14
http://vet.sagepub.com/ Veterinary Pathology Online http://vet.sagepub.com/content/early/2014/01/06/0300985813516642 The online version of this article can be found at: DOI: 10.1177/0300985813516642 published online 6 January 2014 Vet Pathol C. Shilton, G. P. Brown, L. Chambers, S. Benedict, S. Davis, S. Aumann and S. R. Isberg ) in Australia Crocodylus porosus Pathology of Runting in Farmed Saltwater Crocodiles ( Published by: http://www.sagepublications.com On behalf of: Pathologists. American College of Veterinary Pathologists, European College of Veterinary Pathologists, & the Japanese College of Veterinary can be found at: Veterinary Pathology Online Additional services and information for http://vet.sagepub.com/cgi/alerts Email Alerts: http://vet.sagepub.com/subscriptions Subscriptions: http://www.sagepub.com/journalsReprints.nav Reprints: http://www.sagepub.com/journalsPermissions.nav Permissions: What is This? - Jan 6, 2014 OnlineFirst Version of Record >> at University of Sydney on January 13, 2014 vet.sagepub.com Downloaded from at University of Sydney on January 13, 2014 vet.sagepub.com Downloaded from

Pathology of Runting in Farmed Saltwater Crocodiles (Crocodylus porosus) in Australia

Embed Size (px)

Citation preview

http://vet.sagepub.com/Veterinary Pathology Online

http://vet.sagepub.com/content/early/2014/01/06/0300985813516642The online version of this article can be found at:

 DOI: 10.1177/0300985813516642

published online 6 January 2014Vet PatholC. Shilton, G. P. Brown, L. Chambers, S. Benedict, S. Davis, S. Aumann and S. R. Isberg

) in AustraliaCrocodylus porosusPathology of Runting in Farmed Saltwater Crocodiles (  

Published by:

http://www.sagepublications.com

On behalf of: 

Pathologists.American College of Veterinary Pathologists, European College of Veterinary Pathologists, & the Japanese College of Veterinary

can be found at:Veterinary Pathology OnlineAdditional services and information for    

  http://vet.sagepub.com/cgi/alertsEmail Alerts:

 

http://vet.sagepub.com/subscriptionsSubscriptions:  

http://www.sagepub.com/journalsReprints.navReprints:  

http://www.sagepub.com/journalsPermissions.navPermissions:  

What is This? 

- Jan 6, 2014OnlineFirst Version of Record >>

at University of Sydney on January 13, 2014vet.sagepub.comDownloaded from at University of Sydney on January 13, 2014vet.sagepub.comDownloaded from

Article

Pathology of Runting in Farmed SaltwaterCrocodiles (Crocodylus porosus) in Australia

C. Shilton1, G. P. Brown2, L. Chambers1, S. Benedict1, S. Davis1,S. Aumann1, and S. R. Isberg2,3

AbstractExtremely poor growth of some individuals within a birth cohort (runting) is a significant problem in crocodile farming. Weconducted a pathological investigation to determine if infectious disease is associated with runting in farmed saltwater crocodiles(Crocodylus porosus) and to look for evidence of other etiologies. In each of 2005 and 2007, 10 normal and 10 runt crocodiles, withan average age of 5.5 months and reared under identical conditions, were sampled. Laboratory testing included postmortem;histological examination of a wide variety of tissues (with quantitation of features that were noted subjectively to be differentbetween groups); hematology; serum biochemistry (total protein, albumin, globulins, total calcium, phosphorus, and iron);bacterial culture of liver and spleen (2005 only); viral culture of liver, thymus, tonsil, and spleen using primary crocodile cell lines(2007 only); and serum corticosterone (2007 only). The only evidence of infectious disease was mild cutaneous poxvirus infectionin 45% of normal and 40% of runt crocodiles and rare intestinal coccidia in 5% of normal and 15% of runt crocodiles. Bacterial andviral culture did not reveal significant differences between the 2 groups. However, runt crocodiles exhibited significant (P < .05)increases in adrenocortical cell cytoplasmic vacuolation and serum corticosterone, decreased production of bone (osteoporosis),and reduced lymphoid populations in the spleen, tonsil, and thymus. Runts also exhibited moderate anemia, hypoalbuminemia, andmild hypophosphatemia. Taken together, these findings suggest an association between runting and a chronic stress response(hyperactivity of the hypothalamic-pituitary-adrenal axis).

Keywordsadrenocortical hyperplasia, crocodile, farmed, histopathology, inanition, lymphoid atrophy, runting, stress

In the crocodilian farming industry, extremely poor growth of

some hatchlings compared with their birth cohort (runting) is a

major problem worldwide, affecting up to 30% of ani-

mals.3,21,32,33,48,49,55,58 Runting is a cause of major economic

loss, in an industry where juvenile growth and survival are

important economic factors.33

The cause of runting in farmed crocodiles is unknown.

Although various possibilities have been proposed, none have

been thoroughly investigated. Proposed factors associated with

runting include egg incubation conditions;38,69 poor yolk

absorption posthatching;22,32 inappropriate diet or problems

with diet assimilation;11,22,26,55 posthatching environment,

including temperature, stocking density, and behavior of con-

specifics;26,32,39,48,58 and failure to adapt to the captive envi-

ronment.3,64 Poor growth has been found to be clutch related

in that some clutches produce relatively high numbers of

runts.26,33,48,58 However, future crocodile runts are not identifi-

able at hatching, since body size at hatching is a poor predictor

of posthatching growth.26,34,38,69

There is limited information on diseases and pathology exhib-

ited by runt crocodiles. A few studies mention bacteremia,

mycotic dermatitis,64 hepatitis,22,64 atrophy of the liver and

intestine and ascites,21,32 hepatic lipidosis,21,49 or pancreatic

atrophy.22 There has been no in-depth pathological investigation

comparing runt crocodiles with their normal counterparts from

the same cohort and raised under identical conditions. The pur-

pose of this study was to conduct a direct and thorough patholo-

gical comparison between runt and normal crocodiles to rule out

the involvement of infectious disease and to survey for other

lesions or conditions that could suggest a cause for the runting.

Materials and Methods

The study location was a large saltwater crocodile (Crocodylus

porosus) farm 40 km south of Darwin, Australia. The weather

1Berrimah Veterinary Laboratories, Department of Primary Industry and

Fisheries, Northern Territory Government, Berrimah, Northern Territory,

Australia2The University of Sydney, Sydney, New South Wales, Australia3Porosus Pty. Ltd., Noonamah, Northern Territory, Australia

Corresponding Author:

C. Shilton, Berrimah Veterinary Laboratories, Department of Primary Industry

and Fisheries, Northern Territory Government, GPO Box 3000, Darwin, NT

0801, Australia.

Email: [email protected]

Veterinary Pathology1-13ª The Author(s) 2014Reprints and permission:sagepub.com/journalsPermissions.navDOI: 10.1177/0300985813516642vet.sagepub.com

Veterinary Pathology OnlineFirst, published on January 6, 2014 as doi:10.1177/0300985813516642

at University of Sydney on January 13, 2014vet.sagepub.comDownloaded from

in this region is tropical, with a ‘‘wet’’ season having high

humidity and monsoonal rains, from November to April, and

a ‘‘dry’’ season, characterized by low humidity and no rain,

from May to October. The daily maximum temperature is

33�C during both seasons, but the average daily minimum is

24�C in the wet season compared with 18�C in the dry season.

Saltwater crocodiles nest in northern Australia from late

October to April, corresponding to a hatching period from Feb-

ruary to June. Eggs were obtained both from the wild (13 runt

and 10 normal crocodiles; average estimated embryo age at

collection was 21 days) and from captive females (7 runt and

10 normal crocodiles; average estimated embryo age was 6

days). At the farm, eggs were incubated at 32�C and 99% to

100% humidity until hatching. Each hatchling was identified

to clutch of origin by scute cutting.30 The hatchlings used in

this study originated from 35 clutches (1 from each clutch,

except for 5 clutches from which 2 study animals originated

from each clutch).

