood Pressure and Heart Rate During Hemorrhage

ΔMAP

(mm

Hg)

Δ SB

P (m

m H

g)

10

0

-10

-20

-30

-401

CONTROL

97

PHemorrhage<0.0001

53 11 13 15

*#

*#

*#

10

0

-10

-20

-301 97

PHemorrhage<0.0001

53 11 13 15

Δ DB

P (m

m H

g)

10

0

-10

-20

-301 97

PHemorrhage<0.0001

53 11 13 15

ΔHR

(Bea

ts/m

in)

40

30

20

10

0

-101 97

PHemorrhage<0.0001

53 11 13 15

ΔMAP

(mm

Hg)

10

0

-10

-20

-30

-401

CKD

97

PHemorrhage<0.0001

53 11 13 15

*#

*#

*#

Δ SB

P (m

m H

g)10

0

-10

-20

-301 97

PHemorrhage<0.0001

53 11 13 15

Δ DB

P (m

m H

g)

10

0

-10

-20

-301 97

PHemorrhage<0.0001

53 11 13 15

ΔHR

(Bea

ts/m

in)

40

20

30

10

0

-101 97

PHemorrhage<0.0001

53 11 13 15

A B

D

F

C

E

HG

CKD-intact-2mths CKD-intact-5mthsCKD-RDN-5mthsCKD-RDN-2mths

control-intact-2mthscontrol-RDN-2mths

control-intact-5mthscontrol-RDN-5mths

Time (mins) During Hemorrhage Time (mins) During Hemorrhage

Data show change (D) from baseline for MAP (A, B), SBP (C, D), DBP (E, F), and HR (G, H). *p < 0.05 comparing intact with RDN at 2 months

post-procedure; #p < 0.05 comparing intact with RDN at 5 months post-procedure. DBP ¼ diastolic blood pressure; SBP ¼ systolic blood

pressure; other abbreviations as in Figure 1.

J A C C V O L . 6 9 , N O . 8 , 2 0 1 7 Singh et al.F E B R U A R Y 2 8 , 2 0 1 7 : 9 5 1 – 6 4 Renal Denervation Impairs Responses to Hemorrhage

957

FIGURE 4 Mean Arterial Pressure, Heart Rate, and Plasma Renin Activity in Response to Hemorrhage

100

90

80

70

60

50100806040

*#

*# *# *#

*#

*#*#

*#

*#

*#*# *#

*#

*#

*#

*#*#

*#*#

20B 120 140 160 180

CONTROLM

AP (m

m H

g)A

100

90

80

70

60

50100806020 40

40

B 120 140 160 180

CKDB

150

100

50100806020B 120 140 160 180

D

40

6

2

4

0100806020B 120 140 160 180

F

150

100

5010080604020B 120 140 160 180

HR (B

eats

/min

)

C

8

4

6

2

010080

Time (mins) Time (mins)604020B 120 140 160 180

PRA

(ng/

ml/h

)

E

CKD-intact-2mths CKD-intact-5mthsCKD-RDN-5mthsCKD-RDN-2mths

control-intact-2mthscontrol-RDN-2mths

control-intact-5mthscontrol-RDN-5mths

Data at baseline (B), end of hemorrhage (arrowhead), and over 180 min post-hemorrhage for MAP (A, B), HR (C, D), and PRA (E, F). *p < 0.05

comparing intact with RDN at 2 months post-procedure; #p < 0.05 comparing intact with RDN at 5 months post-procedure. Abbreviations as

in Figure 1.

Singh et al. J A C C V O L . 6 9 , N O . 8 , 2 0 1 7

Renal Denervation Impairs Responses to Hemorrhage F E B R U A R Y 2 8 , 2 0 1 7 : 9 5 1 – 6 4

958

earlier stages of hypertension or in younger patientsto prevent onset of hypertension may have merit.Consistent with previous observations (21), re-ductions in PRA in response to RDN in both controland CKD groups were observed. However, this fall inPRA did not produce a marked fall in BP in thecontrol-RDN group, and other compensatory mech-anisms likely accounted for this.

