clinical disease and laboratory abnormalities in

7/31/2019 Clinical Disease and Laboratory Abnormalities In

1/22

35

Journal of Wildlife Diseases, 39(1), 2003, pp. 3556 Wildlife Disease Association 2003

CLINICAL DISEASE AND LABORATORY ABNORMALITIES IN

FREE-RANGING DESERT TORTOISES IN CALIFORNIA (19901995)

Mary M. Christopher,1,5 Kristin H. Berry,2 Brian T. Henen,3,4 and Kenneth A. Nagy3

1 Department of Pathology, Microbiology & Immunology, School of Veterinary Medicine, University of CaliforniaDavis,Davis, California 95616, USA2 22835 Calle San Juan de Los Lagos, Moreno Valley, California 92553, USA

3 Department of Organismic Biology, Ecology, and Evolution, and Laboratory of Biomedical and EnvironmentalSciences, University of CaliforniaLos Angeles, Los Angeles, California 90024, USA4 Current address: Department of Zoology, University of the Western Cape, Private Bag X17, Bellville, 7535,South Africa5 Corresponding author (email: [email protected])

ABSTRACT: Desert tortoise (Gopherus agassizii) populations have experienced precipitous de-clines resulting from the cumulative impact of habitat loss and human and disease-related mor-tality. Diagnosis of disease in live, free-ranging tortoises is facilitated by evaluation of clinical signsand laboratory test results but may be complicated by seasonal and environmental effects. Thegoals of this study were: 1) to describe and monitor clinical and laboratory signs of disease inadult, free-ranging desert tortoises at three sites in the Mojave Desert of California (USA) be-tween October 1990 and October 1995; 2) to evaluate associations between clinical signs andhematologic, biochemical, serologic, and microbiologic test results; 3) to characterize diseasepatterns by site, season, and sex; and 4) to assess the utility of diagnostic tests in predictingmorbidity and mortality. Venous blood samples were obtained four times per year from tortoisesof both sexes at the Desert Tortoise Research Natural Area (DTNA), Goffs/Fenner Valley, andIvanpah Valley. Tortoises were given a physical examination, and clinical abnormalities were grad-ed by type and severity. Of 108 tortoises, 68.5% had clinical signs of upper respiratory tractdisease consistent with mycoplasmosis at least once during the study period. In addition, 48.1%developed moderate to severe shell lesions consistent with cutaneous dyskeratosis. Ulcerated orplaque-like oral lesions were noted on single occasions in 23% of tortoises at Goffs and 6% oftortoises at Ivanpah. Tortoises with oral lesions were significantly more likely than tortoises with-out lesions to have positive nasal cultures for Mycoplasma agassizii (P0.001) and to be dehy-drated (P0.0007). Nine tortoises had marked azotemia (blood urea nitrogen [BUN] 100 mg/dl) or persistent azotemia (BUN 6376 mg/dl); four of these died, three of which had necropsyconfirmation of urinary tract disease. Laboratory tests had low sensitivity but high specificity in

assessing morbidity and mortality; there was marked discrepancy between serologic and cultureresults for M. agassizii. Compared with tortoises at other sites, tortoises at DTNA were morelikely to be seropositive for M. agassizii. Tortoises at Goffs were significantly more likely to havemoderate to severe shell disease, oral lesions, positive nasal cultures for M. agassizii, and increasedplasma aspartate aminotransferase activity. The severe disease prevalence in Goffs tortoises likelycontributed to the population decline that occurred during and subsequent to this study.

Key words: Chelonian, clinical chemistry, cutaneous dyskeratosis, desert tortoise, hematology,Mojave Desert, Gopherus agassizii, Mycoplasma agassizii, mycoplasmosis, shell disease, upperrespiratory tract disease.

INTRODUCTION

Over the past 25 yr, dramatic declines

have been observed in many desert tor-toise (Gopherus agassizii) populations inthe Mojave and Colorado deserts of thesouthwestern United States (US Fish and

Wildlife Service, 1994; Berry and Medica,1995; Berry, 1997). These declines havebeen attributed to the cumulative impactsof human activities, predation, habitat lossand degradation, and disease (US Fish and

Wildlife Service, 1994). In April 1990, de-

sert tortoise populations north and west ofthe Colorado River were listed as threat-

ened by the US Fish and Wildlife Service.At about the same time, the Bureau ofLand Management initiated several re-search projects in response to developingdisease problems in desert tortoises, in-cluding mycoplasmosis (upper respiratorytract disease [URTD]; Jacobson et al.,1991; Brown, et al., 1994, 1999; Homer etal., 1998) and shell disease (cutaneous dys-keratosis; Jacobson et al., 1994; Homer et


2/22

36 JOURNAL OF WILDLIFE DISEASES, VOL. 39, NO. 1, JANUARY 2003

al., 1998). A health profile program wasestablished to determine reference inter-

vals for lab oratory data for clinicallyhealthy desert tortoises at various seasonsand geographic sites (Christopher et al.,1999; Berry and Christopher, 2001). Base-line data would be useful for identifyingabnormal laboratory results, which mayhelp diagnose stress or disease in tortoises(Jacobson et al., 1991, 1994; OConnor etal., 1994).

In the course of the health profile study,two-thirds of samples were obtained fromtortoises that were clinically healthy at thetime of sampling, but one-third of samples

were from tortoises with clinical signs ofdisease or abnormal laboratory test results(Christopher et al., 1999). Reference in-

tervals were constructed using laboratoryvalues only from those sampling periodswhen tortoises had negative serologic andculture results for mycoplasmosis and noor minimal clinical signs of URTD. Tor-toises with severe or persistent clinical ab-normalities or laboratory abnormalities

were not included in the reference popu-lation. The purpose of the present study

was to characterize the prevalence, sever-ity, and types of clinical signs of disease in

these same three populations of free-rang-ing desert tortoises and to determine theirassociation, if any, with alterations in lab-oratory data and mortality. In addition,shells were evaluated retrospectively forlesions of cutaneous dyskeratosis. The lon-gitudinal design of the study facilitatedmonitoring of clinical and laboratory ab-normalities over time, including years hav-ing below and above average rainfall.

MATERIALS AND METHODS

Sites

Samples were obtained from tortoises atthree sites in the Mojave Desert of California(USA): the Desert Tortoise Research NaturalArea (DTNA) in the western Mojave Desert(3510N, 11810W; elevation 869945 m);Goffs/Fenner Valley (Goffs) in the eastern Mo-

jave Desert, at the boundary with the northernColorado Desert (3452N, 11510E; elevation700 m); and Ivanpah Valley (Ivanpah), in the

western corner of the northeastern Mojave De-sert (3539N, 11515W; elevation 866914 m).Long-term (19611996), annual, and winterprecipitation data from these sites were ob-tained from the National Oceanic and Atmo-spheric Administration (19891995).

Animals, sampling periods, and physical findings

Study populations and sampling periodswere previously described (Christopher et al.,1999). Briefly, 10 adult males and 10 adult fe-males at each site were captured, marked, andfitted with radio transmitters (AVM model SB-2, AVM Instrument Co., Livermore, California,

with 48-cm copper antenna, powered by a lith-ium C-cell) in 1989. Tortoises were identifiedby a combination of letters (site and sex) andnumbers. Tortoises that died or disappeared

were replaced, when possible, with tortoises ofthe same sex and given sequential, new num-bers. Samples were collected from tortoises atall sites four times per year: late winter (Feb-ruaryMarch, just prior to emergence from hi-bernation), spring (MayJune, during a time ofpeak activity and forage), summer (JulyAu-gust, a hot, typically dry season), and fall (Oc-tober, just prior to hibernation). Data were or-ganized by hydrologic year (fall through sum-mer seasons; October 1September 30). Study

years included 199091, 199192, 199293,199394, 199495, and a final sample in fall1995.

Precautions were taken to prevent transmis-sion of pathogens within and among study sites.

Each tortoise was handled with a fresh pair ofdisposable gloves, and items of equipmenttouching the tortoise were disinfected with asodium hypochlorite solution (0.175%) or eth-anol (70%) immediately after use. Precautions

were taken to assure the tortoise did not touchor rest upon the field workers limbs, clothing,or equipment without protective covering.

Body weight using a top-loading balance andcarapace length at the midline (MCL) using al-cohol-sterilized calipers were measured eachtime a tortoise was sampled. In addition, tor-toises were given a complete physical exami-nation, using field guidelines described else-

where (Berry and Christopher, 2001). Briefly,the following features were evaluated: beak andnares, forelegs, breathing, eyes, chin (mental)glands, integument, posture/behavior, oral cav-ity, shell, evidence of trauma (e.g., bite wounds,abrasions), and presence of ectoparasites. Ticks

were sent to Dr. James Oliver (Georgia South-ern University, Institute of Arthropodology andParasitology, Statesboro, Georgia, USA) foridentification. Retrospective, blinded examina-tion (by KB) of color transparencies (35 mm


3/22

CHRISTOPHER ET AL.DISEASE ABNORMALITIES IN FREE-RANGING DESERT TORTOISES 37

slides) of the carapace and plastron was doneto further evaluate shell disease. Full-frame im-ages of the head, carapace, and plastron weretaken when the tortoise was first entered in thestudy and again in the spring of each year. Pho-tographs also were taken of any physical ab-normality or lesion, when noted. For each tor-toise, the severity of shell lesions was graded as

none, mild, moderate, or severe, based on thedegree of discoloration, lifting laminae, flaking,and amount of bone exposed on the carapaceand plastron (Berry and Christopher, 2001).Distribution of shell lesions was categorized asaffecting 0%, 10%, 1140%, or 40% of theplastron or carapace. An overall grade of none,mild, moderate, or severe shell disease wasbased on the combined assessment of carapaceand plastron. Shell lesions also were classifiedas active, if current lesions were noted, orchronic, if signs of regression or healing wereseen, such as healthy laminae at the seams ofscutes.

