Kemp's ridley turtles (Lepidochelys kempii) are native to the United States and Mexico and are the world's most endangered species of marine turtle.1 Adults of this species generally live in the Gulf of Mexico, but juveniles often range along the northeastern coast of the United States during the summer.2 If juveniles fail to migrate to warmer waters in autumn, they are exposed to increasingly cold water temperatures and may become cold stunned. Cold-stunning events have been reported for sea turtles from many temperate localities in the United States and Europe, most often occurring at water temperatures less than approximately 15°C (59°F).3–8 Data from these events indicate that 35% to 85% of cold-stunned turtles are already dead when found stranded on beaches.6,7,9 The reasons for these variable mortality rates have not been determined, but mortality rate is likely affected by the duration and degree of hypothermia, presence of preexisting illness, fitness of the population, species, and other factors.
Among those cold-stunned sea turtles that are hospitalized after being found alive, overall successful rehabilitation and release rates of approximately 60% to 70% have been reported.7,8 Most deaths during hospitalization occur during the first few days because of physiologic derangements secondary to hypothermia.8,10,11 Far fewer deaths occur after the first few days of hospitalization. For example, 1 report8 described an 88% success rate for rehabilitation of cold-stunned sea turtles that survived the first 3 days of hospitalization. Morbidity and death of cold-stunned sea turtles after the first week of hospitalization are most commonly due to secondary pathological conditions such as bacterial and fungal pneumonia, chronic renal failure, osteomyelitis, sepsis, and, in the authors’ experience, mycobacteriosis.8,12,13
In Massachusetts, a coordinated rescue-and-rehabilitation program for cold-stunned sea turtles has been in operation for approximately 20 years. Volunteers and staff of the Massachusetts Audubon Society patrol beaches of Cape Cod from October through January to recover stranded turtles, which are then transported to the New England Aquarium for medical care. Details of the triage and medical management of cold-stunned turtles at the New England Aquarium have been published.8 Briefly, turtles are gradually warmed over several days to a body temperature of 24° to 25°C (75° to 77°F) and treated for dehydration, metabolic derangements, cardiorespiratory depression, and concurrent pathological conditions.8 Assessment of health is obtained over time by serial examinations, observation, and a variety of diagnostic tests. Turtles remain in rehabilitation until they can be released into warmer waters, which is often 6 to 8 months after stranding, depending on geographic location. In some instances, turtles are transferred to secondary care facilities to complete rehabilitation.
In general, most serious bacterial infections of reptiles, including sea turtles, are caused by gram-negative aerobic organisms, which are normal flora of the reptilian oral cavity, skin, and digestive system and the marine environment.8,14 However, in recent years, infections caused by gram-positive bacteria of the genus Enterococcus have been increasingly recognized during rehabilitation of cold-stunned Kemp's ridley turtles, and Enterococcus faecalis is now the most commonly cultured bacteria in cases of septicemia and osteomyelitis in sea turtles at the New England Aquarium. Enterococcus spp are gram-positive bacterial cocci that are normally found in the gastrointestinal tract of many vertebrate species and in the environment. They can cause serious infections in humans, other mammals, and birds and are a common cause of nosocomial infections.15–17 Enterococcus spp have been cultured from the feces, skin, and cloaca of healthy reptiles.18–20 However, reports21–24 of Enterococcus spp infections in reptiles are limited to a few case reports in individual animals. The purpose of the study reported here was to evaluate methods used for diagnosis and medical management of infections caused by Enterococcus spp in hospitalized Kemp's ridley turtles.
Materials and Methods
Medical management and rehabilitation of sea turtles were conducted with authorization of the US Department of the Interior Fish and Wildlife Service and the National Oceanic and Atmospheric Administration National Marine Fisheries Service.
Case selection—Medical records for Kemp's ridley turtles that were admitted alive to New England Aquarium between November 2006 and December 2011 were reviewed retrospectively, including records for turtles that had been initially hospitalized at the New England Aquarium and later transferred for secondary care to the National Marine Life Center, Buzzards Bay, Mass; the University of New England, Biddeford, Me; the National Aquarium, Baltimore, Md; and Sea World Orlando, Orlando, Fla. Turtles were included in the study if a bacteriologic culture of blood or other tissue yielded an isolate of Enterococcus spp during hospitalization or postmortem examination.
Medical records review—Data retrieved from the medical record of each case included morphometric data, physical examination findings, water quality data, antimicrobial medication history, history of medications administered IV, day of diagnosis (defined as the date when a sample yielding a positive culture result had been collected), clinical signs at diagnosis, microbiological results, sources of positive culture results, hematologic and plasma biochemical data, cytologic and histopathologic results, radiographic findings, time to first negative culture result, duration of treatment, results of subsequent cultures, and case outcome.
Sample collection—Bacteriologic culture of blood, bacteriologic culture of tissue, tissue biopsy, necropsy, and histologic evaluation were used to diagnose bacterial infections. Investigations were conducted when indicated on the basis of clinical signs, physical examination findings, clinicopathologic results, or radiographic abnormalities. Blood samples were obtained for bacteriologic culture after venipuncture sites were aseptically prepared with sterile povidone iodine and isopropyl alcohol–infused gauze pads. One milliliter of blood was collected from the external jugular vein with a 1.5- to 2.5-cm (5/8- to 1-inch), 23- to 25-gauge needle attached to a 1- to 3-mL syringe. After venipuncture, the needle was replaced with a new needle, and blood was immediately transferred into brain-heart infusion brotha for bacterial culture. Bacterial culture samples of various tissues were also obtained by fine-needle aspiration, tracheal wash, and tissue biopsy. Such samples were transferred to a sterile culture swab or a sterile glass vial. Culture specimens from the New England Aquarium, University of New England, and National Marine Life Center were transported on ice to a commercial veterinary diagnostic laboratory,b and cultures were initiated within 6 to 18 hours. Cultures from the National Aquarium and Sea World Orlando were performed at their respective internal diagnostic laboratories. Following initial diagnosis, cultures were repeated at roughly 2- to 4-week intervals to monitor response to treatment.
Isolation and identification of microbes—Details of the microbiological methods that were used for the majority of cases (n = 43), including all samples from the New England Aquarium, University of New England, and National Marine Life Center, have been described.20 Similar methods were used for National Aquarium and Sea World Orlando samples. Cultures were conducted at both 25° and 35°C for isolation of aerobic and anaerobic bacteria and at 22° to 25°C for fungal culture. Various subcultures and selective media were used when indicated.20
Antimicrobial susceptibility testing—Antimicrobial susceptibility was determined with commercially available colorimetric susceptibility cardsc and Kirby-Bauer agar disk diffusion, following Clinical and Laboratory Standards Institute performance standards.25 Selection of antimicrobials for antimicrobial susceptibility testing was determined by the diagnostic laboratory on the basis of whether the isolate was a gram-negative or gram-positive organism as well as bacterial species-specific standard operating procedures of the diagnostic laboratory and Clinical and Laboratory Standards Institute standards.25 For reasons unrelated to this study, standard operating procedures were modified slightly in laboratories during the years included in this study, such that the specific antimicrobials included in Enterococcus spp antimicrobial susceptibility panels varied (eg, elimination of ampicillin and inclusion of amoxicillin).
