Prognostic value of packed cell volume and blood glucose concentration in 954 client-owned chelonians

Violaine A. Colon 1Tai Wai Small Animal & Exotic Hospital, Tai Wai, Hong Kong.

Search for other papers by Violaine A. Colon in
Current site
Google Scholar
PubMed
Close
 DVM, MSc
and
Nicola Di Girolamo 2Department of Veterinary Clinical Sciences, College of Veterinary Medicine, Oklahoma State University, Stillwater, OK 74078.

Search for other papers by Nicola Di Girolamo in
Current site
Google Scholar
PubMed
Close
 DMV, PhD

Click on author name to view affiliation information

Abstract

OBJECTIVE

To evaluate the prognostic value of PCV and blood glucose concentration in chelonians presented for veterinary care and to develop risk categories on the basis of the interaction of these analytes.

ANIMALS

954 client-owned chelonians (34 genera).

PROCEDURES

Medical records of 1,059 client-owned chelonians presented to 2 veterinary institutions between 2014 and 2018 were reviewed. Logistic regression models were developed to evaluate factors associated with death, including PCV and blood glucose concentrations.

RESULTS

There were 954 chelonians (34 genera) for which the data required to be included in the analysis were available. Both PCV and blood glucose concentration were significant prognostic indicators of death. Odds of death for chelonians with severe anemia (PCV, < 10%) and moderate anemia (PCV, 11% to 20%) were 6.8 times (adjusted odds ratio [aOR], 6.8; 95% CI, 3.8 to 12.3) and 1.6 times (aOR, 1.6; 95% CI, 1.01 to 2.7), respectively, the odds of death for chelonians with PCV within reference range. Odds of death for chelonians with severe hypoglycemia (< 30 mg/dL), moderate hyperglycemia (91 to 150 mg/dL), and severe hyperglycemia (> 181 mg/dL) were 5.3 times (aOR, 5.3; 95% CI; 2.4 to 11.4), 3 times (aOR, 3.0;95% CI, 1.4 to 6.3), and 4.3 times (aOR, 4.3; 95% CI, 2.4 to 7.6), respectively, the odds of death for chelonians with blood glucose concentration within reference range. Five risk categories were identified on the basis of PCV and blood glucose concentration.

CONCLUSIONS AND CLINICAL RELEVANCE

Derangements in PCV and blood glucose concentration in client-owned chelonians were associated with increased odds of death. On the basis of these results, more aggressive diagnostic testing and treatments may be indicated in chelonians with similar alterations.

Abstract

OBJECTIVE

To evaluate the prognostic value of PCV and blood glucose concentration in chelonians presented for veterinary care and to develop risk categories on the basis of the interaction of these analytes.

ANIMALS

954 client-owned chelonians (34 genera).

PROCEDURES

Medical records of 1,059 client-owned chelonians presented to 2 veterinary institutions between 2014 and 2018 were reviewed. Logistic regression models were developed to evaluate factors associated with death, including PCV and blood glucose concentrations.

RESULTS

There were 954 chelonians (34 genera) for which the data required to be included in the analysis were available. Both PCV and blood glucose concentration were significant prognostic indicators of death. Odds of death for chelonians with severe anemia (PCV, < 10%) and moderate anemia (PCV, 11% to 20%) were 6.8 times (adjusted odds ratio [aOR], 6.8; 95% CI, 3.8 to 12.3) and 1.6 times (aOR, 1.6; 95% CI, 1.01 to 2.7), respectively, the odds of death for chelonians with PCV within reference range. Odds of death for chelonians with severe hypoglycemia (< 30 mg/dL), moderate hyperglycemia (91 to 150 mg/dL), and severe hyperglycemia (> 181 mg/dL) were 5.3 times (aOR, 5.3; 95% CI; 2.4 to 11.4), 3 times (aOR, 3.0;95% CI, 1.4 to 6.3), and 4.3 times (aOR, 4.3; 95% CI, 2.4 to 7.6), respectively, the odds of death for chelonians with blood glucose concentration within reference range. Five risk categories were identified on the basis of PCV and blood glucose concentration.

CONCLUSIONS AND CLINICAL RELEVANCE

Derangements in PCV and blood glucose concentration in client-owned chelonians were associated with increased odds of death. On the basis of these results, more aggressive diagnostic testing and treatments may be indicated in chelonians with similar alterations.

Request for high-quality medical care of client-owned chelonians is growing. Currently, there is limited knowledge on prognostic factors in chelonians, with the few available studies including only wild animals.1–6 In wild Kemp's ridley turtles (Lepidochelys kempii), several analytes have been used to predict the likelihood of survival, including blood gas parameters (ie, Pco2 and Po2), blood pH values, and potassium, chloride, calcium, phosphorus, and uric acid concentrations.1–3 In a study4 that included multiple species of sea turtles, high blood activity of aspartate aminotransferase and creatine kinase and high concentrations of creatinine and uric acid were also associated with death.4 A large retrospective study5 of wild turtles presented to a university wildlife clinic found that head injury, coelomic breach, myiasis, and carapace injury on the cranial aspect of the midline are significant predictors of death. A study6 of 46 wild freshwater turtles identified blood lactate concentration and PCV as negative prognostic factors.6 It is unclear whether the use of these data to guide prognosis for client-owned chelonians is appropriate because, compared with client-owned chelonians, wild chelonians have a completely different range of disorders and reasons for hospital admission.

