The giraffe kidney tolerates high arterial blood pressure by high renal interstitial pressure and low glomerular filtration rate

The giraffe kidney tolerates high arterial blood pressure by

high renal interstitial pressure and low glomerular

filtration rate

M. Damkjær,1 T. Wang,2 E. Brøndum,3 K. H. Østergaard,4 U. Baandrup,4 A. Hørlyck,5

J. M. Hasenkam,6 M. Smerup,6 J. Funder,6 N. Marcussen,7 C. C. Danielsen,8 M. F. Bertelsen,9

C. Grøndahl,9 M. Pedersen,10 P. Agger,10 G. Candy,11 C. Aalkjær3,12 and P. Bie1

1 Department of Cardiovascular and Renal Research, University of Southern Denmark, Odense, Denmark

2 Department of Biological Sciences, Institute of Zoophysiology, Aarhus University, Aarhus, Denmark

3 Department of Physiology, Institute of Biomedicine, Aarhus University, Aarhus, Denmark

4 Centre for Clinical Research, Hjørring/Department of Clinical Medicine, Aalborg University, Denmark

5 Department of Radiology, Aarhus University Hospital, Aarhus, Denmark

6 Department of Cardiothoracic and Vascular Surgery, Institute of Clinical Medicine, Aarhus University Hospital, Aarhus, Denmark

7 Department of Clinical Pathology, University of Southern Denmark, Odense, Denmark

8 Department of Anatomy, Institute of Biomedicine, Aarhus University, Aarhus, Denmark

9 Center for Zoo and Wild Animal Health, Copenhagen Zoo, Copenhagen, Denmark

10 MR Research Centre, Institute of Clinical Medicine, Aarhus University Hospital, Aarhus, Denmark

11 Department of Physiology and Medicine, University of the Witwatersrand, Johannesburg, South Africa

12 Department of Biomedicine, University of Copenhagen, Copenhagen, Denmark

Received 15 April 2015,

revision requested 11 May 2015,

revision received 17 May 2015,

accepted 18 May 2015

Correspondence: P. Bie, MD,

DMSc, Institute of Molecular

Medicine, University of Southern

Denmark, 21 J. B. Winsloews Vej,

DK-5000 Odense, Denmark.

E-mail: [email protected]

Abstract

Background: The tallest animal on earth, the giraffe (Giraffa camelopar-

dalis) is endowed with a mean arterial blood pressure (MAP) twice that of

other mammals. The kidneys reside at heart level and show no sign of

hypertension-related damage. We hypothesized that a species-specific evo-

lutionary adaption in the giraffe kidney allows normal for size renal hae-

modynamics and glomerular filtration rate (GFR) despite a MAP double

that of other mammals.Methods: Fourteen anaesthetized giraffes were instrumented with vascular

and bladder catheters to measure glomerular filtration rate (GFR) and

effective renal plasma flow (ERPF). Renal interstitial hydrostatic pressure

(RIHP) was assessed by inserting a needle into the medullary parenchyma.

Doppler ultrasound measurements provided renal artery resistive index

(RI). Hormone concentrations as well as biomechanical, structural and his-

tological characteristics of vascular and renal tissues were determined.Results: GFR averaged 342 � 99 mL min�1 and ERPF 1252 � 305 mL

min�1. RIHP varied between 45 and 140 mmHg. Renal pelvic pressure

was 39 � 2 mmHg and renal venous pressure 32 � 4 mmHg. A valve-like

structure at the junction of the renal and vena cava generated a pressure

drop of 12 � 2 mmHg. RI was 0.27. The renal capsule was durable with

a calculated burst pressure of 600 mmHg. Plasma renin and AngII were

2.6 � 0.5 mIU L�1 and 9.1 � 1.5 pg mL�1 respectively.Conclusion: In giraffes, GFR, ERPF and RI appear much lower than

expected based on body mass. A strong renal capsule supports a RIHP,

which is >10-fold that of other mammals effectively reducing the net filtra-

tion pressure and protecting against the high MAP.

© 2015 Scandinavian Physiological Society. Published by John Wiley & Sons Ltd, doi: 10.1111/apha.12531 1

Acta Physiol 2015

Keywords blood pressure, glomerular filtration, interstitial hydrostatic

pressure, renal plasma flow, renin.

The mean arterial blood pressure (MAP) of giraffes is

approximately twofold higher than that of other

mammals. The high pressure is needed to ensure a

normal mammalian cerebral perfusion pressure in the

erect posture where the head is situated high above

the heart due to the long neck (Goetz 1955, Goetz &

Keen 1957, Goetz et al. 1960, Van Citters et al.

1966, 1968, 1969, Brondum et al. 2009). In contrast,

the kidneys are located at heart level and hence

continuously exposed to arterial pressures of

200–250 mmHg. In humans, hypertension is one of

the most prominent risk factors for chronic kidney dis-

ease (CKD). The major contribution of hypertension to

CKD is both a direct effect in causing hypertensive

nephrosclerosis and an indirect effect by accelerating

the progression of a number of distinct renal diseases

(Barri 2006, Hill 2008). However, the giraffe kidney

shows no sign of hypertension-related damage despite a

continuous exposure to a high arterial pressure.

Given the exquisite sensitivity of the mammalian

kidney to hypertension (Feld et al. 1990, Syme 2011),

the giraffe kidney provides an intriguing possibility to

understand how natural selection has favoured protec-

tive mechanisms in response to high renal perfusion

pressures. Nevertheless, there is a paucity of experi-

mental data available on the structure and function of

the giraffe kidney, except for an anatomical descrip-

tion by Maluf (2002) who, rather surprisingly, found

no obvious differences to other mammals, including

ruminants (Maluf 1994) and humans. Even the media/

lumen ratio of the renal arteries did not differ from

other similar-sized mammals, such as the closely

related okapi (Maluf 2002) or the hook-lipped rhinoc-

eros (Maluf 1994).

The glomerular filtration rate (GFR) in the kidney

is determined by the balance between the hydrostatic

filtration pressure and the opposing oncotic pressure,

the so-called Starling forces:

GFR ¼ Kf½ðPGC � PBSÞ � rðpGC � pBSÞ�;where Kf is the product of capillary permeability and

surface area; PGC is the glomerular capillary pressure,

PBS is Bowman’s space hydrostatic pressure, r is the

reflection coefficient, pGC is the glomerular capillary

oncotic pressure, and pBS is the oncotic pressure

within Bowman’s space. For giraffes to achieve a

mammalian PGC, it would require an extraordinary

high vascular resistance somewhere between the renal

artery and the glomerulus; otherwise, single-nephron

GFR (snGFR) would be much higher than that in

other mammals. Interspecific analyses reveal that GFR

scales similarly to metabolism with body mass,

whereas snGFR is relatively independent of body size

(Singer 2001). The mass of the giraffe kidney relative

to body mass is no different than other mammals

(Maluf 1994, 2002), which implies that they have

either a high GFR (i.e. high snGFR with a normal

number of glomeruli), or a functional adaption that

normalizes the Starling forces across the glomerulus

despite the high renal arterial pressure.

To provide the first physiological studies on the gir-

affe kidney, we measured GFR and renal blood flow

(ERPF) using classical clearance techniques. Further-

more, the renal vasculature and blood flows were

visualized using ultrasound allowing for an estimation

of the resistive index (RI) (Platt et al. 1989, Tublin

et al. 2003). Biomechanical properties of the kidney

and renal vessels and capsule were determined and

kidney structure was studied with both histology and

diffusion tensor magnetic resonance imaging. Finally,

plasma hormone concentrations were determined by

immunoassay.

