- Journal List
- J Vet Diagn Invest
- v.30(3); 2018 May
- PMC6505802
As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsem*nt of, or agreement with, the contents by NLM or the National Institutes of Health.
Learn more: PMC Disclaimer | PMC Copyright Notice
J Vet Diagn Invest. 2018 May; 30(3): 423–429.
Published online 2018 Feb 13. doi:10.1177/1040638718757819
PMCID: PMC6505802
PMID: 29436286
Duane T. March,1 Kimberly Vinette-Herrin, Andrew Peters, Ellen Ariel, David Blyde, Doug Hayward, Les Christidis, and Brendan P. Kelaher
Author information Copyright and License information PMC Disclaimer
Associated Data
- Supplementary Materials
Abstract
To improve understanding of pathophysiologic processes occurring in green seaturtles (Chelonia mydas) stranded along the east coast ofAustralia, we retrospectively examined the hematologic and biochemical bloodparameters of 127 green turtles admitted to 2 rehabilitation facilities, DolphinMarine Magic (DMM) and Taronga Zoo (TZ), between 2002 and 2016. The predominantsize class presented was small immature animals (SIM), comprising 88% and 69% ofadmissions to DMM and TZ, respectively. Significant differences in bloodprofiles were noted between facility, size, and outcome. Elevated levels ofaspartate aminotransferase (AST) and heterophils were poor prognostic indicatorsin animals from TZ, but not DMM. SIM animals at both institutions had lowerprotein levels than large older (LO) animals. SIM animals at DMM also had lowerhematocrit and monocyte concentration; SIM animals at TZ had lower heterophilcounts. Urea was measured for 27 SIM animals from TZ, but the urea-to-uric acidratio was not prognostically useful. Strong correlations were seen between ASTand glutamate dehydrogenase (GDH; r = 0.68) and uric acid andbile acids (r = 0.72) in the 45 SIM animals from DMM in whichadditional analytes were measured. χ2 contingency tests showed thatthe most recently published reference intervals were not prognostically useful.A paired t-test showed that protein levels rose and heterophilnumbers fell in the 15 SIM animals from TZ during the rehabilitation process.Our results indicate that further work is required to identify reliableprognostic biomarkers for green turtles.
Keywords: Anemia, biochemistry, cachexia, green sea turtles, hematology, immunosuppression, kidney, liver
Introduction
The green sea turtle (Chelonia mydas) is 1 of 7 species of marineturtles found in circumtropical regions.8 Hunting these animals for their meat and harvesting their eggs has seen greenturtle populations decline drastically.18 More recent anthropogenic threats include entanglement in fishing gear,9 marine debris,30,31 and ocean climate change.20 In response to declining populations, green turtles and the beaches on whichthey nest are protected in many countries, and global conservation priorities havebeen established.35 Following the introduction of these protections, some populations havedemonstrated promising signs of recovery.6
Despite protection from key threats, there has been an increase in the reporting ofdisease in green sea turtles.36 These diseases include a range of infectious and non-infectious conditionssuch as fibropapillomatosis,11 spirorchidiasis,17 coccidiosis,22 gastrointestinal disorders,17 cachexia, and a condition described as “hepatorenal insufficiency.”14 Many of the infectious pathogens involved in these diseases have been presentin the marine environment for millions of years.23 Environmental factors may be contributory in places where the incidence ofdisease is higher (e.g., polluted environments).3,34
To optimize green sea turtle rehabilitation and characterize disease, there is a needto accurately assess the health of individual stranded animals and ascertain thecausal factors associated with these stranding events. To date, health assessmenttechniques are subjective, given that there is poor correlation between clinicalsigns and histologic disease,17 and that quantitative assessment of body condition appears to be inconsistentwith the subjective appearances of health.16 To address the shortcomings associated with clinical examination, numerousstudies have examined hematologic and biochemical analytes in free-ranging greenturtles, resulting in the creation of reference intervals (RIs) to define healthyindividuals; however, the resulting RIs can differ in different parts of theworld.2,7,16,28 Other modalities such asplasma protein electrophoresis13,15,24 and assessment of thecorrelation between survival and biochemical analytes in sea turtles undergoingrehabilitation are being investigated.25
We examined archived hematologic and biochemical data from 2 rehabilitation hospitalsover 14 y to investigate 1) the prognostic reliability of the most recentlypublished local RIs for this species; 2) pathophysiologic changes within strandedanimal blood profiles, and 3) changes that occurred within these blood profilesduring rehabilitation.