Scute condition (from 35 mm transparen-cies) was used to determine the approximateage class of tortoises, based on the followingcriteria (Berry and Woodman, 1984): 1) very

young adult (1225 yr), full or partial areola oneach scute, growth rings surrounding areolaover 90% of carapace; 2) young adult (2550

yr), most areola worn away, growth rings chip-ping and flaking (normal wear) but distinct andin good condition; 3) middle-aged adult (5070

yr), many growth rings worn away, flaking, andchipping, maximal scute thickness; 4) old adult(70 yr), depressions on surfaces of eight or

fewer scutes, scutes thinning; 5) very old adult,depressions on most scutes involving entirescute, most growth rings gone, thin laminae.

Serology and microbiology

Beginning in winter 1992, plasma samplesfrom all tortoises were sent to Dr. Mary Brown(University of Florida, Gainesville, Florida,USA) where they were tested for antibodies toMycoplasma agassizii using an enzyme linkedimmunosorbent assay (ELISA), as previouslydescribed (Schumacher et al., 1993). Serologicresults were classified as positive or negative;suspect results were considered negative un-less the tortoise seroconverted to a positive titer(see Brown et al., 1999 for a complete descrip-tion of serologic results). Nasal flushes were ob-tained by introducing 0.5 ml of trypticase soybroth (BBL Microbiology Systems, Cockeys-

ville, Maryland, USA) into the anterior naresusing a sterile open-end tomcat catheter. Afterflushing the nasal passages, aspirates wereplaced in sterile cryovials, frozen at 196 C,and transferred to a veterinary diagnostic lab-

oratory (APL Laboratories, Las Vegas, Nevada,USA) within 24 hr. Cultures for Pasteurella tes-

tudinis and mixed flora were done using bloodagar (trypticase soy agar with 5% sheep redblood cells RBC), chocolate agar (incubated at35 C in 5% CO2), and MacConkey agar (in-cubated at 35 C in ambient atmosphere) plates.One blood agar plate was incubated at 35 C in

a 5% CO2 atmosphere, and another at 25 C inambient atmosphere. Cultures were examineddaily for 7 days. Nasal flush samples also werecultured for M. agassizii as previously de-scribed (Brown et al., 1994). Nasal flushes werediscontinued after fall 1994 due to lack of fund-ing.

Blood samples and analysis

Blood collection, processing, and analyticalmethods were previously described (Christo-pher et al., 1999). Briefly, sterile technique wasused to obtain 4.0 to 4.7 ml blood from the

jugular vein. Aliquots were placed in tubes con-taining lithium heparin for whole blood analysisand for separation of plasma. Air-dried smears

were fixed in methanol and stained with mod-ified Wrights stain. Microhematocrit centrifu-gation was used to determine packed cell vol-ume (PCV) both in the field and in the labo-ratory. Frozen plasma (196 C) and chilled

whole blood samples (4 C) were transported tothe laboratory within 24 hr for analysis.

A complete blood count included PCV, RBCcount, white blood cell (WBC) count, hemo-globin (Hb) concentration, mean cell volume(MCV), mean cell Hb (MCH), mean cell Hbconcentration (MCHC), differential WBCcount, and qualitative evaluation of blood cellmorphology (Raskin, 2000). Clinical chemistrytests included glucose, blood urea nitrogen(BUN), creatinine (beginning in spring 1993),uric acid, total protein, albumin, calcium, phos-phorus, total bilirubin, direct bilirubin, triglyc-erides (beginning in spring 1993), bile acids(beginning in spring 1993), cholesterol, iron(beginning in summer 1993), magnesium (be-ginning in spring 1994) sodium, potassium,chloride, and total carbon dioxide (TCO2) con-centrations and alkaline phosphatase (ALP), as-

partate aminotransferase (AST), and alanineaminotransferase (ALT) activities. Globulins,albumin/globulin ratio, anion gap, osmolality,sodium/potassium ratio, and indirect bilirubin

were calculated.

Statistical analysis

Statistical analysis was done using a comput-er-assisted program (StatView, Abacus Con-cepts, Inc., Berkeley, California). Cut-off valuesfor each analyte were established by examina-


4/22


FIGURE 1. Distribution of tortoises by age-sizeclass and sex at three sites in the Mojave desert. Solidbar males; shaded bar females.

tion of previously published reference intervalsfor site, season, and sex (Christopher et al.,1999) and examination of scatterplots of all val-ues. Crude mortality rate (%) at each site wascalculated as the number of tortoises founddead divided by the total number of tortoisessampled at that site per hydrologic year. Aver-age annualized mortality rate (%) was the total

mortality rate divided by 5 yr.Tortoises were categorized by sex, site, age

class, status at the end of the study (live, dead),year of death, presence and severity of shelldisease (none to mild, moderate to severe),amount of P. testudinis and mixed flora growth(none to light, moderate to heavy), presence orabsence of marked azotemia (BUN100 mg/dl), persistent azotemia (increased BUN in twoor more consecutive seasons), bite marks, ticks,oral lesions, clinical dehydration, clinical signsof URTD, seropositivity for M. agassizii, andculture-positivity for M. agassizii. For calcula-tion of sensitivity, specificity, and predictive val-ues, tortoises were categorized as affected (pos-itive, present) or unaffected (negative, absent)based on cut-off values for the following labo-ratory abnormalities: anemia (decreased PCV,Hb, or RBC count), leukocytosis, heterophilia,heteropenia, lymphocytosis, lymphopenia, ba-sopenia, increased iron concentration (hyper-ferremia), increased AST activity, hyperglobu-linemia, hypertriglyceridemia, hypophosphate-mia, hypomagnesemia, any electrolyte abnor-mality, increased bile acids concentration,hyperbilirubinemia, and azotemia. Some ab-normalities affected too few tortoises for statis-

tical analysis. Contingency tables and Chi-square (2) analyses were used to evaluate as-sociations among variables. Quantitative differ-ences in some values (e.g., numbers of positiveELISA tests by site) were determined usingStudents t-test and analysis of variance (AN-OVA). Data were reported as meanSD unlessotherwise indicated. Significance was definedas P0.05. Sensitivity, specificity, and predic-tive values were calculated for laboratory andclinical variables that were significantly associ-ated (based on 2 analysis) with death, sero-positivity, culture positivity for M. agassizii, ormoderate to severe cutaneous dyskeratosis.

RESULTS

A total of 1,071 observations weremade, and blood samples obtained, from108 tortoises. Tortoises were sampled9.96.9 times (median8; range121samples) over the 5 yr (21 sample) period.These included 295 observations on 40tortoises (23 males, 17 females) at DTNA;

393 observations on 35 tortoises (13 males,22 females) at Goffs; and 383 observationson 33 tortoises (18 males, 15 females) atIvanpah. Sixty-three of the 108 tortoises

were sampled one to six times, 21 tortoiseswere sampled seven to 12 times, 23 tor-

toises were sampled 1318 times, and onetortoise was sampled at each of the 21sampling periods. There was no significantdifference in the number of male and fe-male tortoises sampled each season at eachsite. There was a significant difference(225.24, P0.0014) in age distributionof tortoises at the three sites, with thehighest percent of very old tortoises (65%)at Ivanpah, and the highest percentage of

very young tortoises (70%) at the DTNA

(Fig. 1).The study period included both norms

and extremes of annual and winter precip-itation amounts (see Christopher et al.,1999 for detailed rainfall data). Annualprecipitation was at or above long-termnorms in 199091, below long-term normsin 199394, and above long-term norms in199192, 199293, and 199495 at allthree sites. Winter precipitation was belowlong-term norms in 198990 (the year pre-

ceding the study period) and 199394 atall three sites.

Mortality

Of 108 tortoises evaluated, 58 (53.7%)were known survivors as of fall 1995 (Table1). Thirty-two tortoises disappeared and

were not relocated for the remainder ofthe study. Several tortoises were missingfor one or more sampling periods but were


5/22


TABLE 1. Numbers of tortoises entered into and exiting the study each year.

SiteFall

1990 199091 199192 199293 199394 199495Fall

1995 Total % at siteAnnualized

mortality

DTNANo. enteredNo. found deadNo. missingb

13

323

802

21a

1

305

82a

3

301

405

15

12.5%37.5%

2.5%

GOFFSNo. enteredNo. found deadNo. missing

16

401

422

304

36c

1

400

11a

0

3598

25.7%22.9%

5.1%

IVANPAHNo. enteredNo. found deadNo. missing

15

621

503

101

401

221

002

3349

12.1%27.3%

2.4%

ALL SITESNo. enteredNo. found deadNo. missing

44

1345

1727

616

1066

1443

413

1081832

16.7%29.6%

3.3%

a

One tortoise assumed dead because radiotransmitter was found.b Unable to locate at this time and for the remainder of the study.c Five tortoises were found dead in July 1994.

found again at a later date, and thus re-mained in the study. Goffs had about twicethe crude and annualized mortality rate ofDTNA and Ivanpah. All nine Goffs tor-toises that died were female (27.37,P0.0066); whereas, females accountedfor 40% of deaths at the DTNA and 50%

of deaths at Ivanpah (Table 2). Six of ninetortoises at Goffs died in 199394. Eightof nine deaths at the DTNA and Ivanpahoccurred the year following a dry winter(i.e., in 199091 and 199495).

At the DTNA, two of five tortoises thatdied (D40M, I8M) had urolithiasis withazotemia, and clinical signs of URTD (Ta-ble 2). Tortoise D40M was one of only twotortoises at the DTNA (along with D25M)

with moderate (versus none to mild) shell

disease; four of nine Goffs tortoises andthree of four Ivanpah tortoises that diedalso had moderate to severe shell disease.Tortoise D7M had severe peritonitis andmultiple bladder wall abscesses, presum-ably related to previous cystocentesesdone as part of a research study (Klaassen,1991; Peterson, 1994). Eleven of 18 tor-toises that died had no clinical or labora-tory abnormalities in the season preceding

death, including eight of nine Goffs tor-toises. Necropsy results for tortoise G30Flisted predation and acute bacterial pneu-monia as the cause of death (Homer et al.,1998).