Runt crocodiles were not identifiable at hatching. On croco-

dile farms, normal management practice involves ‘‘grading’’

animals to ensure minimal size variation within each pen to

reduce intraspecific aggression and maximize equal opportu-

nity to access food. Grading on our study farm commences

when disparity in growth starts to become apparent, at approx-

imately 2 months of age, and is performed as required but at

least every month within each pen. Thus, through systematic

grading, the runt crocodiles used in this study were continually

maintained in pens of similar sized animals and were not

selected from pens of larger animals whereby exogenous fac-

tors such as competition for food or thermoregulatory ability

could be a cause of the runting. Runt crocodiles were reared

under identical husbandry conditions to normal crocodiles of

the same birth cohort, with pens containing crocodiles of either

of the 2 groups interspersed in the same shed. Stocking density

in all pens was approximately 14 animals/m2. Within each

completely enclosed shed, pens were shaded, of concrete

construction, with shallow water (approx. 70% of pen area;

30–50 cm deep) at one end with a gradual ascent to a dry

feed-deck (approx. 30% of pen area) at the other end. Hide-

boards were provided within each pen. Water temperature in

the pens was kept at approximately 32�C. The diet consisted

of finely minced red meat (horse or buffalo) fortified with

2% vitamin-mineral mix that included vitamin D3 (25 000

IU/kg; Monsoon Crocodile Premix, Darwin, Australia) and

1.5% calcium carbonate. The animals were fed, in excess, 5

times per week until they were 3 months of age when they were

fed every second day. Food was dispersed evenly over the land

area in the afternoon and left overnight. Crocodiles were not

observed for individual food intake, and measurement of

remaining food as an indicator of how much the crocodiles in

each pen were eating was impractical, due to dispersion of the

food into the water by the crocodiles. Crocodiles were fasted

for 48 hours prior to sampling. Ten runt and 10 normal croco-

diles were sampled in each of 2 study years: on November 8 to

11, 2005 (case Nos. 1–20) and on July 10 and 12, 2007 (case

Nos. 21–40). Sampling of animals was done when the grading

process had resulted in entire pens containing subjectively

obvious runt crocodiles, which was at approximately 7 and 5

months of age in 2005 and 2007, respectively. The only differ-

ence between the ‘‘runt’’ and ‘‘normal’’ animals was their body

size and condition; all animals used in the study were subjec-

tively judged to be bright, responsive, and active.

The initial suite of laboratory testing for 2005 samples was

designed to be relatively broad to maximize the chance of

detection of any disease or condition that could be related to

runting. The purpose of the second sample in 2007 was to sub-

stantiate potentially significant findings from the initial 2005

sample, and therefore testing in 2007 differed slightly and was

generally not as broad (specific differences in testing between

the years is detailed in relevant sections below).

Necropsy and Histopathology

Following blood sampling (see below), crocodiles were eutha-

nized by overdose (80 mg/kg) of pentobarbitone sodium

injected into the dorsal tail vein (Lethabarb euthanasia

injection, 325 mg/ml; Virbac Animal Health, Milperra, NSW,

Australia). A full necropsy was performed on all crocodiles. In

2005, liver weights were taken on all crocodiles (not repeated

in 2007 due to no difference between groups in 2005; see

Results).

Tissues were fixed in 10% neutral buffered formalin,

processed in standard fashion for histological examination, and

5-mm sections stained with hematoxylin and eosin (HE). Sec-

tions of bone were decalcified in 10% neutral buffered formalin

with 9% formic acid prior to processing. Sections trimmed for

histological examination were standardized with respect to

location in, and orientation of, the organ/tissue. Organs/tissues

examined only in 2005 were heart, lung, trachea, kidney, gall-

bladder, esophagus, stomach, coelomic fat body, thyroid gland,

skin, femorotibial joint, spinal cord, brain, and eye. These tissues

did not display notable differences between runt and normal cro-

codile groups (see Results) and therefore were not examined

in the second set of samples taken in 2007. Organs/tissues

examined in both years were pituitary gland, adrenal gland,

spleen, thymus, tonsil, bone (mid-sagittal section of proximal

tibial metaphysis, including bone marrow), skeletal muscle,

liver, pancreas, duodenum, jejunum, colon, and gonad. Histo-

logical examination of the gonad was used to determine sex.

Sections of parathyroid gland were available for examination

in 4 normal and 9 runt crocodiles. Perls’s stain for ferric iron

was used to identify the nature of the green-brown pigment

present in splenic macrophages.9

All histology slides were examined by one person (C.S.)

unaware as to whether the tissues were from a normal or runt

crocodile. Following initial screening of all slides, aspects of

tissues that subjectively seemed to differ among crocodiles

were quantified. Features that were scored as 0 (none), 1 (mild),

2 (moderate), or 3 (marked) were degree of cytoplasmic vacuo-

lation of adrenocortical (interrenal) cells and hepatocytes,

amount of zymogen in pancreatic acinar cells, and amount of

globular green-brown pigment in splenic macrophages. Bone

2 Veterinary Pathology

at University of Sydney on January 13, 2014vet.sagepub.comDownloaded from

marrow total percent cellularity and the relative proportion of

erythroid cells (compared with myeloid cells) were subjec-

tively estimated.

Digital images were taken for the following histological

features (MicroPublisher 3.3; Q Imaging, Surrey, BC, Canada)

and the feature measured using a program calibrated to the

magnification of the microscope (analySIS Five; Soft Imaging

System GmbH, Munster, Germany). Skeletal muscle fiber

width was measured by taking the average width of 10 fibers

in the same region of cranial thigh muscle in each crocodile.

In 2005, the muscle was sectioned transversely across the

fibers, while in 2007, the muscle was sectioned longitudinally.

The amount of the primary spongiosa in the proximal tibial

growth zone was quantified by determining the proportion of

the total width of mineralizing cartilaginous trabeculae imme-

diately beneath the growth zone compared with the width of the

medulla at the same level. The area of periarteriolar lymphoid

sheaths was measured by taking the area of each of 10 lym-

phoid sheaths, minus the area of the arteriole each surrounded,

and calculating an average. The 10 sheaths were the first 10

discrete sheaths encountered in a transect of the histological

section starting from the capsule moving toward the center at

the greatest diameter of a section taken of the mid-region of the

spleen. The amount of lymphoid tissue in the tonsils was

measured as a proportion of the area of the tonsil occupied

by lymphocytes compared with the total area of the folds of the

tonsil observed at low power in a standard complete transverse

section taken from the mid-tonsil. Thymus tissue was quanti-

fied by measuring the total area of thymus lobes in a transverse

section containing all the tissues bound by the mediastinum at

the level of the proximal primary bronchi. To take into account

the smaller overall size of tissues of runt crocodiles, the area of

the thymus lobes is presented as a proportion of the area of the

adjacent primary bronchus.

Ancillary Diagnostic Testing

Blood was sampled from the occipital venous sinus and the

initial 0.5 ml of blood placed into EDTA anticoagulant for hema-

tology, with the remainder placed into serum separator gel tubes

(BD Vacutainer; Becton Dickinson, Franklin Lakes, NJ). EDTA

was used as the anticoagulant since in our experience, it does not

cause hemolysis in saltwater crocodiles and is not potentially

associated with leukocyte and platelet clumping, and thus inac-

curate cell counts, as has been reported with lithium heparin

anticoagulated blood.5 Blood samples were obtained 2 to 4 hours

after removal from the pen for crocodiles in 2005 and within 3

minutes of removal from the pen, between 0800 and 0900 in the

morning for crocodiles in 2007. The strict timing of blood sam-

pling for 2007 was to minimize any effects of circadian rhythm

or acute handling stress on serum corticosterone level.23,37,42,44

Serum corticosterone was measured in the 2007 samples in an

effort to expand on the apparent histological difference noted

in adrenal glands between the runt and normal crocodile groups

noted in 2005 (see Results). Frozen serum was not available to

retrospectively test the 2005 crocodiles. Serum corticosterone

was measured using a high-sensitivity enzyme immunoassay

according to kit directions (Corticosterone HS EIA; Immuno-

diagnostics Systems Ltd, Boldon, Tyne & Wear, UK). Values

that exceeded the 20-ng/ml upper limit of kit accuracy were set

at this limit for statistical analysis.

Selected clinical pathological parameters were determined

to assist interpretation of aspects of the gross and/or histo-

pathology. Hematology included total and differential white

blood cell counts conducted using a hemocytometer with an

eosinophil Unopette system (Becton-Dickinson, Rutherford,

NJ) and blood smear examination according to standard reptile

protocol.6,10 Packed cell volume was determined by centrifuga-

tion of blood in microhematocrit tubes. Serum biochemical

parameters were determined on an automated analyzer (Kone-

lab 20; Thermo Electron, Victoria, Australia). Parameters mea-

sured (followed in parentheses by a brief analytical basis for the

measurement provided by Thermo Electron) were total protein

(biuret method), albumin (bromcresol green dye binding

method), globulins (by subtraction of albumin from total

protein), total calcium (reaction with the metallochromogen

Arsenazo III), inorganic phosphorus (formation of phosphomo-

lybdate and subsequent reduction to molybdenum blue), and

iron (hydroxylamine hydrochloride reduction and subsequent

reaction with liquid ferrozine).

Bacterial culture of liver and spleen was performed on all

crocodiles in 2005 but not in 2007 because of a lack of signif-

icant bacteriological differences between the 2 groups in 2005

(see Results). Culture was performed on samples obtained

aseptically during necropsy. To collect samples aseptically, the

skin of the crocodile was cleaned with 100% ethanol and

incised using scalpel and forceps that had been sterilized by

dipping them in 100% ethanol and flaming them over a Bunsen

burner. Following the skin incision and reflection of the skin,

the instruments were again sterilized and the coelom opened.

