SCINTIGRAPHIC AND RADIOGRAPHIC EVALUATION OF APPENDICULAR

SKELETAL LESIONS IN COLD-STUNNED KEMP’S RIDLEY SEA TURTLES

MAURICIO SOLANO, CHARLES INNIS, CYNTHIA SMITH, CONSTANCE MERIGO, ERNEST SCOTT WEBER III

Osteolytic appendicular skeletal lesions in eight-stranded, cold-stunned Kemp’s ridley sea turtles (Lepidochelys

kempii) were evaluated using radiography and skeletal scintigraphy. Radiographic studies were performed

monthly in most animals. Follow-up scintigraphy was performed 45–120 days after the initial exams in six

turtles. Radiographically, lesions slowly progressed from an early osteolytic process contained to either the

proximal or distal end of long bones, to a later stage characterized by thickening of the affected bone, sclerosis,

and remodeling of the lesion borders. In seven turtles, the initial scintigrams were characterized by at least one

focus of abnormal radiopharmaceutical uptake that correlated with a lytic site noted in radiographs. In five

turtles, scintigraphic lesions were characterized by asymmetric radiopharmaceutical uptake rather than by

increased intensity of uptake. Scintigraphic studies obtained more than 4 months after the appearance of clinical

and radiographic signs had minimal, if any, abnormal radiopharmaceutical uptake, despite the persistence of

abnormal radiographic findings. Skeletal scintigraphy is an effective method for more precisely determining if

and when these animals can be returned to the wild. Animals were released if normal radiopharmaceutical

uptake was seen during initial examination, or if decreased uptake was noted between serial examinations. In

four of the turtles, resolution of abnormal scintigraphic findings permitted an objective decision to discontinue

antibiotic and antifungal therapy. Seven of the eight turtles were released after correlation of the clinical signs

with the imaging findings. Radiographs, however, are still needed to facilitate the correct identification of lesions

with scintigraphy. Veterinary Radiology & Ultrasound, Vol. 49, No. 4, 2008, pp 388–394.

Key words: bone lysis, bone scan, cold stunning, Kemp’s ridley sea turtle, osteomyelitis, radiographs,

skeletal scintigraphy, 99mTc-diphosphonate.

Introduction

THE KEMP’S RIDLEY sea turtle (Lepidochelys kempii) is

the smallest and rarest of the seven known species of

sea turtles, with only 2000–3000 adult females likely re-

maining.1 Although adults are generally found along the

coast of the southeastern United States and Gulf of Mex-

ico, juveniles frequent the northeastern coast of the United

States during summer.2,3 Juveniles that do not leave north-

ern waters in autumn are susceptible to severe hypo-

thermia, or cold-stunning, as water temperatures rapidly

drop.4–12 Cold-stunned turtles are frequently found

stranded on the beaches of Massachusetts and New York

from late November to late December when water tem-

perature drops below 121C (541F).4–6,9 Stranded turtles are

collected by a network of volunteers and transported to

regional rehabilitation centers.

Many abnormalities, including osteomyelitis, have been

documented during the rehabilitation of cold-stunned

Kemp’s ridley sea turtles.13–20 These conditions are believed

secondary to dehydration, malnutrition, poor perfusion, and

immunosuppression that occur due to the hypothermia. In

some instances, stranded turtles with osteomyelitis will have

periarticular swelling, abrasions and open wounds, and

lameness. In other instances, these signs develop only after

several weeks to months of rehabilitation.

Lameness in sea turtles is typically characterized by no-

ticeably reduced use of the affected limb, with pain and

reduced range of motion during physical examination. Ra-

diography is a standard modality to assess skeletal abnor-

malities of turtles, and findings are critical in determining

whether a turtle can be released. With osteomyelitis in

turtles, there are often osteolytic lesions of the metaphysis

and epiphysis of long bones of the extremities. These

lesions may persist for many months despite resolution of

clinical signs, making the decision of when to release the

A preliminary report of results from three patients cited in this reportwas presented at the 2000 Conference of the American Association of ZooVeterinarians and the International Association for Aquatic AnimalMedicine.

Address correspondence and reprint requests to Mauricio Solano, at theabove address. E-mail: mauricio.solano@tufts.edu

Dr. Smith’s current address is United States Navy Marine MammalProgram, Space and Naval Warfare Systems Center San Diego, 53560Hull Street, San Diego, CA 92152.

Dr. Weber’s current address is University of California School ofVeterinary Medicine, 2108 Tupper Hall, Davis, CA 95616.