Clinical signs

Three tortoises (D7M, D9M, I9M) hadgeneral signs of illness, including poorbody condition, lethargy, and weakness.One tortoise died (D7M, described above)and the other two tortoises went missing.Tortoise D9M was dehydrated, had hypo-glycemia (glucose 22 mg/dl, reference val-ues 40169 mg/dl), a low basophil count(26/l; reference values, 623,575/l), andlymphopenia (26/l; reference values, 632,746/l), and was seropositive for M.

agassizii. Tortoise I9M also was dehydrat-ed, and subsequently developed markedleukocytosis (40,000 WBC/l; reference

values, 149610,924/l), heterophilia(20,000 heterophils/l; reference values,7197,159/l) and lymphocytosis (16,000lymphocytes/l; reference values, 632,746/l). An additional 16 tortoises(14.8%) were clinically dehydrated, as ev-idenced by decreased skin elasticity or


6/22


TABLE2

.

Clin

icalsignso

fdiseasean

dlaboratoryabnormal

itiesin

18torto

isesthat

died

.

Tortoise

no.

No.

observations

Found

dead

ELISA

Culture

Necropsy

Clinicalandlaboratory

abnormalitiesinthe

seasonprecedingdeath

Earlierclinicalandlaboratoryabnormalities

(inchronologicalorder)

D7M

2

Spring

91

NDa

ND

Periton

itis

,

bladderab-

scessesb

De

hydrate

d,

weak

,azotem

ia

(251mg

/dl)

,hyperosmo

lality

(408m

Osm

/kg

)

A

zotem

ia(217mg

/dl)

,azurop

hilia(1,7

10/l)

D16F

3

Summer

91

ND

ND

ND

None

N

one

D30Fc

2

Winter

93

ND

None

N

one

D40Md

4

Spring

95

Uro

lith

iasise

Eyesswo

llen

,mucus

ineyes,

azotem

ia(71mg

/dl)

,1

po

lychromas

ia,

hypomagne-

sem

ia(3

.0mg

/dl)

D

arkbrownth

ickurine

,hyperb

ilirubinem

ia(0

.6mg

/dl)

,azo-

tem

ia(25an

d76mg

/dl)

,increased

iron

(292g

/dl)

D41Mc

3

Summer

95

ND

Hypertrig

lyceri

dem

ia(431mg

/

dl)

M

ucus

inrig

hteye

,hypomagnesem

ia

(3.0

mg

/dl)

G21F

5

Summer

92

ND

ND

ND

None

B

ubbles

fromnares,weteye

lids,

dehy

dration

,hyperp

hosp

ha-

tem

ia(9

.1mg

/dl)

,highan

iongap

(41mmo

l/l)

G23Fd

16

Fal

l95

ND

Righteyecrusty

R

aspy,cl

icking

breat

hing

,mucus

inbo

theyes,

de

hydrate

d,

swo

lleneye

lids,

drie

dmucus

ineyes,azotem

ia(46mg

/dl)

,

heterop

hilia(8

,500/l)

,hyperkalem

ia(8

.6mmo

l/l)

,ora

lu

l-

cers

,fou

l-sme

llingora

lcavity

,hyperkalem

ia(7

.7mmo

l/l)

,

meta

bo

licac

idosis

(tota

lCO

210mmo

l/l)

G27F

13

Summer

94

ND

None

M

ucus

ineye

,heteropen

ia(364/l)

,azotem

ia(57mg

/dl)

,

highan

iongap

(37mmo

l/l)

,inpoorcon

dition

,highAST

(170U/l),hyperc

hlorem

ia(144mmo

l/l)

,hyperkalem

ia(8

.2

mmo

l/l)

G30F

1

Winter

92

ND

ND

Pneumon

ia,

pre

dat

ione

None

N

otapp

lica

ble

G31Fd

9

Summer

94

ND

None

O

ralp

laques,

highAST(158U/l),wat

eryeye

,highAST(341

U/l),ora

lu

lcer,

hyperg

lycem

ia(223

mg

/dl)

,ticks

G32F

9

Summer

94

ND

None

H

yperosmo

lality

(416m

Osm

/kg

),azotem

ia(112mg

/dl)

,hy-

perp

hosp

hatem

ia(6

.9mg

/dl)

,azotem

ia(194mg

/dl)

,hyper-

chlorem

ia(145mmo

l/l)

,1

po

lychromas

ia,

heterop

hilia

(8,4

22/l)

,hyperc

hlorem

ia(149mm

ol/l)

G34Fd

3

Spring

94

ND

None

1

po

lychromas

ia,

highRBCcount

G35F

5

Summer

94

ND

None

1

po

lychromas

ia,

highRBCcount,h

ypertrig

lyceri

dem

ia

(782mg

/dl)

G36Fd

5

Summer

94

ND

None

H

ighRBCcount


7/22


TABLE2

.

Continue

d.

Tortoise

no.

No.

observations

Found

dead

ELISA

Culture

Necropsy

Clinicalandlaboratory

abnormalitiesinthe

seasonprecedingdeath

Earlierclinicalandlaboratoryabnormalities

(inchronologicalorder)

I8Md

17

Winter

95

Uro

lith

iasise

Azotem

ia(306mg

/dl)

,hyperos-

mo

lality

(421m

Osm

/kg

)

H

eteropen

ia(109/l)

,anem

ia(PCV10%

,hemog

lobin2

.7

mg

/dl)

,heteropen

ia(570/l)

,lymph

open

ia(20/l)

,hetero-

philia(12

,008/l)

,w

heezingraspyb

reat

h,

hypop

hosp

hate-

mia(0

.8mg

/dl)

,increase

dcreat

inine

(0.5

mg

/dl)

,azotem

ia

(46mg

/dl)

,hypop

hosp

hatem

ia(0

.5mg

/dl)

,azotem

ia(140

mg

/dl)

,hypop

hosp

hatem

ia(0

.5mg/

dl)

I12F

3

Summer

91

ND

ND

ND

None

N

one

I28Fd

1

Summer

91

ND

ND

ND

Hyperp

hosp

hatem

ia(9

.5mg

/

dl)

N

otapp

lica

ble

I42Md

2

Spring

95

ND

Hypop

hosp

hatem

ia(0

.2mg

/dl)T

icks

aNDindicatesnotdone;URTD,upperrespiratorytractdisease;AST

,aspartateaminotransferase;RBC

,redbloodcell.

bPresumedsecondarytoprevious

cystocentesis.

cBitemarksnotedonshellatallobservations.

dThesetortoisesalsohadmoderatetomarkedshelldisease.

eCompletenecropsyresultsreportedinHomeretal.,

1998

.


8/22


FIGURE 2. Number (for each site) and mean per-cent of all tortoises showing clinical signs of diseaseduring a 5 yr period. General signs of illness includedlethargy, weakness, and dehydration. Some tortoises

with signs of upper respiratory tract disease (URTD)also had general signs of illness. All tortoises with orallesions had clinical signs of URTD at some point inthe study.

sunken eyes; all but four of these also hadclinical signs of URTD.

Seventy-four tortoises (68.5%) had signsof URTD at some point during the 5 yrstudy period (Fig. 2), including puffy,swollen eyes and eyelids (25.5%), wet or

watery eyes (11.6%) or nose (6.0%), crust-ed eyes and eyelids (16.2%), mucus or oth-er discharge in the eyes (12.4%) or in the

nares (2.2%), and audible respiratorysounds (4.3%). Clinical signs did not differby site, and many tortoises had more thanone sign of URTD. Nine tortoises (threeat DTNA, five at Goffs, one at Ivanpah)had abnormal respiratory sounds, includ-ing wheezing, rasping, and inspiratoryclicks. Weakness, lethargy, or dehydration

was noted in 54% (Goffs), 39% (Ivanpah),and 32% (DTNA) of tortoises with clinicalsigns of URTD. Twenty-two percent of

tortoises had clinical signs of disease onone occasion, 30% had signs two or threetimes, and 19% showed clinical signs fouror more times. Clinical signs of URTD

were seen more often in 199091 andagain in 199495, compared with other

years (274.05, P0.0001). Both of theseyears were normal to wet years that fol-lowed a year of below-normal rainfall.Clinical signs were significantly more likely

to be observed at the time of emergencefrom hibernation in late winter (215.19,P0.0017) and were least likely to be seenin summer.

Eight tortoises at Goffs (23% of theGoffs population) and two tortoises atIvanpah (6% of the Ivanpah population)had oral lesions in 199293 (Table 3). Thelesions were ulcerated or covered with

whitish-yellow plaques and were observedon the top and bottom of the oral cavity,as well as on or along the side of thetongue. Two Goffs tortoises with oral le-sions had concurrent signs of URTD; oth-er tortoises with oral lesions had signs ofURTD, often severe, at other sampling pe-riods. The oral cavity of two tortoisessmelled foul. Tortoises with oral lesions

were significantly more likely to have pos-itive nasal cultures for M. agassizii(216.73, P0.0001); i.e., 70% of tortois-es with oral lesions were culture-positiveat some time during the study. Four tor-toises had multiple positive cultures and30% were seropositive for M. agassizii.Five tortoises (50%) with oral lesions hadincreased plasma enzyme activity (AST,ALT, and/or ALP), and eight of 10 weredehydrated.

