Methods

The African penguin colony evaluated included birds managed at one aquarium in an outdoor habitat consisting of a terrestrial area with natural stone and concrete and a freshwater pool. Birds were fed a mixture of capelin (Mallotus villosus), herring (Clupea harengus), silversides (Menidia menidia), lake smelt (Osmerus mordax), Northern anchovy (Engraulis mordax), and squid (Illex argentinus), supplemented with a multivitamin (Mazuri Small Bird Supplement).

The study population of clinically healthy birds included 29 individuals (14 males and 15 females). Ages ranged between 1 and 36 years old (median, 13 years old). Nonmolting penguins with normal physical examinations and blood analysis within published reference intervals (including a CBC and biochemistry panel) were utilized to establish the reference intervals for GLDH.¹¹ Only routine medications such as primaquine for antimalarial prophylaxis (dosed once weekly) in 16/29 birds and a glucosamine/chondroitin supplement in 1 bird were accepted into the study.

All penguins at one institution were reviewed over a 19-year period (2005 to 2024) to identify birds with liver disease. Twelve mortalities occurred over the period evaluated, and all histopathological reports were from the University of Connecticut Medical Diagnostic Laboratory. Seven birds were found to have liver abnormalities as either the cause of death (n = 2) or a comorbidity (5). A single archived plasma sample was evaluated for each of the clinically healthy birds, and 2 to 4 different time points of archived plasma samples were evaluated from the birds with liver disease. The samples from clinically healthy birds were free from moderate and marked hemolysis and lipemia. All samples had been archived in a −80 °C freezer before evaluation. All samples were processed at the University of Miami, Avian and Wildlife Laboratory within 1 run. Testing was completed on a Vitros 5600 analyzer (Ortho Clinical Diagnostics) using reagents purchased from Randox. The analyzer was maintained by standard quality assurance, and control procedures and tests were performed per manufacturer guidelines.

Statistical analyses were carried out using IBM SPSS Statistics, version 28, and MedCalc, version 22.014. Data were normal in distribution by the D’Agostino-Pearson test (P = .51). The robust method was used after the Box-Cox transformation to establish GLDH reference intervals for healthy penguins via the American Society for Veterinary Clinical Pathology guidelines.¹² In clinically healthy birds GLDH was compared between sexes using a Mann-Whitney U, and a Spearman correlation was utilized to investigate the relationship between age and weight. In birds with liver disease, GLDH was compared to ALT and AST, all from the same blood sample, with a Spearman correlation. To evaluate the potential effects of plasma storage time, GLDH activities were compared to days frozen in both the clinically healthy and liver disease birds using a Spearman correlation. In all cases, α = 0.05.

Results

In 29 clinically healthy penguins, GLDH ranged from 0 to 12.14 U/L, the mean was 6.01, and the reference interval was 0 to 13.55 U/L (0 [90% CI, 0 to 0.6] to 13.6 [90% CI, 11.7 to 15.3]; Figure 1). Two birds’ blood samples were found to have significant hemolysis, so a blood sample from another time point corresponding to clinical health was tested to develop reference intervals. The prevalence of liver disease in the deceased penguin population was 58% (7/12). Hepatic histopathological findings from the 7 penguins with liver disease included (Table 1) hemosiderosis (n = 4; 3 mild, 1 severe), hepatitis (2), metastatic neoplasia (2; melanoma, proventriculus carcinoma), hepatic necrosis (2), lymphoplasmacytic infiltrates (1), and hepatic sinusoidal yolk emboli (1). Four birds with confirmed liver disease had GLDH activities that were at or exceeded the established reference interval at a minimum of 1 time point evaluated (Figure 2). Specific birds with hepatopathies and their respective elevated GLDH activities included (1) disseminated metastatic melanoma (GLDH, 13.88 U/L, 15.34 U/L, and 61.95 U/L); (2) acute severe hepatic necrosis with hemorrhage (GLDH, 26.57 U/L); (3) multifocal marked necrotizing heterophilic hepatitis with a hepatic cyst (GLDH, 13.54 U/L); and (4) hepatic sinusoidal yolk emboli, minimal to mild hepatic necrosis, with severe hepatic hemosiderosis (GLDH, 16.43). Three birds with liver disease and GLDH activities within the newly established reference interval at the time points evaluated included (1) proventricular carcinoma with metastasis to the liver and hepatic extramedullary lymphomyelopoesis, mild to marked with proliferation or bile ductules and fibrosis, and mild hemosiderosis; (2) mild chronic lymphoplasmacytic hepatitis with mild multifocal hemosiderin; and (3) lymphoplasmacytic infiltrates with Kupffer cell hemosiderosis and mild perivascular extramedullary hematopoiesis.