PFGE—In 2007, 13 E faecalis isolates from 11 turtles were evaluated by PFGE with SmaI restriction enzyme digestion and Staphylococcus aureus 519 as a control, following standard techniques.26,27
Histologic evaluation—Histologic evaluations were conducted with described methods28 at 1 of 2 veterinary diagnostic laboratories,d,e by 1 of 2 board-certified veterinary pathologists with considerable experience in histologic evaluation of sea turtles.
Hematologic evaluation and plasma biochemical analysis—Hematologic and plasma biochemical evaluations were conducted for samples obtained at the New England Aquarium, University of New England, and National Marine Life Center at a commercial veterinary diagnostic laboratoryb as described.29
Water quality data—Water chemical and water microbiological testing were performed at each facility and included analysis of water temperature, salinity, pH, and ammonia, nitrite, and nitrate concentrations. For monitoring efficacy of water disinfection at the New England Aquarium, concentrations of Enterococcus sp were determined with a commercially available test kit.f Water quality data ranging from at least 10 days up to 1 month prior to diagnosis were included in this study. For results that were greater than or less than limits of detection, the minimum or maximum possible values were used for data analysis, respectively.
Data analysis—Antimicrobial susceptibility and MIC data for initial Enterococcus spp isolates were collated, and the percentage of susceptible, intermediate, and resistant isolates was calculated for each antimicrobial. For consistency of methodology, only antimicrobial susceptibility data that were available from a single laboratoryb were included (n = 36 cases).
Hematologic and plasma biochemical values were categorized on the basis of time of blood sample collection (initial vs convalescent) and the final outcome of the turtle (survived vs died). Initial values were measured from blood samples collected closest in time and prior to the diagnosis of Enterococcus spp infection. Convalescent values of surviving turtles were measured from the last available blood samples obtained after completion of antimicrobial treatment, prior to release to the wild.
Statistical analysis—Analysis was performed with commercially available software.g Initial hematologic and plasma biochemical data for turtles that survived were compared with initial data for turtles that died by ANOVA for data that were normally distributed and by the Kruskal-Wallace test for data that were not normally distributed. Initial hematologic and plasma biochemical data for turtles that survived were compared with their convalescent data by the paired t test for data that were normally distributed and by the Wilcoxon signed rank test for data that were not normally distributed. A Pearson χ2 test was used to assess whether mortality rate differed for turtles affected by only Enterococcus infection vs polymicrobial infection. Significance was set at P < 0.05.
Results
Three hundred two live Kemp's ridley sea turtles were admitted to the hospital at the New England Aquarium from November 2006 to December 2011. Of these turtles, 212 (70%) were successfully rehabilitated and released to the wild. Ninety (30%) turtles died during hospitalization, with 53 (59%) of 90 deaths in the first 72 hours and 60 (67%) deaths in the first week of hospitalization. Thirty (33%) deaths occurred after 10 days of hospitalization (range, 10 to 186 days). Turtles were treated prophylactically within 1 to 6 days after admission with ceftazidime (22 mg/kg [10 mg/lb], IM, q 3 d) on the basis of pharmacokinetic data for ceftazidime in this species30 and the high prevalence of gram-negative bacterial infections in cold-stunned sea turtles.
Fifty turtles had infections caused by Enterococcus spp during hospitalization between January 2007 and March 2012. Morphometric data and body temperatures at admission for these turtles were summarized (Table 1). Sex was not determined for most turtles because external sexually dimorphic characteristics were not apparent in these juvenile turtles. Thirty-five cases were identified at the New England Aquarium, 7 cases at the National Marine Life Center, 5 cases at the National Aquarium, 2 cases at Sea World Orlando, and 1 case at the University of New England. Cases were identified from at least 18 independent aquatic systems in at least 7 hospital locations. Systems ranged in volume from 1,600 to 80,000 L. Water quality data were available for 42 cases (Table 2).
Weight, straight carapace length, and body temperature at admission for 50 cold-stunned Kemp's ridley turtles (Lepidochelys kempii) with Enterococcus spp infections during hospitalization.
| Variable | Mean ± SD | Median (range) |
|---|---|---|
| Weight (kg) | 2.7 ± 0.8 | 2.6 (1.1–4.7) |
| Straight carapace length (cm) | 26.5 ± 3.0 | 26.9 (19.8–32.8) |
| Body temperature at admission (°C) | 12 ± 2 | 12 (8–17) |
Water quality data for 42 Kemp's ridley turtles with Enterococcus spp infection during hospitalization.
| Variable | Mean ± SD | Median (range) | Desired range (New England Aquarium) |
|---|---|---|---|
| Temperature (°C) | 24.7 ± 0.9 | 24.4 (18.6 to 28.0) | 23 to 26 |
| pH | 8.0 ± 0.06 | 8.0 (7.7 to 8.4) | 7.7 to 8.3 |
| Salinity (g/L) | 31 ± 0.5 | 31 (25 to 37) | 28 to 35 |
| Ammonia (μg/L) | 190 ± 146 | 100 (10 to 2,563) | < 100 |
| Nitrite (mg/L) | 0.6 ± 0.4 | 0.2 (0 to 9.1) | < 0.3 |
| Nitrate (mg/L) | 13.8 ± 7.9 | 4.5 (0 to 228) | < 50 |
| Enterococcus (CFUs/100 mL) | 994 ± 1,626 | 97 (10 to > 48,392) | < 103 |
CFU = Colony-forming unit.
Enterococcus spp infection was diagnosed in 43 live turtles, and in 7 dead turtles by postmortem examination. Clinical abnormalities at the time of diagnosis included radiographic evidence of pneumonia (n = 28 turtles), anorexia or poor appetite (22), appendicular joint swelling and lameness (12), signs of lethargy (11), radiographic evidence of osteomyelitis (11), radiographic evidence of excessive gastrointestinal tract gas (7), moribund condition (4), persistent hypoglycemia (4), shell fractures (4), dermal wounds (3), and subcutaneous masses (2).