Materials and methods
All data examined in our study were analyzed retrospectively, following thecollection of samples by veterinarians, acting under the authority of the VeterinaryPractice Act 2003 (https://goo.gl/k54Shx). The collection of samples was part ofroutine clinical investigation to assess the health of stranded green sea turtlesadmitted to rehabilitation hospitals at Dolphin Marine Magic (DMM; Coffs Harbour,NSW, Australia) and Taronga Zoo (TZ; Sydney, NSW), between 2002 and 2016.
All of the turtles in our study were considered clinically “unhealthy” at the time ofpresentation, based on their appearance and behavior. Blood samples were collectedfrom the dorsal occipital sinus of each turtle within 7 d of presentation, followingpreviously described methods using a 5-mL syringe with a 21-gauge, 2.5-cm needle.27 Blood samples were transferred into sterile vacutainers containing lithiumheparin and refrigerated until analysis. All analyses occurred within 24 h of samplecollection. Each sample was assessed during collection for any evidence of lymphaticcontamination, and if such contamination was suspected, the sample was discarded andrecollected.
Blood samples collected from DMM were submitted to Laverty Vetnostics Pathology(North Ryde, NSW), a National Association Testing Authorities–accredited commercialpathology laboratory. Samples collected from TZ were analyzed on-site at the TarongaWildlife Hospital. The same staff was present at both laboratories for the durationof the study.
Both facilities used a microhematocrit centrifuge to measure hematocrit (Hct)following centrifugation (5 min, 6,000 × g). Blood smears were madefrom the heparinized blood within 1 h of collection, and the total white blood cell(TWBC) and differential WBC counts were calculated.33 For samples collected at DMM, this was achieved via a leukocyte estimate anddifferential from the blood smear,19 and for samples collected at TZ, via an improved Neubauer hemocytometer.26 Where anemia was noted, red blood cell morphology and reticulocyte countswere performed to classify the anemia as regenerative or nonregenerative. Theinstrumentation used for the biochemical analysis varied between laboratories. Tomeasure uric acid, glucose, aspartate aminotransferase (AST), protein, and creatinekinase (CK), samples from DMM were analyzed by 1 of 2 analyzers (Modular [Pmodules], Modular Evo [P modules], Roche Diagnostics, Basel, Switzerland; Cobas 8000[c502 and c702 modules], Roche Diagnostics) as equipment was upgraded partwaythrough the study. Samples collected from TZ were analyzed in-house (VetScanVS2analyzer, REM Systems, Sydney, Australia). Urea was only measured from TZ samples(Reflotron, DTS Diagnostics, Sydney, Australia).
Permutational analysis of variance (PERMANOVA)4 across blood profiles demonstrated significant mean differences betweenlaboratories (p < 0.01; Fig. 1). All further analysis was conductedseparately on data from DMM and data from TZ. Based on the distribution of size,data were divided into previously described age classes based on curved carapacelength (CCL). Animals with CCL <65 cm were considered small immature (SIM)animals, and animals with a CCL >65 cm were considered large older (LO) animalsfor further analysis.17
Figure 1.
Box and whisker plots of the interquartile range and outliers for eachanalyte measured at Dolphin Marine Magic (DMM) and Taronga Zoo (TZ).
PERMANOVA multivariate analyses using 2 orthogonal factors (size class: SIM vs. LO;outcome: released vs. deceased) were used to compare the similarity of bloodprofiles between factors for each facility. PERMANOVA univariate analyses were alsoundertaken to investigate the significance of specific hematologic and biochemicalresponse variables. Further PERMANOVA analyzes were carried out on 27 SIM turtlesfrom TZ in which urea was measured to test whether the urea-to-uric acid ratiodiffered with outcome. All PERMANOVA analyses were performed using the softwarePRIMER 7.0.11 (http://www.primer-e.com/) and used Euclidean distances derived fromdata that were normalized using the Normalize Variable function inthe software package.
Hypotheses about the relationships among the analytes that have the potential toarise from more than one tissue were tested with a Pearson correlation coefficient(r) for 45 SIM animals from DMM. This included the analysis ofarchived results for the additional analytes (alanine aminotransferase, ALT;alkaline phosphatase, ALP; bile acids; cholesterol; and glutamate dehydrogenase,GDH), which were measured on either the Roche Modular Evo or the Roche Cobas 8000depending on the time period within the study that the samples were analyzed.