View Full Size

Box and whisker plot of plasma glutamate dehydrogenase (GLDH) in clinically healthy African penguins (Sphensicus demersus; n = 29) in managed care.

Citation: American Journal of Veterinary Research 86, 5; 10.2460/ajvr.24.12.0410

• Download Figure
• Download figure as PowerPoint slide

View Full Size

African penguins (S demersus) with confirmed liver disease at time of death with 1 or more elevations in GLDH activities over time (n = 4). Time points before death may include routine health examinations. The dashed line indicates the upper limit of the newly established reference interval in clinically healthy penguins (0–13.55 U/L).

Citation: American Journal of Veterinary Research 86, 5; 10.2460/ajvr.24.12.0410

• Download Figure
• Download figure as PowerPoint slide

Sex (r = 0.235; 90% CI, −4.94 to 0.78), age (r = −0.07; 90% CI, −0.382 to 0.256), and weight (r = 0.172; 90% CI, −0.157 to 0.467) had no effect on GLDH in clinically healthy birds. Glutamate dehydrogenase correlated significantly with ALT (r = 0.496, P < .01) and AST (r = 0.476, P < .01) in birds with liver disease. The storage time of plasma samples did not have any statistically significant effects on GLDH in clinically healthy or liver disease birds (P < .581).

Discussion

Reference intervals were established in clinically healthy African penguins and were found to be similar to what has been reported in canines, felines, and equids.¹³ No outliers were identified in the clinically healthy penguin population for plasma GLDH. However, in a previous publication¹⁰ evaluating blood chemistry data in Humboldt penguins (Spheniscus humboldti), the upper reference limit reported was significantly higher (37 U/L) compared to the current study in African penguins (13.55 U/L). Species differences could account for the variation seen between studies and may underscore the need for species-specific reference intervals. However, additional considerations for differences seen between avian species studied include sample collection site (dorsal coccygeal vein vs jugular vein in the present study), hemolysis, storage (next-day processing vs archived at −80 °C in the current study), sample size, and differences in test implementation and analytic methodology.

Glutamate dehydrogenase appears to be clinically useful for detecting liver disease in avians as in mammals, since penguins with confirmed hepatopathies demonstrated GLDH activities exceeding the newly established reference interval for clinically healthy birds. However, it is important to note that there are also time points in birds with liver disease confirmed at the time of death that did not have elevated GLDH (Table 1). In the bird with metastatic melanoma, elevated GLDH started 7.5 months before death (GLDH, 13.88 U/L at 225 days before death, 15.34 U/L at 134 days before death, and 61.95 U/L at 86 days before death); however, interestingly, GLDH was within in the normal reference interval (9.12 U/L) 3 days before death. On gross pathological evaluation, the majority of liver tissue had been effaced by the neoplasm, so the severe tissue loss could have accounted for the difference in plasma GLDH activities. Three other birds with liver disease and elevated GLDH, at the time point sampled closest to death, included (1) multifocal, marked necrotizing and heterophilic hepatitis (GLDH, 13.54 U/L) 68 days before death; (2) minimal to mild hepatic necrosis, severe hepatic hemosiderosis, and hepatic sinusoidal yolk emboli (GLDH, 16.43 U/L) 1 day before death; and (3) acute severe hepatic necrosis with hemorrhage (GLDH, 26.57 U/L) 1 day before death. The remainder of the time points sampled for these birds were far before the time of death, ranging from 79 to 670 days premortem, since most samples in the archive were often from routine blood analyses during preventative wellness examinations; therefore, such blood samples likely represented times of apparent health before disease and not necessarily a false negative result. In addition, 1 bird with acute severe hepatic necrosis with hemorrhage had normal blood chemistry data 3 days before death (GLDH, 0 U/L; ALT, 50 U/L; and AST, 178 U/L) and elevated GLDH 1 day before death (26.57 U/L), which may underscore how acute onset the bird’s hepatic disease was. It is also noteworthy that GLDH has previously been believed to be a late-stage liver disease indicator; however, in the present study, elevated activities were seen up to 225 days before death, which may demonstrate its value before end-stage disease.