Ticks (one to five per tortoise) were ob-served on seven Goffs tortoises and eightIvanpah tortoises in 199394 and spring1995. Ticks were located on the carapacenear or under the transmitter (n8), andon the plastron (n3), neck (n5), tail(n2), eyelids (n1), foreleg (n1) or un-specified sites (n3). All ticks were Orni-

thodorus sp. Ticks were significantly morelikely to be observed on tortoises that hadoral lesions in the previous year (25.32,

P0.021). Draining chin glands were not-ed in six male and two female tortoises insummer (n3), fall (n2), and winter(n3). Three tortoises with draining chinglands had concurrent clinical signs ofURTD.

Shell lesions

Shell lesions were consistent with thosepreviously described for cutaneous dysker-


9/22


TABLE3

.

Clin

icalsignsan

dlaboratoryabnormal

ities

in10tortoisesw

ithora

llesions.

Tortoise

no.

Season

Serology

Culture

Concu

rrent

abnorm

alities

Otherclinicalandlaboratoryabnormalities

inthe5-yearperiod

Outcome

G14M

Winter

93

a

Ocu

larmucu

s

Severeupperand

lowerresp

iratorysigns,ticks,

increase

dASTb

andALTb

,hyperg

lycem

ia,

azotem

ia,

heteropen

ia,hy

pera

lbu-

minem

ia

Alive

G22F

Winter

93

Sun

ken

,dull

eyes,ocu

lar

mucus,we

ak

Severeupperresp

iratorysigns,

increase

dALPb

,hyper

biliru

bi-

nem

ia,

hyperph

osp

hatem

ia,

hyper

kalem

ia

Missing

G23F

Summer

93

Fou

l-sme

llingora

lcavity

,

hyperkalem

ia

Severeupperand

lowerresp

iratorysigns,

leu

kocytosis

,hetero-

philia

,azotemia

,meta

bo

licac

idosis

Dead

G24M

Winter

92

None

Mildupperrespiratorysigns,

hyperb

iliru

binem

ia,

polychroma-

sia,

basop

hilia,

lymp

hopen

ia

Missing

G25M

Winter

93

c

None

Severeupperresp

iratorysigns,po

lychromas

ia,

increase

dAST

,

azotem

ia,

hypoosmo

lality

Missing

G26F

Spring

93

a

Increase

dAS

T

Upperresp

iratory

signs,azotem

ia,

meta

bo

licac

idosis,

hyperka-

lem

ia,

increase

dALP,hypoch

lorem

ia

Alive

G29F

Winter

93

a

Hypouricem

ia

Upperresp

iratory

signs,ticks,po

lychromas

ia,h

yperb

ilir

ubinem

ia

Alive

G31F

Summer

93

Hyperg

lycem

ia

Mildupperrespiratorysigns,ticks,

increase

dAST

Dead

I5M

Winter

93

Fou

l-sme

llingora

lcavity

Upperresp

iratory

signs,ticks,anem

ia,

heteropen

ia,lym

phope-

nia

,increasedA

LT

,hyperferrem

ia,

hypomagnesemia

,hypo-

phosp

hatem

ia

Alive

I11M

Winter

93

None

Genera

lan

duppe

rresp

iratorysigns,ticks,

leu

kocytosis,

hetero-

philia

Alive

aTortoisehadtwopositiveculture

s.

bAST

aspartateaminotransferase;

ALT

alanineaminotransferase;

ALP

alkalinephosphatase.

cTortoisehadthreepositivecultures.


10/22


FIGURE 3. Changes over the 5 yr study period inthe number of tortoises with (A) moderate to severeshell lesions and (B) shell lesions affecting 10% ofthe carapace or plastron.

atosis (Jacobson et al., 1994; Homer et al.,1998); however, histopathologic confirma-tion was not done. Distinct white discol-ored areas also were observed on the shellsof many tortoises at all sites, especiallyduring winter and in wet years.

Shell lesions were observed in 88 tortoises(85.4%) and were significantly more severe

with increasing tortoise age (221.40,P0.0003). Severity of carapace lesions wasnone (n14), mild (n54), moderate(n21), and severe (n1), with significantlyfewer DTNA tortoises having carapace le-sions, compared with other sites (240.24,P0.0001). Distribution of carapace lesionswas 0% (n14), 10% (n37), 1140%(n26), and 30% (n12). Severity of plas-tron lesions was none (n10), mild (n46),

moderate (n35), or severe (n8), with sig-nificantly more Goffs tortoises having mod-erate to severe plastron disease (249.12,P0.0001). Distribution of plastron lesionswas 0% (n10), 10% (n44), 1140%(n23), or 40% (n22). Twenty-seven(77%) Goffs tortoises had moderate to se-

vere plastron lesions, which in 18 tortoisesaffected 40% of the plastron, significantlymore than at other sites (severity249.12;distribution 254.18; P0.0001). Sixty-four

percent of carapace lesions and 83% of plas-tron lesions were active; remaining lesions

were chronic or showed evidence of recov-ery. Overall, for carapace and plastron com-bined, 28 of 35 (79%) tortoises at Goffs hadmoderate or severe shell lesions, significant-ly more (236.06, P0.0001) than at Ivan-pah (18 of 33, 56%) and DTNA (two of 40,6%). Hyperglobulinemia (27.63,P0.0221), positive M. agassizii cultures(25.98, P0.0001), and oral lesions

(2

4.30, P0.0380), all seen primarily intortoises at Goffs, were significantly associ-ated with shell disease.

Annual (spring) shell data were com-pared on a subset of 19 tortoises (five fromDTNA, nine from Goffs, five from Ivan-pah), which were observed over the entire5 yr period. There was a significant in-crease in the number of tortoises withmoderate to severe plastron disease over

time (222.88, P0.0288, Fig. 3). In ad-dition, the number of tortoises with activecarapace lesions increased from six to 13(3274%).

Traumatic lesions were observed on 45(41.7%) tortoises. Old shell or gular trau-ma or scars were observed on 25 tortoises.Shell chips, cracks, or scratches were ob-served on 18 tortoises. Eight of 18 shells

with bite marks were seen in DTNA tor-toises. Other traumatic lesions includedmissing scales (n9), soft tissue abrasions(n8, five at Ivanpah), missing or dam-aged nails or feet (n6), and abnormalgrowths (n3). Traumatic lesions in gen-

eral were more likely to be seen in tor-toises with anemia (26.66, P0.0098).

Serology and microbiology

One hundred and seven (of 834) plasmasamples from 27 tortoises (25% of all tor-toises) were positive for antibodies to M.

agassizii. Eight tortoises were not tested.Tortoises at DTNA were significantly morelikely to be seropositive than tortoises atGoffs (29.29, P0.0096, Fig. 4). Sero-

positive DTNA tortoises also were morelikely to have 2 positive tests (median3.5, range 116), compared with tortoisesat Goffs (median 1.5, range 12) and Ivan-pah (median 1.0, range 15). About thesame percent of seropositive (78%) and se-ronegative tortoises (75%) had clinicalsigns of URTD. Three of four seropositiveGoffs tortoises had concurrent clinicalsigns of URTD compared with 68.8% of


11/22


FIGURE 4. Number and percent of tortoises ateach site that tested positive for M. agassizii by se-rology or nasal culture on one or more occasions be-tween 1990 and 1995.

seropositive DTNA tortoises and no sero-positive tortoises at Ivanpah.

Thirty-six (of 844) nasal cultures from22 tortoises (20% of all tortoises) werepositive for M. agassizii. No tortoises atDTNA were culture-positive for M. agas-

sizii compared with 39.4% of Goffs tor-toises and 28.1% of Ivanpah tortoises (Fig.4). Eight Goffs tortoises and one Ivanpahtortoise had 2 positive cultures (rangetwo to six). Male tortoises were signifi-

cantly more likely to be culture-positivethan female tortoises at Goffs (25.27,P0.0217). Forty-seven percent of Goffstortoises with positive mycoplasma cul-tures were heteropenic (500 cells/l) in199091. Fifteen of 22 tortoises (68%)

with positive nasal cultures for M. agassiziiwere seronegative for M. agassizii. Thenumber of positive cultures more thandoubled between 199091 (n5, 2.7%)and 199394 (n14, 39%).

One hundred fifty-five (of 798) nasalcultures from 67 tortoises (62% of all tor-toises) had moderate to heavy growth of P.

testudinis, which was significantly moreprevalent (27.34, P0.0255) in Ivanpahtortoises (23.7%), than in Goffs (19.4%) orDTNA (14.5%) tortoises. Tortoises atIvanpah with moderate to heavy P. testu-

dinis growth had significantly (t2.03,P0.0430) higher heterophil counts

(4,0774,837/l versus 3,1672,634/l)than tortoises from other sites. At all sites,P. testudinis isolation was significantlymore likely (20.18, P0.0001) to occurin spring (37%) and winter (21.7%) and

was significantly less likely in 199394 (adry year) compared with other years(237.91, P0.0001). Tortoises withmoderate to heavyP. testudinis growth hadsignificantly higher globulin concentra-tions (2.450.66 g/dl; t3.67, P0.0003)compared with tortoises with little or no P.

testudinis growth (2.230.65 g/dl). Mod-erate to heavy growth of mixed bacterialflora was found in 389 (of 633) culturesfrom 94 tortoises, with 59% of heavygrowth in winter and spring. Mixed floragrowth was less in 199394 compared with

other years (2121.48, P0.0001).

Hematology and clinical chemistry

Hematologic abnormalities were ob-served in 53 (49%) tortoises at least onceduring the study period (Table 4). A highernumber of DTNA tortoises had lympho-cytosis in spring 1995 (213.44,P0.0021) compared both with spring ofother years and with tortoises at othersites. Significantly more 27.41,

P0.0246) tortoises at the DTNA (n10)and Ivanpah (n14) had leukocytosis orheterophilia compared with tortoises atGoffs (n4). Tortoises I13F and I22M hadheterophilia (range, 8,38013,391 cells/l)on two occasions, 1214 mo apart. TortoiseI11M had marked heterophilia (34,960/l). Tortoise D33M had heterophilia onthree occasions, concurrent with lympho-cytosis or monocytosis. Basophilia was lessfrequently observed (n5) and was more

likely to occur in spring. Seven Goffs tor-toises had heteropenia. Of tortoises withheteropenia, lymphopenia, or basopenia,19 of 21 were observed in 199091, pri-marily in winter and spring. Tortoises withhematologic abnormalities were signifi-cantly more likely to show clinical signs ofURTD (24.54, P0.033) compared

with tortoises without hematologic abnor-malities.