Basal GFR and RBF were not affected by RDN in thecontrol sheep, in contrast to observations in normo-tensive swine in which RDN increased RBF (22).However, in the swine, RBF was measured underanesthesia. In the CKD-RDN sheep, at 5 monthspost-procedure, significant increases in GFR and RBFof similar magnitude were observed; thus, filtrationfraction was not affected. The elevation in GFR was

FIGURE 5 Renal Function in Response to Hemorrhage

CKD-intact-2mths CKD-intact-5mthsCKD-RDN-5mthsCKD-RDN-2mths

control-intact-2mthscontrol-RDN-2mths

control-intact-5mthscontrol-RDN-5mths

GFR

(ml/m

in/b

w)

3

2

1

0B

CONTROL

10080604020 120 140 160 180

RBF

(ml/m

in/b

w)

20

15

10

5

0B

3

2

1

0B

CKD

10080604020 120 140 160 180

20

15

10

5

0B 10080604020 120 140 160 180

0.3

0.2

0.1

0.0B 10080604020 120 140 160 180

15

10

5

0B 10080604020 120 140 160 180

10080604020 120 140 160 180

Filtr

atio

n Fr

actio

n

0.3

0.2

0.1

0.0B 10080604020 120 140 160 180

RVR

(mm

Hg/

ml/m

in/b

w)

Time (mins) Time (mins)

15

10

5

0B 10080604020 120 140 160 180

*#

*# *#

*#

*#*#

*# *#*# *#

*#

*#*#

*# *#

*#*#

*#*#

*#

*#

*#

*#*#

A B

D

F

C

E

HG

Data presented at baseline (B), end of hemorrhage (arrowhead), and over 180 minutes post-hemorrhage for GFR (A, B), RBF (C, D), filtration

fraction (E, F), and RVR (G, H). *p < 0.05 comparing intact with RDN at 2 months post-procedure; #p < 0.05 comparing intact with RDN at

5 months post-procedure. Abbreviations as in Figures 1 and 2.

J A C C V O L . 6 9 , N O . 8 , 2 0 1 7 Singh et al.F E B R U A R Y 2 8 , 2 0 1 7 : 9 5 1 – 6 4 Renal Denervation Impairs Responses to Hemorrhage

959

TABLE 2 Maximal Variable Changes in Response to Hemorrhage

Control-Intact (n ¼ 6) Control-RDN (n ¼ 8) CKD-Intact (n ¼ 7) CKD-RDN (n ¼ 7)

2 Months 5 Months 2 Months 5 Months 2 Months 5 Months 2 Months 5 Months

DMAP, mm Hg �10 � 1 �10 � 1 �15 � 2* �15 � 6* �11 � 3 �11 � 2 �19 � 9* �17 � 5*

DSBP, mm Hg �15 � 2 �13 � 3 �14 � 3 �15 � 2 �14 � 3 �14 � 2 �18 � 3 �16 � 4

DDBP, mm Hg �10 � 1 �9 � 4 �14 � 2* �14 � 1* �10 � 2 �10 � 2 �21 � 9* �18 � 4*

DHeart rate, beats/min 39 � 6 38 � 5 29 � 3* 24 � 4* 45 � 11 41 � 5 27 � 7* 26 � 8*

%DPRA 221 � 62 240 � 178 125 � 25* 93 � 50* 103 � 31† 139 � 51† 63 � 27* 86 � 38*

%DGFR �63 � 4 �59 � 16 �37 � 11* �40 � 10* �61 � 6 �55 � 6 �34 � 10* �37 � 10*

%DRBF �45 � 11 �46 � 7 �34 � 6* �36 � 4* �36 � 7 �40 � 9 �20 � 4* �21 � 5*

%DRVR 89 � 34 85 � 27 34 � 15* 37 � 8* 44 � 10† 53 � 36 5 � 3* 4 � 7*

%DFF �25 � 6 �26 � 10 �9 � 7* �13 � 8* �28 � 10 �23 � 3 �11 � 5* �11 � 6*

%DUNaV �83 � 8 �82 � 7 �49 � 10* �47 � 17* �79 � 6 �78 � 6 �36 � 19* �37 � 17*

%DUFR �59 � 15 �62 � 9 �56 � 14 �62 � 16 �58 � 8 �58 � 8 �48 � 16 �49 � 9

Values are mean � SD. Maximal change (D) or percentage change (%D) calculated by normalizing trough or peak responses during hemorrhage to baseline. For all variables, thetrough or peak occurred at either end of hemorrhage, 40 or 60 min post-hemorrhage. *p < 0.01, comparing RDN with intact groups within age. †p<0.05, comparing intactgroups within age, from Sidak’s post-hoc analysis.