Glucose is the main source of energy in reptiles as in mammals,7 and its metabolism is tightly regulated by several hormones.8 Blood glucose derangements in mammals, both hyperglycemia and hypoglycemia, have been associated with higher mortality rates in previous studies.9–11

Erythrocytes have a similar function in reptiles as in mammals.12 Although intraspecific and interspecific variations are common, anemia in reptiles is generally defined as a PCV < 20%, while polycythemia is defined as a PCV > 40%.13 Similar to mammals, reptiles may have regenerative and nonregenerative anemias. Common causes for regenerative anemia are hemorrhage (acute blood loss from trauma or parasitism), hemolytic diseases (including oxidative damage from zinc toxicosis), hemoparasitic or infectious disease, macroangiopathic disease, and marked hypophosphatemia.13,14 On the basis of experimental studies, it is assumed that maximum reticulocytosis occurs about 3 to 5 weeks following the onset of anemia.15 Because reptilian RBCs have a longer life span (from 600 to 800 days), compared with mammalian RBCs,13,16,17 nonregenerative anemias are characterized by a less obvious degree of polychromasia.

To the authors’ knowledge and on the basis of extensive searches of the published literature, no published studies exist on prognostic factors for client-owned chelonians. Because both blood glucose concentration and PCV are routinely measured in client-owned chelonians, evaluating their prognostic potential may improve patient care. The objectives of the present study were to evaluate whether blood glucose concentration and PCV had a prognostic value in client-owned chelonians presented for veterinary care and to develop risk categories on the basis of the interaction of these analytes.

Materials and Methods

Study design and outcomes

A retrospective case series and prognostic study were performed by collecting data from chelonians examined at 2 related institutions (ie, Tai Wai Small Animal & Exotic Hospital, Hong Kong, and Island Exotics, Hong Kong) between July 31, 2014, and December 31, 2018. Medical records from Island Exotics were included from the date of its June 1, 2016, opening. The institutions have the same ownership and similar policies related to cases and case management, and some veterinarians worked in both hospitals. The primary outcome of the present study was the associations between blood glucose concentration and PCV and 7-day mortality rate.

Data collection and inclusion criteria

The electronic medical record database of each institution was searched by 1 investigator (VAC) for the following keywords: chelonian, turtle, terrapin, tortoise, and reptile. The medical record was reviewed for each chelonian seen during the period of interest for which radiology or blood analysis was performed. Chelonians that had both PCV and blood glucose measurements available were included in the study. All data were entered manually in an electronic spreadsheeta by the same investigator. The study flow is depicted (Figure 1).

Figure 1—
Figure 1—

Flowchart of medical record review for inclusion of eligible client-owned chelonians in a study of the prognostic value of PCV and blood glucose concentration in chelonians.

Citation: Journal of the American Veterinary Medical Association 257, 12; 10.2460/javma.257.12.1265

Variables extracted

The following variables were extracted from the medical records: age, sex, genus, species habitat, husbandry, diet, current food intake, body weight, attending clinician, blood glucose concentration, PCV, survival time, and status (alive or dead) at 7 days. Before veterinary consultation, a detailed history was collected for each animal as per hospital guidelines and entered in the medical records by veterinary assistants. The history included details on age, sex (including method of sexing and whether the animal had ever laid eggs), species, food intake, husbandry, and diet. Body weight was obtained with different scales depending on the size of the chelonian. Habitat of each species was classified as primarily aquatic or terrestrial, depending on the natural environment for the species. Food intake was classified as normal appetite, hyporexia, or anorexia, on the basis of owner perception. Anorexia was defined as no food ingested for the previous 24 hours. Hyporexia was defined as a smaller quantity of food or a smaller variety of food ingested than normal. For anorectic chelonians, the last day the animal had eaten was calculated on the basis of owner indications. For analytical purposes, if the chelonian had a normal appetite or hyporexia, the last day the animal had eaten was entered as 0. Survival time was calculated between the time of venipuncture and time of death. Survival rate at 7 days was calculated from the medical record data. Animals were classified as deceased if the animals died or were euthanized in the 7 days following the initial blood sample collection and alive if the animal was alive at discharge from hospital or at recheck examination. As per hospital policies for animals that had been sedated, their owners were contacted 48 hours after the procedure. The only blood values extracted for this study were the first available for each individual animal. Data on any repeated hematologic test were not extracted from the medical record. To establish a database for further studies on client-owned chelonian health, data on other variables were extracted from the medical records but were not considered in the present study.

Blood glucose concentration and PCV measurements

Blood sample collection was performed on conscious or sedated animals depending on clinician preference and individual case indications. Animals were not routinely fasted before blood sample collection. As per hospital guidelines, blood was preferably collected from the jugular vein, and only in exceptional cases, was blood obtained from different sites. Blood samples were immediately placed in lithium heparin tubes and analyzed shortly after collection with an in-house analyzerb with an avian-reptile rotor.18 A recommended modified hexokinase method19 was used with the in-house analyzer to estimate glucose concentration and was proven to have good agreement with a laboratory analyzer in chelonians.18 Packed cell volume was determined in duplicate with two 0.8-mm heparinized capillary tubes,c and results were averaged. The microhematocrit tubesc were centrifuged at 1,300 × g for 5 minutes.d,e

Statistical analysis

Continuous variables were summarized as either means or medians and SDs, depending on their distribution. The Shapiro-Wilk test was used to analyze continuous data for normality. For categorical variables, the percentages of patients in each category were calculated.