In addition to having been treated with ceftazidime within several days after hospitalization, 27 of the 50 turtles with Enterococcus spp infection had been initially treated empirically with fluconazole (21 mg/kg [9.5 mg/lb], SC, loading dose, followed by 10 mg/kg [4.5 mg/lb], SC, q 5 d) on the basis of pharmacokinetic data for fluconazole in this species30 and the high prevalence of fungal pneumonia in cold-stunned sea turtles. Fourteen of the 50 turtles had been treated by IV administration with various medications during the first week of hospitalization, including 50% dextrose (n = 13), doxapram (4), atropine (2), and furosemide (1).
Enterococcus spp infection was diagnosed a median of 24 days (range, 10 to 177 days; mean ± SD, 36 ± 29.7 days) after hospitalization. For turtles that survived (n = 34), Enterococcus spp infection was diagnosed a median of 39 days (range, 12 to 177 days; mean ± SD, 44 ± 31.7 days) after hospitalization. For turtles that died (n = 16), Enterococcus spp infection was diagnosed a median of 17 days (range, 10 to 59 days; mean ± SD, 19 ± 11.5 days) after hospitalization. Turtles had Enterococcus spp infections diagnosed by bacteriologic culture of blood (n = 30), bone or joint (7), postmortem lung tissue (4), both blood and tracheal wash (2), both blood and bone (4), both blood and postmortem lung tissue (2), and dermal granuloma (1).
Species identification was performed for most but not all Enterococcus spp isolates. Initial isolates from each case were identified as E faecalis (36 cases), an Enterococcus sp (13 cases), and Enterococcus faecium (1 case). Pulsed-field gel electrophoresis identified 4 infecting E faecalis strains, including PFGE types 1 (n = 4), 2 (6), 3 (2), and 4 (1).
Positive results of bacteriologic culture of bone or joint tissue were obtained from fine-needle aspirates of the elbow joint (n = 5), carpal joint (2), and metacarpal-phalangeal joint (1) and biopsy specimens of phalanges (2) and carapace bone (1). These samples were obtained from joints or bones with radiographically osteolytic lesions, most commonly involving appendicular skeletal epiphyseal surfaces (n = 10). Five turtles had cytologic results for fine-needle aspirates of joint tissue from which Enterococcus spp were isolated. Results generally indicated mixed heterophil and mono-nuclear cell inflammation, and gram-positive cocci were seen in 1 case. Histologic evaluations were available for 3 bone biopsy specimens from which Enterococcus spp were isolated, revealing marked heterophilic osteomyelitis of the phalanx; severe heterophilic periostitis, cellulitis, and chondronecrosis of the phalanx and surrounding tissue; and severe heterophilic and necrotizing osteomyelitis of bone of the carapace. In one phalangeal biopsy specimen, gram-positive cocci were observed, whereas in the other, cocci with equivocal Gram stain results were observed. In bone of the carapace, bacterial cocci or coccobacilli with equivocal Gram stain results were observed.
Of the 16 turtles that died, 7 died before specific treatment was initiated (ie, the diagnosis of Enterococcus spp infection was made after death). For the 8 turtles that died after specific treatment for Enterococcus spp infection was initiated, the median day of death was 21 days (range, 1 to 84 days; mean ± SD, 30 ± 30 days) after treatment initiation. One turtle was excluded from this calculation as an extreme outlier, having died from an unrelated cause 175 days after treatment initiation.
Results of postmortem histologic evaluation were available for 13 cases and included lesions compatible with sepsis caused by a gram-positive coccus as well as lesions generally observed in cold-stunned turtles and lesions that were considered incidental.28 Intralesional or intravascular gram-positive cocci were seen in foci of steatitis and fat necrosis (n = 6), hepatic necrosis (3), and cholecystitis (1). Other lesions that were suspected of being bacterial in origin, even though bacteria were not observed in tissue section, included meningitis (n = 3) and severe chronic osteoarthritis with chondronecrosis (1). Lesions consistent with the general sequelae of cold stunning included renal hyaline or coarse granular cast formation (n = 7), renal tubular necrosis (5), renal calculi and renal mineralization (3), adrenal cell hyperplasia (3), and nephron ectasia (2).11 Lesions that were considered incidental included gastrointestinal tract parasitic granuloma formation (n = 7), hepatic parasitic granuloma formation (3), hepatic hemosiderosis (2), and rhabdomyolysis (4), the latter of which could have been an exertional or agonal change.
Twenty-five initial cultures from 24 cases simultaneously yielded other bacterial or fungal species with 1 other organism isolated in 16 cases, 2 other organisms isolated in 5 cases, and 3 to 5 other organisms isolated in 4 cases. Other organisms that were identified on initial bacteriologic culture of blood included Hafnia alvei (n = 3), Vibrio alginolyticus (2), Citrobacter braakii (2), Citrobacter freundii (2), a Citrobacter sp (1), and a Pseudomonas sp (1). Other organisms that were isolated on initial bacteriologic culture of bone included a Pseudomonas sp (n = 4), a Citrobacter sp (1), and Escherichia coli (1). Other organisms that were isolated from bacteriologic culture of initial tracheal wash fluid included a Citrobacter sp (n = 1), unidentified nonenteric gram-negative rods (1), Pseudomonas putida (1), V alginolyticus (2), Vibrio fluvialis (1), Vibrio parahaemolyticus (1), and Vibrio cholerae (1). Other organisms that were isolated from initial bacteriologic cultures of lung tissue included C freundii (n = 3), a Paecilomyces sp (3), H alvei (2), an unidentified nonenteric gram-negative rod (1), Chromobacterium violaceum (1), a Bacillus sp (1), a Bacteroides sp (1), a Corynebacterium sp (1), Mycobacterium chelonae (1), and Candida guilleromondii (1). Among these 24 turtles with polymicrobial infections, 10 turtles died and 14 survived. The mortality rate of turtles with polymicrobial infection was not different from the mortality rate of turtles with only Enterococcus spp infection (P = 0.16).
After initial diagnosis, an Enterococcus sp was isolated a second time from individual cases via bacteriologic culture of blood (n = 12), bone (4), lung tissue (1), and both blood and skin mass (2). Other organisms that were isolated from second bacteriologic culture of blood included V alginolyticus (n = 4), Shewanella algae (2), C braakii (1), Enterobacter amnigenus (1), and E coli (1). Of the second bacteriologic cultures of bone with positive results, 2 were cultures from forelimb aspirates (1 from the humeral head and 1 from the distal portion of the ulna). Other organisms that were grown from the second bacteriologic culture of bone included a Pseudomonas sp (n = 1) and C braakii (1). Other organisms that were isolated from the second bacteriologic cultures of a skin mass included a Citrobacter sp (n = 2), H alvei (1), a Vibrio sp (1), and a Fusarium sp (1). Enterococcus spp were isolated a third time in 6 turtles by bacteriologic culture of blood (n = 4), elbow joint aspirate (1), or both blood and skin mass (1). An Enterococcus sp was isolated a fourth time in 2 turtles by bacteriologic culture (n = 1) and stifle joint aspirate (1).