CKD ¼ chronic kidney disease; DBP ¼ diastolic blood pressure; FF ¼ filtration fraction; SBP ¼ systolic blood pressure; other abbreviations as in Table 1.

Singh et al. J A C C V O L . 6 9 , N O . 8 , 2 0 1 7

Renal Denervation Impairs Responses to Hemorrhage F E B R U A R Y 2 8 , 2 0 1 7 : 9 5 1 – 6 4

960

consistent with a study in which RDN also led toincreased GFR in patients with CKD (8). Importantly,in our study, RDN did not cause a decline in GFR,similar to findings in humans (7,8), supporting theconcept that RDN may be a safe and effective treat-ment for hypertensive CKD in humans. However,future studies should continue to assess renal func-tion over time.

The kidneys maintain extracellular fluid volume,and hence BP, by modulating sodium excretion, andthe renal nerves contribute to this response (23). At2 months post-procedure in CKD-RDN sheep,enhanced sodium excretion occurred without signif-icant changes in GFR and RBF, consistent with ob-servations in experimental hypertension modelsfollowing surgical denervation (20) and in patientswith resistant hypertension following RDN (24).However, sodium excretion in these studies mighthave reflected an increase in dietary intake, whichwas not assessed. Nevertheless, it is likely that a shiftin pressure natriuresis/diuresis associated withremoval of renal nerves contributed to the observedfall in BP following RDN.

Additionally, we observed decreases in heart ratein the control and CKD groups at both ages post-RDN.This decline might be independent of BP, because BPwas decreased only in the CKD-RDN animals. Ourfindings were consistent with those in resistant hy-pertensive populations following RDN (25) in whichheart rate was reduced following RDN and might be aconsequence of the loss of renal afferent sym-pathoexcitatory input following RDN, leading to aglobal reduction in sympathetic outflow (26).

EFFECT OF RDN ON RESPONSE TO HEMORRHAGE.

The normal response to hemorrhage is a fall in BP,causing a reflex increase in sympathetic activity,which drives an increase in heart rate, total peripheralresistance (TPR; including RVR, leading to a reduc-tion in GFR), renin release, and sodium excretion,which together promote BP maintenance (12,27). Therelative importance of each mechanism depends onthe species and vascular bed examined and the timecourse of the protocol used to induce hemorrhage(28). However, each mechanism (neural, humoral,and renal) contributes importantly to BP maintenancein response to blood loss. In control-intact animals,the fall in BP was restored to baseline levels within 60min following blood loss, comparable with responsesin humans (27) and sheep (29,30) to similar bloodvolume depletion. Similar changes in BP and renalfunction in response to hemorrhage were observed inthe CKD-intact sheep, but recovery of these variableswas slower than in control-intact animals. This sug-gested that the CKD-intact animals exhibited bluntedcompensatory responses to hemorrhage, potentiallybecause of existing renal impairment in this low-renin disease model (16).

In contrast, during blood loss, the decreases inMAP were faster and greater in both RDN groups. Themaximal decrease in MAP was about 15 to 48 mm Hg,reaching a trough of about 50 to 70 mm Hg in the RDNgroups. From this low point in BP, there was littlerecovery during the ensuing 180 min in both controland CKD animals that underwent RDN. The exagger-ated MAP response in RDN groups was due to agreater fall in DBP, because SBP was not different in

CENTRAL ILLUSTRATION Renal Denervation: Effect on Renal Function and Blood Pressure

A. RDN effectively lowered blood pressure and improved kidney function

B. RDN caused a greater fall in blood pressure following blood loss

MAP

(mm

Hg)

100

80

90

70

****

GFR

(ml/m

in/b

w)

2.5

1.5

2.0

1.0

control-intact CKD-intact CKD-RDNcontrol-RDN

Trou

gh B

P (m

m H

g)

100

60

80

40MAP DBP SBP

CONTROL

**

**

*CKD

Trou

gh B

P (m

m H

g)

100

60

80

40MAP DBP SBP

***

***

***

****

Singh, R.R. et al. J Am Coll Cardiol. 2017;69(8):951–64.