Univariable binary logistic regression analyses with 1 predictor were used to identify any relevant predictor of death. Predictors tested were age, sex, body weight, habitat (terrestrial or aquatic), PCV, and blood glucose concentration. Odds ratios, aOR, and 95% CI were used to quantify the strength of these associations. The PCV and blood glucose concentration were first inserted in the models as continuous variables. Furthermore, the variables were binned into 5 and 7 intervals, respectively. The variable PCV was binned into five 10% categories starting from 0 (0 to 10, 11 to 20, 21 to 30, 31 to 40, and ≥ 41). The variable blood glucose concentration was binned into seven 30 mg/dL categories starting from 0 (0 to 30, 31 to 60, 61 to 90, 91 to 120, 121 to 150, 151 to 180, and ≥ 181). The binned categories were cross-tabulated with mortality rate. Because neither PCV nor blood glucose concentration had a linear association with mortality rate, but mortality rate increased at their extremes, these binned categories were used in further models. Multivariable logistic regression models were subsequently developed to evaluate whether the associations between PCV and blood glucose concentration and mortality rate persisted after adjustment for other patient characteristics. Predictors demonstrated in the univariable analysis to be prognostically important at a value of P < 0.15 were entered into the final model. In a further model, the interaction of PCV and blood glucose concentration was included together with the other potential confounders to identify specific categories of risk. To avoid overfitting of the models, no more than 1 predictor variable was included for 10 events (ie, deaths).20 Subgroup logistic regression analyses that included only primarily terrestrial or primarily aquatic chelonians were performed. The Hosmer-Lemeshow statistic and Nagelkerke R2 were used to assess goodness-of-fit of the models.

Data were analyzed with commercial software.f Two-tailed values of P < 0.05 were considered significant.

Results

Patient summary

On the basis of findings in the medical record search, 1,059 chelonians were eligible for inclusion in the data set, of which 954 chelonians had the data required to be included in the data analysis. None of the continuous variables were distributed normally; therefore, medians were reported. Thirty-four genera, including 49 different species of chelonians, were included (Table 1), with the most common species presented being red-eared sliders (Trachemys scripta elegans) that accounted for 43% (407/954) of the total. Of the included chelonians, 73% (696/954; 95% CI, 70.1% to 75.7%) were primarily aquatic and 27% (258/954; 95% CI, 24.3% to 29.9%) were primarily terrestrial. Almost half (49% [467/954]; 95% CI, 45.8% to 52.1%) of the chelonians were females, 27% (258/954; 95% CI, 24.3% to 29.9%) were males, and 24% (229/954; 95% Cl, 21.4% to 26.8%) were of unknown sex. Median ± SD age and body weight were 9.0 ± 9.6 years (range, 0.2 to 61 years) and 1.05 ± 3.86 kg (2.31 ± 8.51 lb; range, 0.04 to 50.0 kg [0.09 to 110.0 lb]), respectively. Of the included chelonians, 15.2% (145/954) died within 7 days of the initial clinical examination.

Table 1—

Frequency of the 12 most common species included in a study of the prognostic value of PCV and blood glucose concentration in chelonians, accounting for 88% of the patients.

SpeciesCount (n)Percentage of total
Red-eared slider (Trachemys scripta elegans)40742.7
African-spurred tortoise (Centrochelys sulcata)9610.1
Chinese three-striped box turtle (Cuora trifasciata)737.7
Leopard tortoise (Stigmochelys pardalis)616.4
Reeve's turtle (Mauremys reevesii)616.4
Indian star tortoise (Geochelone elegans)323.4
Chinese striped-neck turtle (Mauremys sinensis)282.9
Yellow-margined box turtle (Cuora flavomarginata)252.6
Pig-nosed turtle (Carettochelys insculpta)181.9
Radiated tortoise (Astrochelys radiata)181.9
Red-footed tortoise (Chelonoidis carbonarius)121.3
Common musk turtle (Sternotherus odoratus)101.0
Total84188.3

PCV and blood glucose concentration

Median ± SD (range) PCV was 27 ± 1.0% (1% to 53%). Median PCV of aquatic chelonians was 29 ± 9.3% (1% to 53%) and of terrestrial chelonians was 25 ± 8.1% (1% to 50%). Median PCV of chelonians that died was 25 ± 11.2% (1% to 53%) and of chelonians that survived to 7 days from blood sample collection was 28 ± 8.3% (1% to 52%).

Overall, median ± SD (range) of blood glucose concentration was 88 ± 94.9 mg/dL (3 to 700 mg/dL). Median blood glucose of aquatic chelonians was 85 ± 97.6 mg/dL (3 to 700 mg/dL) and of terrestrial chelonians was 95 ± 87.5 mg/dL (5 to 692 mg/dL). Median blood glucose of chelonians that died was 94 ± 126.4 mg/dL (3 to 650 mg/dL) and of chelonians that survived 7 days from blood sample collection was 87 ± 82.7 mg/dL (6 to 700 mg/dL).