We used χ2 contingency tests to assess the prognostic capacity of the RIscurrently accepted as describing healthy green sea turtles in Australian waters16 by evaluating whether the proportion of surviving animals was greater forthose within the existing local RIs16 compared to those outside these RIs for DMM and TZ animals separately. Pairedt-tests were used to evaluate the hematologic and biochemicalchanges that occurred during rehabilitation. These analyses included responsevariables from 15 SIM animals from TZ, in which blood was collected and analyzedwithin the first 7 d of presentation and again after a minimum of 28 d ofrehabilitation.
Results
Included in our study were 56 animals from DMM and 71 animals from TZ, with SIManimals representing 88% and 69% of admissions, respectively. The survival rate atDMM and TZ was 65% and 27%, respectively, for SIM animals, and 29% and 18%,respectively, for LO animals. A significant difference (p <0.05) was seen between the blood profiles of released versus deceased animals at TZ(Table 1). Animalsthat were released initially had lower heterophil counts (p = 0.02)and levels of AST (p = 0.03). No significant differences regardingoutcome were seen at DMM for multivariate or univariate analyses (Table 2). At both DMM andTZ, SIM animals had lower levels of protein (p < 0.01) and(p < 0.01), respectively. At DMM, SIM animals also had lowerlevels of monocytes (p < 0.01) and Hct (p =0.04), whereas at TZ, SIM animals had lower heterophil counts (p =0.01). There were no significant interactions between the factors at eitherfacility.
Table 1.
Hematologic and biochemical results, by green sea turtle size class, outcome,and facility.
| SIM | LO | |||||||
|---|---|---|---|---|---|---|---|---|
| Released | Deceased | Released | Deceased | |||||
| DMM | TZ | DMM | TZ | DMM | TZ | DMM | TZ | |
| n | 32 | 13 | 17 | 36 | 2 | 4 | 5 | 18 |
| TWBC (× 109/L) | 17.9 | 6.1 | 13.7 | 8.1 | 19.0 | 8.0 | 12.4 | 8.6 |
| 1.7 | 0.9 | 1.7 | 1.0 | 8.0 | 1.7 | 3.4 | 0.9 | |
| Heterophils (× 109/L) | 13.2 | 2.8 | 9.8 | 4.2 | 10.2 | 14.0 | 5.0 | 3.9 |
| 1.4 | 0.5 | 1.5 | 0.6 | 1.9 | 9.2 | 1.7 | 0.6 | |
| Lymphocytes (× 109/L) | 2.8 | 1.3 | 2.2 | 1.8 | 1.5 | 2.1 | 2.2 | 2.3 |
| 0.5 | 0.3 | 0.3 | 0.2 | 0.9 | 0.6 | 0.6 | 0.4 | |
| Monocytes (× 109/L) | 1.4 | 1.9 | 1.3 | 1.7 | 6.6 | 0.9 | 5.3 | 2.0 |
| 0.3 | 1.0 | 0.4 | 0.4 | 5.6 | 0.2 | 2.0 | 0.5 | |
| Hct (L/L) | 0.23 | 0.33 | 0.24 | 0.34 | 0.31 | 0.31 | 0.34 | 0.32 |
| 0.0 | 0.0 | 0.0 | 0.0 | 0.1 | 0.0 | 0.0 | 0.0 | |
| Uric acid (µmol/L) | 0.2 | 0.1 | 0.3 | 0.2 | 0.2 | 0.2 | 0.2 | 0.1 |
| 0.1 | 0.0 | 0.0 | 0.0 | 0.1 | 0.0 | 0.1 | 0.0 | |
| Glucose (mmol/L) | 4.2 | 5.8 | 2.2 | 6.2 | 5.8 | 5.9 | 3.7 | 7.8 |
| 0.7 | 0.8 | 0.5 | 0.9 | 2.9 | 1.4 | 1.4 | 1.1 | |
| AST (µkat/L) | 3.2 | 4.2 | 5.1 | 6.6 | 3.8 | 2.2 | 5.0 | 5.1 |
| 0.3 | 0.7 | 0.9 | 0.8 | 0.4 | 0.7 | 1.3 | 0.5 | |
| Protein (g/L) | 23.9 | 30.0 | 24.2 | 33.3 | 47.0 | 40.8 | 50.8 | 52.7 |
| 1.5 | 1.7 | 1.7 | 2.6 | 7.0 | 6.8 | 4.6 | 3.1 | |
| CK (µkat/L) | 74 | 77 | 107 | 156 | 111 | 83 | 168 | 119 |
| 19 | 26 | 24 | 46 | 35 | 34 | 133 | 45 | |
Open in a separate window
AST = aspartate aminotransferase; CK = creatine kinase; DMM = DolphinMarine Magic; Hct = hematocrit; LO = large older turtles; SIM = smallimmature turtles; TWBC = total white blood cells; TZ = Taronga Zoo.Numbers in italics are 1 standard error of each mean.