(A) Renal denervation (RDN) in sheep with hypertensive chronic kidney disease (CKD) significantly reduced blood pressure (mean arterial pressure [MAP]) and

increased renal function (glomerular filtration rate [GFR]) compared with intact sheep. (B) However, RDN significantly impaired responses to hemorrhage, resulting in

a greater fall in blood pressure (BP) during blood loss. Data are from 5 months post-RDN or sham procedure (2-month results were similar). DBP ¼ diastolic blood

pressure; SBP ¼ systolic blood pressure.

J A C C V O L . 6 9 , N O . 8 , 2 0 1 7 Singh et al.F E B R U A R Y 2 8 , 2 0 1 7 : 9 5 1 – 6 4 Renal Denervation Impairs Responses to Hemorrhage

961

the intact groups. TPR was not measured in ourstudy. However, in conscious sheep, increases inheart rate and TPR both contribute to MAP mainte-nance during blood loss, and TPR remains signifi-cantly elevated in the recovery period post-hemorrhage (31,32). In the present study, the greaterfall in DBP suggested that the increase in TPR wasattenuated following RDN. Therefore, the greater fallin MAP following blood loss was likely due to both anattenuated TPR and heart rate response in the RDNgroups, which may have resulted from the lack ofefferent renal sympathetic nerves or an indirect effect

of the loss of renal afferent nerves resulting inreduced cardiac sympathetic nerve activity (32,33).

Following RDN, the reduced increase in PRA andRVR will each have contributed to the greater fall, andlack of recovery, of BP in response to hemorrhage.The reduced increase in RVR will be due partly to theabsence of the efferent renal sympathetic nerves, butthis is likely insufficient on its own to account for thefull effect on BP. It has been demonstrated previouslythat increased vasoconstriction occurs in mostvascular beds, except cerebral and coronary, inresponse to hemorrhage (28). It is therefore possible

Singh et al. J A C C V O L . 6 9 , N O . 8 , 2 0 1 7

Renal Denervation Impairs Responses to Hemorrhage F E B R U A R Y 2 8 , 2 0 1 7 : 9 5 1 – 6 4

962

that the greater fall in DBP is due to a global reductionin vasoconstriction, due to the reduced increasein PRA and global reduction in vasoconstrictioncaused by loss of renal afferent sympathoexcitation.These findings also demonstrated the relativeimportance of renal nerves in the physiologicalresponse to hemorrhage and suggested thatdecreased arterial pressure is not a major stimulus forrenin release during hemorrhage. This aligns withevidence that points to sympathetic pathways pre-dominating over moderate changes in arterial pres-sure in regulation of renin release (34). This suggeststhat in a situation of critical trauma, subjects whohave undergone RDN have a greater risk for enteringthe decompensatory phase of hemorrhage morerapidly than nondenervated subjects.

This attenuated ability to maintain arterial pres-sure in response to blood loss may limit using RDN inclinical practice. However, to date, there have beenno reports of adverse events related to blood lossfrom the Global Symplicity Registry of more than2,500 patients (35,36). Our findings should be viewedin context of clinical findings in which RDN in pa-tients with resistant hypertension has not beenshown to affect chronotropic competence during ex-ercise (37,38) or cause orthostatic dysfunction duringtilting (39), though it should be noted that theseconstitute a more modest challenge than the hemor-rhage we used. In support of our findings, the renalresponse to moderate lower body negative pressurein renal transplantation patients (thus denervated)has been shown to be impaired (40).

STUDY LIMITATIONS. Strengths of our study werethat we used the same radiofrequency catheter sys-tem and algorithm as in patients, and the studieswere all in conscious animals, avoiding confoundinginfluences of anesthesia. Also, in studies of RDN inhypertensive patients, these patients were treatedwith variable doses and combinations of antihyper-tensive drugs that likely mask RDN’s effects. Incontrast, the sheep in this study were not treatedwith antihypertensive drugs.

A limitation of our study was that we have no directevidence of the degree of renal nerve ablation ach-ieved or whether regrowth occurred. However, usingan identical RDN technique in sheep, we previouslydemonstrated complete denervation of the afferentand efferent renal nerves 2 weeks after RDN, usingimmunohistochemistry for tyrosine hydroxylase andcalcitonin gene–related peptide and measurement oftissue norepinephrine levels (41), generating confi-dence that we achieved a significant degree ofRDN here. Previously, we also reported evidence of