Received November 14, 2007; accepted for publication January 31, 2008.doi: 10.1111/j.1740-8261.2008.00387.x

From the Cummings School of Veterinary Medicine, Tufts University,200 Westboro Road, North Grafton, MA 01536 (Solano) and the NewEngland Aquarium, Central Wharf, Boston, MA 02110 (Innis, Smith,Merigo, Weber).

388

turtle difficult. Because the delay between resolution of

clinical and radiographic signs, nuclear scintigraphy has

been performed in a small number of turtles to provide

additional information about the status of the bone lesions.

The objectives of this work were to (1) define a skeletal

scintigraphic technique for Kemp’s ridley sea turtles; (2)

assess the value of nuclear scintigraphy in the management

of turtles with radiographic evidence of osteomyelitis; and

(3) document the radiographic appearance of skeletal

lesions over time in this species.

Materials and Methods

Between 1994 and July 2006, 565 live cold-stunned sea

turtles were evaluated by the New England Aquarium re-

habilitation program. Affected turtles weighed between 2.1

and 21kg (mean 6.3 kg). All turtles were treated with stan-

dard medical management for cold-stunned sea turtles,

including gradual warming, fluid therapy, nutritional sup-

port, antibiotics, and antifungal therapy.21 All turtles were

radiographed within 5 days of admission, and thereafter at

variable intervals as determined by the clinical status. In

most turtles where osteolytic lesions were noted radio-

graphically, blood, synovial fluid, and/or bone aspirates

were analyzed for aerobic and anaerobic bacteria, and

fungi. Synovial fluid or bone cytology was performed in

four turtles. Bone biopsies were not performed.

Sixteen bone scans were performed between 1999 and

2006 in eight turtles with radiographic evidence of osteoly-

tic lesions. Initial skeletal scintigraphy was performed 1–5

months (mean 2.2 months) after detection of radiographic

lesions and clinical signs. Five turtles had follow-up scinti-

graphy 1.5–5 months (mean 3.0 months) after the initial

study. One turtle had a follow-up scan 50 months after the

initial scan, and two turtles had no follow-up scan.

Scintigraphy was conducted at a room temperature of 21–

241C (70–741F). Sedation was required for the bone phase

of the scintigram in six turtles. Two turtles were sedated with

medetomidine� and ketamine.w Four turtles were sedated

with propofol.z Two turtles did not require sedation.

In five turtles, scintigrams were acquired using a 75

photomultiplier tube, 12-in. field of view planar gamma

camera systemy mounted on a mechanical gantry and in-

terfaced with a Microdot imager for the production of film

images. In three turtles, scintigrams were acquired with a

55 photomultiplier tube, rectangular large field of view

planar gamma camera systemz mounted on a mechanical

gantry interfaced with a dedicated computerk for image

postprocessing. All turtles were in ventral recumbency with

the camera below the patient. 99mTc-HDP (111–370MBq)

was injected intravenously into the left or right dorsal cer-

vical sinus. Acquisition settings for each of the phases of

the bone scan varied between animals. For the vascular

phase, dynamic acquisition studies ranging from 1 s per

frame for a total of 125 frames to 5 s per frame for a total

of 32 frames were taken. Image acquisition was started at

the time of injection. The soft tissue phase acquisition set-

tings also varied between animals. In three sedated turtles,

a dynamic acquisition study of 60 s per frame for eight

frames was acquired while static acquisition images be-

tween 500,000 and 800,000 counts per image were acquired

in five turtles. For the bone phase, static images of 500,000

counts each were acquired in all turtles 3 h after radio-

pharmaceutical injection. Radiation safety practices were

in compliance with the license for authorization for the use

of radionuclide approved by the Nuclear Regulatory Com-

mission and the State of Massachusetts. All animals were

transported and housed according to guidelines of the

Institutional Animal Care and Use Committee of NEAQ,

with permission of the United States Fish and Wildlife

Service and the Massachusetts Division of Fish and Wild-

life (Mass Wildlife).

Results

Visible swelling and lameness of affected joints was first

noted between Day 1 and Day 164 of rehabilitation

(mean¼Day 59). Affected turtles had reduced use and re-

duced range of motion of affected limbs. Palpation of

affected joints elicited a pain response. Two turtles had ra-

diographic evidence of pneumonia and one turtle had a

concurrent fracture of the right orbit. No concurrent health

problems were noted in the reminder of the turtles. Cytol-

ogy of synovial fluid was performed in four turtles with all

exams revealing varying degrees of mononuclear inflamma-

tion. Blood cultures performed in six turtles were negative.

Culture of synovial fluid was performed in three turtles with

only one turtle growing an organism (Actinobacter wolffi).

Bone culture was performed in one turtle that revealed En-

terococcus fecalis, coagulase positive, and Staphylococcus sp.