Three birds with liver disease did not demonstrate changes in GLDH at the time points evaluated, specific pathologies (Table 1) included (1) proventricular carcinoma with hepatic metastasis, (2) mild chronic lymphoplasmacytic hepatitis, and (3) mild lymphoplasmacytic infiltrates with hemosiderosis. Several factors have been reported to affect GLDH and may account for differences seen with various etiologies or pathological processes. Glutamate dehydrogenase is a large enzyme that requires significant hepatocyte damage to cause increased activities in serum or plasma. Pathological processes such as liver necrosis may have a propensity for affecting GLDH, as seen with Pacheco disease in New World parrots.³ In the present study, the 3 penguins that had necrotic hepatopathies had elevated GLDH; however, the 2 penguins that had hemosiderosis, hepatitis, or mild inflammatory infiltrates did not have changes in GLDH. Only 1 of the 2 metastatic neoplasm cases had increased GLDH. Therefore, mineral or cellular infiltrates in the hepatic tissue may not be as reliable in affecting GLDH as processes that cause hepatocellular damage. Finally, the timing of blood sampling in relation to hepatocyte damage can affect GLDH activities, since the enzyme’s half-life has been found to be short in other species (dog, 8 hours; bovine, 14 hours; and humans, 18 hours),^5,13 although the half-life in penguins is unknown. Overall, GLDH can be utilized to rule in liver disease; however, normal results cannot conclusively rule out liver disease based on these findings.

In the present study, neither weight, age, nor sex were correlated to GLDH in African penguins. However, in a previous publication¹⁰ in Humboldt penguins, females had significantly increased baseline values as compared to males (90% percentiles: male, 14.6 U/L; female, 32.5 U/L).

The clinically significant prevalence of liver disease as determined from histopathological reports in the zoological institution’s population evaluated in the present study (58% [7/12]) included both hepatopathies as a comorbidity (5 birds with inflammatory, hemosiderin, or neoplastic infiltrates) or as a primary cause of death (2 birds from hepatic necrosis). In a previous study² of mortality trends in African penguins, both gastrointestinal and hepatic diseases were reported together (6% [22/323 birds]) making direct comparison challenging. Overall, the prevalence of liver disease in African penguins underscores the value of noninvasive methodologies for detecting liver disease in the species.

Glutamate dehydrogenase activities showed significant positive correlations with ALT and AST activities in birds with liver disease (P < .01) in the present study. While each enzyme can serve as an indicator of liver disease, AST, ALT, and LDH also increase with extrahepatic diseases involving muscle, kidney, bone, and/or erythrocytes.³ Previous studies^7,8 in other avian species have shown that the only significant activities of GLDH are present in hepatocytes, making it more specific for liver disease than the other enzymes. Glutamate dehydrogenase is particularly useful in differentiating muscle damage from hepatic injury, which also differs from the other enzymes.¹⁴ However, the tissue distributions of GLDH in African penguins are currently unknown and warrant further study. In humans with liver damage, GLDH was shown to undergo rapid elimination, while ALT remained elevated; therefore, GLDH may serve as a real-time monitor for active liver injury.^15,16

The quantitation of GLDH can be affected by hemolysis, lipemia, and biliverdinemia. Samples with moderate and marked hemolysis were excluded in the present study. Glutamate dehydrogenase has been shown to decrease over time when stored at −70 °C,¹⁷ and in the present study for liver disease birds, archived samples were stored at −80 °C for more than 6 months; therefore, the reported data could have been affected. However, no statistically significant effect on GLDH activities was seen in the current study when storage time was compared to GLDH in all birds, so if affected, it may have been uniformly or only minimally affected.

Reference intervals were established in clinically healthy African penguins. Glutamate dehydrogenase appears to be a clinically useful predictor of liver disease in avians as in mammals. However, this liver biomarker is able to rule in liver disease, although it cannot definitively rule it out. The prevalence of liver disease was clinically significant in the population evaluated, underscoring the value of antemortem testing for the medical management of the species. The free-ranging African penguin population is facing continued threats to its survival, and the rehabilitation of ill birds supports the ongoing conservation of the species. These data serve to expand the understanding of noninvasive diagnostics in liver disease in penguins, which may ultimately contribute to the advancement of care and conservation of this at-risk species.

Disclosures

The authors have nothing to disclose. No AI-assisted technologies were used in the composition of this manuscript.

Funding

The authors have nothing to disclose.