Antimicrobial susceptibility and MIC data for initial Enterococcus spp isolates from 38 turtles were summarized (Table 3). Antimicrobials were administered on the basis of antimicrobial susceptibility and MIC results, whether the turtle accepted orally administered medication voluntarily with food, whether multiple bacteria were cultured, and clinician preference. The most commonly used antimicrobials and dosages were ampicillin (20 to 30 mg/kg [9 to 14 mg/lb], IM, q 24 h), amoxicillin–clavulanic acid (30 mg/kg, PO, q 24 h), amoxicillin (30 mg/kg, PO, q 24 h), penicillin (20,000 U/kg [9,000 U/lb], IM, q 48 h), amikacin (5 to 10 mg/kg [2.3 to 4.5 mg/lb], IM or IV, q 72 h), enrofloxacin (10 to 20 mg/kg, PO, SC, or IM, q 48 to 72 h), gentamicin (2.5 mg/kg [1.1 mg/lb], IM, q 72 h), and ceftazidime (22 mg/kg, IM, q 72 h). The most commonly used successful single course of treatment was a combination of a penicillin and an aminoglycoside (n = 16), including amikacin and amoxicillin–clavulanic acid or ampicillin (11), gentamicin and amoxicillin–clavulanic acid (1), and amikacin and penicillin (1). For 3 cases, treatment was initiated with amikacin and amoxicillin–clavulanic acid or ampicillin, but gentamicin was later substituted for amikacin because of a nationwide amikacin shortage. Other successful single courses of treatment included amoxicillin only (n = 3), ampicillin only (2), ampicillin and ceftazidime (2), and ampicillin and enrofloxacin (1). Negative results of bacteriologic culture of blood (n = 31) were first obtained a median of 30 days (range, 8 to 86 days) after the start of treatment (mean ± SD, 36 ± 23 days). The median treatment duration for 24 turtles successfully treated with a single course of treatment was 60 days (range, 36 to 135 days; mean ± SD, 70 ± 25 days).
Antimicrobial susceptibility and MIC (μg/mL) data for 38 Enterococcus spp isolates from cold-stunned Kemp's ridley sea turtles.
| Antimicrobial | No. of isolates for which drug was assessed | Percentage of isolates susceptible (median MIC) | Percentage of isolates with intermediate susceptibility (median MIC) | Percentage of isolates resistant (median MIC) |
|---|---|---|---|---|
| Penicillin | 25 | 96 (4) | 0 | 4 (> 16) |
| Ampicillin | 9 | 89 (0.5) | 0 | 11 (> 16) |
| Amoxicillin | 29 | 100 (< 2) | 0 | 0 |
| Amoxicillin–clavulanic acid | 13 | 100 (NA) | 0 | 0 |
| Imipenem | 13 | 100 (< 1) | 0 | 0 |
| Vancomycin | 25 | 100 (< 2) | 0 | 0 |
| Gentamicin synergy | 37 | 97 (NA) | 0 | 3 (NA) |
| Ciprofloxacin | 13 | 92 (NA) | 8 (NA) | 0 |
| Enrofloxacin | 29 | 55 (0.5) | 41 (1) | 4 (> 2) |
| Marbofloxacin | 13 | 38 (1) | 62 (2) | 0 |
| Azithromycin | 13 | 69 (NA) | 31 (NA) | 0 |
| Erythromycin | 13 | 69 (< 0.25) | 31 (1.5) | 0 |
| Doxycycline | 13 | 85 (NA) | 0 | 15 (NA) |
| Tetracycline | 22 | 82 (< 1) | 0 | 18 (> 16) |
| Chloramphenicol | 38 | 95 (≤ 4) | 0 | 5 (> 32) |
NA = Not assessed.
Seven turtles had treatment modifications because of persistently positive results of Enterococcus spp cultures. Details of the individual courses of treatment for these 7 animals were beyond the scope of this study. However, as an example, treatment was modified and later repeated for 1 turtle because of persistent E faecalis infection, concurrent Citrobacter spp infection, and later recurrent E faecalis infection. This turtle was initially treated for 126 days with penicillin (24 days), followed by ampicillin (56 days), amikacin (75 days), and enrofloxacin (102 days), resulting in 2 bacteriologic cultures of blood with negative results during treatment. However, a bacteriologic culture of blood obtained 2 months after completing treatment again yielded E faecalis, and the turtle was then successfully treated with amikacin (42 days) and ampicillin (42 days).
Two turtles had Enterococcus spp isolated from initial bacteriologic cultures of blood with positive results but had no clinical signs, and no treatment was prescribed. One of these had a second positive result 1 month later but subsequently had 2 negative results 2 and 3 months after the first positive culture result. The second turtle had 2 negative results of bacteriologic culture of blood 5 days and 1 month after the positive culture result. Among the 32 surviving turtles that were treated with antimicrobials, final negative culture results were obtained a median of 33 days (range, 11 to 124 days) after completion of treatment for 18 cases (mean ± SD, 37 ± 24 days). Fourteen turtles did not have additional cultures performed after completion of treatment.
Hematologic and plasma biochemical results that met the definition of initial data were available for 22 turtles that survived and 14 turtles that died, and convalescent data were available for those 22 surviving turtles (Table 4). For turtles that survived, initial samples were collected a median of 9 days (range, 0 to 23 days; mean ± SD, 9 ± 7 days) prior to diagnosis of infection. For turtles that died, initial samples were collected a median of 7 days (range, 0 to 32 days; mean ± SD, 9 ± 8 days) prior to diagnosis of infection. Convalescent data were collected a median of 71 days (range, 3 to 171 days; mean ± SD, 73 ± 51 days) after completion of antimicrobial treatment. Compared with turtles that survived, turtles that died had significantly higher plasma activities of alanine aminotransferase, aspartate aminotransferase, and creatine kinase; higher plasma concentrations of glucose, sodium, and uric acid; higher anion gap; and lower plasma calcium concentrations. Compared with initial values, convalescent values for 22 turtles that survived were lower for plasma activities of aspartate aminotransferase, creatine kinase, and lactate dehydrogenase and plasma concentrations of cholesterol and uric acid. Compared with initial values, convalescent values were greater for plasma concentrations of albumin, total protein, and potassium; anion gap; and relative and absolute eosinophil count.