Instruments were sterilized a third time prior to sampling tis-

sue. In addition, yolk was cultured from case No. 1 (runt) and

case No. 11 (normal) with enlarged internal yolk sac remnants

and a swab of the subcutis from the swollen forelimb of case

No. 16 (normal). For these samples, the tissue was entered

aseptically as described above and sampled using sterile swabs.

Bacterial culture was performed using standard veterinary bac-

teriology phenotypic and biochemical techniques. Briefly,

samples were homogenized and plated onto sheep blood agar

(Oxoid Australia, Thebarton, Australia) and MacConkey agar

(Oxoid Australia, Thebarton, Australia) and incubated at

35�C for 48 hours. The bacterial isolates were initially charac-

terized using Gram’s stain, colony morphology, and relevant

preliminary tests, including oxidase and catalase, and then the

appropriate commercial kits were used for speciation (api 20

Strep, api Coryne, bioMerieux, Marcy-l’Etoile, France;

Microbact Gram-Negative Identification System, Oxoid Ltd,

Basingstoke, Hants, UK).

Virus isolation was attempted on liver, thymus, tonsil, and

spleen from all crocodiles in 2007 using primary crocodile liver

and kidney cell lines developed at Berrimah Veterinary Labora-

tories.50 Culture was not performed on samples from 2005 since

Shilton et al 3

at University of Sydney on January 13, 2014vet.sagepub.comDownloaded from

the crocodile cell lines had not been developed yet and frozen

stored samples from 2005 were no longer available. In 2007,

samples were stored at –70�C until processing. Briefly, tissue

samples were homogenized, clarified by centrifugation, and the

supernatant filtered through a 0.45-mm filter. The filtered super-

natant was then inoculated into 25-cm2 flasks containing conflu-

ent primary cell line monolayers. Each sample was inoculated

into 2 different cell lines. The flasks were incubated at 28�C and

examined for viral growth every 3 days for 21 days. The cultures

were then passaged into fresh flasks for another 21 days and

examined every 3 days. This process was repeated a final time,

and if no growth was observed, the culture was deemed negative.

Viral growth was recognized by a cytopathic effect characterized

by loss of confluence of the cell monolayer, rounding up of cells,

and, in some cases, syncytia formation.50 The virus isolations

were part of a larger study investigating the viral flora of salt-

water crocodiles,50 and further characterization of isolated

viruses was beyond the scope of this study.

Statistical Analyses

To generate a single continuous variable that described overall

body size relative to age (a growth index), we first conducted

a principal component analysis to combine 2 measures of body

length (snout-vent length and vent-tail tip length) with body

weight. The resultant first principal component (PC1) incorpo-

rated 99% of the variation in the 3 measures and was used as

a single measure of overall body size. This PC1 body size

measure was then corrected for crocodile age by calculating the

residuals from a linear regression of PC1 on age. These residual

values form the growth index, as they provide a composite mea-

sure of how large each crocodile was for its age. There was no

overlap in this value for crocodiles assigned to the normal and

runt groups, verifying its validity in distinguishing runts from

normal crocodiles (Table 1). This continuous variable was used

as an explanatory covariate in all statistical analyses. However,

for ease of interpretation, summary statistics of basic measure-

ments are presented as 2 groups: normal and runt crocodiles.

Response variables that were measured as ordinal categories

(eg, degree of cytoplasmic vacuolation of adrenocortical cells)

were analyzed using logistic regression with year and growth

index as explanatory variables. Measurements made on a

continuous scale (eg, hematology parameters, size of splenic

periarteriolar lymphoid cuffs) were analyzed using multiple

regression, with year and growth index as explanatory

variables.

Table 1. Body Size, Age, Sex, Hematological, and Biochemical Parameters for All Normal and Runt Crocodiles in 2005 (Case Nos. 1–20) and2007 (Case Nos. 21–40).a

Parameter

Normal Crocodile Runt Crocodile

2005 2007 2005 2007

Total length, cm*# 56.0 + 1.5 (47.0, 64.0) 48.4 + 1.2 (43.0, 55.0) 35.4 + 0.4 (33.0, 37.5) 34.4 + 0.5 (32.5, 37.5)Body weight, g*# 460 + 40 (251, 743) 286 + 24 (185, 393) 74 + 8 (47, 110) 64 + 2 (52, 74)Age, d# 225 + 5 (208, 254) 118 + 1 (115, 121) 197 + 12 (136, 256) 125 + 4 (114, 142)Growth index,*b 0.9 + 0.2 (–0.1, 2.0) 0.8 + 0.1 (0.2, 1.4) –1.1 + 0.1 (–1.3, –0.7) –0.6 + 0.03 (–0.7, –0.5)Sex, male, female, unknown#c 8, 0, 2 8, 2, 0 9, 1, 0 6, 4, 0Hematology

Total white blood cell count,�109/L#

9.4 + 1.4 (3.1, 17.6) 12.1 + 1.0 (4.1, 15.9) 6.6 + 0.7 (3.1, 10.0) 13.1 + 2.0 (4.5, 21.6)

Heterophils, �109/L 4.1 + 0.8 (0.3, 9.1) 3.4 + 0.2 (2.0, 4.8) 3.4 + 0.3 (2.1, 5.2) 3.2 + 0.3 (2.1, 5.4)Lymphocytes, �109/L# 4.0 + 1.0 (1.0, 12.0) 7.5 + 0.7 (1.9, 9.7) 2.5 + 0.5 (0.9, 4.8) 9.2 + 1.8 (1.1, 17.3)Monocytes, �109/L 1.1 + 0.3 (0.3, 3.6) 0.9 + 0.2 (0.1, 2.1) 0.7 + 0.2 (0.1, 1.4) 0.7 + 0.2 (0.2, 1.7)Eosinophils, �109/L 0.1 + 0.04 (0, 0.4) 0.1 + 0.06 (0, 0.4) 0.1 + 0.04 (0, 0.5) 0.01 + 0.01 (0, 0.1)Packed cell volume, L/L*# 0.23 + 0.004 (0.2, 0.24) 0.18 + 0.002 (0.17, 0.19) 0.15 + 0.01 (0.09, 0.19) 0.12 + 0.01 (0.1, 0.14)

Serum biochemistryTotal protein, g/L*# 43.9 + 1.9 (31, 54) 38.7 + 1.1 (33, 43) 32.5 + 1.3 (27, 39) 32.2 + 1.2 (25, 38)Albumin, g/L*# 18.6 + 0.4 (16, 20) 16.5 + 0.4 (15, 18) 11.6 + 0.7 (8, 14) 10.1 + 0.5 (8, 13)Globulins, g/L 25.7 + 1.6 (16, 35) 22.2 + 1.0 (17, 27) 20.8 + 0.6 (18, 24) 21.9 + 1.2 (14, 27)Total calcium, mmol/L*#d 2.8 + 0.05 (2.5, 3.0) 2.7 + 0.05 (2.4, 2.9) 2.5 + 0.04 (2.3, 2.7) 2.3 + 0.02 (2.2, 2.5)Phosphorus, mmol/L* 1.6 + 0.1 (0.9, 2.3) 1.5 + 0.1 (1.3, 1.8) 1.1 + 0.1 (0.9, 1.5) 1.2 + 0.3 (0.7, 4.1)Calcium/phosphorus ratio* 1.8 + 0.2 (1.2, 3) 1.8 + 0.1 (1.5, 2.2) 2.3 + 0.1 (1.7, 2.7) 2.5 + 0.2 (0.6, 3.2)Iron, mmol/L# 11.9 + 4.5 (1, 39) 1.8 + 0.4 (1, 5) 3.9 + 0.8 (1, 7) 1.4 + 0.3 (1, 4)

aSummary statistics are presented by group, although statistics were performed using the growth index as the independent variable with year as a covariate. Exceptfor sex, values are mean + SE with minimum and maximum in parentheses. Statistically significant differences (at least P < .05 level) between runt and normalgroups are signified with an asterisk (*); differences between years of the project are signified with a hash symbol (#). All differences between groups were observedin both years and vice versa. N ¼ 10 for normal and runt groups in each year of the study.bGrowth index is a summary statistic indicating how well grown a crocodile is for its age. It was generated by combining snout-vent length, vent-tail tip length, andbody weight into 1 principal component to indicate body size, then correcting this value for body age by linear regression and using the resultant residual values toform the growth index.cThere were significantly more females in 2007 in both groups.dTotal calcium was not significantly different between the runt and normal crocodile groups when albumin level was controlled for statistically.

4 Veterinary Pathology

at University of Sydney on January 13, 2014vet.sagepub.comDownloaded from

Since 40% to 45% of total calcium is transported in serum

bound to protein (principally albumin), serum albumin level

may influence total calcium level.1,51 To correct for this, we

regressed total calcium on albumin measures. The residual

values from this regression were then compared between the

runt and normal crocodiles using analysis of covariance.