Table 2.
PERMANOVA multivariate and univariate results for facilities, green turtlesize class, and outcome.
| Facility | Size | Outcome | |||
|---|---|---|---|---|---|
| DMM | TZ | DMM | TZ | ||
| Blood profiles (multivariate) | <0.01 | <0.01 | 0.01 | 0.49 | <0.05 |
| TWBC | <0.01 | 0.95 | 0.41 | 0.33 | 0.53 |
| Heterophils | <0.01 | 0.20 | 0.01 | 0.25 | 0.02 |
| Lymphocytes | 0.04 | 0.58 | 0.15 | 0.91 | 0.48 |
| Monocytes | 0.57 | <0.01 | 0.75 | 0.91 | 0.53 |
| Hct | <0.01 | 0.04 | 0.45 | 0.73 | 0.73 |
| Uric acid | 0.06 | 0.78 | 0.46 | 0.32 | 0.06 |
| Glucose | <0.01 | 0.36 | 0.51 | 0.08 | 0.39 |
| AST | <0.01 | 0.61 | 0.10 | 0.23 | 0.03 |
| Protein | <0.01 | <0.01 | <0.01 | 0.69 | 0.05 |
| CK | 0.42 | 0.44 | 0.87 | 0.38 | 0.46 |
Open in a separate window
AST = aspartate aminotransferase, CK = creatine kinase; DMM = DolphinMarine Magic; Hct = hematocrit; TWBC = total white blood cells; TZ =Taronga Zoo.
For the 27 SIM animals from TZ in which the urea-to-uric acid ratio was measured,there was no significant difference between animals that lived and animals that died(p = 0.96). For the 45 SIM animals from DMM in which additionalbiochemical analytes were measured, strong relationships where seen between AST andGDH (r = 0.68), and uric acid and bile acids (r =0.72; Supplementary Table 1).
χ2 contingency tests comparing the survival rate of LO animals in andoutside of the RIs was not possible because the survival rate of these animals wasvery low. For SIM animals, there was no significant difference for any analyte inthe proportion of animals that died between the turtles within and outside of theRIs currently accepted as describing healthy animals (Supplementary Table 2). During the rehabilitation period,significant increases were demonstrated in protein (p < 0.01)and decreases in heterophils (p = 0.03) using a pairedt-test (Table 3) for 15 SIM animals at TZ.
Table 3.
Hematologic and biochemical results for 15 SIM animals at the Taronga Zoocollected at 7 and 28 d after admission.
| 7 d | 28 d | p value | |
|---|---|---|---|
| TWBC (× 109/L) | 7.7 | 6.3 | 0.19 |
| 0.9 | 1.1 | ||
| Heterophils (× 109/L) | 3.3 | 1.8 | 0.03 |
| 0.5 | 0.3 | ||
| Lymphocytes (× 109/L) | 2.4 | 3.4 | 0.23 |
| 0.5 | 0.7 | ||
| Monocytes (× 109/L) | 1.8 | 1.1 | 0.39 |
| 0.8 | 0.3 | ||
| Hct (L/L) | 0.27 | 0.29 | 0.26 |
| 0.0 | 0.0 | ||
| Uric acid (µmol/L) | 0.1 | 0.0 | 0.07 |
| 0.0 | 0.0 | ||
| Glucose (mmol/L) | 5.7 | 7.1 | 0.19 |
| 0.9 | 0.4 | ||
| AST (µkat/L) | 5.6 | 6.0 | 0.82 |
| 0.8 | 1.1 | ||
| Protein (g/L) | 23.7 | 37.6 | <0.01 |
| 1.8 | 2.7 | ||
| CK (µkat/L) | 107 | 52 | 0.19 |
| 29 | 26 |
Open in a separate window
AST = aspartate aminotransferase; CK = creatine kinase; Hct = hematocrit;TWBC = total white blood cells. Numbers in italics are 1 standard errorof each mean.