Materials and methods

Experiments were conducted on two separate occasions

in Hammanskraal (South Africa) where we studied

juvenile giraffes (14 males and 1 female; 1 to 4 years

old) that had been caught in Namibia and South Africa

several months before the experiments. Body mass and

height were 478 � 22 kg and 3.5 � 0.1 m (range

380–654 kg, and range 3.1–3.9 m), respectively, during

the first campaign and 480 � 40 kg and 3.6 � 0.1 m

(range 350–570 kg and 3.4–3.8 m) during the second

campaign. Leucocytes, haemoglobin concentration and

plasma electrolytes measured on freshly collected blood

immediately prior to experiments were within normal

limits.

The experimental protocol was approved by the

Danish Animal Experiments Inspectorate (Ministry of

Food, Agriculture and Fisheries) as well as the Animal

Use and Care Committee at University of Pretoria

(South Africa), and the experiments were supervised

by local ethical committee members.

© 2015 Scandinavian Physiological Society. Published by John Wiley & Sons Ltd, doi: 10.1111/apha.125312

Renal function in the giraffe · M Damkjær et al. Acta Physiol 2015

Animal handling and anaesthesia

Upon overnight deprivation of food and water, the

giraffe was pre-medicated by remote injection

(Daninject, Børkop, Denmark) of medetomidine

(5.5 lg kg�1) and guided to a chute where a halter

and blindfold was mounted. Anaesthesia was then

induced by intravenous injection of etorphine

(6.5 lg kg�1) and ketamine (0.65 mg kg�1) rendering

the animal recumbent within 3–7 min, whereupon a

cuffed endotracheal tube (i.d. 20 mm) was inserted

for assisted ventilation with oxygen using a demand

valve (Hudson RCI, Research Triangle Park, NC,

USA). A supplementary dose of ketamine (0.2

mg kg�1, i.v.) was administered to allow placing the

giraffe in right lateral recumbence.

Anaesthesia was maintained by continuous infu-

sion of a-chloralose (15 mg mL�1 i.v.; University

of Copenhagen, Department of Pharmacy, Copen-

hagen, Denmark) at 30 mg kg�1 h�1 decreasing to

20 and 15 mg kg�1 h�1 after 72 min and 140 min,

respectively, and then gradually to 3 mg kg�1 h�1

over the next 7–8 h depending on reflexes and

breathing pattern. Heart rate, electrocardiogram,

rectal temperature and end-expiratory CO2 pres-

sure (PETCO2) were monitored using a portable

monitor (Mindray PM9000Vet; E-Vet, Haderslev,

Denmark), and both arterial and venous blood

gases were measured every 10 min (GEM Premier

3500; Instrumentation Laboratory, Bedford, MA,

USA).

Instrumentation and indicator infusions

Following local infiltration by lidocaine (2%; SAD,

Copenhagen, Denmark), vascular sheaths were placed

in the carotid artery and jugular vein at the base of

the neck and secured to the skin. Blood pressures were

determined by inserting tip-transducer catheters

through the sheaths (5F Micro-Tip SPC 350, range

50–400 mmHg; Millar Instruments, Houston, TX,

USA). Both the catheters in the carotid artery and the

jugular vein were advanced towards the heart for

measurements of MAP and central venous pressure

(CVP). Signals from both pressure transducers were

recorded with a Biopac MP100 data acquisition sys-

tem (Biopac Systems, Goleta, CA, USA) at 200 Hz,

and heart rate (HR) was derived from the pulsatile

pressure signal.

Additional sterile catheters (Becton Dickinson, Hel-

singborg, Sweden) were placed in the right and left

jugular vein. One was used for continuous infusion of

sinistrin and p-aminohippurate (PAH) and the other

for blood sampling. Due to the urethral anatomy of

male giraffes, it proved difficult to insert bladder

catheters through the urethra without making a peri-

neal incision. Through a small incision in the urethra,

a Foley catheter (R€usch Ch. 14; Teleflex Medical, Ker-

nen, Germany) was inserted into the bladder for con-

tinuous urine sampling.

GFR was calculated from the clearance of sinistrin

(sinistrin, Inutest�; Fresenius Kabi, Graz, Austria). A

bolus was injected at least 1.5 h before the start of

urine collection. The size was determined under the

assumptions (i) that the volume of distribution of sini-

strin equals the extracellular volume (ECF), (ii) that

the ECF constitutes 20% of body mass and (iii) that

the analysis of sinistrin in plasma is optimal at a con-

centration of 0.2 mg mL�1. After bolus injection, a

continuous infusion was started to match expected

renal sinistrin clearance under the assumption of a

GFR of 1 mL min�1 kg�1. Effective renal plasma flow

(ERPF) was measured by the clearance of PAH

(Merck, Westpoint, PA, USA). At least 1.5 h before

the start of the experiment, a bolus of 10 mg kg�1

was injected i.v., followed by a continuous infusion at

0.125 mg min�1 kg�1.

Ultrasound investigations

In 8 giraffes, the kidneys were localized via transrec-

tal exploration, and an ultrasound transducer

(curved array 8C-RS, Vivid i; GE Healthcare,

Brøndby, Denmark) was placed transrectally directly

over the surface of the kidney, the aorta and the

vena cava. B-mode and colour Doppler images and

films were taken from the junction of the kidney

vessels to the aorta and vena cava, respectively, and

from the kidney tissue. Doppler curves from the in-

trarenal arteries at the level of the interlobar/arcuate

arteries were obtained. RI was calculated as (peak

systolic velocity – end-diastolic velocity)/peak systolic

velocity (Platt et al. 1989). For comparison, similar

ultrasound and Doppler examinations and corre-

sponding blood pressure measurements were per-

formed on 3 adult awake cows at the Copenhagen

Zoo, Denmark.

Experimental protocol

The experimental protocols differed between the two

experiments.

Experiment 1: After instrumentation, giraffes were

placed in the vertical position using a custom-built

sling. After at least 1.5 h of equilibration, the blad-

der was flushed with 200 mL of sterile water, and

urine was collected over two 30-min periods. Just

before the end of each sampling period, the bladder

was flushed with 200 mL sterile water. Blood sam-

ples were obtained at the beginning and the end of

© 2015 Scandinavian Physiological Society. Published by John Wiley & Sons Ltd, doi: 10.1111/apha.12531 3

Acta Physiol 2015 M Damkjær et al. · Renal function in the giraffe

each period. Samples were centrifuged immediately

at 4 °C and plasma separated and stored at

�20 °C. Plasma remained frozen until analysis, but

not always at target temperature.

Experiment 2: After instrumentation, giraffes were

placed in right lateral recumbency. The bladder,

renal pelvis, ureter, renal vein and renal interstitial

pressures where then recorded over 30 min (see

below for details).

Plasma hormone concentrations

Plasma renin concentration (PRC) was determined by

the antibody trapping method of Poulsen and Jørgen-

sen (Poulsen & Jorgensen 1974) modified as by

(Kjolby et al. 2005). Results are reported as milli-

international units (mIU L�1) based on the WHO

International Standard (ref. no. 68–356; National

Institute for Biological Standards and Control, Hert-

fordshire, UK), which was measured in each assay of

plasma samples.