Results

Necropsy

The runt and normal crocodile groups differed significantly in

total length and body weight in both years, with runt crocodiles

being, on average, 33% shorter and 82% lighter than normal

crocodiles (Fig. 1, Table 1). There was no significant difference

in the ages of runt vs normal crocodiles; however, both groups

were older in 2005. The growth index differed significantly

between the runt and normal crocodile groups, with no overlap

in range, confirming significantly smaller body size for age

(ie, poorer growth) in runt crocodiles. Overall, there was no dif-

ference in sex ratio between the 2 groups, but in 2007, there

were more females in both groups than there had been in

2005 (Table 1). There was no significant difference in the ori-

gin of the eggs (wild vs captive nest) between the 2 groups.

Visually, runt crocodiles exhibited reduced muscling, pro-

minence of bony protuberances, reduced amount of adipose tis-

sue in the base of the tail, and markedly reduced size of the

coelomic fat body (which was not grossly visible in many of

the runts). These subjective necropsy findings indicate poor

body condition in runt compared with normal crocodiles. Com-

pared with normal crocodiles, gallbladders of runt crocodiles

were relatively large and distended with bile. Bones were sub-

jectively judged to be of comparable strength between the 2

groups. Grossly, lobes of the thymus were generally unapparent

Figure 1. Normal crocodile Nos. 21 to 25 (left) and runt crocodile Nos. 26 to 30 (right), sampled July 10, 2007.

Shilton et al 5

at University of Sydney on January 13, 2014vet.sagepub.comDownloaded from

in runt crocodiles while they were usually obvious in normal

crocodiles. The tonsils of runt crocodiles had less prominent

folds than those of normal crocodiles. Case No. 1 (runt) and

case No. 11 (normal) had 1-cm diameter yolk sac remnants pro-

truding from the outer wall of the mid-jejunum. The remainder

of the crocodiles either had a small (1–2 mm diameter) yolk sac

remnant or the remnant was not grossly appreciable. One nor-

mal crocodile (case No. 16) had moderate diffuse enlargement

and edema involving 1 forelimb. Except for small fecal pellets

in the rectum, the gastrointestinal tracts were largely empty in

both groups. There were rare to occasional 0.3- to 1.0-mm

round white cutaneous foci, typical of mild superficial poxvirus

lesions, in case Nos. 2, 6, 13, 14, 27, 29, 30, and 32 (8 runt cro-

codiles) and case Nos. 5, 9 to 12, 16, and 38 to 40 (9 normal

crocodiles). Liver weight as a percentage of body weight aver-

aged 2.4% in both the runt and normal crocodile groups.

Histopathology

Except for features that are expanded upon below, there were

no differences in histological appearance of organs/tissues

among the crocodiles. The significance of findings between

normal and runt crocodiles applies to both years of the project

unless otherwise stated.

There were rare sporulated coccidia oocysts in the mucosal

epithelium of the jejunum and/or colon in case Nos. 2, 6, and 15

(runts) and case No. 25 (normal). Histologically, the poxvirus

lesions appeared as discrete slightly raised foci of markedly

hypertrophic epithelium containing large eosinophilic intracy-

toplasmic inclusion bodies, typical of mild crocodile poxvirus

lesions.33,35 Histologically, the single swollen leg present in

case No. 16 (normal) exhibited marked subcutaneous and intra-

muscular edema, associated moderate heterophil and macro-

phage infiltration, and numerous macrophages containing

abundant intracytoplasmic bacterial cocci.

Statistics for histological features that were quantified are

presented in Table 2. Degree of vacuolation of adrenocortical

cells was significantly greater in runt compared with normal

crocodiles and slightly greater in both groups in 2007 (Figs.

2, 3). The degree of hepatocyte vacuolation was slightly but

significantly less in runt crocodiles and less in both groups in

2005. The degree of pancreatic zymogen was significantly

decreased in runt crocodiles. Skeletal muscle fiber width was

significantly less in runt crocodiles compared with normal cro-

codiles in both years. Muscle fiber width was significantly less

in both groups in 2007, presumably artifactually, due to the

muscle being sectioned longitudinally in 2007, compared with

transversely in 2005. Bone marrow cellularity was reduced in

runt compared with normal crocodiles with no difference in the

proportion of erythroid cells between the 2 groups. There was

an increase in the proportion of erythroid cells in the bone mar-

row in both groups in 2005.

Proximal tibial bone growth zones differed subjectively in

histological appearance between runt and normal crocodiles.

While the growth zones of runt crocodiles retained all normal

histological components with no additional lesions, runts had

generally narrower and less distinct zones of proliferating,

maturing, and hypertrophic chondrocytes and decreased

production of primary spongiosa (Figs. 4, 5). Quantitation of

the amount of primary spongiosa confirmed decreased produc-

tion of bone at the growth zone in runt compared with normal

crocodiles (Table 2).

Runt crocodiles had significantly reduced lymphoid popula-

tions compared with normal crocodiles in all tissues in which

this was evaluated, including, spleen (Figs. 6, 7), tonsil, and

thymus (Figs. 8, 9). Tonsils had a greater lymphoid component

in both groups in 2005. The amount of green-brown pigment in

splenic macrophages was greater in runt compared with normal

crocodiles. The majority of this pigment was strongly positive

with Perls’s stain for ferric iron (Fig. 7, inset).

Ancillary Testing

In 2007 (testing was not performed in 2005), mean serum cor-

ticosterone levels were significantly higher in runt (mean [SE],

17.1 [1.1] ng/ml; range, 10.2–20) compared with normal croco-

diles (mean [SE], 9.4 [2.3] ng/ml; range, 0.5–20).

Hematologically, there were no significant differences

between runt and normal crocodiles in white blood cell para-

meters, although both groups had significantly higher total

white blood cell and lymphocyte counts in 2007. Mean packed

cell volume was significantly lower in runt compared with nor-

mal crocodiles and significantly lower in both groups in 2007

(Table 1). Examination of blood smears did not reveal evidence

of hemoparasites in crocodiles in either group. There were low

to moderate numbers of polychromatophilic red blood cells in

peripheral blood smears in 11 runt and 17 normal crocodiles.

Serum biochemistry (Table 1) revealed significantly lower

total protein and albumin in runt compared with normal croco-

diles in both years, and these parameters were higher in both

groups in 2005. Serum total calcium and phosphorus were sig-

nificantly lower in runt compared with normal crocodiles, with

total calcium being slightly but significantly higher in all cro-

codiles in 2005. However, when the residuals of a regression

of total calcium vs albumin were compared between the runt

and normal crocodile groups, there was no significant differ-

ence, indicating that the difference in total calcium between the

2 groups was attributable to their differing serum albumin lev-

els. The relatively greater decrease in phosphorus in runt croco-

diles compared with total calcium resulted in significantly

higher total calcium to phosphorus ratio in runt compared with

normal crocodiles (Table 1). There were no statistically signif-

icant differences between runt and normal crocodiles in serum

iron, although this parameter exhibited substantial variation

within the groups and was on average higher in both groups

in 2005.

Bacterial culture results in 2005 were negative except as

follows: Streptococcus agalactiae was isolated in 3 normal cro-

codiles (from the liver and spleen of case No. 4, the liver of

case No. 8, and the swollen forelimb of case No. 16). Edward-

siella tarda was isolated from the enlarged yolk sac remnant of

case No. 11 (normal). Corynebacterium sp was isolated from

6 Veterinary Pathology

at University of Sydney on January 13, 2014vet.sagepub.comDownloaded from

Tab

le2.

His

tolo

gica

lFe

ature

sQ

uan

tifie

dfo

rA

llN

orm

alan

dR

unt

Cro

codile

sin

2005

(Cas

eN

os.

1–20)

and

2007

(Cas

eN

os.