Initial and convalescent hematologic and plasma biochemical data from cold-stunned Kemp's ridley turtles with Enteroccous spp infection.
| Initial | ||||||
|---|---|---|---|---|---|---|
| Survived (n = 22) | Died (n = 14) | Convalescent (n = 22) | ||||
| Variable | Mean ± SD | Median (range) | Mean ± SD | Median (range) | Mean ± SD | Median (range) |
| Alkaline phosphatase (U/L) | 609 ± 885 | 149 (33–2,875) | 1,045 ± 1,396 | 353 (71–3,937) | 181 ± 176 | 142 (43–886) |
| Alanine aminotrasferase (U/L)* | 9 ± 9 | 6 (2–35) | 46 ± 59 | 11 (3–182) | 15 ± 17 | 8 (1–52) |
| Asparate aminotrasferase (U/L)*† | 506 ± 368 | 367 (101–1,438) | 1,143 ± 1,120 | 583 (315–4,105) | 308 ± 214 | 185 (89–760) |
| Creatine kinase (U/L)*† | 19,995 ± 27,228 | 10,962 (603–94,602) | 54,186 ± 48,624 | 40,313 (35–120,000) | 3,012 ± 2,453 | 2,308 (461–11,637) |
| Lactate dehydrogenase (U/L)† | 14,643 ± 17,128 | 10,040 (1,889–77,153) | 21,704 ± 21,527 | 14,575 (4,995–83,511) | 3,774 ± 2,514 | 3,380 (472–9,780) |
| γ-Glutamyltransferase (U/L) | 3 ± 3 | 3 (0–11) | 3 ± 2 | 3 (0–7) | 3 ± 2 | 2 (1–8) |
| Albumin (g/dL)† | 1.1 ± 0.3 | 1.1 (0.7–1.6) | 1.0 ± 0.4 | 1.0 (0.0–1.6) | 1.3 ± 0.2 | 1.4 (1.0–1.7) |
| Total protein concentration (g/dL)† | 3.1 ± 0.6 | 3.2 (2.1–4.3) | 2.8 ± 0.6 | 2.8 (1.8–4.0) | 3.4 ± 0.4 | 3.5 (2.6–4.3) |
| Globulin (g/dL) | 2.0 ± 0.4 | 2.0 (1.4–2.8) | 1.8 ± 0.3 | 1.8 (1.2–2.4) | 2.1 ± 0.3 | 2.1 (1.6–2.7) |
| BUN (mg/dL) | 107 ± 52 | 107 (21–210) | 79 ± 68 | 54 (1–200) | 123 ± 30 | 120 (77–179) |
| Cholesterol (mg/dL)† | 379 ± 122 | 388 (164–614) | 320 ± 115 | 283 (195–543) | 206 ± 99 | 187 (94–486) |
| Glucose (mg/dL)* | 117 ± 26 | 118 (53–166) | 180 ± 91 | 159 (64–341) | 117 ± 10 | 117 (96–139) |
| Calcium (mg/dL)* | 6.7 ± 0.9 | 6.4 (5.1–8.5) | 6.0 ± 0.9 | 5.7 (5.0–7.9) | 7.1 ± 1.0 | 6.9 (5.9–9.4) |
| Bicarbonate (mEq/L) | 29 ± 4 | 28 (24–39) | 24 ± 9 | 27 (6–34) | 29 ± 5 | 28 (20–40) |
| Phosphorus (mg/dL) | 7.6 ± 1.7 | 7.9 (4.8–11.0) | 7.4 ± 1.4 | 6.9 (5.6–9.3) | 7.5 ± 2.4 | 7.4 (4.1–12.0) |
| Chloride (mEq/L) | 119 ± 4 | 119 (109–126) | 117 ± 7 | 116 (105–129) | 117 ± 4 | 117 (110–124) |
| Potassium (mEq/L) † | 3.3 ± 0.6 | 3.2 (2.4–4.8) | 3.4 ± 0.6 | 3.3 (2.3–4.3) | 3.9 ± 0.4 | 3.8 (3.0–4.6) |
| Sodium (mEq/L)* | 153 ± 3 | 152 (148–161) | 158 ± 5 | 159 (150–166) | 153 ± 4 | 154 (144–160) |
| Uric acid (mEq/L) *† | 1.8 ± 2.8 | 0.6 (0.3–10.1) | 6.2 ± 5.5 | 4.1 (0.5–19.7) | 0.5 ± 0.2 | 0.5 (0.3–0.8) |
| Anion gap (mEq/L)* † | 8.6 ± 3.8 | 8.5 (2.0–15.0) | 16.7 ± 7.5 | 16.5 (7.0–29.0) | 11.6 ± 5.3 | 11.5 (2.0–20.0) |
| WBC count (cells/μL) | 10,177 ± 9,449 | 7,350 (1,800–36,000) | 11,550 ± 7,494 | 12,800 (1,100–23,200) | 6,382 ± 2,832 | 6,300 (1,800–12,400) |
| Hct (%) | 30 ± 6 | 32 (16–39) | 33 ± 5 | 33 (21–40) | 30 ± 3 | 30 (26–37) |
| Heterophil (%) | 65 ± 14 | 69 (37–88) | 65 ± 14 | 70 (40–87) | 59 ± 15 | 62 (29–84) |
| Lymphocyte (%) | 28 ± 14 | 25 (10–60) | 23 ± 16 | 18 (4–56) | 33 ± 15 | 33 (10–64) |
| Monocyte (%) | 6 ± 6 | 4 (0–19) | 12 ± 11 | 13 (1–37) | 5 ± 5 | 4 (0–18) |
| Eosinophil (%)† | 0 ± 2 | 0 (0–7) | 0 ± 0 | 0 (0–1) | 3 ± 3 | 3 (0–11) |
| Heterophil count (cells/μL) | 7,235 ± 7,386 | 4,782 (814–26,640) | 7,959 ± 5,569 | 8,520 (462–16,472) | 3,731 ± 1,781 | 3,398 (957–7,252) |
| Lymphocyte count (cells/μL) | 2,379 ± 1,938 | 1,950 (270–7,920) | 2,136 ± 1,900 | 1,408 (572–6,496) | 2,090 ± 1,534 | 1,666 (171–5,580) |
| Monocyte count (cells/μL) | 552 ± 579 | 237 (0–2,050) | 1,425 ± 1,402 | 715 (11–3,876) | 285 ± 339 | 174 (0–1,364) |
| Eosinophil count (cells/μL)† | 6 ± 17 | 0 (0–68) | 21 ± 46 | 0 (0–128) | 199 ± 217 | 138 (0–630) |
Initial values are significantly (P < 0.05) different between turtles that survived and turtles that died.
Convalescent values are significantly (P < 0.05) different from initial values.