Peptide hormone concentrations in plasma were

measured by radio-immunoassay after extraction

(Kjolby et al. 2005). The AngII immunoreactivity of

extracts was measured using a highly specific rabbit

antibody (AB-5-030682). The detection limit was

11 pg mL�1, and the mean recovery of unlabelled

aldosterone was 89%. For Arg vasopressin (AVP)

determinations, a highly specific rabbit antibody

(AB3096) was used. The detection limit of AVP was

0.15 pg mL�1, and the mean recovery of unlabelled

vasopressin was 67%. Antibody against atrial natri-

uretic peptide (ANP) was purchased from Peninsula

Laboratories (RAS8798). The detection limit of the

assay was 1.6 pg mL�1, and the mean extraction

recovery of unlabelled ANP added to plasma was

71%. Hormone data are not corrected for incomplete

recovery.

Renal interstitial, venous and pelvic pressures

Renal interstitial pressure was measured in both cam-

paigns. A standard 21-G, fluid-filled needle (Becton

Dickinson, Helsingborg, Sweden) was inserted trans-

rectally or via laparotomy into the kidney parenchyma

and connected to a pressure transducer placed at heart

level. Pressures were recorded on a portable monitor

(Mindray PM9000Vet). At insertion, a sharp increase

to a stable baseline pressure was observed. Flushing

with small volumes of saline (0.1–0.3 mL) caused

transient increases with pressure returning to baseline

within 10–20 s. Baseline pressure was taken as RIHP.

Finally, the needle was flushed with 100 lL Evans

blue solution to mark the position of the needle tip

(Fig. 1).

This protocol was identical in the two experiments

except that in experiment 2, animals were placed in

right lateral recumbency. For renal pelvis, ureter and

bladder pressures, a perineal incision was made, and

the urethra dissected, and through a small incision, a

fibre-optic scope was inserted (MAF; Olympus, Bal-

lerup, Denmark) to locate the ureter ostia. A side-

mounted tip-transducer catheter (SPC350; Millar

Instruments) was advanced up each ureter and pres-

sure recorded 5 cm above the ostium ureteris. After

10 min, the catheter was advanced slowly up the ure-

ter, during which regular peristaltic pressure waves

were observed. Just before wedging, the peristaltic

waves were no longer present. We interpreted this as

indicative of the pressure catheter being within the

renal pelvis. Subsequently, a left flank incision was

made and the left (intraperitoneally situated) kidney

exposed. The hilus was dissected, the renal vein identi-

fied, and a vascular sheath inserted in the renal vein

as close to the kidney as possible. Through the vascu-

lar sheath, a side-mounted tip-transducer catheter

(SPC350; Millar Instruments) was introduced and

renal vein pressure recorded. The catheter was then

slowly advanced into the abdominal vena cava during

pressure recording. Signals from the pressure transduc-

ers were recorded with a Biopac MP100 data acquisi-

tion system (Biopac Systems, Goleta, CA, USA) at

200 Hz.

Tissue biomechanical analysis

Segments of renal vessels and of strips of renal cap-

sules were obtained at autopsy. The biomechanical

analyses included cross-sectional area, A; ultimate (i.e.

maximal) load, F (N); ultimate stress, F/A (MPa); and

ultimate strain (fractional increase in length). The

maximum mechanical strength (ultimate tensile

Figure 1 Hemisected giraffe kidney. Arrows indicate: (EBT)

Evans blue trace left by the needle after recording of renal

interstitial hydrostatic pressure, (VP) vascular process, (C)

cortex, (M) medulla and (P) pelvis.

© 2015 Scandinavian Physiological Society. Published by John Wiley & Sons Ltd, doi: 10.1111/apha.125314

Renal function in the giraffe · M Damkjær et al. Acta Physiol 2015

strength) of collagen was calculated by dividing the

maximum load by (mg of collagen per mm original

strip length) (N 9 mm per mg collagen).

Tissue structural analyses

Upon completion of the physiological measurements,

the anaesthetized giraffes were killed by exsanguina-

tion. Both kidneys were removed within <1 h and per-

fused with 4% formaldehyde at a pressure of

�150 mmHg, immersed in 4% formaldehyde for 48 h

and subsequently flushed with phosphate-buffered sal-

ine (PBS). Sections of the abdominal aorta and vena

cava were dissected from 15 cm above to 15 cm

below the branching of the corresponding renal ves-

sels. The vessels were placed in 4% formaldehyde for

48 h and transferred to PBS. For stereological exami-

nation, the samples were embedded in paraffin and

sectioned (2 lm slices) for staining with haematoxy-

lin–eosin (H&E), Elastic Van Gieson and Masson’s

trichrome. Volume fraction of muscle tissue was esti-

mated using the point count tools of NEWCAST stereol-

ogy software (Visiopharm, Hørsholm, Denmark).

From the two kidneys, tissue for histological analysis

was sampled. Large mounted sections including both

cortex and medulla were sampled from one kidney,

whereas sections from the other kidney focused on ure-

ter, renal pelvis, pelvic and vascular extensions, attach-

ment of the capsule to the renal sinus and large vessels

in cortex and medulla. Large mounted sections were

cut at a thickness of 3–4 lm and stained with H&E

and PAS. Ordinary sections were also cut at 3–4 lmand stained with H&E, PAS, Masson’s trichrome, We-

igert’s elastin and PAS/alcian blue at pH 2.7. On these

sections, immunohistochemistry (Ventana system) was

performed for actin and a-smooth muscle actin.

To visualize geometric directions of pathways, we

performed diffusion tensor imaging (DTI) with MRI

on a giraffe kidney with a PHILIPS 1.5 T Achieva sys-

tem (Colchester, UK). The kidney was placed in the

magnet oriented with the long axis (upper to lower

pole) parallel to the axis of the main magnetic field,

and a surface receiver-coil was used for data recep-

tion. A diffusion tensor imaging sequence was applied

using 32 different diffusion-weighted directions. A

number of voxels of interest are selected from this

three-dimensional matrix, and based upon the charac-

teristics of the primary eigenvectors, the algorithm

then calculates any possible ‘track’, or pathway, which

passes through the chosen voxels of interest.

Statistical analysis

All values are presented as mean � standard error of

the mean (SEM). Comparisons were performed by

one-way analysis of variance (ANOVA) for repeated

measures. In case of significant differences, post hoc

Dunnett’s test was performed (Ludbrook 1998). Sta-

tistical calculations were performed with GRAPHPAD

PRISM (GraphPad Software, San Diego, CA, USA). Dif-

ferences were considered significant at P < 0.05; ‘n.s.’

denotes numerical deviation which is not statistically

significant.

Results

The anaesthesia allowed surgical preparation of the

giraffe followed by observation over 6–8 h. During

the measurements of renal clearance, mean arterial

blood pressure ranged from 168 � 7 to 242 � 8

mmHg and heart rate from 34 � 2 to 43 � 3 bpm.

Unless otherwise is stated in the text, results

represent the mean of observations obtained from nine

animals.

Renal function

Renal plasma flow (ERPF) was estimated from the

clearance of PAH, and the PAH concentration in

plasma was 210 � 3 lmol L�1 during the clearance

periods, resulting in a calculated ERPF of 1252 � 305

and 1286 � 212 mL min�1 (n.s.) during the first and

second period respectively. When normalized to total

body weight (BW), ERPF was 2.6 � 0.5 mL kg�1

BW min�1.

Glomerular filtration rate (GFR) was measured as

the clearance of sinistrin with stable plasma sinistrin

concentrations of 0.15 � 0.01 mg mL�1 during the

clearance periods. GFR measured 342 � 99 mL

min�1 in the first period and 348 � 67 in the second.