Radiographic lesions were polyostotic with as low as

two sites per animal to as many as 19 sites per animal. No

mid-diaphyseal lesions were noted in any turtle. Two dis-

tinct types of bone lesions, characterized as early and late

stage, were noted on radiographs made at the time of the

initial scintigram. Early stage lesions were characterized by

a purely osteolytic process restricted to either the proximal

or distal end of the epiphysis, physis, and metaphysis of

long bones. Borders of these early lesions were poorly de-

fined with a long transition zone (Fig. 1). Late stage lesions

�Domitor, Pfizer, Exton, PA.wKetaset, Fort Dodge, IA.zPropofol, butler Healthcare Co., Irvine, CA.yPlanar gamma camera system, Siemens Gammasonics Inc., Des

Plaines, IL.zPlanar gamma camera system, IS2 Medical Systems Inc., Ottawa,

Canada. kSegami Corporation, Columbia, MD.

389SKELETAL SCINTIGRAPHY OF SEATURTLESVol. 49, No. 4

were characterized by the thickening of the affected bone,

sclerosis, and remodeling of the lesion borders (Figs. 2 and

3). With the exception of six lesions, soft tissue swelling

centered at the joint was apparent in all early lesions

(Fig. 3A). Larger lytic foci tended to remain void of new

bone formation. In four turtles, lesions were exclusively of

Fig. 1. Dorsopalmar radiograph left front flipper. Multiple well-definedbone lesions (arrowheads) are noted. Lesions are mostly lytic with no boneremodeling and are affecting the growth plates of long bones. The lyticprocess is commonly seen communicating with the joint space as noted in theelbow joint.

Fig. 2. Dorsopalmar radiograph right front flipper. A late stage bonelesion is noted at the level of the proximal interphalangeal joint of the thirddigit (arrow). There is bone remodeling of the distal end of the first andproximal end of the second phalanges. The affected bones are mildly thick-ened and there is well-defined sclerosis surrounding rounded lesion borders.A smaller, mostly lytic, lesion is detected at the level of the fifth carpal boneand proximal end of the metacarpal bone of the fifth digit (arrow head). Thiscarpal lesion shows no bone sclerosis or remodeling of its borders andtherefore is considered to be at an earlier stage of presentation than the largerlesion in the third digit. The metallic opacities overlying the distal phalanx ofthe first digit are skin artifacts (dirt).

390 SOLANO ET AL. 2008

the early stage, while in four turtles, lesions were a com-

bination of early and late stages. In one turtle, healing

progressed to the formation of a separate bone fragment

that increased in size for the first 3 months after the lesion

appeared. This fragment decreased in size and opacity over

a 6-month period (Fig. 3B). Only one turtle had a well-

defined periosteal reaction, which appeared 3 months into

the rehabilitation period. This reaction subsequently in-

creased in size.

In seven turtles, the initial scintigram had at least one

focus of abnormal radiopharmaceutical uptake, which cor-

responded to a lytic radiographic site (Figs. 4 and 5). In

one turtle, none of the visible radiographic lesions were

abnormal in the bone phase. All radiographically visible

lesions were scintigraphically abnormal in three turtles. In

five turtles, scintigraphic lesions were detected by the

asymmetric appearance of the radiopharmaceutical uptake

rather than by an increase in intensity of the uptake. The

intensity of abnormal uptake varied from mild to severe,

regardless of the size of the radiographic lesion. In two

turtles, the soft tissue phase was characterized by increased

radiopharmaceutical uptake that corresponded to lytic sites

noted on radiographs. None of the vascular phase studies

were abnormal. In three turtles, the radiopharmaceutical

uptake in follow-up scans was decreased or unchanged

despite persistent lytic lesions present on radiographs. In

two turtles, radiopharmaceutical uptake decreased or re-

mained unchanged which correlated well with radiographic

healing. Five of six scintigraphic studies obtained more

than 4 months after initial appearance of clinical and ra-

diographic abnormalities were normal.

Seven of the eight turtles were subsequently released. In

the other, the formation of unrelated bone lesions and a

coelomic granulomata prevented release. In turtles where

there was abnormal radiopharmaceutical uptake, systemic

antibiotic and antifungal therapy was continued until

scintigraphic and/or clinical signs resolved.

Discussion

With the exception of the lack of periosteal new bone

formation, the lesions followed the typical reaction of bone

to osteomyelitis.22–26 However, because no biopsies were

taken, definitive histopathologic confirmation could not be

made. Articular gout has been reported as a lytic lesion of

reptiles,27 however, there was no clinical evidence of such

in any of the turtles. A syndrome with radiographic

changes similar to those seen in these turtles, and causing

delayed osteolysis of the phalangeal epiphyses, has been

described in human frostbite patients.28–31 In this syn-

drome, radiographic changes may not be visible until

several months after the insult.