21–41).

a

Feat

ure

Norm

alC

roco

dile

Runt

Cro

codile

2005

2007

2005

2007

Adre

noco

rtic

alce

llva

cuola

tion*#

b1.0

+0.0

(1,1)

1.1

+0.1

(1,2)

2.0

+0.2

(1,3)

2.2

+0.1

(2,3)

Skel

etal

musc

lefib

erw

idth

,mm

*#31.3

+1.3

(25.6

,40.1

)11.3

+0.3

(9.4

,12.9

)12.2

+0.9

(8.9

,17.3

)6.8

+0.3

(5.2

,8.6

)H

epat

ocy

teva

cuola

tion*#

b1.7

+0.2

(1,2)

2.0

+0.0

(2,2)

1.3

+0.2

(0,2)

1.9

+0.1

(1,2)

Pan

crea

tic

zym

oge

n*b

3.0

+0.0

(3,3)

3.0

+0.0

(3,3)

2.0

+0.3

(1,2)

2.9

+0.1

(2,3)

Sple

nic

mac

rophag

egr

een-b

row

npig

men

t*b

1.0

+0.2

(0,2)

0.9

+0.3

(0,2)

2.5

+0.2

(2,3)

2.9

+0.2

(2,3)

Bone

mar

row

per

cent

cellu

lari

ty*

46+

2(4

0,60)

53+

2(5

0,60)

32+

4(2

0,50)

25+

3(1

5,40)

Bone

mar

row

eryt

hro

idpro

port

ion

#0.5

1+

0.0

3(0

.33,0.7

5)

0.4

5+

0.0

3(0

.33,0.5

0)

0.5

3+

0.0

2(0

.50,0.6

6)

0.3

9+

0.0

3(0

.25,0.5

0)

Pri

mar

ysp

ongi

osa

pro

port

ion

ofm

edulla

ryw

idth

intibia

lgr

ow

thzo

ne*

#0.4

9+

0.0

2(0

.37,0.5

6)

0.4

1+

0.0

2(0

.31,0.5

2)

0.2

8+

0.0

3(0

.16,0.4

3)

0.2

5+

0.0

2(0

.17,0.3

2)

Sple

nic

peri

arte

riola

rly

mph

oid

cuff

mea

nar

ea,

mm2*

19

078+

1560

(11

507,26

052)

15

952+

2114

(9328,32

816)

9736+

1431

(2763,16

755)

11

727+

1220

(8177,20

441)

Tonsi

lar

eapro

port

ion

popula

ted

by

lym

phocy

tes*

#0.6

7+

0.0

3(0

.52,0.8

2)

0.4

4+

0.0

4(0

.23,0.6

5)

0.2

9+

0.0

7(0

.00,0.6

6)

0.2

1+

0.0

6(0

.00,0.4

9)

Thym

us

lobe

area

aspro

port

ion

ofpri

mar

ybro

nch

us

area

*3.6

+0.6

(1.8

,7.2

)3.7

+0.7

(1.4

,7.7

)0.6

+0.3

(0.1

,2.9

)0.5

+0.1

(0.0

5,1.2

)

a Sum

mar

yst

atis

tics

are

pre

sente

dby

group,a

lthough

stat

istics

wer

eper

form

edusi

ng

agr

ow

thin

dex

asth

ein

dep

enden

tva

riab

lew

ith

year

asa

cova

riat

e.V

alues

are

mea

n+

SEw

ith

min

imum

and

max

imum

inpar

enth

eses

.St

atis

tica

llysi

gnifi

cantdiff

eren

ces

(P<

.05

leve

l)bet

wee

nnorm

alan

dru

ntgr

oups

are

sign

ified

with

anas

teri

sk(*

);diff

eren

ces

bet

wee

nye

ars

oft

he

pro

ject

are

sign

ified

with

ahas

hsy

mbol(

#).

All

diff

eren

ces

bet

wee

nye

ars

wer

eobse

rved

inboth

norm

alan

dru

nt

groups

and

vice

vers

a.N¼

10

for

norm

alan

dru

nt

groups

inea

chye

arofth

est

udy.

bFe

ature

was

grad

edas

0(n

one)

,1

(mild

),2

(moder

ate)

,or

3(m

arke

d).

c Gro

wth

index

isa

sum

mar

yst

atis

tic

indic

atin

ghow

wel

lgro

wn

acr

oco

dile

isfo

rits

age.

Itw

asge

ner

ated

by

com

bin

ing

snout-

vent

lengt

h,v

ent-

tail

tip

lengt

h,a

nd

body

wei

ght

into

1pri

nci

pal

com

ponen

tto

indic

ate

body

size

,th

enco

rrec

ting

this

valu

efo

rbody

age

by

linea

rre

gres

sion

and

usi

ng

the

resu

ltan

tre

sidual

valu

esto

form

the

grow

thin

dex

.

7 at University of Sydney on January 13, 2014vet.sagepub.comDownloaded from

the liver and spleen of case No. 14 (runt) and Morganella

morganii from the liver and Salmonella sp from the enlarged

yolk sac remnant in case No. 1 (runt).

Cytopathic effect in viral culture suggestive of viral growth

was observed in 2007 from the liver and tonsil of case No. 22

(normal) and from the tonsil of case Nos. 21, 24, and 25 (nor-

mal). No cytopathic effect in viral culture was detected from

runt crocodiles.

Discussion

Runt crocodiles in this study exhibited markedly smaller body

size and weight for their age compared with normal crocodiles,

indicating very poor growth as noted in other stud-

ies.3,21,32,48,49,55,58 In addition, multiple parameters indicated

inanition in runt crocodiles, including their moderately

decreased serum albumin, smaller skeletal muscle fiber width,

and decreased pancreatic zymogen. Liver weight as a percent-

age of body weight was the same in runt and normal crocodiles,

suggesting that a significant degree of hepatic atrophy was not

present in runt crocodiles. Histologically, runt crocodiles had

slightly decreased hepatocyte cytoplasmic vacuolation com-

pared with normal crocodiles. In reptiles, mild to moderate

hepatocyte vacuolation, usually due to lipid, is commonly pres-

ent in reptiles in good body condition and is considered within

physiological normal, rather than a degenerative change.30,59

Thus, the mildly decreased hepatocyte vacuolation in runt com-

pared with normal crocodiles likely reflects their generalized

paucity of body fat and is another parameter supporting inani-

tion in the runt crocodiles.

Runt crocodiles had significantly lower packed cell volumes

than normal crocodiles. The anemia in runts was nonregenera-

tive, being unaccompanied by evidence of erythroid hyperpla-

sia in bone marrow or evidence of regeneration in peripheral

Figure 2. Adrenal gland; normal crocodile (case No. 36). Mild vacuolation of the corticosterone-secreting cells of the interrenal cords (deli-neated by double-headed arrow). Hematoxylin and eosin (HE). Figure 3. Adrenal gland; runt crocodile (case No. 32). Marked vacuolation of thecorticosterone-secreting cells of the interrenal cords (delineated by double-headed arrow). HE. Figure 4. Proximal tibial growth zone; normalcrocodile (case No. 5). Note layer of proliferating chondrocytes (P) and abundant primary spongiosa (PS). Decalcified, HE. Figure 5. Proximaltibial growth zone; runt crocodile (case No. 3). Note indistinct, poorly cellular layer of proliferating chondrocytes (P) and paucity of primaryspongiosa (PS). Decalcified, HE.

8 Veterinary Pathology

at University of Sydney on January 13, 2014vet.sagepub.comDownloaded from

blood beyond the mild to moderate red blood cell polychroma-

sia that was observed in both runt and normal crocodiles. This

may be at least in part due to the tendency of young reptiles to

exhibit a greater degree of polychromasia than adults.5 Serum

iron was inconsistent between the 2 years, with 2005 crocodiles

having significantly higher serum iron levels than 2007 croco-

diles regardless of whether they were runt or normal animals.

Both groups also had lower packed cell volumes in 2007, pos-

sibly reflecting lower serum iron. In the runt crocodiles, there

appeared to be sequestration of iron in the form of splenic

hemosiderin. Splenic hemosiderosis occurs with starvation or

other catabolic states in all species and, in reptiles, is generally

considered secondary to other conditions, such as chronic

inanition.46 Thus, it is likely that inanition is the main contribu-

tor to the anemia in runt crocodiles.62,67

One of the main objectives of this study was to investigate

whether chronic infectious disease could be associated with

runting. Despite a thorough examination of all organ systems

in runt vs normal crocodiles, there were no gross or histological

lesions to suggest a significant presence of infectious disease.

The only parasite detected in the study was rare sporulated

oocysts in the intestinal epithelium of 3 runt crocodiles and 1

normal crocodile, consistent with the Goussia-like coccidia

described in saltwater crocodiles in Australia.25,40 The pres-

ence of this parasite as a contributing factor to runting is

dubious, given that the coccidia were evident in only 3 of 20

runt crocodiles, and then only rarely in the intestinal epithelium

with no associated histological lesions.

Bacterial culture of 2 filtering organs (liver and spleen) was

conducted to detect bacterial sepsis, since this may be difficult

Figure 6. Spleen; normal crocodile (case No. 12). Large discrete lymphoid cuffs (arrows) surround arterioles (A). Hematoxylin and eosin (HE).Figure 7. Spleen; runt crocodile (case No. 13). Relatively small lymphoid cuffs (arrows) surround arterioles (A). Note yellow-brown tinge tored pulp due to pigment within macrophages. HE. Inset: Pigment within macrophages stains positive for ferric iron indicative of hemosiderin.Perls’s stain. Figure 8. Thymus; normal crocodile (case No. 12). Thymus lobes are densely populated with lymphocytes with clear distinctionbetween cortex (C) and medulla (M). For perspective, a section of the primary bronchus (lumen, B) and thyroid gland (T) is noted. Same mag-nification as Fig. 9. HE. Figure 9. Thymus; runt crocodile (case No. 1). Thymus lobes (arrows) are small with poor distinction between cortexand medulla. For perspective, a section of the primary bronchus (lumen, B) and thyroid gland (T) is noted. Same magnification as Fig. 8. HE.