Association between PCV, blood glucose concentration, and death

In the unadjusted analyses, PCV (OR, 0.95; 95% CI, 0.93 to 0.97; P < 0.001) and blood glucose concentration (OR, 1.004; 95% CI, 1.002 to 1.005; P < 0.001) were both significantly associated with increased odds of death (Figure 2). Age had a weak, nonsignificant association with increased odds of death (OR, 1.01; 95% CI, 0.99 to 1.03; P = 0.11). Other predictors tested (body weight, sex, and habitat) were not associated with death (Table 2). Because both PCV and blood glucose concentration were associated with death in a nonlinear fashion, further models included the variables after binning. The final model, therefore, included age, PCV (binned), and blood glucose concentration (binned). After multivariable adjustment, the association between PCV, blood glucose concentration, and death persisted. The odds of death for chelonians with severe anemia and moderate anemia were 6.8 times (aOR, 6.8; 95% CI, 3.8 to 12.3; P < 0.001) and 1.6 times (aOR, 1.6; 95% CI, 1.01 to 2.7; P = 0.04), respectively, the odds of death for chelonians with PCV values within reference range. The odds of death for chelonians with severe hypoglycemia, moderate hyperglycemia, and severe hyperglycemia were 5.3 times (aOR, 5.3; 95% CI, 2.4 to 11.4; P < 0.001), 3 times (aOR, 3.0; 95% CI, 1.4 to 6.3; P = 0.004), and 4.3 times (aOR, 4.3; 95% CI, 2.4 to 7.6; P < 0.001), respectively, the odds of death for chelonians with blood glucose concentrations within reference range. The final logistic regression model was well fitted (Hosmer-Lemeshow test, P = 0.17; Nagelkerke R2 = 0.16) and accounted for age (aOR, 1.02; 95% CI, 1.00 to 1.04; P = 0.01). Subgroup analyses, including only terrestrial or aquatic chelonians, generally confirmed the robustness of the findings but highlighted a lower effect of hyperglycemia in terrestrial chelonians (Table 3).

Figure 2—
Figure 2—

Scatterplot of status (died vs alive) 7 days after blood collection for client-owned chelonians presented at 2 veterinary hospitals in relationship with blood glucose (BG) concentration and PCV. Notice the higher distribution of chelonians that died within 7 days from blood collection in the peripheral areas of the plot. Blood glucose is on a logarithmic scale. Points represent animals that died within 7 days. Circles represent animals alive at 7 days.

Citation: Journal of the American Veterinary Medical Association 257, 12; 10.2460/javma.257.12.1265

Table 2—

Results of univariable and multivariable logistic regression analyses to determine association of individual factors with 7-day mortality rate (death within 7 days after blood collection) in the included population of chelonians.

VariableAliveDeadOR (95% CI)P valueaOR (95% CI)Adjusted P value
Blood glucose      
 Reference range (61–90 mg/dL)258 (90.8)26 (9.2)ReferenceNAReferenceNA
 Severe hypoglycemia (< 30 mg/dL)27 (58.7)19 (41.3)7.0 (3.4–14.2)< 0.0015.3 (2.4–11.4)< 0.001
 Moderate hypoglycemia (31–60 mg/dL)147 (89.1)18 (10.9)1.2 (0.6–2.3)0.541.1 (0.6–2.2)0.68
 Mild hyperglycemia (91–150 mg/dL)238 (88.5)31 (11.5)1.3 (0.7–2.2)0.361.5 (0.8–2.6)0.18
 Moderate hyperglycemia (151–180 mg/dL)48 (77.4)14 (22.6)2.9 (1.4–5.9)0.0043.0 (1.4–6.3)0.004
 Severe hyperglycemia (181 mg/dL)91 (71.1)37 (28.9)4.0 (2.3–7.0)< 0.0014.3 (2.4–7.6)< 0.001
PCV      
 Reference range (21%–40%)608 (88.9)76 (11.1)ReferenceNAReferenceNA
 Severe anemia (< 10%)32 (51.6)30 (48.4)7.5 (4.3–13.0)< 0.0016.8 (3.8–12.3)< 0.001
 Moderate anemia (11%–20%)134 (81.7)30 (18.3)1.8 (1.1–2.8)0.0131.6 (1.01–2.7)0.04
Polycythemia (> 41%)35 (79.5)9 (20.5)2.1 (0.9–4.4)0.0661.7 (0.7–3.8)0.18
Age (y)11.3 (9.6)12.5 (9.5)1.0 (0.9–1.0)0.101.01 (0.9–1.03)0.11
Weight (kg)2.0 (4.14)1.85 (2.69)0.9 (0.9–1.0)0.921.01 (0.9–1.05)0.63
Sex      
 Female347 (74.3)120 (25.7)1.1 (0.7–1.7)0.630.7 (0.5–1.1)0.10
 Male208 (80.6)50 (19.4)0.9 (0.5–1.5)0.681.0 (0.7–1.6)0.84
 Undetermined183 (79.9)46 (20.1)ReferenceNAReferenceNA
Habitat      
 Aquatic536 (77.0)160 (23)1.1 (0.7–1.5)0.670.9 (0.6–1.3)0.67
 Terrestrial202 (78.3)56 (21.7)ReferenceNAReferenceNA

Continuous data are reported as mean ± SD. Binary data are reported as number of observed events (percentage).

NA = Not applicable.

Table 3—

Results of multivariable logistic regression analysis to determine association of PCV and blood glucose concentration with 7-day mortality rate in aquatic and terrestrial chelonians.