Discussion
The most recently published biochemical and hematologic RIs for green sea turtles in Australia16 did not provide prognostic information for clinicians. This may be because ofthe inclusion of animals with subclinical disease in the collection of data fromseemingly healthy free-ranging groups, given that previous studies have shown littlecorrelation between the clinical presentation of an animal and the histologicfindings at postmortem.14 Although rigorous statistical techniques have been used to address thisproblem and eliminate outliers prior to calculating RIs,16,28 the presence of widespreadsubclinical disease in a sampled population may not generate outliers. This issupported by the postmortem detection of spirorchidiasis in the majority ofAustralian green sea turtles,17 including a prevalence rate of 98% from the site from which the data for thecurrent RIs were collected.21 Alternatively, the poor prognostic capabilities of the current RIs seen inour study may be the result of secondary disease processes such as immunosuppressionand dehydration occurring in parallel, and effectively masking primary diseaseprocesses such as infection and/or inflammation and hypoproteinemia and/or anemia,respectively.
The observed differences between laboratories may be the result of differentanalytical techniques or may reflect differing disease processes in green seaturtles regionally. The elevated glucose in TZ animals compared to DMM animals islikely to reflect the increased time between sample collection and sample analysisthat would occur in samples from DMM, as they were not analyzed on-site and thelithium heparin vacutainer would not conserve glucose.12 Given that different techniques were used to assess the TWBC and differentialwhite cell count between facilities, the origin of the observed differences isunclear. However, the significant difference between Hct at each facility, despitethe use of the same methodology, may indicate that regional differences do occur. TZis located adjacent to more temperate waters, and environmental factors such aswater temperature may play an important role.
Overall, the occurrence of disease predominantly in SIM animals was consistent withprevious studies.11,17 The reason for the over-representation of this demographic isunknown; previous studies have suggested that it could be the result of increasedexposure to parasites because of habitat use or immunological naivety.17 However, it is also possible that the size structure of the turtlespresenting is representative of the local population.11 SIM animals recruit to the neritic feeding grounds5 following a pelagic period known as the “lost years,”29 and it is possible that environmental components could play a role in thedevelopment of disease.
Despite the observation of both regenerative and non-regenerative anemias in animalsundergoing rehabilitation, the Hct of stranded animals was not significantlyassociated with survival and did not change significantly during rehabilitation.This finding contradicts previous work that reported lower Hct values in strandedanimals compared with that of healthy free-ranging animals (Work T, et al. Causes ofgreen turtle (Chelonia mydas) morbidity and mortality in Hawaii.Proc Ann Sea Turtle Symp; March 1997; Orlando, FL). However, there is a markeddifference between Hct levels accepted as healthy in Australia16,37 compared toother regions.2,7,28 This may be a reflection ofthe prevalence of spirorchidiasis in Australian waters,17 given that infection with this parasite in loggerhead turtles(Carretta caretta) has been shown to induce lesionshistologically similar38 to the inflammation-associated anemia found in humans infected with schistosomes.10
The observed monocytosis in LO animals at DMM, relative to SIM animals, may reflect adifferent disease process affecting these animals or differing immunocompetence inthe older animals. The bulk of the LO animals in the DMM data set were presented tocare during a regionally significant coccidiosis outbreak, and the relativemonocytosis in this age class may represent the cell-mediated inflammation that isassociated with these acute infections.
The correlation between uric acid and bile acids may represent the presence ofconcurrent renal insufficiency and compromised hepatic function. This is anobservation that had been made and described as hepatorenal insufficiency in aprevious study.14 Compromised cardiac output, as can occur secondary to systemic diseaseprocesses such as cachexia, may result in generalized hypoperfusion. This would leadto hepatorenal insufficiency and potentially generate the observed changes. Thestrong correlation between AST and GDH supports the hypothesis that elevated ASTobserved in deceased TZ animals is related to hepatic compromise as opposed toskeletal muscle damage. However, care should be taken with this interpretation,given that a temporal lag in the AST elevation after muscle damage could lead to aloss of correlation with CK.