Clinicians used observations of behavior, appetite, physical examination, and the suite of information described to monitor response to treatment. In addition to negative culture results, typical signs of improvement in affected turtles included improvement of appetite, increased activity, resolution of lameness and joint swelling, radiographic remodeling of affected bone,22 and return of hematologic and plasma biochemical results to reference ranges. All turtles that survived were released to the wild after a median of 250 days (range, 150 to 471 days; mean ± SD, 242 ± 52 days) of hospitalization.
Discussion
Consistent with previous reports, the overall survival rate for 302 hospitalized cold-stunned Kemp's ridley turtles was 70%, and most deaths occurred within the first week of hospitalization.7,8 Most deaths that occurred after the first week of hospitalization were associated with Enterococcus spp infections, with fewer deaths associated with fungal infection, chronic renal failure, other bacterial infections, and an intestinal foreign body. In total, Enterococcus spp infections were identified in 17% of hospitalized Kemp's ridley sea turtles during this 6-year period. Clinicians were prompted to pursue diagnostic testing in these cases because of clinical problems that persisted or developed despite general medical management (eg, anorexia, signs of lethargy, lameness, joint swelling, and radiographic abnormalities). Diagnosis of enterococcal infection was made most often by bacteriologic culture of blood, but also by bacteriologic culture of tissues of the respiratory tract, skeletal system, and a skin lesion. In some cases, the validity of diagnosis was supported by identical culture results from > 1 tissue source as well as histopathologic findings.
Bacteremia, septicemia, and osteomyelitis caused by Enterococcus spp result in substantial morbidity and mortality rates in mammals and birds.15–17,31 Risk factors for development of Enterococcus spp infections in humans include serious underlying diseases, long hospital stays, renal insufficiency, neutropenia, hospitalization in intensive care units, and use of third-generation cephalosporins.15,31 Many of these factors were relevant to the hospitalized Kemp's ridley turtles. Results of previous studies11,28,32 indicate that after cold-stunning, Kemp's ridley turtles are often affected by life-threatening metabolic and respiratory tract disorders, lesions in 1 or more organ systems, reduced thyroid function, and potentially immunosuppressive plasma concentrations of corticosterone. Abnormally low body temperatures in reptiles can also reduce immune function.33 The initial body temperatures of turtles in this study were quite low but were similar to those reported in general for cold-stunned sea turtles27 and included a range of temperatures. Thus, it did not appear that turtles that developed Enterococcus spp infection uniformly had exceptionally low body temperatures, compared with other turtles. Owing to the generally high prevalence of gram-negative bacterial infections in debilitated reptiles, suitable pharmacokinetic profiles in several reptile species, and favorable clinical experiences, ceftazidime is widely used in the medical management of sea turtles.8,30 However, it is possible that the widespread use of ceftazidime or other third-generation cephalosporins could be contributing to the development of Enterococcus spp infections. Clinicians should consider this possibility when selecting antimicrobials for use in such cases.
The earliest diagnosis of Enterococcus spp infection in this study was made on day 10 of hospitalization. It remains unknown whether some cold-stunned Kemp's ridley turtles arrived at the hospital already infected by Enterococcus spp, given that none of the turtles had bacteriologic culture of blood performed on the day of admission. This is an obvious and important question for future investigation. In some cases, diagnosis of Enterococcus spp infection was not made until months into rehabilitation. In many of these later cases, the turtles had been clinically normal for weeks to months prior to development of clinical problems. Because cultures were not performed earlier, it cannot be determined when infections developed in these cases. Clinical observations suggested that these later diagnoses were made because of infections that developed later in rehabilitation, rather than failure to diagnose infection earlier in rehabilitation.
At the time of first diagnosis of Enterococcus spp infection, cultures revealed polymicrobial infections in 24 cases, including several gram-negative bacterial pathogens. Fourteen of the 24 turtles with polymicrobial infections survived, indicating that many polymicrobial infections were successfully managed. Furthermore, there was not a significant difference between the mortality rate of polymicrobial infection cases and those cases with only Enterococcus spp infection. Most Enterococcus spp isolates were identified to the species level as E faecalis, with a single isolate identified as E faecium, and a minority of isolates identified only to the genus level. Although it could not be determined from medical records why species identification was not completed for some cases, experience suggests that this was often due to failed communication between clinicians and diagnostic laboratories. For future cases, it is recommended that species identification and molecular strain identification be pursued to increase our understanding of which species in the Enterococcus genus and which strains are associated with infection in turtles.
In 2 cases, E faecalis was isolated on bacteriologic culture of blood of healthy turtles as part of general surveillance by a diligent clinician. These turtles were not treated with antimicrobials, and both turtles eventually had negative results of bacteriologic culture of blood. Nonpathological bacteremia has been reported in several reptile and fish species.20,34,35 It is possible that nonpathological bacteremia due to Enterococcus spp may occur in Kemp's ridley turtles and that bacteremia may resolve because of general improvement in health status during rehabilitation. Obtaining bacteriologic culture of blood from free-living healthy individuals of this species could be informative. In most cases with positive culture results, other clinical abnormalities were present that supported a diagnosis of septicemia rather than nonpathological bacteremia.
Histopathologic examination for 13 cases revealed intralesional or intravascular bacterial cocci in a variety of tissues. Combined with simultaneous culture of Enterococcus sp from these cases, histopathologic evidence supported the diagnosis of Enterococcus spp infection. In other cases, histopathologic lesions were suspected of being of bacterial origin (eg, meningitis), although bacteria were not seen. Histologic evaluation often does not reveal bacteria in cases of confirmed Enterococcus spp septicemia or visceral infection.36 Thus, the failure to histopathologically identify bacteria in some turtles does not negate the culture-based diagnosis of infection. As expected on the basis of the history of these turtles, other observed histopathologic lesions were typical of those generally seen as sequelae of cold stunning or were incidental lesions typical of free-ranging sea turtles (eg, parasitic granulomas).28
Hematologic and plasma biochemical data from this study indicated turtles that died were affected by dehydration, reduced renal function, nonspecific cellular injury, and deranged metabolic states, consistent with previous reports11,29 for cold-stunned individuals of this species. Although we cannot state that these findings are specifically related to Enterococcus spp infection, clinicians should consider Enterococcus spp infection in turtles with persistent hematologic and biochemical abnormalities despite medical management. It is possible that Enterococcus spp infection contributed to these derangements or that these derangements made the turtles more susceptible to Enterococcus spp infection. In turtles that survived, convalescent data revealed that a wide variety of variables had returned to reference ranges, consistent with results of previous studies,29 indicating improved general health concurrent with treatment.