Lesions were classified as early and late stage reac-

tions.22–26 This classification remains subjective, which

relies primarily in grouping multiple radiographic findings

and their changing appearance between serial radiographs.

This classification does not assign a specific timeline to an

isolated radiographic change. A typical early stage lesion

(Figs. 1, 3A, and 4) was characterized by a purely osteolytic

geographic-like pattern with a short transition zone, which

may or may not be surrounded by perilesional (mostly

periarticular) soft tissue swelling. As expected, an early

stage lesion was likely to be associated with abnormal

increased radiopharmaceutical uptake (Fig. 5).32,33 The

majority of lesions at the time of initial radiographs were

early lesions. An accurate age of the lesions could not be

Fig. 3. Dorsopalmar radiograph right front flipper. (A) The typical ap-pearance of early bone lesions. (B) A follow-up radiograph made 6 monthslater. Soft tissue swelling (arrowheads) surrounds two lytic lesions associatedwith the proximal interphalangeal joint of digits 2 and 3. Lesions are mostlylytic and lack new bone formation. A pathologic fracture associated with theproximal third of the middle phalanx of the second digit is also noted. Awell-defined joint body (arrow) remains 6 months after the initial presen-tation. The lytic sites have remodeled as indicated by sclerosis associatedwith the affected articular surfaces and thickening of the affected bones. Softtissue swelling is no longer present. The appearance of the lesions in (B) istypical of a late stage presentation.

Fig. 4. Front flipper scintigram and dorsopalmar radiograph of the leftfront flipper. There is a well-defined area of moderate radiopharmaceuticaluptake associated with the left elbow (arrowhead). On the radiograph, thelesion (arrowhead) is noted as a large well-defined lytic area, at the earlystage. To achieve adequate visualization of the digits during the scintigrams,the forelimbs are placed as flat as possible to the gamma camera and leadscreens covering the body and neck are used (asterisks).

391SKELETAL SCINTIGRAPHY OF SEATURTLESVol. 49, No. 4

determined with certainty as the majority of the lesions

were already present at the time of the initial radiographs.

All early stage reactions were painful and exhibited soft

tissue swelling. A late-stage lesion (Figs. 2 and 3B) was

characterized by sclerosis, thickening of the bone, and

blunted lesion borders. The area of lysis did not fill with

new bone over time and there was no evidence of periosteal

new bone formation surrounding the area. Late stage le-

sions were likely to have no abnormal radiopharmaceutical

uptake in spite of persistent radiographic abnormalities.

The absence of a periosteal reaction and the somewhat

slower rate of healing in this species were two important

characteristics that differentiate this population of animals

from other species.22,25,26,34 With the exception of one

turtle, in which surgery to debride a surrounding abscess

was performed, there was a lack of periosteal new bone

formation associated with the lytic site even when other

signs of healing such as sclerosis, thickening, and blunting of

the lesion borders were present. In general, the radiographic

signs of healing occurred at a slower rate than in other

species. One turtle had evidence of healing over a 17-month

period. These results are consistent with previous observa-

tions that document the relatively slow healing of chelonian

bone in comparison with that in mammals.19,27,35,36

Skeletal scintigraphy is a useful diagnostic test in the

evaluation of appendicular skeletal lesions of Kemp’s rid-

ley sea turtles. The dosage of 37MBq/kg of body weight,

used in several turtles reported here, provided images of

diagnostic count densities. Scintigrams of adequate count

density were also obtained at a dosage as low as 20MBq/

kg of body weight. The three classic phases of a bone scan

were noted in these turtles.37 The vascular phase did not

seem to be useful in the assessment of the bone lesions in

this particular disease. We varied the acquisition protocols

in an attempt to identify the most useful scanning tech-

nique. We recommend a frame rate of 1 s per frame for the

first 2min starting at the time of radiopharmaceutical in-

jection. At this setting, the injection site, heart, and major

vessels are clearly visible. Accurate determination of the

length of time that the radiopharmaceutical remains in the

interstitial space was beyond the scope of this investigation.

However, all scans in which a dynamic acquisition study

was performed during the soft tissue phase revealed radio-

pharmaceutical in soft tissues at 10–20min after radio-

pharmaceutical injection, which is similar to other

species.38,39 Subjectively, there was no difference between

dynamic or static mode soft tissue images. However, it is

more practical to acquire the soft tissue scans in static

mode, as sedation was not required. Hence, for soft tissue

scans, static, count-based, whole body images between

500,000 and 800,000 counts are recommended, 8–15min

after radiopharmaceutical injection.