Given the potential for dehydration, the interpretation of protein levels varies inrelation to an animal’s clinical condition. Hypoproteinemia has been demonstrated inanimals that are in poor health.1,16 This finding is supported byour study, given the significant increase in protein levels that occurred duringrehabilitation, which ultimately rose to normal levels.1,2,7,16,28,32,37
Supplementary Material
Supplementary material:
Click here to view.(711K, pdf)
Acknowledgments
We thank the staff of Dolphin Marine Magic, with particular thanks to KieranMarshall, and the staff at Taronga Wildlife Hospital, with particular thanks to theLaboratory Manager, Paul Thompson.
Footnotes
Declaration of conflicting interests: The authors declared no potential conflicts of interest with respect to theresearch, authorship, and/or publication of this article.
Funding: DT March acknowledges the support of an Australian Post-Graduate Award, funded bythe Australian Federal Government.
References
1. Aguirre AA, Balazs GH.Blood biochemistry values of green turtles, Cheloniamydas, with and without fibropapillomatosis.Comp Haematol Int2000;10:132–137. [Google Scholar]
2. Aguirre AA, et al.Adrenal and hematological responses to stress injuvenile green turtles (Chelonia mydas) with and withoutfibropapillomas. Physiol Biochem Zool1995;68:831–854. [Google Scholar]
3. Aguirre AA, Lutz PL.Marine turtles as sentinels of ecosystem health: isfibropapillomatosis an indicator?EcoHealth2004;3:275–283. [Google Scholar]
4. Anderson M, et al.PERMANOVA+ for PRIMER: Guide to Software andStatistical Methods. Plymouth, UK:PRIMER-E,2008. [Google Scholar]
5. Arthur KE, et al.Ontogenetic changes in diet and habitat use ingreen sea turtle (Chelonia mydas) lifehistory. Marine Ecol Prog Ser2008;362:303–311. [Google Scholar]
6. Balazs GH, Chaloupka M.Thirty-year recovery trend in the once depleted Hawaiian greensea turtle stock. Biol Conserv2004;117:491–498. [Google Scholar]
7. Bolten AB, Bjorndal KA.Blood profiles for a wild population of greenturtles (Chelonia mydas) in the southern Bahamas:size-specific and sex-specific relationships. JWildl Dis1992;28:407–413. [PubMed] [Google Scholar]
8. Bowen BW, et al.Global population structure and natural historyof the green turtle (Chelonia mydas) in terms ofmatriarchal phylogeny. Evolution1992:865–881. [PubMed] [Google Scholar]
9. Bugoni L, et al.Marine debris and human impacts on sea turtlesin southern Brazil. Marine Pollut Bull2001;42:1330–1334. [PubMed] [Google Scholar]
10. Butler SE, et al.Mechanism of anemia in Schistosomamansoni–infected school children in westernKenya. Am J Trop Med Hyg2012;87:862–867. [PMC free article] [PubMed] [Google Scholar]
11. Chaloupka M, et al.Cause-specific temporal and spatial trends ingreen sea turtle strandings in the Hawaiian Archipelago(1982–2003). Marine Biol2008;154:887–898. [Google Scholar]
12. Chan A, et al.Effectiveness of sodium fluoride as apreservative of glucose in blood. Clin Chem1989;35:315–317. [PubMed] [Google Scholar]
13. Cray C, et al.Altered in vitro immune responses in greenturtles (Chelonia mydas) withfibropapillomatosis. J Zoo Wildl Med2001;32:436–440. [PubMed] [Google Scholar]
14. Flint M, et al.Clinical and pathological findings in greenturtles (Chelonia mydas) from Gladstone, Queensland:investigations of a stranding epidemic.EcoHealth2015;12:298–309. [PubMed] [Google Scholar]
15. Flint M, et al.Establishment of reference intervals for plasmaprotein electrophoresis in Indo-Pacific green sea turtles, Cheloniamydas. Conserv Physiol2015;3:cov037. [PMC free article] [PubMed] [Google Scholar]
16. Flint M, et al.Development and application of biochemical andhaematological reference intervals to identify unhealthy green sea turtles(Chelonia mydas). VetJ2010;185:299–304. [PubMed] [Google Scholar]