12/22


TABLE4

.

Hemato

log

icabnormal

ities

infree-r

ang

ing

deserttortoises

between

1990an

d1995

.

Analyte

Cut-offvalues

Rangeofabnormalvalues

No.oftortoises

DTNA

Goffs

Ivanpah

Total

Comments

PCVa(%)or

hemog

lobin

(Hb

;g

/dl)

RBCa(106/l)

MCVa(fl)

PCV18

21(male

)

PCV15

.0(female

)

Hb4

.01

.0

197

.0

PCV16

19%(male

)

PCV8.5%(female

)

Hb1

.53

.6

1.41

.9

97

.11

83

.2

7b

1 2

1 3 3

2b

0 0

10b

4 5

Fourtorto

ises

had

both

low

PCVan

dlow

Hb

AllGo

ffs

were

inspring

93

Allinfall

91an

d95

MCHa(pg

)

MCHCa(%)

WBCa(/

l)

Heterop

hils

(/

l)

40

.0

18

.0

12

,000

8

,000

500

28

.83

9.0

14

.61

7.9

12

,0004

0,0

00

8,0

002

0,0

00

734

20

1 210b

7b

0

6 3 4c

2c

7c,

d

0 014b

10b

3

7 528b

19b

10

Most

inspring

93

Allinspr

ing

Allbuton

ein1990

91

Lymp

hocytes

(/

l)

Basop

hils

(/

l)

5

,000

20(30summer)

5

,000

20(30summer)

5,0

781

6,0

00

02

6

5,2

831

7,7

60

112

6

9c

4 1 2

0 2 2 1

1 7b

2 1

10

13 5 4

AllDTNA

were

inspring

95

Most

in1

990

91

Usual

lyin

spring

Allin199

0

91

Azurop

hils

(/

l)

Eosinop

hils

(/

l)

Monocytes

(/

l)

Tota

l

1

,500

1

,000

800

1,5

702

,505

1,7

39

8473

,149

4 0 252b

2b

0 238b

0 1 142b

6b

1 5115b

Allinwin

teror

fall

aPCV

packedcellvolume;

RBC

redbloodcells;MCV

meancellv

olume;

MCH

meancellhemoglobin;M

CHC

meancellhemoglobinconcentration;

WBC

whiteblood

cells.

bSometortoiseshadmultipleabnormalvalues.

cSignificantlydifferentthanother

sites(P0

.03).

dThreetortoisesalsowereleukopenic

.


13/22


Thrombocytes were clumped or consid-ered adequate in all specimens. Parasites ofRBCs were not observed in any tortoise.Slight (1) polychromasia was observed innine tortoises at the DTNA and 11 tortoisesat Goffs in winter or spring. Moderate (2)to marked (3) polychromasia was observedin six DTNA tortoises in winter 1995, with-out anemia. In winter 1993, tortoise D26Fhad marked, microcytic, hypochromic, re-generative anemia (PCV8.5%, Hb1.5mg/dl, MCV113.9 fl, MCH39.0 pg,MCHC15.9 g/dl, 2 polychromasia), withleukocytosis (WBC18,610/l) and hetero-philia (17,307/l). This tortoise remainedanemic in 199394 (Hb3.2 g/dl), had ba-sophilia (5,283 cells/l) in 199495, and wasseropositive for M. agassizii for 4 consecu-

tive years.Biochemical abnormalities were ob-

served in 61 (56%) tortoises at least onceduring the study period (Table 5). Tortois-es at Goffs were significantly more likelyto have increased AST activity (29.23,P0.0099) and electrolyte abnormalities(212.74, P0.0017) compared with tor-toises at other sites. In addition, three offour tortoises with increased bile acids orbilirubin concentrations were from Goffs.

Bile acids concentrations were high inthree Goffs tortoises in summer 1994, co-incident with the death of five other Goffstortoises. Tortoises from DTNA weremore likely to have low magnesium values(26.25, P0.0440) than tortoises at oth-er sites, especially in winter 1995. Hypo-phosphatemia (27.39, P0.0066) andhypomagnesemia (29.97, P0.0028)

were more likely to occur in male tortoisescompared with female tortoises, and nei-

ther of these abnormalities were observedin Goffs tortoises.Six tortoises had marked, persistent azo-

temia (BUN100 mg/dl); four of these(D07M, G32F, I08M, I36M) also were hy-perosmolar and three of these four werefound dead by the next sampling period(Fig. 5). Tortoises D7M and I8M had anecropsy diagnosis of urologic disease, aspreviously described. The BUN values in

two tortoises (I36M, G32F) with severeazotemia decreased to within the refer-ence interval after three to six samplingperiods, respectively; tortoise G32F wasfound dead about 2 yr later. Concurrentlaboratory abnormalities in tortoises withsevere azotemia included hypophosphate-mia (n3), anemia (n3), hyperchloremia(n2), and heterophilia (n2). Three ad-ditional tortoises (I2M, D36M, D40M)had moderate but persistent azotemia(range 6376 mg/dl) for two to three con-secutive seasons beginning in summer orfall 1994 (a dry period) and persistingthrough hibernation in 1995 (a wet win-ter). Eleven additional tortoises at Goffsand one at DTNA had mild transient in-creases in BUN (range 1777 mg/dl).

Dehydrated tortoises, i.e., those with clin-ical and biochemical (azotemia or hyperbil-irubinemia) evidence of dehydration wereconsidered as a group. Thirty-six tortoises

were dehydrated on at least one occasion,including 11 at DTNA, 16 at Goffs, and nineat Ivanpah. Compared with hydrated tor-toises, dehydrated tortoises had significantlyhigher BUN (57.69.0 vs. 12.20.5 mg/dl),uric acid (5.60.3 vs. 4.70.1 mg/dl), totalbilirubin (0.160.02 vs. 0.120.01 mg/dl),

osmolality (312.46.0 vs. 284.80.8 mOsm/kg), sodium (150.61.9 vs. 143.20.4mmol/l), and chloride (118.81.6 vs.112.40.4 mmol/l) concentrations. Dehy-dration more often affected male tortoises(210.35, P0.0001) and primarily oc-curred in 199091 (DTNA, Goffs) and199495 (DTNA, Ivanpah). Dehydrated tor-toises had a greater prevalence of oral le-sions (211.40, P0.0007) and positive na-sal cultures for M. agassizii (212.30,

P0.0005), but were not more likely to beseropositive (20.008, P0.9268).

Sensitivity, specificity, and predictive values

Significant associations between labora-tory and clinical variables were tested fortheir ability to detect or predict outcome(death, mycoplasmosis, oral lesions, mod-erate to severe shell disease). Marked orpersistent azotemia had low sensitivity but


14/22


TABLE5

.

Clin

ical

biochemicalabnormal

ities

infree-r

ang

ing

deserttorto

ises

between

1990an

d1995.

Analyte

Cut-offvalues

Rangeof

abnormalvalues

No.

oftortoises

DTNA

Goffs

Ivanpah

Total

Comments

Enzymes

Alkal

inep

hosp

hatase

(U/l)

Alan

ineam

inotransferase

(U/l)

Aspartateam

inotransferase(U

/l)

125

10

150

13

5

852

1

4

87

17

0

612

1 1a

1

2 2a

7a,

b

0 3 1

3 6a

9a

Allinsea

sonsotherthanspring

AllGoffs

were

inspring

Organ

ics

Bileac

ids

(g

/ml)

Biliru

bin

,tota

l(mg

/dl)

Bloo

durean

itrogen

(mg

/dl)

Bloo

durean

itrogen

(mg

/dl)c

600

.4

100

100

63.0

186

.6

0.5

0

.6

1

7

77

10

8

373

1 1 3a

3a

3 311 1

a

0 0 1a

2a

4 415a

6a

Go

ffswere

insummer

1994

Go

ffswere

insummer

1992

See

Figure

5

Cho

lestero

l(mg

/dl)

Globu

lins

(g/dl)d

Glucose

(mg

/dl)

Tota

lprote

in(g/dl)

Trig

lyceri

des

(mg

/dl)

Seasonal

lydepen

dent

4

.0

2005

.4

164

31

2

543

4.0

4

.7

20

2

293

5.6

6

.3

25

5

554

0 0 0 0 5

2 2a

2 2a

0

1 2a

5 2a

1

3 4a

7 4a

6

Females

on

ly

Ivanpah

inw

inter

93

Maleson

lyinspring

95

Uricac

id(mg/

dl)

Minerals

Iron

(g

/dl)

Magnesium

(mg

/dl)

Phosp

horus

(mg

/dl)

1

.3

150

3

.0

0

.5to

0.8

mg

/dl

0

1.0

18

9

292

2.7

3

.0

0.2

0

.7

2 1 6a

4

2 2 0b

0b

0 2 2a

7a

4 5 8a

11a

Most

inw

inter

95

Maleson

ly

Electro

lytes

Osmo

lality

(mg

/kg

)e

Osmo

lality

(mg

/kg

)

An

iongap

(mmo

l/l)

Chlori

de

(mmo

l/l)

400

235

29

141

40

0

481

16

8

234

3

2

49

14

4

150

1 0 0 0

1 2 4 2a

2a

1 0 3a

4a

3 4 5a

Allinsummer

91

Potass

ium

(mmo

l/l)

Tota

lCO

2(mmo

l/l)

Otherf

Tota

l

7

.5

14

7.6

8

.9

3

11

0 0 3a

33a

4a

3 6a

63a

3a

1a

3a

42a

7a

4a

12a

138a

Summer

or

fall

aSometortoiseshadmultipleabnormalvalues.

bSignificantlydifferentfromother

sites(P0

.05).

cTwotortoiseswithincreasedbloodureanitrogenalsohadmildhypercreatinemia(0

.5mg/dl)

.

dThesetortoisesalsohadincrease

dtotalproteinconcentration.

eThesetortoisesalsohadazotemia.

fIndividualtortoiseshadmildhyp

o-

(1)orhyperalbuminemia(1);mildhypoglobulinemia(1);mildhypoglycemia(2);

mildhypochloremia(2);mildhyponatrem

ia(2);mildhyperuricaemia

(2);andmildhypocalcemia(1).