When normalized to BW, GFR was 0.7 � 0.1 mL

min�1 kg�1 BW. Filtration fraction was calculated as

GFR/ERPF and measured 0.3 � 0.1 and 0.3 � 0.0 in

the two periods respectively.

The timed collection of urine allowed the

determination of the rates of excretion of electrolytes. In

the first period, the rate of excretion of Na+ (NaEx) was

2769 � 793 lmol min�1 and in the second 2157 �752 lmol min�1 (n.s.). The corresponding rates of

excretion of K+ (KEx) were 1900 � 295 lmol min�1

and 1629 � 88 lmol min�1, respectively, in the two

periods (n.s.). The fractional sodium excretion was

5.8 � 3.4% and 4.6 � 2.0% (n.s.) and the fractional

potassium excretion 1.5 � 0.3 and 1.2 � 0.2 (n.s.).

Renal interstitial hydrostatic, renal vein, renal pelvis,

ureter and bladder pressures

During the first campaign, we measured renal interstitial

hydrostatic pressure in three animals lifted to the

© 2015 Scandinavian Physiological Society. Published by John Wiley & Sons Ltd, doi: 10.1111/apha.12531 5

Acta Physiol 2015 M Damkjær et al. · Renal function in the giraffe

upright position to be 146 � 7 mmHg; the concomitant

MAP at heart level was 190 � 20 mmHg. In all cases,

post-mortem examinations attested that the needle tip

had been located deep in the renal medulla. During the

second campaign, the measurements were repeated on

five giraffes under very similar conditions except that

the animals were in right lateral recumbency; in this ser-

ies, we obtained average pressures of 47 � 3 mmHg at

a concomitant MAP of 200 � 15 mmHg.

Renal vein pressure was 32 � 4 mmHg. The mean

pressure drop across the valve separating the renal

vein from the abdominal cava (Fig. 2) was 12 � 2

mmHg. Renal pelvic pressure was 39 � 2 mmHg.

The mean ureter pressure 5 cm proximal to the

ostium was 24 � 4 mmHg. Pelvic and ureteral pres-

sures were measured at bladder pressures of 8 �2 mmHg. All of these measurements were taken in the

second campaign.

Concentrations of hormones in plasma

Plasma renin concentration was measured using stan-

dard incubation conditions, and we secured that gir-

affe plasma generated angiotensin II immunoreactivity

as in other species. At the end of the first clearance

period, PRC was 2.6 � 0.5 mIU L�1 and 2.4 � 0.5

mIU L�1 by the end of the last period. Plasma

angiotensin II immunoreactivity was 9.1 � 1.5. The

concentrations of arginine vasopressin varied from

97 � 62 to 57 � 35 pg mL�1 in the first and second

period respectively. The immunoreactivity of atrial

natriuretic peptide was 123 � 18 pg mL�1. Plasma

noradrenaline was only determined in the first period

where the mean value was 1.1 � 0.9 pmol mL�1. For

overview of results, see Table 1.

Ultrasound investigations

Doppler curves from the intrarenal arteries (Fig. 3c)

yielded a RI of 0.26 � 0.07. Similar examination of 3

cows showed a RI of 0.64 � 0.03. All the RI results

together with MAP from the time of kidney ultra-

sound examination with a regression line are shown

in Figure 4. There was a highly significant inverse

correlation (P < 0.001) between the MAP and RI.

B-mode and colour Doppler pictures and films showed

live morphology of the kidney (Fig. 3a,b) with no visi-

ble pelvis and calyceal system. Dynamic Doppler

images showing blood flow through the renal vein

valve into the abdominal vena cava were also

obtained. These images showed that flow from the

renal vein into the cava was slow, intermittent and

turbulent (Video S1).

Biomechanics of renal capsule and vessels

The renal capsule was strong and anisotropic, being

stronger in the transverse than in the longitudinal

direction (Fig. 5a, Table 2). In direct comparison with

the cow, the giraffe renal capsule was twofold thicker

and had a much higher density (some fivefold larger

collagen content per unit area) and a correspondingly

high ultimate load (increased about sixfold). However,

when normalized to collagen content, capsular

strength of giraffes and cows was similar (i.e. normal-

ized ultimate strength, Table 2). The maximal pres-

sure that the renal capsule can resist (burst pressure)

was estimated from Young–Laplace law (DP = T(1/

r1 + 1/r2), DP is pressure difference, T wall tension

and r1 and r2 radii of orthogonal curvatures.); for the

giraffe kidney, the estimated burst pressure was

600–650 mmHg, and merely 125 mmHg for the cow.

There were no conspicuous differences between the

mechanical properties of renal vessels from giraffes

and cows except the higher ultimate strain in the renal

0 50 100 150 200 2500

10

20

30

40

mm

Hg

Renal vein

Caudal vena cava

Time (s)

Figure 2 Continuous pressure recording initially in the renal

vein; the pressure catheter is then advanced 5 cm so that it

enters the caudal vena cava (occurring at the arrow). This

illustrates the pressure drop occurring due to the valve at the

junction of the renal vein and the caudal cava. Oscillations

in venous pressure are caused by artificial ventilation.

Table 1 Immunoreactivity of giraffe plasma in conventional

radio-immunological assays, plasma renin being determined

the rate of generation of angiotensin I

Plasma hormone Period 1 Period 2

Renin (mIU L�1) 2.6 � 0.5 2.4 � 0.4

Angiotensin II (pg mL�1) 9.1 � 1.5 10.1 � 2.1

Arg vasopressin (pg mL�1) 97 � 62 57 � 35

Atrial natriuretic peptide

(pg mL�1)

123 � 18 95 � 15

Noradrenaline (pmol mL�1) 1.1 � 0.9 –

Values are mean �SEM = 9. No significant differences

between the clearance periods.

© 2015 Scandinavian Physiological Society. Published by John Wiley & Sons Ltd, doi: 10.1111/apha.125316

Renal function in the giraffe · M Damkjær et al. Acta Physiol 2015

artery of giraffes (Fig. 5b, Table 3) and a lower un-

strained luminal radius compared with cattle

(2.26 � 0.05 and 3.29 � 0.14 mm respectively). The

lower luminal radius and the higher ultimate strain in

the giraffe artery could be explained by a higher elas-

tin content (% of dry weight) (30.3 � 1.8 vs.

16.7 � 0.7 in cattle, Table 4), giving rise to elastic

recoil and the prolonged low load increment during

deformation.

Regional and renal morphology

The walls of caudal vena cava were very variable. For

the first 10 mm of the vena cava after the confluence

of the iliac veins, the volume fraction of medial

smooth muscle was 0.39 � 0.10 (n = 7) (Fig 6a).

From about 10 mm below to 10 mm above the

entrance of the renal vein, the vena cava wall was

thick with a volume fraction of medial smooth muscle

of 0.62 � 0.05 (n = 6; Fig. 5). Near the right atrium

(10–30 mm), the wall was thin with a volume fraction

of medial smooth muscle cells of 0.26 � 0.04 (n = 6;

Fig. 6). Within this vein, therefore, the increase in

wall thickness was associated with an increase in vol-

ume fraction of smooth muscle.