Analysis of water chemical analysis and water microbiological results indicated that the water quality of hospital tanks was generally good. Median values for all variables were within reference ranges of the involved institutions and published guidelines, indicating that turtles were most often exposed to desired environmental conditions.37 However, examination of ranges and means as well as individual turtle data indicated that reference ranges for many variables (eg, nitrogenous wastes and bacterial concentrations) were transiently exceeded in many cases. For example, 27 of 37 turtles for which data were available resided in systems that transiently exceeded the maximum desired range for bacterial concentrations. Although it is possible that exposure to such conditions may have influenced the development of infection, it must be noted that most turtles hospitalized in these same systems did not develop Enterococcus spp infections. General guidelines for environmental variables of captive sea turtles are extrapolated from requirements of marine fish and recommendations for other species (eg, beach closure limits for humans). However, no controlled study has assessed the effects of exposing sea turtles to various environmental concentrations of ammonia, bacteria, and other commonly measured variables. These issues should be the subject of further investigation, and epidemiological studies are needed to assess the influence of these potential risk factors during hospitalization of sea turtles. The origins of the Enterococcus spp isolates that caused infections in these cases remain unknown. Documentation of Enterococcus spp infection in numerous cohorts of turtles, aquatic systems, hospital locations, institutions, and laboratories provided evidence that a single source of infection was unlikely. Results of PFGE analysis of a limited number of clinical isolates revealed that multiple PFGE types were present, also supporting this conclusion. On the basis of available data, it is likely that multiple independent infections occurred because of bacteria that originated from the environment or from turtles’ resident bacterial flora, under the influence of a variety of risk factors.
In humans affected by serious Enterococcus spp infections, treatment commonly includes penicillin or ampicillin used in combination with an aminoglycoside for synergistic effect.38 In the turtles described in this study, this combination was found to be the most successful single course of treatment. It remains unclear at this time whether it is important to use an aminoglycoside for synergistic effect in the treatment of Enterococcus spp infection in turtles, but clinical experience suggests that it is safe to do so. There are presently no pharmacokinetic studies for penicillin, ampicillin, amoxicillin, amoxicillin–clavulanic acid, or aminoglycosides in marine turtles, and thus, treatment recommendations are based on clinical experiences in recent years, with consideration of pharmacokinetic data for other species. At present, the protocol used at the New England Aquarium for management of Enterococcus spp infections in Kemp's ridley turtles includes ampicillin (30 mg/kg, IM, q 24 h) for turtles that are not eating voluntarily, which is later changed to amoxicillin–clavulanic acid (30 mg/kg, PO, q 24 h) once turtles are eating consistently. In many cases, at clinician discretion or owing to polymicrobial infection, amikacin (10 mg/kg, IM, q 3 d) is used concurrently, with due consideration of potential nephrotoxicity. Treatment is continued for at least 1 month beyond negative culture results, with monitoring of bacteriologic culture of blood every 2 to 4 weeks.
One notable deficiency in the management of 14 cases described in this study was the failure of clinicians to perform cultures after completion of antimicrobial treatment. This appears to have been based on presumed treatment success resulting from resolution of clinical signs. Although clinicians used a variety of clinical data, clinical judgment, and observations over time to assess the health of turtles after treatment, it would have been informative to confirm that culture results remained negative after completion of antimicrobial treatment. We currently recommend that at least 1 negative result of bacteriologic culture of blood be obtained at least 1 month after completion of antimicrobial treatment.
In the present study, 50 cases of Enterococcus spp infection were documented during the rehabilitation of cold-stunned Kemp's ridley sea turtles. Diagnosis was made most commonly by bacteriologic culture of blood but also by bacteriologic culture of bones, joints, respiratory tract, and skin lesions. Affected turtles had clinical abnormalities such as anorexia, signs of lethargy, lameness, and radiographic evidence of osteomyelitis and pneumonia. Treatment was effective in most cases. Clinicians should be aware of the possibility of Enterococcus spp infections in debilitated marine turtles and should pursue appropriate diagnostic testing when indicated.
ABBREVIATIONS
| MIC | Minimum inhibitory concentration |
| PFGE | Pulsed-field gel electrophoresis |
BBL Septicheck BHI, Beckton, Dickinson & Co, Sparks, Md.
IDEXX Laboratories, North Grafton, Mass.
VITEK Colorimeter Susceptibility Cards, bioMerieux Inc, Durham, NC.
Connecticut Veterinary Medical Diagnostic Laboratory, Department of Pathobiology and Veterinary Science, College of Agriculture and Natural Resources, University of Connecticut, Storrs, Conn.
Marine Animal Disease Laboratory, College of Veterinary Medicine, University of Florida, Gainesville, Fla.
Enterolert, IDEXX Laboratories Inc, Westbrook, Me.
PASW Statistics, version 18.0, SPSS Inc, Chicago, Ill.
References
1. Márquez-Millan R, Burchfield PM, Díaz-Flores J, et al. Status of the Kemp's ridley turtle, Lepidochelys kempii. Chelonian Conserv Biol 2005; 4: 761–766.
2. Morreale SJ, Standora EA. Western North Atlantic waters: crucial developmental habitat for Kemp's ridley and loggerhead sea turtles. Chelonian Conserv Biol 2005; 4: 872–882.
3. Morreale SJ, Meylan AB, Sadove SS, et al. Annual occurrence and winter mortality of marine turtles in New York waters. J Herpetol 1992; 26: 301–308.
4. Still BM, Griffin CR, Prescott R. Climatic and oceanographic factors affecting daily patterns of juvenile sea turtle cold-stunning in Cape Cod Bay, Massachusetts. Chelonian Conserv Biol 2005; 4: 883–890.
5. Witherington BE, Ehrhart LM. Hypothermic stunning and mortality of marine turtles in the Indian River lagoon system, Florida. Copeia 1989; 696–703.
6. Bentivegnal F, Breber P, Hochscheid S. Cold stunned loggerhead turtles in the south Adriatic Sea. Mar Turtle Newsl 2000; 97: 1–3.
7. Gerle E, DiGiovanni R, Pisciotta R. A fifteen year review of cold-stunned sea turtles in New York waters, in Proceedings. 18th Int Sea Turtle Symp 2000; 222–224.
8. Wyneken J, Mader DR, Weber ES, et al. Medical care of seaturtles. In: Mader D, ed. Reptile medicine and surgery. 2nd ed. St Louis: Elsevier, 2006; 972–1007.
9. Turnbull BS, Smith CR, Stamper MA. Medical implications of hypothermia in threatened loggerhead (Caretta caretta) and endangered Kemp's ridley (Lepidochelys kempii) and green (Chelonia mydas) sea turtles, in Proceedings. Joint Conf Am Assoc Zoo Vet Int Assoc Aquat Anim Med 2000; 31–35.
10. Stacy NI, Innis CJ, Hernandez JA. Development and evaluation of three mortality prediction indices for cold-stunned Kemp's ridley sea turtles (Lepidochelys kempii). Cons Physiol [serial online]. 2013; 1. Available at: conphys.oxfordjournals.org/content/1/1/cot003.short?rss=1. Accessed Nov 3, 2013.