Only larger lesions had abnormal radiopharmaceutical

uptake during the soft tissues phase. It is not known

whether this uptake is due to early bone uptake or true soft

tissue disease.40 For the bone phase, static images acquired

2–3h after radiopharmaceutical injection provided good

depiction of skeletal anatomy. Count-based scintigrams of

500,000 counts or higher for the entire body, and 150,000–

200,000 counts or higher if only one limb was imaged,

resulted in adequate identification of phalanges. Care

should be taken to ensure symmetry between the limbs

during positioning to facilitate comparison between con-

tralateral normal limbs. Flippers should also be positioned

as flat and as close as possible to the detector, otherwise

the phalanges will not be discernible and lesions could be

missed. As noted in other species with open growth plates,

Kemp’s ridley sea turtles had a higher count density in the

metaphyses and epiphyses of the long bones.41

No detectable abnormal radiopharmaceutical uptake was

noted in any lesion considered to be at the late stage of

healing. Of 37 early stage lesions, 11 lesions in one turtle were

not detected. We believe this was the result of poor posi-

tioning of the flippers in relationship to the detector. In ad-

dition, two early stage lesions were not detected in another

turtle. These lesions were characterized by concurrent thick-

ening and sclerosis; therefore, they may have been at a more

advanced stage of healing than the typical early lesion. Size of

the area of interest in this turtle may have also played a role,

as the lesions were associated with the small distal phalanges.

The presence or absence of radiopharmaceutical uptake

at sites of radiographically visible lesions was important for

determining therapy for individual turtles. Despite radio-

graphically persistent large bone defects that remain

void of new bone, animals were considered releasable if

normal radiopharmaceutical uptake was seen during initial

examination, or if decreased uptake was noted between

serial examinations. For these turtles, we hypothesize that

the bone has ceased osteoclastic and osteoblastic activity

Fig. 5. Ventral forelimb bone scintigram and dorsopalmar radiograph ofa left front flipper. There is a focal area of mild radiopharmaceutical uptakeassociated with the distal interphalangeal joint of the third digit (black ar-rowhead). The lesion corresponds to the lytic area identified by the whitearrowhead. The radiographic lesion is at the early stages of presentation asindicated by the lack of remodeling of the affected bones and the purely lyticreaction present. The area of moderate radiopharmaceutical uptake associ-ated with the right second digit (arrow) corresponds to the pathologicalfracture presented in Figure 3A.

392 SOLANO ET AL. 2008

despite persistent abnormal radiographic findings. Even

though histopathologic examination was not performed,

clinical experience has supported this hypothesis, as the

turtles have shown concurrent resolution of joint swelling,

pain, and lameness. In four of the turtles, resolution of

abnormal scintigraphic findings permitted an objective de-

cision to discontinue antibiotic and antifungal therapy. We

believe, however, that radiographs are still needed to fa-

cilitate the correct identification of lesions with scinti-

graphy. The intensity of the radiopharmaceutical uptake

did not always correlate well with the size of the radio-

graphic lesion, as some of the larger lytic lesions had only

asymmetric uptake when compared with the contralateral

normal limb, rather than increased uptake.

In mammals with septic synovitis or osteomyelitis, cultures

of blood, synovial fluid, synovial membrane, and bone are

often negative, and diagnosis is based on a combination of

supportive clinical, laboratory, and imaging findings.42–47 In

two turtles, bacteria were isolated from the affected bone or

joint, while blood cultures were negative in six turtles. As all

turtles were being treated with antimicrobials at the time of

needle aspirates, it is possible that culture results may have

been affected. Unfortunately, synovial or bone cultures were

not performed in four turtles, and histopathologic analysis

was not performed in any turtle. Bacterial and fungal osteo-

myelitis in Kemp’s ridley sea turtles has been previously re-

ported.14,19,20,48 Most cold-stunned turtles presented for

rehabilitation are treated with broad-spectrum antibiotic

and antifungal medications in light of the high prevalence of

sepsis and pneumonia.14,21 This therapy is likely also effective

at treating some osteomyelitic lesions, which may explain the

clinical improvement of these patients, and the negative

blood, synovial fluid, and bone culture results in some tur-

tles. However, it is unclear at this time whether the osteolytic

lesions of the Kemp’s ridley sea turtles always represent

osteomyelitis or whether the lesions may be due to sterile

necrosis secondary to hypothermia.28–31

Surgical biopsy, histopathologic examination, and cul-

ture of affected bone may be useful in determining the

etiology of these lesions. Histopathologic examination

would likely provide significant insight into the nature of

these lesions. However, in our experience, and in one pre-

viously published report, surgical management and biopsy

of Kemp’s ridley osteolytic bone lesions has exacerbated

clinical signs and prolonged rehabilitation.20 Such compli-

cations are to be avoided when dealing with an endangered

species. In most turtles, medical management as discussed

in this manuscript is effective. To date, all affected turtles

have survived, thus gross necropsy and histopathologic

examination of affected bones have not been performed.