We relied on the terminology of Maluf (Maluf

2002) to describe the giraffe kidney. The microscopic

structure of the cortex was very similar to the human

kidney cortex; occasionally, the glomeruli were close

to the wall of the cortical radial arteries. The medul-

lary rays seemed broader than in the human counter-

part and to split the cortex with proximal tubules in

thin elongated bundles (Fig. 7a). The outer stripe of

the outer zone of renal medulla (OSOMZ) was thin,

whereas the inner stripe of the outer zone of renal

medulla (ISOMZ) was broad. In the functioning kid-

ney, the ISOMZ and the inner medulla will be

exposed to urine in the pelvis via the pelvic exten-

sions, which also contain the vascular processes. The

vascular processes were lined with urothelium and the

adjacent medullary zones were lined with a single

layer of columnar or cuboidal cells. Under this mono-

cellular layer, a-smooth muscle actin was apparent

and surrounded the adjacent vascular capillary bun-

dles (Fig. 7b). The vascular processes consisted of in-

terlobar blood vessels. These blood vessels were

situated in fibrous tissue (Fig. 7c). However, in some

part of the vascular processes, the interlobar blood

vessels were located in a less cellular connective tissue

positive for alcian blue at pH 2.7 (Fig. 7d). This con-

nective tissue appears similar to the mucoid, mesen-

(a) (b)

(c) (d)

Figure 3 Ultrasound with Doppler of

the giraffe kidney. (a) B-mode picture.

The pelvi-calyceal system not visible

(compressed). Vascular process (arrow).

(b) Colour Doppler showing flow in the

vessels. (c) Spectral Doppler with low

resistive index (RI). The vertical axis to

the left shows the Doppler shift in Khz

and the vertical axis to the right shows

the corresponding flow velocity of the

blood cells calculated from the Doppler

equation. The horizontal axis shows

time in seconds. (d) Renal vein (RV)

junction with vena cava. Note the valve

(arrow).

MAP (mmHg)

Res

istiv

e in

dex

0 50 100 150 200 2500.0

0.2

0.4

0.6

0.8

1.0

Figure 4 Plotted is corresponding value of resistive index

(RI) and mean arterial blood pressure (MAP) in 8 giraffes

(●), 3 cows (■), pigs (Rawashdeh et al. 2000) (▼) and the

standard reference value for health human subjects (Tublin

et al. 2003) (○). The solid line shows the correlation analysis

giving an R2 value of 0.77 (P < 0.001), and dotted lines indi-

cate 95% confidence interval.

© 2015 Scandinavian Physiological Society. Published by John Wiley & Sons Ltd, doi: 10.1111/apha.12531 7

Acta Physiol 2015 M Damkjær et al. · Renal function in the giraffe

chymal connective tissue found in the human umbilical

cord and in the human vitreous body of the eyeball.

The inner medulla consisted of the collecting ducts and

very prominent vascular bundles. A dense connective

tissue was found under the 5- to 6-cell-thick urotheli-

um.

The geometric directions of pathways in the kidneys

were visualized by directional colour coding using

Table 2 Biomechanical characteristics of renal capsules from giraffes (n = 7) and cows. Data from each giraffe are represented

by averages of 3 longitudinal and 3 transverse specimens from one kidney. The results for cow are based on 14 specimens of dif-

ferent orientation sampled from both kidneys

Renal capsule Thickness (mm)

Collagen

(mg cm�2)

Ultimate

strength (N)

Ultimate

stress (MPa)

Normalized ultimate

strength (N 9 mm per

mg collagen)

Giraffe

Combined 0.202 � 0.011 4.25 � 0.24 15.1 � 1.0 14.6 � 0.6 69.0 � 2.3

Longitudinal 0.218 � 0.015 4.41 � 0.36 13.0 � 1.4 11.5 � 1.0 56.8 � 4.5

Transverse 0.187 � 0.008 4.09 � 0.18 17.2 � 1.0 17.7 � 0.8 81.2 � 3.4

Cow 0.081 � 0.006 0.88 � 0.11 2.5 � 0.2 5.9 � 0.5 58.1 � 4.6

Data are mean �SEM.

(a)

(b)

Figure 5 Biomechanical properties. (a)

Biomechanics of renal vessels from gir-

affes and cattle (one cow and one young

bull). Mean load–strain curves and the

mean parameters (ultimate strain, ulti-

mate load) are shown. Error bars �SEM

(for cattle load–strain curves plotted

without error bars). (b) Biomechanics of

renal capsule from giraffes and cows.

Mean load–strain curves and the mean

parameters (ultimate strain, ultimate

load) are shown. Error bars �SEM.

© 2015 Scandinavian Physiological Society. Published by John Wiley & Sons Ltd, doi: 10.1111/apha.125318

Renal function in the giraffe · M Damkjær et al. Acta Physiol 2015

diffusion tensor imaging (Fig. 8). The centripetal

alignment of these tracks probably represents drainage

of the medullary tubules and collecting ducts into the

pelvis.

Discussion

We demonstrate that giraffes have lower values of

GFR and ERPF than those of similar-sized mammals,

whilst filtration fraction resembles other mammals.

The low GFR is remarkable considering the very high

pressure of the arterial blood perfusing the kidneys,

which would be expected to increase filtration pres-

sure and hence cause a very high GFR. This somewhat

counterintuitive findings appears to be, at least par-

tially, explained the very high interstitial hydrostatic

pressure within the Bowman capsule, a feature made

possible by two unique adaptions: i) a thick and dura-

ble renal capsule that can sustain the high pressure

and ii) a valve structure at the junction of the renal

vein with the vena cava that allows for high renal

venous pressure.

Renal function

The GFR of 0.7 mL kg�1 min�1 of a 500-kg giraffe is

approx. 40% lower than predicted value of around

1.2 mL kg�1 min�1 from the scaling relationships

provided by Singer (2001). In horses with a body mass

ranging from 436 to 682 kg, reported GFR values

range from 1.68 to 2.69 mL kg�1 min�1 (Wilson

Table 3 Biomechanical characteristics of renal vessels in giraffe and cattle. Data from each animal are represented by the aver-

age of 3 ring samples from each vessel

Renal vessel Species Cross-sectional area (mm2) Ultimate load (N) Ultimate stress (MPa) Ultimate strain (�)

Vein Giraffe 2.13 � 0.12 1.16 � 0.11 0.56 � 0.06 0.421 � 0.036

Cattle 1.99 � 0.47 1.51 � 0.69 0.89 � 0.55 0.394 � 0.046

Artery Giraffe 2.78 � 0.06 3.76 � 0.31 1.36 � 0.12 0.970 � 0.033

Cattle 2.49 � 0.42 2.67 � 0.86 1.04 � 0.17 0.702 � 0.002

Data are mean �SEM (n = 10 giraffes and 2 for cattle).

Table 4 Table s-CC3. Elastin and collagen content in renal

vessels in giraffe and cattle

Vessel Species Elastin (%) Collagen (%)

V. renalis Giraffes 12.3 � 1.4 52.6 � 2.2

Cattle 17.0 � 8.0 41.7 � 6.2

A. renalis Giraffes 30.3 � 1.8 44.3 � 1.5

Cattle 16.7 � 0.7 30.9 � 5.4

Data are mean �SEM per cent of dry defatted weight. Each

animal is represented by the average of 2–5 ring samples

from each vessel (n = 10 giraffes and 2 for cattle).

(a) (b)

(c)

Figure 6 Caval vein histology: Sections taken at three different locations along the caval vein of the giraffe. (a) Bar graph show-

ing statistically significant differences in the volume fraction of smooth muscle cells 3 cm below R. atrium, 1 cm above renal

vein and 1 cm above iliac vein. (*P < 0.05). (b) Masson trichrome staining of vena cava inferior (smooth muscle tissue is red,

collagen tissue, blue). B1: 1 cm above the iliac region, B2: 1 cm above the renal vein and B3: 3 cm below the Right atrium. In

B3, a band of slightly stained smooth muscle cells are seen under the intima (↑). (c) Sections from the same locations in another

giraffe, but in a smaller magnification.