Shilton et al 9

at University of Sydney on January 13, 2014vet.sagepub.comDownloaded from

to detect histologically and is a common cause of morbidity and

mortality in juvenile crocodiles.3,41 Bacterial isolates were

limited to a total of 5 species in 2 runt crocodiles and 4 normal

crocodiles, with no particular pattern to suggest a relationship

with runting.

Histological evidence of virus infection was limited to mild

cutaneous poxvirus infection present in several crocodiles, both

normal and runts. Virus isolation was attempted for multiple

internal organs in all crocodiles in 2007 to further substantiate

lack of virus involvement and because of the association of

viral infection with atrophy of lymphoid tissues and stunted

growth in other species. Examples include mammalian

parvoviruses and pestiviruses,47 as well as retroviruses of the

reticuloendotheliosis virus group in poultry.18 However, our

isolation attempts yielded only a cytopathic effect suggestive

of 5 isolates from 4 normal crocodiles. This, as well as the lack

of histological lesions such as necrosis or inflammation that

could substantiate a significant virus infection, suggests that

viral infection was not related to runting in our study.

A second objective of the study, to look for histological clues

of other disease processes, revealed some associations with runt-

ing and could help suggest an alternate etiology to infectious dis-

ease. One notable finding was the increased cytoplasmic

vacuolation of adrenocortical cells in runt compared with normal

crocodiles in both phases of the study. Variation in the degree of

vacuolation is evidence of altered activity of the adrenocortical

cells in reptilians and is evidence of increased activity of the

cells in mammals.7,24 To further investigate this finding, in

2007, serum level of corticosterone, the main glucocorticoid

stress hormone in crocodilians,37,43 was measured and found

to be significantly elevated in runt crocodiles. Chronic elevation

of corticosteroids results in a general catabolic state7,61 and anor-

exia in crocodiles,32 and it inhibits both growth hormone secre-

tion and action in mammals.31 A negative correlation between

corticosterone level and growth rate has been noted previously

in saltwater crocodiles58,65 and alligators,15 and extremely poor

growth has been experimentally reproduced in alligators by

chronic elevation of corticosterone.43,53

Another notable histological finding in runt crocodiles was

the presence of bone growth zones that had all the components

of normal growth zones but were relatively quiescent and

exhibit decreased production of primary spongiosa. The runt

crocodiles were mildly hypocalcemic and moderately

hypophosphatemic compared with normal crocodiles. Since

approximately 40% to 45% of total calcium is transported in

serum bound to protein (principally albumin), the mild hypo-

calcemia in the runt crocodiles was suspected to be a reflection

of their concurrent substantial hypoalbuminemia (runt

crocodiles had on average 38% lower albumin than normal

crocodiles).1,51 This was confirmed by statistically correcting

for the level of albumin. Measurement of the ionized (unbound,

biologically active) portion of total calcium would have been

an additional parameter we could have used to investigate

serum calcium in this study1,51 but was judged to not be worth

pursuing given the lack of hypocalcemia in the runts once

albumin was taken into account.

The mild to moderate hypophosphatemia in runt crocodiles

is likely at least in part attributable to prolonged lack of dietary

intake, a common cause in reptiles.4 Since runt crocodiles were

offered the same diet as the normal crocodiles, the deficit may

be due to a lack of food intake rather than a deficient diet per se.

Also, increased glucocorticoid hormone has been shown in

humans to result in decreased renal resorption and gastrointest-

inal absorption of phosphate.57 A final possible cause for the

relative hypophosphatemia in runt crocodiles may be that it

is simply reflecting decreased bone formation, as serum

phosphorus tends to be higher in growing animals.1

The formation of endochondral bone in growth zones is

essentially similar in mammals, birds, and reptiles.28 In mam-

mals, abnormal bone growth in young animals due primarily to

severe prolonged lack of dietary phosphorus generally results

in rickets or osteomalacia, characterized by deformities of bone

shape with excessive production of unmineralized osteoid and

cartilaginous matrix,63 lesions that were not evident in runt

crocodiles. The histology of the poor bone growth in the runt

crocodiles is best described as osteoporosis, with relative inac-

tivity of the growth zone and decreased production of bone

being typical findings in mammals or reptiles undergoing

inanition.28,63 Another factor in reduced activity of growth

zones in runt crocodiles may be elevated corticosterone, since

elevated glucocorticoid hormones leads to osteoporosis and

decreased bone formation in some mammals.56,57,63

The decreased lymphoid populations in the major immune

tissues of crocodiles,19 including spleen, tonsil (gut-

associated lymphoid tissue), and thymus in runt crocodiles,

indicate generalized lymphoid hypoplasia or atrophy. Marked

lymphoid involution in reptiles can be seasonal or may result

from stressors such as starvation or disease.10 Corticosterone

has been shown to be associated with lymphoid involution and

immunosuppression in a wide variety of species, including cro-

codilians.43,54,60,70 Despite the lymphoid involution evident in

runt crocodiles, there was no overt evidence of immunosup-

pression in the form of an increased prevalence of infectious

disease. This may be partly related to the sampling procedure,

in that only bright, responsive, and active runt crocodiles were

selected for the study. Runt crocodiles may be predisposed to

eventual morbidity/mortality due to infectious disease, but this

would have gone undetected in the present study and would be

a useful avenue for further research. Perhaps even better would

be research aimed at detecting subtle evidence of immunosup-

pression in runt crocodiles, such as response to phytohemag-

glutinin injection or other immunoassays.20,27

The marked histological difference in lymphoid tissues of

runt vs normal crocodiles was not reflected in numbers of

circulating lymphocytes in this study. This is in contrast to

other studies in reptiles in which tissue lymphoid depletion was

correlated with a decrease in circulating lymphocytes.62

Furthermore, the stress response, characterized by increased

activity of the adrenocortical cells, has been found to be asso-

ciated with a decrease in circulating lymphocytes along with an

increase in heterophils/neutrophils in animals generally,

including crocodilians.5,29,53,56 Despite the histologically

10 Veterinary Pathology

at University of Sydney on January 13, 2014vet.sagepub.comDownloaded from

appreciable adrenocortical hyperplasia and significantly

elevated corticosterone in runt crocodiles in this study, this was

not correlated with alterations in circulating leukocytes com-

pared with normal crocodiles. However, a few other studies

have noted variable or absent alterations in numbers of circulat-

ing lymphocytes and heterophils with stress in crocodilians,

indicating that measurement of circulating leukocytes may be

an unreliable indicator of adrenocortical activity in these

species.43,65

In summary, there is no significant evidence of infectious

disease in runt crocodiles in this study. Rather, there is evi-

dence of inanition, lymphoid atrophy, and quiescent growth

zones in bones. Regarding inanition and poor growth, although

food intake of the runt crocodiles in this study was not mea-

sured, because we found no evidence of diseases or conditions

that might result in poor food conversion, it seems likely that

the runt crocodiles, while being offered equal access to food,

were not eating as much as the normal crocodiles. As discussed

previously, decreased appetite could be a consequence of a

stress response.32 Alternatively, it is possible that the food was

inherently unacceptable to runt crocodiles (eg, they required

the additional stimulation of movement of prey to eat), they

were unable to learn to eat the captive diet, or they had altered

behavior that prevented them from approaching the food (eg,

excessively fearful disposition). Supporting these as possibili-

ties is the finding by some investigators that changing the diet

will ameliorate runting in some circumstances.11,22,26,55 One

avenue for further research into runting in crocodiles could thus

be confirmation that runt crocodiles do eat less than their

normal conspecifics and alteration of the diet or feeding envi-

ronment in an attempt to ameliorate the runting.

Another possibility is that the inanition, lymphoid atrophy,

and quiescent growth zones of bones are a result of a chronic

adrenocortical stress response, which was demonstrated by the

adrenocortical cell vacuolation and elevation of corticosterone

in runt crocodiles. Chronic stress could be a secondary

response to some unknown stressor. For example, one possible

cause of secondary stress could be related to a perceived lack of

suitable food in runt crocodiles, as discussed above. Since runt

crocodiles were reared in pens containing other similarly sized

runts, a chronic stress response is unlikely to be secondary to

aggression from larger pen-mates. Although runt crocodiles

were reared under identical husbandry conditions as normal

crocodiles, it is possible that there could be something inher-

ently different in runt crocodiles that results in them being

stressed by a feature of the husbandry or environment that is

not stressful to normal crocodiles. Further research along these

lines could involve trials altering the features of runt pens, such

as design (eg, provide more hides, variation in water depth),

lowering stocking density, or providing a wider spectrum of

available temperature (eg, provide heat lamps for basking).