 Aquatic speciesTerrestrial species
VariableaOR (95% CI)Adjusted P valueaOR (95% CI)Adjusted P value
Blood glucose    
 Reference range (61–90 mg/dL)ReferenceNAReferenceNA
 Severe hypoglycemia (< 30 mg/dL)4.4 (1.6–12.1)0.0046.7 (1.8–25.2)0.005
 Moderate hypoglycemia (31–60 mg/dL)1.2 (0.6–2.5)0.620.9 (0.2–4.4)0.92
 Mild hyperglycemia (91–150 mg/dL)1.5 (0.8–2.9)0.191.3 (0.4–4.2)0.67
 Moderate hyperglycemia (151–180 mg/dL)3.3 (1.5–7.7)0.0041.8 (0.3–10.5)0.54
 Severe hyperglycemia (≥ 181 mg/dL)4.2 (2.1–8.2)< 0.0014.4 (1.3–14.6)0.02
PCV    
 Reference range (21%–40%)ReferenceNAReferenceNA
 Severe anemia (< 10%)7.2 (3.6–14.6)< 0.0015.6 (1.8–17.6)0.003
 Moderate anemia (11%–20%)1.7 (0.9–3.1)0.061.4 (0.6–3.5)0.42
 Polycythemia (> 41%)1.8 (0.8–4.2)0.181.0 (0.09–11.1)0.97
 Age (y)1.0 (1.0–1.0)0.041.0 (0.9–1.1)0.42

NA = Not applicable

Categories of risk

Chelonians that had blood glucose concentration and PCV within reference ranges had a mortality rate of 6.7% (95% CI, 4.0% to 10.9%). The multivariable model developed to identify specific categories of risk included the interaction between PCV and blood glucose concentration and age as a covariate (Hosmer-Lemeshow test, P = 0.99; Nagelkerke R2 = 0.14). Compared with the odds of death for chelonians with blood glucose concentration and PCV within reference ranges, the odds of death for chelonians with moderate hyperglycemia and polycythemia, severe hypoglycemia and severe anemia, severe hyperglycemia and severe anemia, and mild hyperglycemia and severe anemia were 14 times (aOR, 14.0; 95% CI, 1.2 to 156.2; P = 0.03), 11.8 times (aOR, 11.8; 95% CI, 3.6 to 38.4; P < 0.001), 10.3 times (aOR, 10.3; 95% CI, 2.3 to 47.2; P = 0.003), and 9.9 times (aOR, 9.9; 95% CI, 2.9 to 33.4; P < 0.001) as high, respectively. The other interactions were not significantly associated with increased odds of death in this cohort of patients.

Discussion

In the present study, both blood glucose concentration and PCV were relevant prognostic indicators in client-owned chelonians. Including both factors in the multivariable analysis permitted the development of 5 distinct risk categories for client-owned chelonians. Chelonians at higher odds of death were those that had combinations of severe hypoglycemia and severe anemia, severe hyperglycemia and severe anemia, mild hyperglycemia and severe anemia, and moderate hyperglycemia and polycythemia. Besides these specific risk categories, both blood glucose and PCV were significant independent predictors of death in this large cohort of client-owned chelonians.

Although the present study provided an association between blood glucose concentration and PCV derangements and death, it is currently unclear what is the cause of this association. One possible explanation is that derangements of blood glucose concentration and PCV represent a response to either acute or chronic external stressors. Conclusions of a study6 including traumatized freshwater turtles suggest that blood glucose concentration can be used as a quantitative measurement of stress. Stress can induce a fight-or-flight response that will stimulate the hypothalamic-pituitary-adrenal axis.21 The aim of this acute adrenocortical response is to behaviorally or physiologically eliminate or reduce the impact of potential stressors on an individual and thus promote its survival time.21,22 If this stress becomes chronic, plasma corticosterone concentrations remain elevated, with potential impairment of basic functions such as reproduction, growth, and immunocompetence.23 This adrenocortical response to stress has been identified in reptiles including chelonians21,24 and is believed to trigger increased hepatic glycogen, protein catabolism, and increased gluconeogenesis, with resulting hyperglycemia.25 In a study23 of crocodiles, blood glucose concentration significantly increased with duration of capture stress and had a positive and significant relationship with plasma corticosterone concentration. In cold-stunned sea turtles, both hyperglycemia and hypoglycemia are commonly detected.3 It is possible that after an initial hyperglycemic status of variable duration, stressed reptiles deplete their glycogen sources and develop hypoglycemia. Therefore, the observed increased odds of death with both hyperglycemia and hypoglycemia could represent blood sample collection from patients at different stages of severe external stress.

On the basis of findings in the present study, hyperglycemia in chelonians could be a predictor of death, similar to what has been described in mammals. In humans, hyperglycemia has been associated with a poorer prognosis in patients with diverse clinical presentations.26,27 For example, in patients admitted for myocardial infarction, stress hyperglycemia (120.6 to 198 mg/dL) was associated with increased odds of in-hospital death.28 Interestingly, the magnitude of the effect on odds of death observed in that study28 (OR, 3.9; 95% CI, 2.9 to 5.4) was similar to that detected in the present study (aOR, 4.3; 95% CI, 2.4 to 7.6). In dogs and cats, hyperglycemia has been associated with the severity of head trauma but not with the outcome.29 In rabbits, blood glucose concentrations are considered a prognostic and potentially diagnostic factor. Hyperglycemia has been associated with a poorer outcome in a retrospective study30 on rabbits and has been found to be associated with hyponatremia in a subsequent study.31 A recent study32 on wild sea turtles showed that the animals that died had significantly higher blood glucose concentrations than those that survived, together with other analyte differences. A literature review33 suggested that persistent hyperglycemia is an indicator of poor prognosis.