While clinical follow-up of these patients is generally not

possible after release, one turtle was followed by satellite

tracking for 231 days and for a distance of 2760 km before

failure of the transmitter battery. These data provide in-

direct evidence that the bone lesions had no subsequent

impact on the health of the turtle.

Based on our observations, the radiographic changes as-

sociated with appendicular skeletal lytic lesions in Kemp’s

ridley sea turtles may lag behind the initial insult to bone by

several weeks to months. Once present, however, radio-

graphic changes compatible with an aggressive reaction may

persist for months after resolution of clinical signs. As ex-

pected, skeletal scintigraphy offers a method of evaluating

whether these lytic sites have active bone turnover, and by

association helps to determine whether or not the lesion is

clinically significant. Hence, scintigraphy has been more use-

ful than radiographs alone in determining if and when these

animals can be released. Based on our findings, we recom-

mend that osteolytic radiographic lesions in Kemp’s ridley

sea turtles should be evaluated by skeletal scintigraphy. If a

radiographically visible lytic site is not associated with in-

creased or asymmetric radiopharmaceutical uptake, and

clinical signs are not present, we believe that the lesion has

become inactive and unlikely to change radiographically over

time, and that such patients are candidates for immediate

release, if all other medical problems have resolved.

ACKNOWLEDGMENTS

This study was made possible by the generous technical imagingexpertize of Kathleen Hunt and Karen Johnson, Tufts University,Cummings School of Veterinary Medicine. The authors would also liketo thank past and present staff of the Animal Health Department, andthe Rescue and Rehabilitation Department of the New England Aquar-ium, Wood’s Hole Science Aquarium, and National Marine Life Centerfor obtaining radiographs of these patients, and assisting with thescintigraphic procedures. We also thank the staff and volunteers of theMassachusetts Audubon Society for retrieving stranded turtles from thebeach, and transporting them to the New England Aquarium.

REFERENCES

1. Marquez MR, Burchfield PM, Dıaz FJ, et al. Status of the Kemp’sridley turtle, Lepidochelys kempii. Chel Cons Biol 2005;4:761–766.

2. Lazell JD. New England waters: critical habitat for marine turtles.Copeia 1980;2:290–295.

3. Morreale SJ, Standora EA. Western North Atlantic waters: crucialdevelopmental habitat for Kemp’s ridley and loggerhead sea turtles. ChelCons Biol 2005;4:872–882.

4. Morreale SJ, Meylan A, Sadove SS, Standora EA. Annual occurrenceand winter mortality of marine turtles in New York waters. J Herp1992;26:301–308.

5. Gerle E, DiGiovanni R, Pisciotta RP. A fifteen-year review of cold-stunned sea turtles in New York waters. In: Abreu-Grobois FA, Briseno-Duenas R, Marquez R, Sarti L (compilers): Proceedings of the EighteenthInternational Sea Turtle Symposium, Miami, FL. NOAA Tech. MemoNMFS-SEFSC-436. U.S. Department of Commerce, 2000;222–224.

6. Burke VJ, Standora EA, Morreale SJ. Factors affecting strandings ofcold-stunned juvenile Kemp’s ridley and loggerhead sea turtles in Long Is-land, NY. Copeia 1991;4:1136–1138.

7. Bentivegnal F, Paolo Breber P, Hochscheid S. Cold stunned logger-head turtles in the South Adriatic Sea. Mar Turtle News 2000;97:1–3.

393SKELETAL SCINTIGRAPHY OF SEATURTLESVol. 49, No. 4

8. Still BM, Griffin CR, Prescott R. Climatic and oceanographic factorsaffecting daily patterns of juvenile sea turtle cold-stunning in Cape Cod Bay,Massachusetts. Chel Cons Biol 2005;4:883–890.

9. Still B, Tuxbury K, Prescott R, et al. A record cold stun season in CapeCod Bay, Massachusetts, USA. In: Mosier A, Foley F, Brost B (compilers):Proceedings of the Twentieth Annual Symposium on Sea Turtle Biology andConservation, Miami, FL. NOAA Tech. Memo NMFS-SEFSC-477, USDepartment of Commerce, 2002;205.

10. Dodge KD, Prescott R, Lewis D, Murle D, Merigo C. A review ofcold-stun strandings on Cape Cod Massachusetts from 1979–2003. In: Pro-ceedings of the Twenty Fourth Annual Symposium on Sea Turtle Biologyand Conservation, NOAA Technical Memorandum, in press.