Thus, runt crocodiles could be suffering from a ‘‘maladaptation

syndrome’’ suggested by a few other investigators and recog-

nized as a feature of stress in reptiles.3,12,65

Rather than being a secondary response in individual hatchl-

ings to some aspect of the farming situation, the stress response

could be a primary feature inherent in a runt crocodile. It has

been established in many animals, from fish to birds and mam-

mals, that maternal stress may influence the hypothalamic-

pituitary-adrenal axis of the offspring, resulting in altered beha-

vioral development and/or ill thrift in the offspring for several

months after birth.2,8,13,16,17,36,45 In lizards, treatment of

reproductive females or eggs with corticosterone can affect

hatchling body size, growth rate, behavior (eg, dispersal and

antipredator responses), and even sex.52,66,68 Likewise, in

crocodilians, limited studies suggest that high plasma corticos-

terone in females may result in relatively poor egg quality and

decreased hatchling survival.14,61

The clutch effect in crocodile runting26,33,48,58 suggests that

crocodiles in specific clutches are hatched with a tendency to

become runts, although more research needs to be done on the

degree of clutch effect and variation within a clutch. A clutch

effect could be due to genetics, maternal circumstances, egg

incubation conditions, or, where members of a clutch are reared

together in some degree of isolation from other clutches, their

rearing environment. These considerations are beyond the

scope of this study, given the large number of clutches the

study animals originated from and the mixture of eggs from

wild vs captive nests, but may be valuable avenues for further

research. Exploration of the possible role of genetics in causing

runting would require controlled breeding experiments in cap-

tive pairs over multiple years. Maternal circumstances, includ-

ing size and body condition, geographic location, diet, food

availability, and stress level, would also be useful avenues for

further research. Egg incubation conditions in this study were

uniform and controlled at the farm once eggs were collected but

not necessarily prior to collection, particularly for eggs

originating from wild nests. An effect of incubation conditions

on posthatching growth has been noted in a few studies of

crocodilians,38,69 but its role in runting is an area that requires

further research. Finally, if it is established that there is a strong

clutch effect to runting and predictors discovered, investigation

into either preventative measures or altered rearing practices

for at-risk clutches might ameliorate the runting.

Acknowledgements

This research was approved by the University of Sydney Animal

Ethics Committee (reference number N00/9-2005/3/4204).

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to

the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for

the research, authorship and/or publication of this article: This project

was funded in part by the Australian Government, Rural Industries

Research and Development Corporation and findings briefly summar-

ized in RIRDC Publication No. 09/135 Improving Australia’s Croco-

dile Industry Productivity—Understanding Runtism and Survival.

Shilton et al 11

at University of Sydney on January 13, 2014vet.sagepub.comDownloaded from

References

1. Bohn AA. Laboratory evaluation of electrolytes. In: Thrall MA,

Weiser G, Allison R, et al, eds. Veterinary Hematology and Clin-

ical Chemistry. 2nd ed. Ames, IA: John Wiley; 2012:378–392.

2. Braastad BO. Effects of prenatal stress on behaviour of offspring

of laboratory and farmed mammals. Appl Anim Behav Sci. 1998;

61(2):159–180.

3. Buenviaje GN, Ladds PW, Melville L, et al. Disease-husbandry

associations in farmed crocodiles in Queensland and the Northern

Territory. Aust Vet J. 1994;71(6):165–173.

4. Campbell TW. Clinical chemistry of reptiles. In: Thrall MA,

Weiser G, Allison R, et al, eds. Veterinary Hematology and Clin-

ical Chemistry. 2nd ed. Ames, IA: John Wiley; 2012:599–606.

5. Campbell TW. Hematology of reptiles. In: Thrall MA, Weiser G,

Allison R, et al. Veterinary Hematology and Clinical Chemistry.

2nd ed. Ames, IA: John Wiley; 2012:277–297.

6. Canfield PJ. Characterization of the blood cells of Australian cro-

codiles (Crocodylus porosus [Schneider] and C. johnstoni

[Krefft]). Zbl Vet Med C Anat Histol Embryol. 1985;14:269–288.

7. Capen CC. Endocrine glands. In: Maxie MG ed. Jubb, Kennedy,

and Palmer’s Pathology of Domestic Animals. 5th ed. Vol 3.

Philadelphia, PA: Elsevier Saunders; 2007:325–428.

8. Chadio SE, Kotsampasi B, Papadomichelakis G, et al. Impact of

maternal undernutrition on the hypothalamic-pituitary-adrenal

axis responsiveness in sheep at different ages postnatal. J Endo-

crinol. 2007;192:495–503.

9. Churukian CJ. Pigments and minerals. In: Bancroft JD, Gamble

M, eds. Theory and Practice of Histological Techniques. Philadel-

phia, PA: Churchill Livingston Elsevier; 2008:233–259.

10. Cooper EL, Klempau AE, Zapata AG. Reptilian immunity. In:

Gans C, ed. Biology of the Reptilia. Vol 14, Development A. New

York, NY: John Wiley; 1985:599–678.

11. Davenport J, Grove DJ, Cannon TR, et al. Food capture, appetite,

digestion rate and efficiency in hatchling and juvenile Crocodylus

porosus. J Zool Lond. 1990;220:569–592.

12. Denardo D. Stress in captive reptiles. In: Mader DR, ed. Reptile

Medicine and Surgery. 2nd ed. St Louis, MO: Saunders Elsevier;

2006:119–134.

13. Edwards HE, McIntyre Burnham W. The impact of corticoster-

oids on the developing animal. Pediatr Res. 2001;50(4):433–440.

14. Elsey RM, Joanen T, McNease L, et al. Stress and plasma corti-

costerone levels in the American alligator—relationships with

stocking density and nesting success. Comp Biochem Phys A.

1990;95(1):55–63.

15. Elsey RM, Joanen T, McNease L, et al. Growth rate and plasma

corticosterone levels in juvenile alligators maintained at different

stocking densities. J Exp Zool. 1990;255:30–36.

16. Entringer S, Epel ES, Kumsta R, et al. Stress in intrauterine life is

associated with shorter telomere length in young adulthood. Proc

Natl Acad Sci U S A. 2011;108(33): E513–E518.

17. Eriksen MS, Bakken M, Espmark A, et al. Prespawning stress in

farmed Atlantic salmon Salmo salar: maternal cortisol exposure

and hyperthermia during embryonic development affect offspring

survival, growth and incidence of malformations. J Fish Biol.

2006;69:114–129.

18. Fadley AM, Zaval G, Witter RL. Reticuloendotheliosis. In: Saif

YM, ed. Diseases of Poultry. 12th ed. Ames, IA: Blackwell;

2008:568–588.

19. Finger JW, Isberg SR. A review of innate immune functions in

crocodilians. CAB Rev. 2012;7(67):1–11.

20. Finger JW, Adams AL, Thomson PC, et al. Using phytohaemag-

glutinin to determine immune responsiveness in saltwater croco-

diles (Crocodylus porosus). Aust J Zool. 2013;61(4):301–311.

21. Foggin CM. Disease and disease control on crocodile farms in

Zimbabwe. In: Webb GJW, Manolis SC, Whitehead PJ, eds.

Wildlife Management: Crocodiles and Alligators. Chipping Nor-

ton, Australia: Surrey Beatty & Sons; 1987:351–362.

22. Foggin CM. Diseases of farmed crocodiles. In: Smith GA, Marais

J, eds. Conservation and Utilization of the Nile Crocodile in

Southern Africa: Handbook on Crocodile Farming. Pretoria: Cro-

codilian Study Group of Southern Africa; 1992:107–140.

23. Franklin CE, Davis BM, Peucker SKJ, et al. Comparison of stress

induced by manual restraint and immobilization in the estuarine

crocodile, Crocodylus porosus. J Exp Zool. 2003;298A:86–92.

24. Gabe M. The adrenal. In: Gans C, ed. Biology of the Reptilia. Vol

3, Morphology C. New York, NY: Academic Press; 1970:

263–318.

25. Gardiner CH, Imes GD, Jacobson ER, et al. Sporulated coccidian

oocysts resembling Goussia Labbe, 1896 in the viscera of Nile

crocodiles. J Wildl Dis. 1986;22(4):575–577.

26. Garnett ST, Murray RM. Parameters affecting the growth of the

estuarine crocodile, Crocodylus porosus, in captivity. Aust J Zool.

1986;34:211–223.

27. Grasman KA. Assessing immunological function in toxicological

studies of avian wildlife. Integ Comp Biol. 2002;42:34–42.

28. Haines RW. Epiphyses and sesamoids. In: Gans C, ed. Biology of

the Reptilia. Vol 1, Morphology A. New York, NY: Academic

Press; 1969:81–115.

29. Hakim AH. Corticosteroids and immune systems of non-

mammalian vertebrates: a review. Dev Comp Immunol. 1988;

12:481–494.

30. Hernandez-Divers SJ, Cooper JE. Hepatic Lipidosis. In: Mader

DR, ed. Reptile Medicine and Surgery. 2nd ed. St Louis, MO:

Saunders Elsevier; 2006:806–813.