17. Flint M, et al.Health surveillance of stranded green turtles insouthern Queensland, Australia (2006–2009): an epidemiological analysis ofcauses of disease and mortality. EcoHealth2010;7:135–145. [PubMed] [Google Scholar]
18. Frazier J.Exploitation of marine turtles in the IndianOcean. Hum Ecol1980;8:329–370. [Google Scholar]
19. Fudge AM.Laboratory Medicine: Avian and Exotic Pets.Philadelphia, PA:Saunders,2000. [Google Scholar]
20. Fuentes M, et al.Potential impacts of projected sea-level rise onsea turtle rookeries. Aquat Conserv2010;20:132–139. [Google Scholar]
21. Gordon A, et al.Lesions caused by cardiovascular flukes(Digenea: Spirorchidae) in stranded green turtles (Cheloniamydas). Vet Pathol1998;35:21–30. [PubMed] [Google Scholar]
22. Gordon A, et al.Epizootic mortality of free-living greenturtles, Chelonia mydas, due tococcidiosis. J Wildl Dis1993;29:490–494. [PubMed] [Google Scholar]
23. Herbst LH, et al.Tumor outbreaks in marine turtles are not due torecent herpesvirus mutations. Curr Biol2004;14:R697–R699. [PubMed] [Google Scholar]
24. Kaplan AJ, et al.Development and validation of a competitiveenzyme-linked immunosorbent assay for the measurement of total plasmaimmunoglobulins in healthy loggerhead sea (Caretta caretta)and green turtles (Chelonia mydas).J Vet Diagn Invest2015:28:5–11. [PubMed] [Google Scholar]
25. Li TH, et al.Development of a summarized health index (SHI)for use in predicting survival in sea turtles. PLoSOne2015;10:e0120796. [PMC free article] [PubMed] [Google Scholar]
26. Merck D.Hematological Laboratory Methods. 2nded.Darmstadt, Germany: GITVerlag, 1984. [Google Scholar]
27. Owens DW, Ruiz GJ.New methods of obtaining blood and cerebrospinal fluid frommarine turtles. Herpetologica1980;36:17–20. [Google Scholar]
28. Page-Karjian A, et al.Baseline blood values for healthy free-ranginggreen sea turtles (Chelonia mydas) in PuertoRico. Comp Clin Pathol2014;24:567–573. [Google Scholar]
29. Reich KJ, et al.The ‘lost years’ of green turtles: using stableisotopes to study cryptic lifestages. BiolLett2007;3:712–714. [PMC free article] [PubMed] [Google Scholar]
30. Santos RG, et al.Debris ingestion by juvenile marine turtles: anunderestimated problem. Marine Pollut Bull2015;93:37–43. [PubMed] [Google Scholar]
31. Schuyler Q, et al.Global analysis of anthropogenic debrisingestion by sea turtles. Conserv Biol2014;28:129–139. [PMC free article] [PubMed] [Google Scholar]
32. Snoddy JE, et al.Blood biochemistry of sea turtles captured ingillnets in the lower Cape Fear River, North Carolina, USA.J Wildl Manag2009;73:1394–1401. [Google Scholar]
33. Tyler R, et al.Classification and diagnosis ofanaemia. Comp Haematol Int1996;6:1–16. [Google Scholar]
34. Van Houtan KS, et al.Eutrophication and the dietary promotion of seaturtle tumors. PeerJ2014;2:e602. [PMC free article] [PubMed] [Google Scholar]
35. Wallace BP, et al.Global conservation priorities for marineturtles. PLoS One2011;6:e24510. [PMC free article] [PubMed] [Google Scholar]
36. Ward JR, Lafferty KD.The elusive baseline of marine disease: are diseases in oceanecosystems increasing?PLoS Biol2004;2:e120. [PMC free article] [PubMed] [Google Scholar]
37. Whiting SD, et al.Blood chemistry reference values for twoecologically distinct populations of foraging green turtles, eastern IndianOcean. Comp Clin Pathol2007;16:109–118. [Google Scholar]
38. Wolke R, et al.Spirorchidiasis in loggerhead sea turtles(Caretta caretta): pathology. JWildl Dis1982;18:175–185. [PubMed] [Google Scholar]
Articles from Journal of Veterinary Diagnostic Investigation : Official Publication of the American Association of Veterinary Laboratory Diagnosticians, Inc are provided here courtesy of SAGE Publications