© 2015 Scandinavian Physiological Society. Published by John Wiley & Sons Ltd, doi: 10.1111/apha.12531 9

Acta Physiol 2015 M Damkjær et al. · Renal function in the giraffe

et al. 2009), supporting our vies of a surprisingly low

GFR in giraffes. The comparisons amongst mammals

of different sizes show that mass-specific GFR

decreases with body mass in a similar manner as the

mass-specific rate of metabolism, which has led to the

speculation that GFR matches catabolic bodily pro-

cesses. The low GFR in giraffes could either mean that

they do not obey the overall scaling relationship for

mammals, or that giraffes have low mass-specific

metabolism. These two explanations are not mutually

exclusive and it is noteworthy that the scaling rela-

tionship between GFR and metabolism is based

almost entirely on data from mammals with BW

below 40 kg (Singer & Morton 2000) and that simple

allometric scaling laws remain disputed (Glazier

2005). Nevertheless, the low GFR is consistent with

our measurements of a low cardiac output in giraffes

(Damkjaer et al. 2011), which may indicate a rela-

tively low basal metabolism and emphasizing a the

link between cardiac and renal function (Burggren

et al. 2014). In any event, in relation to the question

of overall kidney function, the low GFR in compari-

son with other mammals is remarkable given their

very high MAP, and warrants further explanation.

From the Starling equation, that describes glomeru-

lar filtration, it would be tempting to speculate that

the glomerular membrane of the giraffe was endowed

with different mechanical properties to enable toler-

ance to a high transmural pressure and a low Kf either

by virtue of low capillary permeability or surface area.

Such a mechanism would presumably result in lower

filtration fractions, which is inconsistent with our

measurements of 0.3, which is in the high end com-

pared to values reported for humans (Frank et al.

(a) (b)

(c) (d)

Figure 7 Renal histology: (a) giraffe

renal cortex. The medullary rays split

the proximal tubules in thin elongated

bundles, Masson’s trichrome stain. (b)

a-smooth muscle actin is present under

the thin lining cells of the medullary

zones facing the pelvic extensions,

a-smooth muscle actin and counterstain

H&E. (c) The interlobar blood vessels of

the vascular extension with surrounding

fibrous tissue, Masson’s trichrome. (d) In

other parts of the vascular processes, the

interlobar blood vessels are located in a

less cellular connective tissue that is

positive for alcian blue at pH 2.7.

(a) (b)

Figure 8 Diffusion tensor imaging with magnetic resonance imaging of giraffe kidney. The fibres in this transverse section show

a centripetal alignment of calculated tracks (a), representing the medullary tubules, and they follow the minor calyces and join

together to form major calyces which in turn drain into the pelvis (b). Track colours are directional colour coatings derived

from the orientation of the diffusion direction relative to the scanner coordinate system. The colours are therefore merely a

visual aid and do not represent any anatomical or physiological properties.

© 2015 Scandinavian Physiological Society. Published by John Wiley & Sons Ltd, doi: 10.1111/apha.1253110

Renal function in the giraffe · M Damkjær et al. Acta Physiol 2015

2009), dogs (Emmeluth et al. 1987), rats (Myers et al.

1975) and mice (Sallstrom & Friden 2013). Also, both

our present study and that of Maluf (Maluf 2002)

show normal microscopic structure of the giraffe cor-

tex and glomeruli, leaving no evidence for peculiar

mechanical properties of the giraffe glomerular mem-

brane. Instead, our measurements of the high renal

interstitial hydrostatic pressure could account for a

reduction in transmural pressure across Bowman’s

capsule that would reduce GFR in spite of the high

MAP.

Biomechanical properties of the renal capsule

With regard to thickness and collagen content, the gir-

affe renal capsule was considerably stronger than the

bovine renal capsule, and the ultimate stress value for

the cow capsule is similar to that of other mammals

(Herbert et al. 1976, Snedeker et al. 2005). Neverthe-

less, the thickness of the renal capsules in our study

(0.20 mm) is even lower than the 0.75 mm reported

by Maluf (Maluf 2002). This difference in thickness

could be age-related as the age of the animals in the

latter study was 18 and 22 years. In any case, the

thicker and stronger capsule in the giraffe can with-

stand high intrarenal pressures as illustrated by the

estimated burst pressure of 600–650 mmHg compared

to around 125 mmHg in cow.

From a physiological point of view, the pressure at

the inflexion point of the load–strain curve is interest-

ing as this point represents the maximum elastic mod-

ulus; straining above this point is supposed to initiate

damage of the collagen structure (Herman 2007). The

inflexion points (strain, load) for the giraffe renal cap-

sule of (0.17, 7.5 N) and for cow (0.15, 1.4 N) allow

calculation of pressures at inflexion point of 350 and

75 mmHg respectively. Also by measure of the inflex-

ion point pressure, the renal capsule of the giraffe is

much stronger than that of the cow.

Ultrasound and Doppler curves

The average renal resistance index (RI) in the giraffes

was 0.26, that is much lower than the values of

around 0.60 normally seen in humans (Tublin et al.

2003) and in the various species included in Figure 4.

The significance of RI in the kidney is not fully

understood (Tublin et al. 2003), and the correlation

with vascular resistance is less distinct than believed

at the time when present experiments were performed

(O’Neill 2014). We have analysed the correlation

between MAP and RI (Fig. 4) and have found a

strong inverse correlation between a high MAP and a

low RI, so the low RI in the giraffe may therefore be

part of a more general trend. This observation is

indeed consistent with the emerging vies that RI may

say more about systemic haemodynamics than renal

resistance (Chirinos & Townsend 2014, Viazzi et al.

2014). But even if there is no strong correlation

between RI and vascular resistance, the low RI does

at least not indicate a high vascular resistance in the

giraffe. If we assume that the values we measured on

the two separate occasions (MAP 190 mmHg, renal

vein pressure 32 mmHg, ERPF 1286 mL min�1 and

assume haematocrit is 0.45) are representative, we can

use this to estimate renal vascular conductance to

11 mL (mmHg min)�1 which is not markedly differ-

ent from values reported for healthy humans subjects

(15–17 mL (mmHg min)�1) (Damkjaer et al. 2014),

but provides no information about the regional resis-

tance, that is whether the pre- and post-capillary resis-

tances have a different distribution in giraffes

compared with other animals.

Visualization by ultrasound of the semilunar valve

(Fig. 9) partially obstructing the inflow from the venal

vein to the cava clearly showed that the function of

the valve was intermittent (Video S1). Direct measure-

ment of the renal venous pressure revealed that on

average, there was a 12-mmHg pressure drop across

the valve from the renal vein to the cava. It is tempt-

ing to speculate that the function of the valve is to

help maintain a high renal venous pressure supporting

a high renal interstitial pressure. We recorded the

interstitial pressure in the medullary region of the kid-

ney at two separate experiments and found it to be

146 � 7 and 47 � 3 mmHg respectively. We cannot

account for the difference between these two data sets

both derived from measurements in several giraffes.

The first set of (high) values was quite consistent over

Figure 9 Post-mortem section through the caudal vena cava

at the level of renal vein entry. A stippled line highlights the

edge of the semilunar valve partially obstructing the inflow

from the renal vein to the cava. The direction of blood is

shown for both vena cava (solid arrow) and out from the

renal vein into the vena cava (stippled arrow).