15/22


FIGURE 5. Plasma urea nitrogen concentration insix tortoises with severe azotemia (BUN100 mg/dl)and three tortoises (D36M, D40M, I02M) with mod-erate but persistent azotemia (6376 mg/dl). Opensymbols indicate tortoises that were alive at the endof the study; closed symbols indicate tortoises that

were found dead; partially open symbols indicate tor-toises missing for the remainder of the study.

high specificity for mortality (Table 6).

Leukocytosis and heterophilia had lowsensitivity but higher specificity for sero-positivity. A cut-off value of 10,000 WBC/l resulted in lower specificity (60%), butincreased sensitivity to 73%; using a cut-off value of 15,000 cells/l, sensitivity waslower (30%), but specificity increased to94%. High AST had low sensitivity buthigh specificity for oral lesions; when tor-toises with high AST, ALT, or ALP activity

were considered together, sensitivity for

oral lesions increased to 60%, with only aslight decrease in specificity (92%); there

was no effect on predictive values. Clinicaldehydration and weakness had relativelyhigh sensitivity (80%) for identifying tor-toises with oral lesions. Positive nasal cul-tures for M. agassizii had relatively highpositive predictive values for tortoises withmoderate to severe shell disease.

DISCUSSION

Desert tortoise populations in the Mo-jave Desert have declined precipitouslyover the last three decades from the cu-mulative impact of habitat loss and humanand disease-related mortality. Diagnosis ofdisease in live, free-ranging tortoises byevaluation of clinical signs and laboratorytest results is complicated by seasonal andenvironmental effects, particularly becausedesert tortoises are capable of profound

physiologic adaptations (Nagy and Medica,1986; Henen et al., 1998). Recent publi-cation of site, season, and sex-specific ref-erence intervals for desert tortoises in theMojave Desert (Christopher et al., 1999)has made possible evaluation of laboratorydata from tortoises with clinical signs ofdisease. In this study, we evaluated thesame three populations of tortoises from

which reference values were derived; how-ever, we focused on laboratory data fromthose tortoises showing clinical and labo-ratory abnormalities and compared diseasepatterns by site, season, and year. Becauseof the low number of tortoises in eachgroup, repeated measures on some tor-toises, marked seasonal variation in theconcentrations of some analytes, and the

tendency for tortoises to have intermittentbut multiple abnormalities over time, com-parisons between affected and unaffectedanimals were helpful in identifying asso-ciations among clinical and laboratory ab-normalities during the 5 yr study period.

While important and useful for identifyingdisease cues in a population, these com-parisons did not necessarily indicate directassociations between laboratory data andspecific disorders. In addition, because of

the large number of comparisons made,the probability for type I errors was in-creased, such that statistical comparisonsshould be interpreted cautiously.

Three major disease abnormalities wereobserved in tortoises in this study: URTD,cutaneous dyskeratosis, and urinary tractdisease. These same disorders predomi-nated in a pathologic study of 24 dead ordying tortoises from the Mojave and Col-orado deserts (Homer et al., 1998). To our

knowledge, the oral lesions seen primarilyin Goffs tortoises have not previously beendescribed in free-ranging tortoises in theMojave, Colorado, or Sonoran deserts.The manifestation, prevalence, and appar-ent impact of these disease conditions dif-fered substantially among sites. Compared

with other sites, mortality, especially of fe-male tortoises, was twice as high at Goffs,

where there was a combined impact of se-


16/22


TABLE6

.

Sensitivity

,spec

ificity,an

dpre

dictiveva

lueo

fse

lectedc

lin

icalan

dlaboratoryabnormal

ities(

allsitescom

bine

d).

Diseaseoutcome

Laboratoryorclinicalabnormality

Pvalue

2

Sensitivity

(%)

Specificity

(%)

PPVa

(%)

NPVb

(%)

Deat

h

Mycop

lasmosis

Seropositivity

Mar

ke

dazotem

ia(BUN

100mg

/dl)

Pers

istentazotem

ia(twoc

onsecutivecaptures)

Leu

kocytosis

Heterop

hiliac

0.0

454

0.0

268

0.0

187

0.0

088

4.0

0

4.9

0

5.5

3

6.8

7

17

22

44

19

97

95

79

97

60

57

46

71

79

80

78

75

Positivenasalcu

lture

d

Ora

lu

lcers

d

Clin

ical

de

hydrationorweakness

Hyperg

lobu

linem

ia

Clin

ical

de

hydrationorweakness

PositivenasalM.agassizii

Highaspartateam

inotransferase

0.0

018

0.0

057

0.0

005

0

.0001

0.0

001

9.7

6

7.6

4

12

.03

16

.73

14

.47

59

14

80

70

40

76

99

73

85

95

38

75

24

32

44

88

82

97

97

94

Mo

deratetosevereshe

lllesio

nse

Electro

lyteabnormal

ities

Hyperb

iliru

binem

ia

PositivenasalM.agassizii

0.0

104

0

.0001

0

.0001

6.3

7

21

.37

15

.98

50

30

40

84

99

94

24

75

86

94

93

62

aPositivepredictivevalue,i.e.,

the

likelihoodthatatortoisewiththatabnormalitywillhavetheoutcomeindicated

.

bNegativepredictivevalue,i.e.,th

elikelihoodthatatortoisewithoutthatab

normalitywillnothavetheoutcomeindic

ated.

cResultsprimarilyreflectIvanpah

tortoises.

dResultsprimarilyreflectGoffsto

rtoises.

eResultsprimarilyreflectGoffsan

dIvanpahtortoises.


17/22


vere shell disease, culture-positive URTD,and oral lesions.

Clinical signs of URTD were similar tothose described previously for mycoplas-mosis (Jacobson et al., 1991; Lederle et al.,1997; Brown et al., 1999). Draining chinglands were not considered a sign ofURTD in this study, but may have beenabnormal when observed in female tor-toises or in winter. Upper respiratory signscan be seen in respiratory diseases otherthan mycoplasmosis (such as the acutepneumonia in tortoise G30F), however,the persistence of clinical signs over yearsin association with seropositivity or posi-tive nasal cultures strongly suggested thatM. agassizii was the most likely cause ofmost if not all of the respiratory disease in

these tortoises.Upper respiratory tract disease in the

DTNA population has evolved from anacute, epidemic disease with high mortal-ity to a chronic disease with variable mor-bidity, low mortality, and a high serocon-

version rate for antibodies to M. agassizii(Brown et al, 1999). Nearly one-half ofDTNA tortoises had serologic evidence ofexposure to M. agassizii, a higher preva-lence than either of the other two sites and

than other desert tortoise populations(Lederle et al., 1997). Tortoises with M.

agassizii infection may remain seropositivewithout overt disease (Brown et al., 1994);however, two-thirds of DTNA tortoiseshad intermittent clinical signs of URTD.Tortoises at DTNA never had positive cul-ture results for M. agassizii, possibly be-cause M. agassizii may be shed from thenasal epithelium (and thus be less likely tobe cultured) by the time antibodies devel-

op and are detected serologically. Theyouth of the DTNA population was con-sistent with loss of up to 90% of adult tor-toises due to URTD (between 1988 and1996), and predation and human interven-tion (between 19791988) (Peterson,1994; Berry, 1997; Brown et al, 1999).Predator attacks continue to be an impor-tant cause of morbidity in DTNA tortoises,based on the prevalence of bite wounds.

Increased frequency of clinical signs in199495 may indicate recrudescence ofURTD in the DTNA population and im-pending spread among younger, more sus-ceptible tortoises.

Lymphocytosis, heterophilia, and azur-ophilia in DTNA tortoises were consistent

with immunologic responses to infection.Lymphocytosis occurred concurrent withresurgence in clinical signs of URTD, per-haps exacerbated by drought the previous

year. The nasal passages of tortoises withmycoplasmosis contain subepitheloid lym-phoid aggregates and heterophilic infil-trates (Jacobson et al., 1991). Mild anemiain a few DTNA tortoises may have beenassociated with chronic infection. Interest-ingly, Goffs tortoises rarely had hemato-

logic evidence of inflammation (e.g., het-erophilia), compared with Ivanpah andDTNA tortoises.

Poor correlation between M. agassiziiculture and serology results suggested thatGoffs tortoises may be infected with a dif-ferent strain or isolate of mycoplasma thantortoises at other sites, one perhaps lesslikely to be detected by the ELISA test.Recent molecular studies of seven Myco-

plasma isolates from clinically ill desert

tortoises revealed differences in 16S ribo-somal RNA nucleotide sequences, al-though serologic crossreactivity occurred(Brown et al., 2001). Discrepancies be-tween serologic and culture results alsocould indicate genetic or environmentaltraits that alter tortoise response to dis-ease. Re-evaluating culture and serologicresults on surviving Goffs tortoises wouldbe useful for determining whether nasalgrowth ofM. agassizii remains high in this

population and/or whether more Goffs tor-toises are now seroconverting. The higherprevalence of positive cultures in male tor-toises is consistent with the increasedbeak-to-beak contact between fightingmales.