31. Hochberg A. Mechanisms of steroid impairment of growth. Horm

Res. 2002;58(suppl 1):33–38.

32. Huchzermeyer FW. Crocodiles: Biology, Husbandry and

Diseases. Oxon, UK: CABI Publishing; 2003.

33. Isberg S, Shilton C, Thomson P. Improving Australia’s Crocodile

Industry Productivity—Understanding Runtism and Survival.

Barton, Australia: Rural Industries Research and Development

Corporation; 2009.

34. Isberg SR, Thomson PC, Nicholas FW, et al. Quantitative analysis

of production traits in saltwater crocodiles (Crocodylus porosus):

II. Age at slaughter. J Anim Breed Genet. 2005;122:370–377.

35. Jacobson ER. Viruses and viral diseases of reptiles. In: Jacobson

ER, ed. Infectious Diseases and Pathology of Reptiles. Boca

Raton, FL: CRC Press; 2007:395–460.

36. Janczak AM, Braastad BO, Bakken M. Behavioural effects of

embryonic exposure to corticosterone in chickens. Appl Anim

Behav Sci. 2006;96:69–82.

12 Veterinary Pathology

at University of Sydney on January 13, 2014vet.sagepub.comDownloaded from

37. Jessop TS, Tucker AD, Limpus CJ, et al. Interactions between

ecology, demography, capture stress, and profiles of corticoster-

one and glucose in a free-living population of Australian fresh-

water crocodiles. Gen Comp Endocr. 2003;132:161–170.

38. Joanen T, McNease L, Ferguson MWJ. The effects of egg incuba-

tion temperature on post-hatching growth of American alligators.

In: Webb GJW, Manolis C, Whitehead PJ, eds. Wildlife Manage-

ment: Crocodiles and Alligators. Chipping Norton, Australia:

Surrey Beatty & Sons; 1987:533–537.

39. Kanui T, Mwenda C, Aulie A, et al. Effects of temperature on growth,

food uptake and retention time of juvenile Nile crocodiles (Crocody-

lus niloticus). Comp Biochem Physiol A. 1991;99(3):453–456.

40. Ladds PW. Pathology of Australian Native Wildlife. Colling-

wood, Australia: CSIRO Publishing; 2009.

41. Ladds PW, Bradley J, Hirst RG. Providencia rettgeri meningitis in

hatchling saltwater crocodiles (Crocodylus porosus). Aust Vet J.

1996;74:397–398.

42. Lance VA, Lauren D. Circadian variation in plasma corticoster-

one in the American alligator, Alligator mississippiensis, and the

effects of ACTH injections. Gen Comp Endocr. 1984;54:1–7.

43. Lance VA, Morici LA, Elsey RM. Physiology and endocrinology

of stress in crocodilians. In: Grigg CG, Seebacher F, Franklin CE,

eds. Crocodilian Biology and Evolution. Chipping Norton,

Australia: Surrey Beatty & Sons; 2000:327–340.

44. Lauren DJ. The effect of chronic saline exposure on the electro-

lyte balance, nitrogen metabolism and corticosterone titer in the

American alligator, Alligator mississippiensis. Comp Biochem

Physiol. 1985;81A(2):217–223.

45. Lesage J, Sebaai N, Leonhardt M, et al. Perinatal maternal under-

nutrition programs the offspring hypothalamo-pituitary-adrenal

(HPA) axis. Stress. 2006;9(4):183–198.

46. Lowenstine LJ, Munson L. Iron overload in the animal kingdom.

In: Fowler ME, Miller RE, eds. Zoo and Wild Animal Medicine,

Current Therapy 4. Philadelphia, PA: Saunders; 1999:260–268.

47. MacLachlan NJ, Dubovi EJ. Fenner’s Veterinary Virology. 4th

ed. San Diego, CA: Academic Press; 2011.

48. Mayer R. Crocodile Farming: Research, Development and On-

Farm Monitoring, 1995–1998. Barton: Australian Government

Rural Industries Research and Development Corporation; 1998.

49. McInerney J. Liver enzymes and pathology in runt crocodiles (C.

porosus). Proc Assoc Reptilian Amphibian Vet. 1994:57–58.

50. Melville L, Davis S, Shilton C, et al. Hunting for Viruses in

Crocodiles: Viral and Endogenous Retroviral Detection and

Characterization in Farmed Crocodiles. Barton: Australian

Government Rural Industries Research and Development Corpo-

ration; 2012.

51. Meuten D. Parathyroid glands and calcium and phosphorus meta-

bolic pathology: Thrall MA, Weiser G, Allison R, et al. Veterin-

ary Hematology and Clinical Chemistry. Ames, IA: John Wiley;

2012:545–568.

52. Meylan S, Belliure J, Clobert J, et al. Stress and body condition

as prenatal and postnatal determinants of dispersal in the com-

mon lizard (Lacerta vivipara). Horm Behav. 2002;42:319–326.

53. Morici LA, Elsey RM, Lance V. Effects of long-term corticoster-

one implants on growth and immune function in juvenile alliga-

tors, Alligator mississippiensis. J Exp Zool. 1997;279:156–162.

54. Origgi FC. Reptile immunology. In: Jacobson ER, ed. Infectious

Diseases and Pathology of Reptiles. Boca Raton, FL: CRC Press;

2007:131–166.

55. Peucker S, Mayer R. Runt observation studies. Crocodile Res

Bull. 1995;1:57–62.

56. Poetker DM, Reh DD. A comprehensive review of the adverse

effects of systemic corticosteroids. Otolaryngol Clin North Am.

2010;43:753–768.

57. Reid IR. Glucocorticoid osteoporosis—mechanisms and manage-

ment. Eur J Endocrinol. 1997;137:209–217.

58. Riese G. Factors Affecting Food Intake and Growth in Captive

Saltwater Crocodile (Crocodylus porosus) [MSc thesis]. Brisbane:

Department of Zoology, University of Queensland, Australia;

1995.

59. Schaffner F. The liver. In: Gans C, Gaunt AS, eds. Biology of the

Reptilia, Volume 19, Morphology G, Visceral Organs. Ithaca, NY:

Society for the Study of Amphibians and Reptiles; 1998:485–531.

60. Schobitz B, Reul JMHM, Holsboer F. The role of the

hypothalamic-pituitary-adrenocortical system during inflamma-

tory conditions. Crit Rev Neurobiol. 1994;8(4):263–291.

61. Smith GA, Marais J. Stress in crocodilians—the impact of nutri-

tion: Proceedings of the 12th Working Meeting of the Crocodile

Specialist Group. Vol 2. Gland, Switzerland: IUCN—The World

Conservation Union; 1994:2–38.

62. Sypek J, Borysenko M. Reptiles. In: Rowley AF, Ratcliffe NA,

eds. Vertebrate Blood Cells. Cambridge, UK: Cambridge Univer-

sity Press; 1988:211–256.

63. Thompson K. Bones and joints. In: Maxie MG, ed. Jubb, Ken-

nedy, and Palmer’s Pathology of Domestic Animals. 5th ed. Vol

1. Philadelphia, PA: Elsevier Saunders; 2007:1–184.

64. Turton JA. Preliminary report on stress in farmed Crocodylus poro-

sus hatchlings. In: Proceedings of the 2nd Regional Meeting of the

Crocodile Specialist Group of the Species Survival Commission of

IUCN—The World Conservation Union. International Union for the

Conservation of Nature and Natural Resources in conjunction with

the Conservation Commission of the Northern Territory, Darwin,

Australia; 1993.

65. Turton JA, Ladds PW, Manolis SC, et al. Relationship of blood

corticosterone, immunoglobulin and haematological values in

young crocodiles (Crocodylus porosus) to water temperature,

clutch of origin and body weight. Aust Vet J. 1997:75(2):114–119.

66. Uller T, Olsson M. Direct exposure to corticosterone during

embryonic development influences behaviour in an ovovivipar-

ous lizard. Ethology. 2006;112(4):390–397.

67. Valli VEO. Hematopoietic system. In: Maxie MG, ed. Jubb, Ken-

nedy, and Palmer’s Pathology of Domestic Animals. 5th ed. Vol

3. Philadelphia, PA: Elsevier Saunders; 2007:107–324.

68. Warner D, Radder RS, Shine R. Corticosterone exposure during

embryonic development affects offspring growth and sex ratios

in opposing directions in two lizard species with environmental

sex determination. Physiol Biochem Zool. 2009;82(4):363–371.

69. Webb GJ, Cooper-Preston H. Effects of incubation temperature

on crocodiles and the evolution of reptilian oviparity. Am Zool.

1989;29:953–971.

70. Webster Marketon JI, Glaser R. Stress hormones and immune

function. Cell Immunol. 2008;252:16–26.

Shilton et al 13

at University of Sydney on January 13, 2014vet.sagepub.comDownloaded from