functional reinnervation at 5.5 and 11 months post-RDN in normotensive sheep, as demonstrated bygradual return of renal responses to electric stimula-tion of the renal nerves with time (15). It is possible thatcoexistence of hypertension, which has been stronglyassociated with sympathetic overactivity andincreased sympathetic innervation density (42), mightalter subsequent nerve regrowth following RDN andaccount for the sustained reduction in BP observed inthis study as well as in some human trials post-RDN(1,12,19). It could be argued that electric stimulationis not equivalent to reflex activation (43), and rein-nervation may not always imply return of normalfunction (40). In the present study, by using hemor-rhage to physiologically induce reflex activation of therenal sympathetic nerves, we observed virtuallyidentical responses at 2 and 5 months post-RDN, sug-gesting little restoration of nerve function. Furtherstudies are warranted to determine if renal nervefunction returns over a longer time frame.

CONCLUSIONS

Radiofrequency catheter ablation of the renal arteriesin a sheep model of hypertensive CKD reduced BP tonormotensive levels and improved renal function outto 5 months post-procedure. Moreover, the reductionin resting BP and the similar responses to hemorrhageat 2 and 5 months post-RDN provided no evidence ofrenal nerve function recovery within this time frame.This suggests that the regrowth of renal nerves intothe kidney, as previously demonstrated, does notimply return of normal function, which may explainthe sustained and long-term effects of RDN. However,attenuated compensatory responses to hemorrhagepresent a potential caveat to widespread applicationof RDN. Although RDN in selected patient pop-ulations, such as those with CKD and hypertension,will likely yield beneficial BP reduction and potentialrenoprotection, it is possible such patients may havea reduced capacity to mount an adequate response tosevere homeostatic challenges, such as hemorrhagicor septic shock.

ACKNOWLEDGMENTS The authors thank AlanMcDonald, Dr. Ross Young, and Lawrence Easton forproviding assistance with surgical procedures. Theauthors thank Lawrence Easton for assistance withlaboratory procedures.

ADDRESS FOR CORRESPONDENCE: Dr. Reetu R.Singh, Cardiovascular Program, Monash BiomedicineDiscovery Institute and Department of Physiology,Monash University, Wellington Road, Clayton VIC,3800 Australia. E-mail: reetu.singh@monash.edu.

PERSPECTIVES

COMPETENCY IN MEDICAL KNOWLEDGE: Radio-

frequency catheter-based RDN is being trialed as an

alternative approach for treatment of resistant hyper-

tension. Initial reports (SYMPLICITY HTN-1 and HTN-2)

demonstrated significant reductions in BP in patients with

resistant hypertension, but the largest, sham control trial

(HTN-3) did not support these findings. Limitations of

HTN-3 have prompted further clinical trials examining the

effectiveness of RDN in reducing BP. Given the contri-

bution of renal sympathetic overactivity to CKD, RDN is

also being trialed in patients with hypertensive CKD.

COMPETENCY IN PATIENT CARE & PROCEDURAL

SKILLS: Catheter-based radiofrequency renal

denervation in a sheep model of hypertensive chronic

kidney disease compromised the compensatory

hemodynamic response to hemorrhage.

TRANSLATIONAL OUTLOOK: Future studies of renal

artery sympathetic denervation should consider

impaired responsiveness to hemodynamic challenge in

the estimation of net clinical benefit.

J A C C V O L . 6 9 , N O . 8 , 2 0 1 7 Singh et al.F E B R U A R Y 2 8 , 2 0 1 7 : 9 5 1 – 6 4 Renal Denervation Impairs Responses to Hemorrhage

963

RE F E RENCE S

1. Krum H, Schlaich M, Whitbourn R, et al.Catheter-based renal sympathetic denervation forresistant hypertension: a multicentre safety andproof-of-principle cohort study. Lancet 2009;373:1275–81.

2. Esler M, Krum H, Sobotka P, et al. Renal sym-pathetic denervation in patients with treatment-resistant hypertension (the Symplicity HTN-2trial): a randomised controlled trial. Lancet 2010;376:1903–9.

3. Bhatt DL, Kandzari DE, O’Neill WW, et al.A controlled trial of renal denervation for resis-tant hypertension. N Engl J Med 2014;370:1393–401.

4. Azizi M, Pereira H, Hamdidouche I, et al.Adherence to antihypertensive treatment and theblood pressure-lowering effects of renal dener-vation in the Renal Denervation for Hypertension(DENERHTN) trial. Circulation 2016;134:847–57.