11. Keller KA, Innis CJ, Tlusty MF, et al. Metabolic and respiratory derangements associated with death in cold-stunned Kemp's ridley turtles (Lepidochelys kempii): 32 cases (2005)2009). J Am Vet Med Assoc 2012; 240: 317–323.
12. Stockman J, Innis CJ, Solano M, et al. Prevalence, distribution, and progression of radiographic abnormalities in the lungs of cold-stunned Kemp's ridley sea turtles (Lepidochelys kempii): 89 cases (2002)2005). J Am Vet Med Assoc 2013; 242: 675–681.
13. Manire CA, Rhinehart HL, Sutton DA, et al. Disseminated mycotic infection caused by Colletotrichum acutatum in a Kemp's ridley sea turtle (Lepidochelys kempii). J Clin Microbiol 2002; 40: 4273–4280.
14. Pare JA, Sigler L, Rosenthal, et al. Microbiology: fungal and bacterial diseases of reptiles. In: Mader D, ed. Reptile medicine and surgery. 2nd ed. St Louis: Elsevier, 2006; 217–238.
15. Cervera C, Almela M, Martínez-Martínez JA, et al. Risk factors and management of gram-positive bacteraemia. Int J Antimicrob Agents 2009; 34 (suppl 4):S26–S30.
16. Orsini JA, Snooks-Parsons C, Stine L, et al. Vancomycin for the treatment of methicillin-resistant staphylococcal and enterococcal infections in 15 horses. Can J Vet Res 2005; 69: 278–286.
17. Devriese LA, Chiers K, De Herd P, et al. Enterococcus hirae infections in psittacine birds: epidemiological, pathological and bacteriological observations. Avian Pathol 1995; 24: 523–531.
18. Mundt JO. Occurrence of enterococci in animals in a wild environment. Appl Microbiol 1963; 11: 136–140.
19. Sheridan BS, Wilson GR, Weldon PJ. Aerobic bacteria from the skin of the rattlesnake, Crotalus atrox. J Herpetol 1989; 23: 200–202.
20. Innis C, Merigo C, Dodge K, et al. Health evaluation of leather-back turtles (Dermochelys coriacea) in the northwestern Atlantic during direct capture and fisheries gear disentanglement. Chelonian Conserv Biol 2010; 9: 205–222.
21. Schröter M, Heckers KO, Rüschoff B, et al. Severe case of spinal osteomyelitis due to Enterococcus sp. in a three-year old rhinoceros horned viper, Bitis nasicornis. J Herpetol Med Surg 2005; 15: 53–56.
22. Solano M, Innis C, Smith C, et al. Scintigraphic and radiographic evaluation of appendicular skeletal lesions in cold-stunned Kemp's ridley sea turtles. Vet Radiol Ultrasound 2008; 49: 388–394.
23. Johnson J. Enteritis in a pair of juvenile Aldabra tortoises, Geochelone gigantea, in Proceedings. Assoc Reptil Amphib Vet 8th Ann Conf 2001; 147–152.
24. Anderson ET, Kennedy-Stoskopf S, Sandy JR, et al. Squamous cell carcinoma with vascular invasion in a diamondback rattlesnake (Crotalus adamanteus). J Zoo Wildl Med 2010; 41: 745–748.
25. Clinical and Laboratory Standards Institute. Performance standards for antimicrobial susceptibility testing; twentieth informational supplement. M100)S20. Wayne, Pa: Clinical and Laboratory Standards Institute, 2010.
26. Baden LR, Critchley IA, Sahm DF, et al. Molecular characterization of vancomycin-resistant enterococci repopulating the gastrointestinal tract following treatment with a novel glycolipodepsipeptide, ramoplanin. J Clin Microbiol 2002; 40: 1160–1163.
27. Baden LR, Thiemke W, Skolnik A, et al. Prolonged colonization with vancomycin-resistant Enterococcus faecium in long-term care patients and the significance of “clearance.” Clin Infect Dis 2001; 33: 1654–1660.
28. Innis C, Nyaoke A, Williams C III, et al. Pathologic and parasitologic findings of cold-stunned Kemp's ridley sea turtles (Lepidochelys kempii) stranded on Cape Cod, MA, 2001)2006. J Wildl Dis 2009; 45: 594–610.
29. Innis CJ, Ravich JB, Tlusty MF, et al. Hematologic and plasma biochemical findings in cold-stunned Kemp's ridley turtles: 176 cases (2001)2005). J Am Vet Med Assoc 2009; 235: 426–432.
30. Innis CJ, Ceresia ML, Merigo C, et al. Single-dose pharmacokinetics of ceftazidime and fluconazole during concurrent clinical use in cold-stunned Kemp's ridley turtles (Lepidochelys kempii). J Vet Pharmacol Ther 2012; 35: 82–89.
31. Jang HC, Lee S, Song KH, et al. Clinical features, risk factors and outcomes of bacteremia due to enterococci with high-level gentamicin resistance: comparison with bacteremia due to enterococci without high-level gentamicin resistance. J Korean Med Sci 2010; 25: 3–8.
32. Hunt KE, Innis C, Rolland RM. Corticosterone and thyroxine in cold-stunned Kemp's ridley sea turtles (Lepidochelys kempii). J Zoo Wildl Med 2012; 43: 479–493.
33. Zimmerman LM, Vogel LA, Bowden RM. Understanding the vertebrate immune system: insights from the reptilian perspective. J Exp Biol 2010; 213: 661–671.
34. Mylniczenko ND, Harris B, Wilborn RE, et al. Blood culture results from healthy captive and free-ranging elasmobranchs. J Aquat Anim Health 2007; 19: 159–167.
35. Hanel R, Heard DJ, Ellis G, et al. Isolation of Clostridium spp from the blood of captive lizards: real or pseudobacteremia? Bull Assoc Reptil Amphib Vet 1999; 9 (2):4–8.
36. Gentry-Weeks C, Estay M, Loui C, et al. Intravenous mouse infection model for studying the pathology of Enterococcus faecalis infections. Infect Immun 2003; 71: 1434–1441.
37. US Department of the Interior Fish and Wildlife Service. Standard permit conditions for care and maintenance of captive sea turtles. Available at: www.fws.gov/northflorida/seaturtles/Captive_Forms/20130213_revised%20_standard_permit_conditions_for_captive_sea_turtles.pdf. Accessed Nov 3, 2013.
38. Caputo GM, White S, Weitekamp MR. Infections due to antibiotic-resistant gram-positive cocci. J Gen Intern Med 1993; 8: 626–634.