11. Witherington WB, Ehrhart L. Hypothermic stunning and mortalityof marine turtles in the Indian river lagoon system, Florida. Copeia1989;2:696–703.

12. Schwartz F. Behavioral and tolerance responses to cold water tem-peratures by three species of sea turtles (Reptilia, Cheloniidae) in NorthCarolina. Florida Mar Res Publ 1978;33:16–18.

13. Turnbull BS, Smith CR, Stamper MA. Medical implications ofhypothermia in threatened loggerhead (Caretta caretta) and endangeredKemp’s ridley (Lepidochelys kempii) and Green (Chelonia mydas) seaturtles. In: Proceedings of the American Association of Zoo Veterinariansand the International Association for Aquatic Animal Medicine, 2000;31–35.

14. Matassa K, Early G, Wyman B, Prescott R, Ketton D, Krum H. Aretrospective study of Kemp’s ridley (Lepidochelys kempii) and loggerhead(Caretta caretta) in the area of the Northeast stranding network and asso-ciated clinical and postmortem pathologies. In: Bjordnal KA, Bolten AB,Johnson DA, Eliazar PJ (eds): Proceedings of the Fourteenth Annual Sym-posium on Sea Turtle Biology and Conservation, Miami, FL. NOAA Tech-nical Memorandum NMFS-SEFSC-351, US Department of Commerce,1994;255–258.

15. Sadove SS, Pisciotta R, DiGiovanni R. Assessment and initial treat-ment of cold-stunned sea turtles. Chel Cons Biol 1998;3:84–87.

16. Pisciotta RP, DurhamK, DiGiovanni R, Sadove SS, Gerle E. A casestudy of invasive medical techniques employed in the treatment of aseverely hypothermic Kemp’s ridley sea turtle. In: Keinath JA, Barnard DE,Musick JA, Bell BA (eds): Proceedings of the Fifteenth Annual Symposiumon Sea Turtle Biology and Conservation, Miami, FL. NOAA TechnicalMemorandum NMFS-SEFSC-387, US Department of Commerce 1996;253–255.

17. Carminati C, Gerle E, Kiehn LL, Pisciotta RP. Blood chemistrycomparison of healthy vs. hypothermic juvenile Kemp’s ridley sea turtles(Lepidochelys kempi) in the New York bight. In: Bjordnal KA, Bolten AB,Johnson DA, Eliazar PJ (compilers): Proceedings of the Fourteenth AnnualSymposium on Sea Turtle Biology and Conservation, Miami, FL. NOAATechnical Memorandum NMFS-SEFSC-351, US Department of Com-merce, 1994;203–207.

18. Smith CR, Hancock AL, Turnbull BS. Comparison of white bloodcell counts in cold-stunned and subsequently rehabilitated loggerhead seaturtles (Caretta caretta). In: Proceedings of the American Association of ZooVeterinarians and the International Association for Aquatic Animal Med-icine, 2000;50–53.

19. Harms CA, Lewbart GA, Beasley J, Beasley K. Medical manage-ment of mixed Nocardial and unidentified fungal osteomyelitis in a Kemp’sridley sea turtle, Lepidochelys kempii. J Herp Med Surg 2002;12:21–26.

20. Greer LL, Strandberg JD, Whitaker BR. Mycobacterium chelonaeosteoarthritis in a Kemp’s ridley sea turtle (Lepidochelys kempii). J WildlifeDis 2003;39:736–741.

21. Wyneken J, Mader D, Weber S, Merigo C. Medical management ofsea turtles. In: Mader DM (ed): Reptile medicine and surgery, 2nd ed. St.Louis: Elsevier, 2006;972–1007.

22. Ta F, Oguz S, Bulut O, Isik AO. Comparison of the diagnosis ofplain radiography ultrasonography and magnetic resonance imaging in earlydiagnosis of acute osteomyelitis experimentally formed on rabbits. Eur JRadiol 2005;56:107–112.

23. Tehranzadeh J, Wong E, Wang F, Sadighpour M. Imaging of osteo-myelitis in the mature skeleton. Radiol Clin North Am 2001;39:223–250.

24. Oudjhane K, Azouz EM. Imaging of osteomyelitis in children. Ra-diol Clin North Am 2001;39:251–266.

25. Reese Hand D, Watkins JP, Clifford CM, Kemper D. Osteomyelitisof the sustentaculum ali in horses: 10 cases (1992–1998). J Am Vet MedAssoc 2001;219:341–345.

26. Erne JB, Walker MC, Strik N, Alleman AR. Systemic infection withGeomyces organisms in a dog with lytic bone lesions. J Am Vet Med Assoc2007;230:537–540.