5. Masuo K, Lambert GW, Esler MD, Rakugi H,Ogihara T, Schlaich MP. The role of sympatheticnervous activity in renal injury and end-stage renaldisease. Hypertens Res 2010;33:521–8.

6. Ravera M, Re M, Deferrari L, Vettoretti S,Deferrari G. Importance of blood pressure controlin chronic kidney disease. J Am Soc Nephrol 2006;17:S98–103.

7. Hering D, Mahfoud F, Walton AS, et al. Renaldenervation in moderate to severe CKD. J Am SocNephrol 2012;23:1250–7.

8. Ott C, Mahfoud F, Schmid A, et al. Renaldenervation preserves renal function in patientswith chronic kidney disease and resistant hyper-tension. J Hypertens 2015;33:1261–6.

9. Esler M. Illusions of truths in the SymplicityHTN-3 trial: generic design strengths but neuro-science failings. J Am Soc Hypertens 2014;8:593–8.

10. Silva JD, Costa M, Gersh BJ, Goncalves L.Renal denervation in the era of HTN-3. Compre-hensive review and glimpse into the future. J AmSoc Hypertens 2016;10:656–70.

11. Schlaich MP, Hering D, Sobotka P, et al. Effectsof renal denervation on sympathetic activation,blood pressure, and glucose metabolism in pa-tients with resistant hypertension. Front Physiol2012;3:10.

12. Malpas SC, Evans RG, Head GA,Lukoshkova EV. Contribution of renal nerves torenal blood flow variability during hemorrhage.Am J Physiol Regul Integr Comp Physiol 1998;274:R1283–94.

13. Lankadeva YR, Singh RR, Moritz KM,Parkington HC, Denton KM, Tare M. Renaldysfunction is associated with a reduced contri-bution of nitric oxide and enhanced vasoconstric-tion after a congenital renal mass reduction insheep. Circulation 2015;131:280–8.

14. Singh RR, Denton KM, Bertram JF, et al.Development of cardiovascular disease due torenal insufficiency in male sheep following fetalunilateral nephrectomy. J Hypertens 2009;27:386–96.

15. Booth LC, Nishi EE, Yao ST, et al. Reinnervationof renal afferent and efferent nerves at 5.5 and 11months after catheter-based radiofrequency renaldenervation in sheep. Hypertension 2015;65:393–400.

16. Singh RR, Jefferies AJ, Lankadeva YR, et al.Increased cardiovascular and renal risk is associ-ated with low nephron endowment in aged fe-males: an ovine model of fetal unilateralnephrectomy. PLoS One 2012;7:e42400.

17. Henegar JR, Zhang Y, De Rama R, Hata C,Hall ME, Hall JE. Catheter-based radiorefrequencyrenal denervation lowers blood pressure in obesehypertensive dogs. Am J Hypertens 2014;27:1285–92.

18. Machino T, Murakoshi N, Sato A, et al. Anti-hypertensive effect of radiofrequency renaldenervation in spontaneously hypertensive rats.Life Sci 2014;110:86–92.

19. Prochnau D, Otto S, Figulla HR, Surber R. Renaldenervation with standard radiofrequency abla-tion catheter is effective in 24-hour ambulatory

blood pressure reduction—follow-up at 1/3/6/12months. Neth Heart J 2016;24:449–55.

20. Salman IM, Sattar MA, Abdullah NA, et al.Renal functional & haemodynamic changesfollowing acute unilateral renal denervation inSprague Dawley rats. Indian J Med Res 2010;131:76–82.

21. Evans RG, Burke SL, Lambert GW, Head GA.Renal responses to acute reflex activation of renalsympathetic nerve activity and renal denervationin secondary hypertension. Am J Physiol RegulIntegr Comp Physiol 2007;293:R1247–56.

22. Tsioufis C, Papademetriou V, Dimitriadis K,et al. Catheter-based renal sympathetic denerva-tion exerts acute and chronic effects on renal he-modynamics in swine. Int J Cardiol 2013;168:987–92.

23. DiBona GF, Esler M. Translational medicine:the antihypertensive effect of renal denervation.Am J Physiol Regul Integr Comp Physiol 2010;298:R245–53.

24. Pöss J, Ewen S, Schmieder RE, et al. Effects ofrenal sympathetic denervation on urinary sodiumexcretion in patients with resistant hypertension.Clin Res Cardiol 2015;104:672–8.