Starvation, hepatobiliary disease, and septicemia have been associated with hypoglycemia in reptiles34,35; however, a chelonian trapped for 11 months without access to food had blood glucose concentrations within reference range when rescued.36 The lack of changes in that animal underlines the high degree of physiological adaptation in chelonians and suggests that it is unlikely that the hypoglycemia observed in the chelonians of the present study was uniquely associated with starvation. The prognostic value of hypoglycemia is well established in mammal medicine. In humans, even just a single episode of severe hypoglycemia (ie, blood glucose concentration < 40 mg/dL) was independently associated with increased odds of death in critically ill patients (OR, 2.28; 95% CI, 1.41 to 3.70).37 In elderly hospitalized patients, hypoglycemia was associated with an increase of the in-hospital 3- and 6-month mortality rate, but hypoglycemia was not an independent predictor for death, implying that it is only a marker of poor health without a direct effect on survival time.38 Blood glucose derangements may also lead to other complications associated with high mortality rates. For example, human patients with severe hypoglycemia were more likely to develop septic shock, compared with patients with no hypoglycemia.39

Packed cell volume is a fundamental value to consider in any vertebrate, and anemia has been demonstrated to be a useful prognostic factor in mammals, including human patients with chronic heart failure.40 Early diagnosis of anemia in reptile patients can help identify diseases before they progress to life-threatening stages.41 Hematocrit has been used to address the severity of fibropapillomatosis in sea turtles; a study42 of green turtles (Chelonia mydas) showed that Hct, hemoglobin concentration, and total protein concentration were inversely related to fibropapillomatosis progression. In the present study, severe anemia (PCV, < 10%) resulted in an almost 50% mortality rate. Contrary to findings in the present study, PCV was not significantly associated with survival rate in 127 green sea turtles (C mydas) stranded in Australia.43 However, in that study,43 only a linear effect of PCV on death was evaluated. If increased odds of death were actually present at both low and high PCVs, as observed in the present study, the effect of PCV on odds of death would have been missed. In freshwater turtles, PCV has been evaluated as a prognostic factor only in turtles with blood lactate concentration < 11.45 mg/dL.6 Turtles with blood lactate concentration > 12.0 mg/dL had a 100% mortality rate. In the present study, a low PCV has been identified as a potential standalone prognostic factor, whereas a high PCV may be a prognostic factor when evaluated in combination with blood glucose concentration.

In the present study, increased odds of death were observed in chelonians with a PCV > 40% when this value was associated with moderate hyperglycemia. Recently, polycythemia has been defined in reptiles as when the PCV exceeds 40%.13 Of the total cohort of chelonians in the present study, 4.7% (44/945) had polycythemia, and this was not evidently associated with the habitat of the species. Because of the design of our study, it was not possible to determine whether the increased PCV was caused by an actual increase in the number of RBCs (absolute polycythemia) or rather a decrease of plasma volume (relative polycythemia). To our knowledge, absolute polycythemia has not been reported in reptiles; however, increased PCV has been reported in terrapins subjected to hyperosmotic stress.25 An increase in PCV has been observed in water-deprived bearded dragons administered high-dose furosemide44 but not in turtles administered lower doses of furosemide.45 The most likely explanation for the association between polycythemia and death in the present study was that it represented an underlying severe dehydration.

Because the present study included actual chelonian patients presented for a multitude of disorders, this was a pragmatic population sample that was more likely to reflect real practice, and this design is inherently associated with certain limitations.

In reptiles, glucose homeostasis is complex and influenced by several factors, including season, sex, health, and body temperature.46 Some of these factors were accounted for in the present study, but cloacal temperature, for example, was not recorded in most physical examinations and not accounted for in the analysis.

In the present study, the time frame for blood sample collection was not consistent between animals and depended on the attending clinician. A study47 of wild turtles recommended that venipuncture be performed within 10 minutes of capture to minimize the effect of stress handling on blood values. Evaluation of the effect of animal transport and handling on blood glucose concentrations revealed a clear and significant effect of transport (mean increase in blood glucose concentration of 33% to 36% after 13 to 26 hours of transport).48 In the present study, potentially biased results for the odds of death could have arisen if chelonians coming from further away (eg, chelonians that were referred from other hospitals) were also the animals that had the more serious conditions. However, considering the size of the country (ie, Hong Kong), the time of transport for most chelonians included in the study was likely < 1 hour. Also, the cited studies47,48 included wild chelonians that are likely more stressed by human handling, compared with the captive animals in the present study.

The time frame between venipuncture and analysis of blood samples was not specified in the medical records of the present study. As per the hospital protocol, blood analysis is performed as soon as blood samples are collected, but a delay could have occurred during an emergency. It has been shown that measured blood glucose centration decreases in samples at ≥ 30 minutes after sample collection.49 In the present study, it was unlikely that analysis of any blood sample was delayed for ≥ 30 minutes; if such a delay did occur, however, its values would have had minimal impact on the overall findings.

At the time of hospital admission of chelonians, owners or caregivers were asked the age of the animal. At the time of data extraction from the medical records, it was impossible to establish whether the owners at the time of consultation were confident about the age they provided for an animal. It is possible that for some chelonians in the present study, age was more of an estimation provided by the owners rather than an exact age.

In conclusion, on the basis of findings in the present study, blood glucose concentration and PCV were prognostic indicators for chelonian patients. Five risk categories were identified in the present study that could be useful for triage of these animals in the future. The findings of the present report should not be used as a standalone tool for making conclusive clinical decisions, but rather as a potential aid to help inform owners of the prognosis and the need for aggressive treatments in chelonians that fall in a high-risk category.