The oral lesions in nearly one-fourth ofGoffs tortoises and two Ivanpah tortoises

were a new disease finding of substantialconcern. Oral lesions were significantly as-


18/22


sociated with positive M. agassizii nasalcultures, and as such may be a uniquemanifestation or complication of myco-plasmosis. Viral infection is another possi-ble cause, because similar diphtheriticplaques have been described in gophertortoises and Mediterranean tortoises withpathogenic herpesvirus infections (Jacob-son, 2000; Origgi et al., 2001). The clus-tering of oral lesions and positive Myco-

plasma cultures in a single dry year, andconcurrent dehydration in tortoises withoral lesions suggest that environmentalstress also may have played a role. Othercauses of oral lesions, such as infectionsecondary to cactus spine trauma (Homeret al., 1998) could not be excluded.

Cutaneous dyskeratosis, most severe in

Goffs tortoises, is characterized by greyish-white, roughened, and flaky areas thatmost severely affect the plastron. It hasbeen associated with high mortality of de-sert tortoises on the Chuckwalla BenchArea of California (Jacobson et al., 1994;Berry, 1997). Necropsy results in a subse-quent study showed a number of tortoisesfrom different locations in the Mojave andColorado deserts with cutaneous dyskera-tosis as well as shell necrosis (Homer et

al., 1998). Jacobson et al. (1994) suggestednutrient deficiency or heavy metal toxico-sis as a cause of cutaneous dyskeratosis.The fact that larger, older tortoises in thisstudy had more severe shell lesions sup-ports a chronic, cumulative problem. The

white discolored areas seen more often inwet seasons and years were suggestive offungal colonization, which has been re-ported previously as causing opportunisticinfection secondary to cutaneous dyskera-

tosis (Homer et al., 1998). Concurrentpositive M. agassizii cultures, oral lesions,and hyperglobulinemia in Goffs tortoisessuggest chronic infection or antigenicstimulation could be a factor in the path-ogenesis of cutaneous dyskeratosis. Differ-ences in the soluble plastron scute pro-teins of desert tortoises with mycoplas-mosis, urolithiasis, and inflammation sug-gest that systemic disease may affect

keratinization (Homer et al., 2001). Of 30tortoises (from 150) remaining on an ad-

jacent study site in Goffs in 2000, mosthave moderate to severe shell disease(Berry, 2000).

Tortoises at Goffs were more likely thanother tortoises to have laboratory evidenceof hepatic disease, including increased bileacids concentration, which are specific forhepatic insufficiency (Divers, 2000a). Twotortoises recently necropsied at Goffs hadliver atrophy, a decrease in the size andnumber of hepatocytes, and in overall he-patic mass (Homer and Berry, pers.comm.), which could result in increasedbile acids concentration. In addition, bileacids concentration was very high (670 g/ml) in a severely dehydrated tortoise (Ber-ry et al., 2002). Hepatic lipidosis, degen-eration, and hemosiderosis are commonlyobserved in tortoises and other reptilessecondary to anorexia, and may have con-tributed to increased AST, ALT, and ALPactivities in tortoises at Goffs. Hyperglob-ulinemia also is a feature of chronic he-patic disease.

Drought is a natural part of the deserttortoises environment, but when com-bined with disease or habitat loss, maycontribute to additional disease problemsand mortality (Peterson 1996). Clinicalsigns of URTD, lower leukocyte counts,positive nasal cultures of M. agassizii, andmild to moderate azotemia were morecommonly seen in 199394, a year of be-low-normal annual and winter precipita-tion. Heteropenia previously has been as-sociated with drought and starvation (Ber-ry et al., 2002). Tortoises entering hiber-nation in a drought year may bephysiologically compromised, becauseclinical signs of URTD and heteropenia

were noted at the time of emergence fromhibernation in 199091 and 199495,

years following a period of drought. Mostdehydrated tortoises and most deaths atthe DTNA and Ivanpah also occurred in199091 and 199495, years following dry

winters. Dehydration in tortoises at Goffs


19/22


may have increased their susceptibilities tothe oral lesions.

Azotemia was the most common labo-ratory abnormality observed. It is likelythat the mild transient azotemia in severaltortoises at Goffs was evidence of physio-logic extreme rather than disease, becauseit occurred in multiple tortoises during dry

years or seasons. Healthy Goffs tortoisesshowed marked physiologic alterations inrelation to periodic drought and summerprecipitation, including sharp changes inelectrolyte, osmolality, BUN, and aniongap values (Christopher et al., 1999).Drought causes urine retention to con-serve water and urea and can cause BUN

values of up to 60 mg/dl in healthy deserttortoises without clinical dehydration

(Christopher et al., 1999).Tortoises with severe azotemia (substan-

tially higher than other tortoises at thesite) or persistent azotemia (even throughrelatively wet seasons) more likely had de-creased renal function due to severe de-hydration (prerenal azotemia, perhaps dueto inability to access water due to diseaseor other barriers), urolithiasis (postrenalazotemia), or renal disease (renal azote-mia). These underlying mechanism of azo-

temia are closely related, especially in tor-toises, and may be difficult to differentiate.A desert tortoise trapped for 11 mo in itsburrow had a BUN of 259 mg/dl and anosmolality of 395 mOsm/kg (Christopher,1999). Severely dehydrated tortoises hadBUN values (range 88.7640 mg/dl), sim-ilar to those seen here (Berry et al., 2002).Clinically dehydrated tortoises in this andother studies often have other laboratoryabnormalities, including anemia (rather

than polycythemia), heteropenia, hyperbil-irubinemia, hyperosmolality, hypernatre-mia, hyperchloremia, hypocalcemia, andhypophosphatemia (Jacobson et al., 1991;OConnor et al., 1994; Christopher, 1999;Berry et al., 2002).

Urolithiasis causes azotemia due to theinability of the kidney to sufficiently ex-crete urea, and is more likely to developin dehydrated tortoises (Berry et al., 2002).

The two tortoises with urolithiasis in thisstudy had BUN values ranging from 71306 in the season preceding death. Inter-estingly, both of these tortoises had mul-tiple incidences of mild but worsening azo-temia (25140 mg/dl) in earlier seasons,likely indicating early development of uro-liths. Uroliths may develop from pro-longed retention of urine and excessivecrystal formation, primarily in dehydratedtortoises, which have thick, sludged urine,as seen in tortoise D40M. Uroliths maycontribute to renal failure by obstructingurination, but do not necessarily causedeath. Several tortoises with moderate tomarked azotemia in this study also had in-flammation, including heterophilia, azuro-philia, or lymphocytosis, which has beenobserved in tortoises with bacterial or pro-tozoal infections of the kidneys (Jackson1991; Homer et al., 1998; Divers, 2000b).

High BUN concentration was a good in-dicator of decreased renal function, butdid not always predict mortality. In tortoiseI36M, an extremely high BUN valuequickly normalized with the onset of win-ter rains, possibly coinciding with availabledrinking water or with the passage of smalluroliths. However, because four of ninetortoises with severe or persistent azote-mia died, and because three of these fourhad necropsy evidence of urolithiasis orbladder infection, underlying diseaseshould be strongly considered in the dif-ferential diagnosis of tortoises with severeor persistent azotemia.

Hypophosphatemia has been describedpreviously in ill tortoises (Jacobson et al.,1991; Christopher, 1999), and is likely dueto decreased food intake. There was nocorrelation between hypophosphatemiaand disease in this study. Low phosphorus

values were usually observed in winter fol-lowing a drought year, consistent withchanges in healthy tortoises (Christopheret el., 1999). Abnormalities in plasma mag-nesium, triglyceride, and iron concentra-tions were seasonal and also may havebeen diet related, but the low number of


20/22


measurements available limited conclu-sions.

Tortoises at Ivanpah had disease typeand severity intermediate between that ofDTNA and Goffs, including intermediateseverity of cutaneous dyskeratosis. LikeGoffs tortoises, a few Ivanpah tortoiseshad oral lesions and positive nasal myco-plasma cultures. Like DTNA tortoises,Ivanpah tortoises often were seropositiveand had inflammatory leukograms. Inflam-mation (heterophilia and hyperglobulin-emia) and clinical signs of URTD weremore likely to be seen in Ivanpah tortoises

with heavy P. testudinis growth. Althoughthe role of P. testudinis in URTD is notfully understood, it may be synergistic withmycoplasmosis or contribute to secondary

infection in tortoises infected with Myco-plasma (Jacobson et al., 1991). Although P.testudinis can be isolated from the nasalcavity of healthy tortoises, certain isolateshave been recovered solely from tortoises

with URTD (Snipes et al., 1995). UnlikeMycoplasma, Pasteurella growth was lesssevere in the drought year.

In summary, URTD, cutaneous dysker-atosis, and oral lesions comprised the ma-

jor disease conditions observed in three

populations of free-ranging Mojave Deserttortoises. Azotemia was the most commonlaboratory abnormality, and, when severeor persistent, it was often associated withunderlying urolithiasis or bladder infec-tion. With few exceptions, routine hema-tologic and biochemical laboratory testshad poor sensitivity and low predictive val-ues in detecting or monitoring disease.Predictive values are affected by the prev-alence of disease, however, and because of

the low numbers of tortoises with specificdiseases, caution must be exercised in in-terpreting these results. The severe prob-lems associated with positive Mycoplasmacultures, oral lesions, dehydration, andmoderate to severe shell disease in the tor-toises at Goffs were of particular concernbecause of their potential for cumulativeimpact on the population. A population ofGoffs tortoises 2 km from our study site,

considered the gold standard for ahealthy population since 1979, has suf-fered recent 9296% decreases in tortoisedensity between 1994 and 2000 (Berry andMedica 1995; Berry 2000). In light ofthese findings, disease and laboratory ab-normalities identified in the present studymay be important clues to the demise ofthe neighboring Goffs population. Furtherinvestigation into different strains of My-

coplasma organisms, histopathology of orallesions, and research into the pathogenesisof shell disease, hepatic disease, and uro-logic disease are critical for better under-standing these problems.