© 2015 Scandinavian Physiological Society. Published by John Wiley & Sons Ltd, doi: 10.1111/apha.12531 11

Acta Physiol 2015 M Damkjær et al. · Renal function in the giraffe

five measurements in four giraffes held upright on a

tilt board by straps around the head and legs; some of

these measurements were taken through a window in

the abdominal wall with the left kidney (in the giraffe

completely covered by peritoneal membrane) immobi-

lized manually. It is possible that physical compression

and/or puncture of arterial vessels may account for

the high pressures at the tip of the needle under these

conditions, but it seems less likely. If we assume that

the ratio of pre- to post-glomerular RVR is similar to

other mammals, that is 50% : 50%, then a RIHP of

146 mmHg seems unreasonable high, whereas in the

case of the recumbent values, this would give a more

reasonable value. However, even the lower value is

substantially above the values reported for other

mammals (Nakamura et al. 1998, Majid et al. 2001).

The renal pelvic pressure was 39 mmHg; as we

assume that these measurements reflect the pressure

inside the renal capsule, the result might also be an

indication of the augmentation of renal interstitial

pressure in this species.

Renal and perirenal morphology

The structure of the caudal vena cava at the entry of

the renal vein was markedly different from sections of

the same vein before and after the region of entry,

indicating a low compliance of the vena cava at the

level of renal vein entry. The functional significance of

this is not clear, but this localized structural change

warrants a detailed description of the longitudinal

pressure profile within the vena cava from the iliac

veins to the atrium. It is possible that the relative stiff-

ness of the vena cava at the entry of the renal veins is

haemodynamically significant by being related to the

functionally important pressure gradient existing at

this location.

The renal cortex was quite similar to the human

renal cortex, except that the cortical proximal

tubules between medullary rays were arranged into

thin elongated bundles. The major difference

between the giraffe kidney and other mammalian

kidneys is the complete absence of cortical tissue

(renal columns) between the pyramids. The pelvic

fluid space is extended to the outer (cortical) level

of the medulla. The functional significance of this

remains unknown, but may be related to unusual

haemodynamic function(s) within the vascular pro-

cesses. In the medulla, the major difference was seen

in the outer stripe of the outer medullary zone that

is scarce bordering to absent. This may indicate a

rearrangement of the tubular structures normally

defining this region (proximal straight tubule and

the most distal portion of the thick ascending loop

of Henle) and warrants further investigation of renal

tubular morphology in this species. Interestingly, a-smooth muscle actin was present beneath the single

cell layer of the medullary zones facing the pelvic

extension. It is likely that the majority of the a-smooth muscle actin is located in the extracellular

space with functional significance for medullary

dynamics.

The functional importance of the presence of what

appears to be mesenchymal connective tissue in part

of the vascular processes is uncertain; similar kinds of

mesenchymal connective tissues are found in the

umbilical cord and in the human vitreous body of the

eyeball. The special connective tissue of the vascular

processes could provide the vascular processes with

unusual biomechanical properties.

Plasma hormones

To our knowledge, measurements of vasoactive hor-

mone concentrations in giraffe plasma have not been

reported before. The measurements in plasma of spe-

cific immunoactivities of hormones known in other

species indicate that the giraffe plasma contained the

peptide hormones towards which the antibodies are

raised, or at least very similar substances. More defini-

tive identification could be obtained, for example, by

mass spectrometry.

Assuming that our assays measured the designated

hormones, the plasma levels were analogous to the

concentrations seen in other species with the exception

of vasopressin concentrations. The latter were high

and highly variable, warranting further studies of the

nature of the reactant(s) and – if arginine vasopressin

were to identified – of the reasons for the high and

variable plasma concentrations obtained under the

present conditions.

Perspectives

We have demonstrated that giraffes have a lower

GFR than similar-sized mammals despite their very

high blood pressures. This remarkable feature

appears to due to very high interstitial pressures

within the kidney, enabled by a thick and strong

capsule and a valve at the renal vein entry into the

abdominal cava. These relatively simple structural

modifications serve to normalize the Starling forces

driving filtration over the Bowman capsule. Thus, in

line with very tight fascia in the legs that effectively

function as an internal ‘g-suit’ (Hargens et al. 1987,

Nilsson et al. 1988), our study proposed a similar

internal g-suit in the kidneys that reduces the pres-

sure gradient across the glomerular membrane and

might protect the giraffe kidney from hypertensive

damage.

© 2015 Scandinavian Physiological Society. Published by John Wiley & Sons Ltd, doi: 10.1111/apha.1253112

Renal function in the giraffe · M Damkjær et al. Acta Physiol 2015

Financial disclosure

The study was supported by the Lundbeck Foundation,

Carlsbergfondet, The Danish Heart Association, the

Aase and Ejnar Danielsen Foundation, The Danish

Research Council, The Danish Cardiovascular Research

Academy, Nyreforeningens Forskningsfond, Fonden til

Lægevidenskabens Fremme, The Faculty of Health

Science and the Faculty of Natural Sciences at Aarhus

University and the Aarhus University Research Founda-

tion. The funders had no role in study design, data col-

lection and analysis, decision to publish or preparation

of the manuscript.

Conflict of interest

There is no conflict of interest.

The expert technical assistance of Bodil Aa. Kristensen in the

field as well as with the assays is greatly appreciated. The

authors appreciate the expert animal handling provided dur-

ing the study by Charles van Niekerk, Martin Krogh, Dean

and Nicky Riley-Hawkins as well as the staff at Wildlife

Assignments International, Hammanskraal, South Africa.

References

Barri, Y.M. 2006. Hypertension and kidney disease: a deadly

connection. Curr Cardiol Rep 8, 411–417.

Brondum, E., Hasenkam, J.M., Secher, N.H., Bertelsen,

M.F., Grondahl, C., Petersen, K.K., Buhl, R., Aalkjaer, C.,

Baandrup, U., Nygaard, H. et al. 2009. Jugular venous

pooling during lowering of the head affects blood pressure

of the anesthetized giraffe. Am J Physiol Regul Integr

Comp Physiol 297, R1058–R1065.

Burggren, W.W., Christoffels, V.M., Crossley, D.A. 2nd,

Enok, S., Farrell, A.P., Hedrick, M.S., Hicks, J.W., Jensen,

B., Moorman, A.F., Mueller, C.A., Skovgaard, N., Taylor,

E.W. & Wang, T. 2014. Comparative cardiovascular phys-

iology: future trends, opportunities and challenges. Acta

Physiol (Oxf) 210, 257–276.

Chirinos, J.A. & Townsend, R.R. 2014. Systemic arterial

hemodynamics and the “renal resistive index”: what is in a

name? J Clin Hypertens (Greenwich) 16, 170–171.

Damkjaer, M., Bertelsen, M.F., Grondahl, C., Hasenkam,

J.M., Wang, T., Brondum, E., Candy, G. & Bie, P.

2011. Low blood volume in the giraffe (Giraffa camelo-

pardalis). FASEB J 25(Meeting Abstract Supplement),

1027.17.

Damkjaer, M., Jensen, P.H., Schwammle, V., Sprenger, R.R.,

Jacobsen, I.A., Jensen, O.N. & Bie, P. 2014. Selective renal

vasoconstriction, exaggerated natriuresis and excretion

rates of exosomic proteins in essential hypertension. Acta

Physiol (Oxf) 212, 106–118.

Emmeluth, C., Eiken, P., Johannessen, A.C. & Bie, P. 1987.

Effects of atrial natriuretic peptides and furosemide in con-

scious dogs. Proc Soc Exp Biol Med 186, 103–112.