Acknowledgments

The authors have no financial interests with companies that manufacture products used in this study or with companies that manufacture competing products.

ABBREVIATIONS

aOR

Adjusted odds ratio

Footnotes

a.

Excel, Microsoft Corp, Redmond, Wash.

b.

VETSCAN VS2 chemistry analyzer, Avian Reptilian Profile Plus, Abaxis Inc, Union City, Calif.

c.

StatSpin heparinized glass tubes, Iris Sample Processing, Chatsworth, Calif.

d.

StatSpin VT, Iris Sample Processing, Chatsworth, Calif.

e.

PrO-Vet Centurion Scientific, Chichester, England.

f.

SPSS Statistics, version 22.0, IBM Corp, Chicago, Ill.

References

  • 1. 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;240:317323.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 2. Stacy NI, Innis CJ, Hernandez JA. Development and evaluation of three mortality prediction indices for cold-stunned Kemp's ridley sea turtles (Lepidochelys kempii). Conserv Physiol 2013;1:cot003.

    • Search Google Scholar
    • Export Citation
  • 3. 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;235:426432.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 4. Li T-H, Chang C-C, Cheng IJ, et al. Development of a summarized health index (SHI) for use in predicting survival in sea turtles. PLoS One 2015;10:e0120796.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 5. Sack A, Butler E, Cowen P, et al. Morbidity and mortality of wild turtles at a North Carolina wildlife clinic: a 10-year retrospective. J Zoo Wildl Med 2017;48:716724.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 6. Savo APH, Zheng Y, Zheng Y, et al. Health status assessment of traumatic injury freshwater turtles. PLoS One 2018;13:e0202194.

  • 7. Zhang Y, Qin C, Yang L, et al. A comparative genomics study of carbohydrate/glucose metabolic genes: from fish to mammals. BMC Genomics 2018;19:246.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 8. Rastogi SC. Metabolism. In: Rastogi SC, ed. Essentials of animal physiology. 4th ed. New Delhi: New Age International, 2007;100123.

    • Search Google Scholar
    • Export Citation
  • 9. Hollis AR, Furr MO, Magdesian KG, et al. Blood glucose concentrations in critically ill neonatal foals. J Vet Intern Med 2008;22:12231227.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 10. Torre DM, DeLaforcade AM, Chan DL. Incidence and clinical relevance of hyperglycemia in critically ill dogs. J Vet Intern Med 2007;21:971975.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 11. Trefz FM, Feist M, Lorenz I. Hypoglycaemia in hospitalised neonatal calves: prevalence, associated conditions and impact on prognosis. Vet J 2016;217:103108.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 12. Stacy NI, Alleman AR, Sayler KA. Diagnostic hematology of reptiles. Clin Lab Med 2011;31:87108.

  • 13. Heatley JJ, Russell KE. Hematology. In: Divers SJ, Stahl SJ, eds. Mader's reptile and amphibian medicine and surgery. 3rd ed. St Louis: Elsevier, 2019;301318.

    • Search Google Scholar
    • Export Citation
  • 14. Hynes B, Girling SJ. Cardiovascular and haemopoietic systems. In: Girling SJ, Raiti P, eds. BSAVA manual of reptiles. 3rd ed. Quedgeley, England: BSAVA Library, 2019;323341.

    • Search Google Scholar
    • Export Citation
  • 15. Dessauer HC. Blood chemistry of reptiles: physiological and evolutionary aspects. In: Gans C, Parsons TS, eds. Biology of the reptilia. London: Academic Press, 1970;172.

    • Search Google Scholar
    • Export Citation
  • 16. Sykes JMIV, Klaphake E. Reptile hematology. Clin Lab Med 2015;35:661680.

  • 17. Frye F. Hematology as applied to clinical reptile medicine. In: Frye FL, ed. Biomedical and surgical aspects of captive reptile husbandry. Vol 1. 2nd ed. Melbourne: Krieger Publishing Corp, 1991;209277.

    • Search Google Scholar
    • Export Citation
  • 18. Di Girolamo N, Ferlizza E, Selleri P, et al. Evaluation of point-of-care analysers for blood gas and clinical chemistry in Hermann's tortoises (Testudo hermanni). J Small Anim Pract 2018;59:704713.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 19. Neeley WE. Simple automated determination of serum or plasma glucose by a hexokinase/glucose-6-phosphate dehydrogenase method. Clin Chem 1972;18:509515.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 20. Peduzzi P, Concato J, Feinstein AR, et al. Importance of events per independent variable in proportional hazards regression analysis II. Accuracy and precision of regression estimates. J Clin Epidemiol 1995;48:15031510.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 21. Moore IT, Jessop TS. Stress, reproduction, and adrenocortical modulation in amphibians and reptiles. Horm Behav 2003;43:3947.

  • 22. Wingfield JC, Ramenofsky M. Hormones and the behavioral ecology of stress. In: Balm PH, ed. Stress physiology in animals. Sheffield, England: Sheffield Academic Press, 1999;151.

    • Search Google Scholar
    • Export Citation
  • 23. Jessop TS, Tucker AD, Limpus CJ, et al. Interactions between ecology, demography, capture stress, and profiles of corticosterone and glucose in a free-living population of Australian freshwater crocodiles. Gen Comp Endocrinol 2003;132:161170.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 24. Fazio E, Medica P, Bruschetta G, et al. Do handling and transport stress influence adrenocortical response in the tortoises (Testudo hermanni)? ISRN Vet Sci 2014;2014.