27. Mader DM, Bennett RA, Funk RS, Fitzgerald KT, Vera R, Her-nandez-Divers SJ. Surgery. InMader DR (ed): Reptile medicine and surgery,2nd ed. St. Louis, MO: Elsevier, 2006;581–630.

28. Hakstian RW. Cold-induced digital epiphyseal necrosis in childhood(symmetric focal ischemic necrosis). Can J Surg 1972;15:168–178.

29. Bhatnagar A, Sarker BB, Sawroop K, Chopra MK, Sinha N, Kash-yap R. Diagnosis, characterization and evaluation of treatment response offrostbite using pertechnetate scintigraphy: a prospective study. Eur J NuclMed 2002;29:173–175.

30. Tishler JM. The soft-tissue and bone changes in frostbite injuries.Radiology 1972;102:511–513.

31. Selke AC. Destruction of phalangeal epiphyses by frostbite. Radi-ology 1969;93:859–860.

32. Lamb CR. Bone scintigraphy in small animals. J Am Vet Med Assoc1987;191:1616–1622.

33. Nutton RW, Fitzgerald Jr RH, Kelly PJ. Early dynamic bone-imag-ing as an indicator of osseous blood flow and factors affecting the uptake of99mTc hydroxymethylene diphosphonate in healing bone. J Bone Joint SurgAm 1985;67:763–770.

34. Henry GA. Fracture healing and complications. In: Thrall DE (ed):Textbook of veterinary diagnostic radiology, 5th ed. St. Louis, MO: Saun-ders Elsevier, 2007;284–305.

35. Mitchell MA. Diagnosis and management of reptile orthopedic in-juries. Vet Clin North Am Exot Anim Pract 2002;5:97–114.

36. Crane SW, Curtis M, Jacobson ER, et al. Neutralization bone-plat-ing repair of a fractured humerus in an Aldabra tortoise. J Am Vet MedAssoc 1980;177:945–948.

37. Schauwecker DS. The scintigraphic diagnosis of osteomyelitis. Am JRadiol 1992;158:9–18.

38. Trout DR, Hornof WJ, Liskey CC, Fisher PE. The effects of re-gional perineural anesthesia on soft tissue and bone phase scintigraphy in thehorse. Vet Radiol 1991;32:140–144.

39. Metcalf MR. Preliminary clinical use of combined blood pool andbone phase radionuclide imaging in dogs. Vet Radiol 1985;26:117–122.

40. O’Callaghan MW. The integration of radiography and alternativeimaging methods in the diagnosis of equine orthopedic disease. Vet ClinNorth Am Equine Pract 1991;7:339–364.

41. Metcalf MR, Sellet LC, Smallwood JE, Thrall DE, Harrington BD.Scintigraphic characterization of the equine fore digit and metacarpophalan-geal region from birth to six months of age. Vet Radiol 1989;30:111–118.

42. Lescun TB, Vasey JR, Ward MP, Adams SB. Treatment with con-tinuous intrasynovial antimicrobial infusion for septic synovitis in horses: 31cases (2000–2003). J Am Vet Med Assoc 2006;228:1922–1929.

43. Bertone AL, McIlwraith CW, Jones RL, Norrdin RW, Radin MJ,Lebel JL. Comparison of various treatments for experimentally inducedequine infectious arthritis. Am J Vet Res 1987;48:519–529.

44. Gupta MN, Sturrock RD, Field M. Prospective comparative studyof patients with culture proven and high suspicion of adult onset septicarthritis. Ann Rheum Dis 2003;62:327–331.

45. Connolly LP, Connolly SA. Skeletal scintigraphy in the multimo-dality assessment of young children with acute skeletal symptoms. Clin NuclMed 2003;28:746–754.

46. Madison JB, Sommer M, Spencer PA. Relations among synovialmembrane histopathologic findings, synovial fluid cytologic findings, andbacterial culture results in horses with suspected infectious arthritis: 64 cases(1979–1987). J Am Vet Med Assoc 1991;8:1655–1661.

47. Shneider RK, Bramlage LR, Moore RM, Mecklenburg LM, KohnCW, Gabel AA. A retrospective study of 192 horses affected with septicarthritis/tenosynovitis. Equine Vet J 1992;24:436–442.

48. Smith CR, Turnbull BS, Osborn AL, Dube K, Johnson KL, SolanoM. Bone scintigraphy and computed tomography: advanced diagnosticimaging techniques in endangered sea turtles. In: Proceedings of the Amer-ican Association of Zoo Veterinarians and the International Association forAquatic Animal Medicine, 2000;217–221.

394 SOLANO ET AL. 2008