Serum haptoglobin concentration and liver enzyme activity as indicators of systemic inflammatory response syndrome and survival of sick calves

Abstract Background Increased concentration of haptoglobin (Hp) in serum is associated with survival of critically ill humans and horses. High serum activity of liver‐derived enzyme is associated with sepsis in children and foals. Hypothesis/Objectives Investigate whether admission serum Hp and glutamic dehydrogenase (GLDH) are associated with systemic inflammatory response syndrome (SIRS) and survival of sick calves. Animals One hundred two calves. Methods Retrospective cross‐sectional study. Electronic medical records from all calves <30 days of age admitted to a teaching hospital for 8 years were reviewed. The signalment, clinicopathological findings, the presence of SIRS, final diagnosis, hospitalization time and outcome were recorded. A Cox proportional hazard ratio (HzR) were calculated to assess the association between clinicopathological variables and survival to discharge. Results Serum Hp concentrations were similar between SIRS (0.29 g/L; range, 0.05‐3.6) and non‐SIRS calves (0.22 g/L; range, 0‐4.2; P = .62). GLDH activity was similar between SIRS (12 U/L; range, 1‐1025) and non‐SIRS calves (9 U/L; range, 2‐137; P = .2). Absent suckle reflex (HzR: 6.44, 95% CI: 1.44‐28.86), heart rate (HR) < 100 beats per minute (bpm; HzR: 12.2; 95% CI: 2.54‐58.62), HR > 140 bpm (HzR: 3.59, 95% CI: 1.05‐12.33), neutrophil count <1.7 × 109/L (HzR: 7.36; 95% CI: 2.03‐26.66) and increased gamma‐glutamyl transferase activity (every 50‐unit, HzR: 1.12; 95% CI: 1.03‐1.21) were predictive of nonsurvival. Conclusions and Clinical Importance The use of Hp and GLDH for prediction of survival in sick calves cannot be recommended at this time.


| INTRODUCTION
Haptoglobin (Hp) is a positive acute phase inflammatory protein that is produced by the liver, and increased production during inflammation is stimulated by interleukin-1, interleukin-6, and tumor necrosis factor α released from macrophages and monocytes at the site of inflammation. 1,2 The anti-inflammatory properties of Hp are associated with its antioxidant activity because hemoglobin binding to Hp prevents heme and iron release, which reduce radical oxygen species generation. 3 Haptoglobin (Hp) improves immune-tolerance via suppressing the proinflammatory mediators and plays a role in preventing bacterial proliferation during bacteremia. 4 In humans, increased serum Hp concentrations are associated with decreased in-hospital mortality because sepsis, whereas an increased risk of mortality occurs in septic subjects with increased concentrations of circulating free-hemoglobin because of low serum Hp concentrations. 5 There is increased concentration of Hp in calves with diarrhea, 6 respiratory disease complex 7,8 and during experimentally induced Salmonella Dublin infection. 9 In those diseases, serum Hp concentration correlates positively with the severity of illness. 6,9 In bovine medicine, Hp concentration is utilized in cattle to assist in the diagnosis and prediction of several diseases (eg, pneumonia, traumatic reticuloperitonitis, metritis), 7 Assessment of hepatic necro-inflammatory activity and cholestasis in large animals requires the measurement of serum activity of liver-derived enzyme including serum aspartate aminotransferase (AST), glutamic dehydrogenase activity (GLDH), sorbitol dehydrogenase (SDH), gamma-glutamyltransferase (GGT), alkaline phosphatase (ALP). In ruminants, GLDH is a sensitive indicator of hepatic necrosis, more specific than AST, and has a better in vitro storage ability than SDH. 13 In children, high serum activity of liver-derived enzyme is commonly identified in hospitalized subjects and increased activity is associated with death. 14 In foals, serum activity of liver-derived enzyme (GGT and SDH) is increased during sepsis, but their concentrations are not associated with survival. 15 Calves are frequently presented to tertiary hospitals with primary diseases associated with systemic alterations such as hypoxia, hypoperfusion, or sepsis that can lead to hepatic injury. However, the association between serum activity of liver-derived enzyme with SIRS or survival during hospitalization of sick calves is yet to be described.  Haptoglobin concentrations were measured by determining the hemoglobin binding capacity of serum, which was quantified against a standard sample. 16 From the blood gas analysis, the blood pH, venous PCO 2 (PvCO2, mm Hg), HCO 3 À (mmol/L), L-lactate (L-Lact, mmol/L) and ion-

| Statistical analysis
Normality of the data was evaluated using the Kolmogorov-Smirnov test. The mean and SD were calculated for normally distributed variables and the median and ranges were determined for nonnormally distributed variables. Numerical variables were compared between groups (eg, SIRS vs non-SIRS and survivors vs nonsurvivors) using a tstudent or Mann-Whitney U-tests, depending on the normality of the data, whereas categorical variables were compared using the Fisher exact or X 2 tests.
A Cox proportional hazard model was constructed to assess the impact of potential predictor variables on the outcome of interest