25. Ukena C, Mahfoud F, Spies A, et al. Effects ofrenal sympathetic denervation on heart rate andatrioventricular conduction in patients with resis-tant hypertension. Int J Cardiol 2013;167:2846–51.

26. Schlaich MP, Sobotka PA, Krum H, Lambert E,Esler MD. Renal sympathetic-nerve ablation foruncontrolled hypertension. N Engl J Med 2009;361:932–4.

27. Stone AM, Stahl WM. Renal effects of hemor-rhage in normal man. Ann Surg 1970;172:825–36.

28. Schadt JC, Ludbrook J. Hemodynamic andneurohumoral responses to acute hypovolemia inconscious mammals. Am J Physiol 1991;260:H305–18.

29. Brandon AE, Boyce AC, Lumbers ER,Kumarasamy V, Gibson KJ. Programming of therenin response to haemorrhage by mild maternal

Singh et al. J A C C V O L . 6 9 , N O . 8 , 2 0 1 7

Renal Denervation Impairs Responses to Hemorrhage F E B R U A R Y 2 8 , 2 0 1 7 : 9 5 1 – 6 4

964

renal impairment in sheep. Clin Exp PharmacolPhysiol 2011;38:102–8.

30. Smith FG, Sener A, Basati R, Abu-Amarah I.Renal responses to hemorrhage are age dependentin conscious sheep. J Appl Physiol 2004;96:131–6.

31. Scully CG, Daluwatte C, Marques NR, et al.Effect of hemorrhage rate on early hemodynamicresponses in conscious sheep. Physiol Rep 2016;4:e12739.

32. Frithiof R, Ramchandra R, Hood SG, May CN.Hypertonic sodium resuscitation after hemorrhageimproves hemodynamic function by stimulatingcardiac, but not renal, sympathetic nerve activity.Am J Physiol Heart Circ Physiol 2011;300:H685–92.

33. Johns EJ. The neural regulation of the kidneyin hypertension and renal failure. Exp Physiol2014;99:289–94.

34. Damkjaer M, Isaksson GL, Stubbe J, Jensen BL,Assersen K, Bie P. Renal renin secretion as regu-lator of body fluid homeostasis. Pflugers Arch2013;465:153–65.

35. Böhm M, Mahfoud F, Ukena C, et al. Firstreport of the Global Symplicity Registry on the

effect of renal artery denervation in patients withuncontrolled hypertension. Hypertension 2015;65:766–74.

36. Kim BK, Bohm M, Mahfoud F, et al. Renaldenervation for treatment of uncontrolled hyper-tension in an Asian population: results from theGlobal Symplicity Registry in South Korea (GSRKorea). J Hum Hypertens 2016;30:315–21.

37. Ewen S,Mahfoud F, Linz D, et al. Effects of renalsympathetic denervation on exercise blood pres-sure, heart rate, and capacity in patients with resis-tant hypertension. Hypertension 2014;63:839–45.

38. Ukena C, Mahfoud F, Kindermann I, et al.Cardiorespiratory response to exercise after renalsympathetic denervation in patients with resistanthypertension. J Am Coll Cardiol 2011;58:1176–82.

39. Lenski M, Mahfoud F, Razouk A, et al. Ortho-static function after renal sympathetic denerva-tion in patients with resistant hypertension. Int JCardiol 2013;169:418–24.

40. Hansen JM, Abildgaard U, Fogh-Andersen N,et al. The transplanted human kidney does notachieve functional reinnervation. Clin Sci (Lond)1994;87:13–20.

41. Booth LC, Schlaich MP, Nishi EE, et al. Short-term effects of catheter-based renal denervationon cardiac sympathetic drive and cardiac barore-flex function in heart failure. Int J Cardiol 2015;190:220–6.

42. Kline R, Mercer P. Functional reinnervationand development of supersensitivity to NE afterrenal denervation in rats. Am J Physiol RegulIntegr Comp Physiol 1980;238:R353–8.

43. Leonard BL, Malpas SC, Denton KM,Madden AC, Evans RG. Differential control ofintrarenal blood flow during reflex increases insympathetic nerve activity. Am J Physiol RegulIntegr Comp Physiol 2001;280:R62–8.

KEY WORDS chronic kidney disease,hypertension, mean arterial pressure, plasmarenin activity, sheep

APPENDIX For extended methods andmaterials, results, and discussion as well assupplementary figures, please see the onlineversion of this article.