Feld, L.G., Cachero, S., Van Liew, J.B., Zamlauski-Tucker,

M. & Noble, B. 1990. Enalapril and renal injury in spon-

taneously hypertensive rats. Hypertension 16, 544–554.

Frank, H., Graf, J., Amann-Gassner, U., Bratke, R., Daniel,

H., Heemann, U. & Hauner, H. 2009. Effect of short-term

high-protein compared with normal-protein diets on renal

hemodynamics and associated variables in healthy young

men. Am J Clin Nutr 90, 1509–1516.

Glazier, D.S. 2005. Beyond the ‘3/4-power law’: variation in

the intra- and interspecific scaling of metabolic rate in ani-

mals. Biol Rev Camb Philos Soc 80, 611–662.

Goetz, R.H. 1955. Preliminary observations on the circula-

tion in the giraffe. Trans Am Coll Cardiol 5, 239–248.

Goetz, R.H. & Keen, E.N. 1957. Some aspects of the cardio-

vascular system in the giraffe. Angiology 8, 542–564.

Goetz, R.H., Warren, J.V., Gauer, O.H., Patterson, J.L. Jr,

Doyle, J.T., Keen, E.N. & Mc, G.M. 1960. Circulation of

the giraffe. Circ Res 8, 1049–1058.

Hargens, A.R., Millard, R.W., Pettersson, K. & Johansen, K.

1987. Gravitational haemodynamics and oedema preven-

tion in the giraffe. Nature 329, 59–60.

Herbert, L.A., Chen, W.C., Hartmann, A. & Garancis, J.C.

1976. Mechanical properties of the dog renal capsule.

J Appl Physiol 40, 164–170.

Herman, I.P. 2007. Physics of the Human Body. Biological

and Medical Physics, Biomedical Engineering. Springer,

Berlin, Heidelberg.

Hill, G.S. 2008. Hypertensive nephrosclerosis. Curr Opin

Nephrol Hypertens 17, 266–270.

Kjolby, M.J., Kompanowska-Jezierska, E., Wamberg, S. &

Bie, P. 2005. Effects of sodium intake on plasma potas-

sium and renin angiotensin aldosterone system in conscious

dogs. Acta Physiol Scand 184, 225–234.

Ludbrook, J. 1998. Multiple comparison procedures updated.

Clin Exp Pharmacol Physiol 25, 1032–1037.

Majid, D.S., Said, K.E., Omoro, S.A. & Navar, L.G. 2001.

Nitric oxide dependency of arterial pressure-induced

changes in renal interstitial hydrostatic pressure in dogs.

Circ Res 88, 347–351.

Maluf, N.S. 1994. Further studies on the kidney of the hook-

lipped African rhinoceros, Diceros bicornis. Anat Rec 238,

38–48.

Maluf, N.S. 2002. Kidney of giraffes. Anat Rec 267, 94–

111.

Myers, B.D., Deen, W.M. & Brenner, B.M. 1975. Effects of

norepinephrine and angiotensin II on the determinants of

glomerular ultrafiltration and proximal tubule fluid reab-

sorption in the rat. Circ Res 37, 101–110.

Nakamura, T., Alberola, A.M., Salazar, F.J., Saito, Y., Ku-

rashina, T., Granger, J.P. & Nagai, R. 1998. Effects of

renal perfusion pressure on renal interstitial hydrostatic

pressure and Na+ excretion: role of endothelium-derived

nitric oxide. Nephron 1998, 104–111.

Nilsson, O., Booj, S., Dahlstrom, A., Hargens, A.R., Millard,

R.W. & Pettersson, K.S. 1988. Sympathetic innervation of

the cardiovascular system in the giraffe. Blood Vessels 25,

299–307.

O’Neill, W.C. 2014. Renal resistive index: a case of mistaken

identity. Hypertension 64, 915–917.

© 2015 Scandinavian Physiological Society. Published by John Wiley & Sons Ltd, doi: 10.1111/apha.12531 13

Acta Physiol 2015 M Damkjær et al. · Renal function in the giraffe

Platt, J.F., Rubin, J.M., Ellis, J.H. & DiPietro, M.A. 1989.

Duplex Doppler US of the kidney: differentiation of

obstructive from nonobstructive dilatation. Radiology 171,

515–517.

Poulsen, K. & Jorgensen, J. 1974. An easy radioimmunologi-

cal microassay of renin activity, concentration and sub-

strate in human and animal plasma and tissues based on

angiotensin I trapping by antibody. J Clin Endocrinol

Metab 39, 816–825.

Rawashdeh, Y.F., Mortensen, J., Horlyck, A., Olsen, K.O.,

Fisker, R.V., Schroll, L. & Frokiaer, J. 2000. Resistive

index: an experimental study of the normal range in the

pig. Scand J Urol Nephrol 34, 10–14.

Sallstrom, J. & Friden, M. 2013. Simultaneous determination

of renal plasma flow and glomerular filtration rate in con-

scious mice using dual bolus injection. J Pharmacol Toxi-

col Methods 67, 187–193.

Singer, M.A. 2001. Of mice and men and elephants: meta-

bolic rate sets glomerular filtration rate. Am J Kidney Dis

37, 164–178.

Singer, M.A. & Morton, A.R. 2000. Mouse to elephant: bio-

logical scaling and Kt/V. Am J Kidney Dis 35, 306–309.

Snedeker, J.G., Niederer, P., Schmidlin, F.R., Farshad, M.,

Demetropoulos, C.K., Lee, J.B. & Yang, K.H. 2005.

Strain-rate dependent material properties of the porcine

and human kidney capsule. J Biomech 38, 1011–1021.

Syme, H. 2011. Hypertension in small animal kidney

disease. Vet Clin North Am Small Anim Pract 41,

63–89.

Tublin, M.E., Bude, R.O. & Platt, J.F. 2003. Review. The

resistive index in renal Doppler sonography: where do we

stand? AJR Am J Roentgenol 180, 885–892.

Van Citters, R.L., Kemper, W.S. & Franklin, D.L. 1966.

Blood pressure responses of wild giraffes studied by radio

telemetry. Science 152, 384–386.

Van Citters, R.L., Kemper, W.S. & Franklin, D.L. 1968.

Blood flow and pressure in the giraffe carotid artery. Comp

Biochem Physiol 24, 1035–1042.

Van Citters, R.L., Franklin, D.L., Vatner, S.F., Patrick, T. &

Warren, J.V. 1969. Cerebral hemodynamics in the giraffe.

Trans Assoc Am Physicians 82, 293–304.

Viazzi, F., Leoncini, G., Derchi, L.E. & Pontremoli, R. 2014.

Ultrasound Doppler renal resistive index: a useful tool for

the management of the hypertensive patient. J Hypertens

32, 149–153.

Wilson, K.E., Wilcke, J.R., Crisman, M.V., Ward, D.L.,

McKenzie, H.C. & Scarratt, W.K. 2009. Comparison of

serum iohexol clearance and plasma creatinine clearance in

clinically normal horses. Am J Vet Res 70, 1545–1550.

Supporting Information

Additional Supporting Information may be found in

the online version of this article:

Video S1. In-vivo B-mode ultrasound recording,

showing the slow, intermittent, turbulant blood flow

from the renal vein into the abdominal cava.

© 2015 Scandinavian Physiological Society. Published by John Wiley & Sons Ltd, doi: 10.1111/apha.1253114

Renal function in the giraffe · M Damkjær et al. Acta Physiol 2015