    • Search Google Scholar
    • Export Citation
  • 25. Martínez Silvestre A. How to assess stress in reptiles. J Exot Pet Med 2014;23:240243.

  • 26. Russel R, Ranjith R, Sartorius B. Admission blood glucose as a prognostic indicator in patients with acute myocardial infarction. Cardiovasc J S Afr 2019;16:118126.

    • Search Google Scholar
    • Export Citation
  • 27. Danisman B, Yilmaz MS, Isik B, et al. Analysis of the correlation between blood glucose level and prognosis in patients younger than 18 years of age who had head trauma. World J Emerg Surg 2015;10:8.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 28. Capes SE, Hunt D, Malmberg K, et al. Stress hyperglycaemia and increased risk of death after myocardial infarction in patients with and without diabetes: a systematic overview. Lancet 2000;355:773778.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 29. Syring RS, Otto CM, Drobatz KJ. Hyperglycemia in dogs and cats with head trauma: 122 cases (1997–1999). J Am Vet Med Assoc 2001;218:11241129.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 30. Harcourt-Brown FM, Harcourt-Brown SF. Clinical value of blood glucose measurement in pet rabbits. Vet Rec 2012;170:674.

  • 31. Bonvehi C, Ardiaca M, Barrera S, et al. Prevalence and types of hyponatraemia, its relationship with hyperglycaemia and mortality in ill pet rabbits. Vet Rec 2014;174:554.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 32. Innis CJ, McGowan JP, Burgess EA. Cold-stunned loggerhead sea turtles (Caretta caretta): initial vs. convalescent physiologic status and physiologic findings associated with death. J Herpetol Med Surg 2019;29:105112.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 33. Stahl SJ. Hyperglycemia. In: Divers SJ, Stahl SJ, eds. Mader's reptile and amphibian medicine and surgery. 3rd ed. St Louis: Elsevier, 2019;13141315.

    • Search Google Scholar
    • Export Citation
  • 34. Campbell T. Clinical pathology in reptiles. In: Mader DR, ed. Reptile medicine and surgery. 2nd ed. St Louis: Elsevier Saunders, 2006;453470.

    • Search Google Scholar
    • Export Citation
  • 35. Raiti P. Endocrinology. In: Divers SJ, Stahl SJ, eds. Mader's reptile and amphibian medicine and surgery. 3rd ed. St Louis: Elsevier, 2019;835848.

    • Search Google Scholar
    • Export Citation
  • 36. Christopher MM. Physical and biochemical abnormalities associated with prolonged entrapment in a desert tortoise. J Wildl Med 1999;35:361366.

    • Search Google Scholar
    • Export Citation
  • 37. Krinsley JS, Grover A. Severe hypoglycemia in critically ill patients: risk factors and outcomes. Crit Care Med 2007;35:22622267.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 38. Kagansky N, Levy S, Rimon E, et al. Hypoglycemia as a predictor of mortality in hospitalized elderly patients. Arch Intern Med 2003;163:18251829.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 39. Waeschle RM, Moerer O, Hilgers R, et al. The impact of the severity of sepsis on the risk of hypoglycaemia and glycaemic variability. Crit Care 2008;12:R129.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 40. Szachniewicz J, Petruk-Kowalczyk J, Majda J, et al. Anaemia is an independent predictor of poor outcome in patients with chronic heart failure. Int J Cardiol 2003;90:303308.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 41. Saggese MD. Clinical approach to the anemic reptile. J Exot Pet Med 2009;18:98111.

  • 42. Hirama S, Ehrhart LM, Rea LD, et al. Relating fibropapilloma tumor severity to blood parameters in green turtles Chelonia mydas. Dis Aquat Organ 2014;111:6168.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 43. March DT, Vinette-Herrin K, Peters A, et al. Hematologic and biochemical characteristics of stranded green sea turtles. J Vet Diagn Invest 2018;30:423429.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 44. Parkinson LA, Mans C. Effects of furosemide administration to water-deprived inland bearded dragons (Pogona vitticeps). Am J Vet Res 2018;79:12041208.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 45. Stephens GA, Robertson F. Renal responses to diuretics in the turtle. J Comp Physiol B 1985;155:387393.

  • 46. Heatley JJ, Russell KE. Clinical chemistry. In: Divers SJ, Stahl SJ, eds. Mader's reptile and amphibian medicine and surgery. 3rd ed. St Louis: Elsevier, 2019;319332.

    • Search Google Scholar
    • Export Citation
  • 47. Cash WB, Holberton RL, Knight SS. Corticosterone secretion in response to capture and handling in free-living red-eared slider turtles. Gen Comp Endocrinol 1997;108:427433.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 48. Hunt KE, Innis CJ, Kennedy AE, et al. Assessment of ground transportation stress in juvenile Kemp's ridley sea turtles (Lepidochelys kempii). Conserv Physiol 2016;4:cov071.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 49. Kerr MG. Carbohydrate metabolism. In: Kerr MG, ed. Veterinary laboratory medicine: clinical biochemistry and haematology. 2nd ed. Oxford, England: John Wiley & Sons, 2008;111126.

    • Search Google Scholar
    • Export Citation
All Time Past Year Past 30 Days
Abstract Views 302 0 0
Full Text Views 1234 1049 190
PDF Downloads 421 212 22
Advertisement