| SIRS and non-SIRS calves
Eighty-four (82%) calves had available data for SIRS classification.
Forty-nine (58%) of 84 calves were considered as having SIRS. Calves with SIRS had a significantly lower temperature than non-SIRS calves; however, the median temperature for both groups was within the normal ranges (SIRS: 38.6 ± 1.16 C and non-SIRS: 39.2 ± 0.7 C; P < .003). The proportion of calves presenting in standing or in lateral recumbency was significantly different between the SIRS group (40% standing and 55% in lateral recumbency) compared with the non-SIRS group (74% standing and 23% in lateral recumbency; P < .001; Table S2).  Table S4).  were significant (Table 3)

| DISCUSSION
In the present study, no significant differences were identified in Hp concentration of calves with or without SIRS. Similarly, admission serum Hp concentration was not associated with survival of hospitalized sick calves. This is in contrast to our hypothesized outcomes, in which we expected associations between serum Hp with severity of the disease (eg, SIRS) or survival in sick calves. We based our hypotheses on previous studies in calves showing that serum or blood concentration of Hp in diarrheic calves, 6 and calves with respiratory disease complex 7,8 correlated with the severity of clinical signs. Higher Hp concentration occurs in calves with fever, obtundation and severe dehydration compared to calves without mild clinical signs. 6 Additionally, in humans with sepsis 5 and horses with colic, 21 plasma Hp are higher in survivors than nonsurvivors. The reasons for the lack of association between serum Hp concentration and severity of illness or survival in our study are unknown. However, similar to our results, no association between Hp concentrations at hospital admission and survival was identified in cats, 22 critically ill foals 23 or humans with sepsis. 24 Our results indicated that in the study sample of calves included in this study, Hp was useful to identify an ongoing inflammatory process; however, its ability to predict severity of disease or survival was poor. Future studies should investigate whether longitudinal monitoring, rather than a single Hp measurement, can better predict survival of sick calves as reported with other acute phase proteins in dogs. 25 The GLDH activity, an indicator of hepatic insult, was not a predictor of survival in the calves included in our study based on multivariable analysis; however, GLDH activity was higher in nonsurviving than surviving calves. Similarly, serum activity of liverderived enzyme (GGT and SDH) is increased in foals with sepsis, but they are not useful predictors of survival. 15 Primary hepatic disease was only reported in 1 calf included in this study, and high GLDH activity was most commonly identified secondary to other diseases, such as sepsis, diarrhea, and pneumonia. Therefore, high GLDH activity likely resulted from hepatic insult associated with hypoperfusion, tissue hypoxia or hematogenous dissemination of bacteria or endotoxin. 26,27 The number of calves with SIRS was higher in the nonsurviving than in the surviving group. In addition, the clinical diagnoses and postmortem examinations of nonsurviving calves indicated that most of the calves died because of sepsis (n = 17/22). This could explain the association of higher percentages of dehydration, L-lactate, and creatinine concentration in nonsurviving calves in the univariable analysis, and the association of low neutrophil count with nonsurvival identified in the multivariable analysis. High GGT activity was also associated with nonsurvival. In contrast, low serum GGT activity (<31 U/L) was reported as a risk factor for nonsurvival of sick neonatal calves with diarrhea. 16 The authors of that study proposed that lack of consumption of colostrum, suggested by the low GGT activity, and therefore failure of transfer of passive immunity (FTPI) could explain the association between low GGT activity and mortality of sick neonatal calves with diarrhea in an age-dependent manner. In our study, it is likely that nonsurviving calves consumed colostrum, as suggested by the high serum GGT activity, but nothing is known about the IgG concentration in the consumed colostrum because the predictor capacity of GGT activity for transfer of passive immunity is poor. 28  might also be a contributing factor to the changes in liver enzyme activity. 18,31 The retrospective nature of our study prevented us from determining the specific source of GGT activity elevation in this group of calves.
Limitations of this study include its retrospective nature, the subsequent lack of standardized data collection, and the lack of categorization of clinical signs and treatments, which prevented assessment of the effect of individual treatment variations on the survival analysis.
Economic constraints could have contributed to the decision to euthanize calves in the nonsurvival group. However, the diagnosis of most euthanized calves had a grave prognosis regardless of the available economic resources. Our study sample was a heterogeneous group of sick calves with different levels of sickness. Thus, we classified the calves as SIRS and non-SIRS, and our results are comparable to those reported previously in hospitalized calves with diarrhea using the same SIRS classification criteria. 16 An additional limitation of our study was that the reference ranges used to dichotomize selected variables of the biochemical profile were based on values for calves 3 to 4 weeks of age or for adult cattle. The rational for choosing these references ranges was that in our study 70% (n = 72) calves were > 7 days of age (mean age 12 days) and biochemical variables of calves >1 week of age are comparable to those obtained from adult animals. 18,31 Similarly, reference ranges for hematological variables of calves >5 days of age are comparable to those from 3 to 4 weeks of age. 19,20,32 Another limitation of our study is that only 38% of potentially eligible sick calves admitted during the study period fulfilled the inclusion criteria which would likely tend to bias the study toward sicker or high-risk septic calves. Therefore, the results of our study should be interpreted taking this limitation into consideration when extrapolating to a different study sample.
In summary, serum Hp concentrations and GLDH activity were not useful in predicting SIRS or nonsurvival in sick, hospitalized calves younger than 30 days of age enrolled in this study. In contrast, suckle reflex quality, HR, neutrophil count, and GGT concentration were predictive of mortality.

ACKNOWLEDGMENT
No funding was received for this study.