ACKNOWLEDGMENTS

This work was funded by the Department of

the Interior, Bureau of Land ManagementContract Nos. B950-C1-0060 (Christopher),YA-651-CT0-340079 (Nagy), and CA950-CT0-47 (APL Laboratories) and National BiologicalService Contract No. 14-48-0006-95-003. Fed-eral and state permits were issued to and ad-ministered by K. Berry. The hard work anddedication of the following individuals madethis work possible: C. C. Peterson, I. A. Girard,J. Parinayakosol, B. S. Wilson, I. R. Wallis, A.Ancel, M. A. Wilson, and C. Tran. The authorsalso wish to thank M. Brown, E. R. Jacobson,

V. Dickinson, J. Conway-Klaassen, and J. Klaas-sen for their contributions to this work.

LITERATURE CITED

BERRY, K. H. 1997. Demographic consequences ofdisease in two desert tortoise populations in Cal-ifornia, USA. In Proceedings: Conservation, res-toration, and management of tortoises and tur-tles, An international conference, J. Van Abbema(ed.). Wildlife Conservation Society Turtle Re-covery Program and the New York Turtle andTortoise Society, pp. 9199.

. 2000. Preliminary report on the spring sur-vey of desert tortoises at the Goffs permanentstudy plot and special project on the effects ofroads. US Geological Survey, Western EcologicalResearch Center, 11 pp.

, AND M. M. CHRISTOPHER. 2001. Guidelinesfor the field evaluation of desert tortoise healthand disease. Journal of Wildlife Diseases 37:427450.

, AND P. MEDICA. 1995. Desert tortoises in theMojave and Colorado deserts. In Our living re-sources: A report to the nation on the distribu-tion, abundance, and health of U.S. plants, ani-mals, and ecosystems, E. T. LaRoe, G. S. Farris,


21/22


C. E. Puckett, P. D. Doran, and M. J. Mac (eds.).US Department of the Interior, National Biolog-ical Service, Washington, D. C. pp. 135137.

, AND A. P. WOODMAN. 1984. Preliminary in-vestigations of shell wear in determining adultage groups in desert tortoises. Appendix 4. InThe status of the desert tortoise (Gopherus agas-

sizii) in the United States. K. H. Berry (ed.). Re-

port to US Fish and Wildlife Service from theDesert Tortoise Council on Order No. 11310008381. A4-1 to A4-54.

, E. K. SPANGENBERG, B. L. HOMER, AND E.R. JACOBSON. In press. Deaths of desert tortoisesfollowing periods of drought and research ma-nipulation. Chelonian Conservation and Biology.

BROWN, M. B., I. M. SCHUMACHER, P. A. KLEIN, K.HARRIS, T. CORRELL, AND E . R . JACOBSON.1994. Mycoplasma agassizii causes upper respi-ratory tract disease in the desert tortoise. Infec-tion and Immunity 62: 45804586.

, K. H. BERRY, I . M. SCHUMACHER, K. A.NAGY, M. M. CHRISTOPHER, AND P. A. KLEIN.1999. Seroepidemiology of upper respiratorytract disease in the desert tortoise in the westernMojave desert of California. Journal of WildlifeDiseases 35: 716727.

, D . R . BROWN, P. A. KLEIN, G . S . MC-LAUGHLIN, I. M. SCHUMACHER, E. R. JACOBSON,H. P. ADAMS, AND J. G. TULLY. 2001. Mycoplas-

ma agassizii sp. nov., isolated from the upper re-spiratory tract of the desert tortoises (Gopherus

agassizii) and the gopher tortoise (Gopherus po-lyphemus). International Journal of Systematicand Evolutionary Microbiology 51: 413418.

CHRISTOPHER, M. M. 1999. Physical and biochemical

abnormalities associated with prolonged entrap-ment in a desert tortoise. Journal of Wildlife Dis-eases 35: 361366.

, K. H. BERRY, I. R. WALLIS, K. A. NAGY, B.T. HENEN, AND C. C. PETERSON. 1999. Refer-ence intervals and physiologic alterations in he-matologic and biochemical values of free-rangingdesert tortoises in the Mojave desert. Journal of

Wildlife Diseases 35: 212238.DIVERS, S. J. 2000a. Reptilian liver and gastrointes-

tinal testing. In Laboratory medicineavian andexotic pets, A. Fudge (ed.). W. B. Saunders, Phil-adelphia, pp. 205209.

. 2000b. Reptilian renal and reproductive dis-ease diagnosis. In Laboratory medicineavianand exotic pets, A. Fudge (ed.). W. B. Saunders,Philadelphia, pp. 217222.

HENEN, B. T., C. C. PETERSON, I. R. WALLIS, K. H.BERRY, AND K. A. NAGY. 1998. Effects of cli-matic variation on field metabolism and water re-lations of desert tortoises. Oecologia 117: 365373.

HOMER, B. L., K. H. BERRY, M. B. BROWN, AND E.R. JACOBSON. 1998. Pathology of spontaneouslyoccurring diseases in free-ranging California de-

sert tortoises (Gopherus agassizii). Journal ofWildlife Diseases 34: 508523.

, C. LI, K. H. BERRY, N. D. DENSLOW, E. R.JACOBSON, R. H. SAWYER, AND J. E. WILLIAMS.2001. Soluble scute proteins of healthy and illdesert tortoises (Gopherus agassizii). AmericanJournal of Veterinary Research 62: 104110.

JACKSON, O. F. 1991. Reptiles, Part One. Chelonians.

In Manual of exotic pets, P. H. Beynon and J. E.Cooper (eds.). British Small Animal VeterinaryAssociation, Cheltenham, Gloucestershire, UK,pp. 221243.

JACOBSON, E. R. 2000. Reptilian viral diagnostics. InLaboratory medicineavian and exotic pets, A.Fudge (ed.). W. B. Saunders, Philadelphia, pp.229235.

, T. J. WRONSKI, J . SCHUMACHER, C. REG-GIARDO, AND K. H. BERRY. 1994. Cutaneous dys-keratosis in free-ranging desert tortoises, Go-

pherus agassizii, in the Colorado desert of south-ern California. Journal of Zoo and Wildlife Med-icine 25: 6881.

, J. M. GASKIN, M. B. BROWN, R. K. HARRIS,C. H. GARDINER, J. L. LAPOINTE, H. P. ADAMS,AND C. REGGIARDO. 1991. Chronic upper respi-ratory tract disease of free-ranging desert tor-toises (Xerobates agassizii). Journal of WildlifeDisease 27: 296316.

KLAASSEN, J. 1991. Necropsy report for male tortoiseD7M. Report from APL Veterinary Laboratories,Las Vegas, Nevada, to U.S. Department of theInterior, Bureau of Land Management, River-side, California, 5 pp.

LEDERLE, P. E., K. R. RAUTENSTRAUCH, D. L. RAK-ESTRAW, K. K. ZANDER, AND J. L. BOONE. 1997.

Upper respiratory tract disease and mycoplas-mosis in desert tortoises from Nevada. Journal of

Wildlife Diseases 33: 759765.NAGY, K. A., AND P. A. MEDICA. 1986. Physiological

ecology of desert tortoises in southern Nevada.Herpetologica 42: 7392.

NATIONAL OCEANIC AND ATMOSPHERIC ADMINIS-TRATION. 19891995. Climatological data forCalifornia, Vol. 93-99, National Climatic DataCenter, Asheville, North Carolina.

OCONNOR, M. P., J. S. GRUMBLES, R. H. GEORGE,L. C. ZIMMERMAN, AND J. R. SPOTILA. 1994. Po-tential hematological and biochemical indicatorsof stress in free-ranging desert tortoises and cap-tive tortoises exposed to a hydric stress gradient.Herpetological Monographs 8: 526.

ORIGGI, F., P. KLEIN, C. ROMERO, AND E. JACOBSON.2001. Transmission study with a tortoise herpes-

virus (THV) in Greek tortoises. Proceedings, De-sert Tortoise Council Symposium, Tucson, Ari-zona, pp. 3133.

PETERSON, C. C. 1994. Different rates and causes ofhigh mortality in two populations of the threat-ened desert tortoise Gopherus agassizii. Biolog-ical Conservation 70: 101108.


22/22


. 1996. Anhomeostasis: Seasonal water andsolute relations in two populations of the deserttortoise (Gopherus agassizii) during chronicdrought. Physiological Zoology 69: 13241358.

RASKIN, R. E. 2000. Reptilian complete blood count.In Laboratory medicineavian and exotic pets,A. Fudge (ed.). W. B. Saunders, Philadelphia, pp.193197.

SCHUMACHER, I. M., M. B. BROWN, E. R. JACOBSON,B. R. COLLINS, AND P. A. KLEIN. 1993. Detec-tion of antibodies to a pathogenic mycoplasma indesert tortoises (Gopherus agassizii) with upperrespiratory tract disease. Journal of Clinical Mi-crobiology 31: 14541460.

SNIPES, K. P., R. W. KASTEN, J. M. CALAGOAN, AND

J. T. BOOTHBY. 1995. Molecular characterizationof Pasteurella testudinis isolated from desert tor-toises (Gopherus agassizii) with and without up-per respiratory tract disease. Journal of WildlifeDiseases 31: 2229.

US FISH AND WILDLIFE SERVICE. 1990. Endangeredand threatened wildlife and plants; Determina-tion of threatened status for the Mojave popu-

lation of the desert tortoise. Federal Register44(63):1217812191.US FISH AND WILDLIFE SERVICE. 1994. Desert tor-

toise (Mojave population) recovery plan. U.S.Fish and Wildlife Service, Portland, Oregon, 73pp.

Received for publication 25 October 2001.

clinical disease and laboratory abnormalities in

Documents