Clinical implications of high liver enzyme activities in hospitalized neonatal foals

Emily F. Haggett Veterinary Medical Teaching Hospital, School of Veterinary Medicine, University of California-Davis, Davis, CA 95616.

Search for other papers by Emily F. Haggett in
Current site
Google Scholar
PubMed
Close
 BVSc, DACVIM
,
K. Gary Magdesian Department of Medicine and Epidemiology, School of Veterinary Medicine, University of California-Davis, Davis, CA 95616.

Search for other papers by K. Gary Magdesian in
Current site
Google Scholar
PubMed
Close
 DVM, DACVIM, DACVECC, DACVCP
, and
Philip H. Kass Department of Population Health and Reproduction, School of Veterinary Medicine, University of California-Davis, Davis, CA 95616.

Search for other papers by Philip H. Kass in
Current site
Google Scholar
PubMed
Close
 DVM, PhD, DACVPM

Click on author name to view affiliation information

Abstract

Objective—To determine whether high liver enzyme activities were negatively associated with outcome in sick neonatal foals as compared with foals that did not have high liver enzyme activities.

Design—Retrospective case-control study.

Animals—147 foals < 30 days old with high γ-glutamyltransferase activity, high sorbitol dehydrogenase activity, or both (case foals) and 263 foals < 30 days old with γ-glutamyltransferase and sorbitol dehydrogenase activities within reference limits (control foals).

Procedures—Medical records were reviewed for information on signalment, physical examination findings, and results of hematologic and serum biochemical analyses performed at the time of admission to a veterinary medical teaching hospital.

Results—Case foals were significantly more likely to die or be euthanized, compared with control foals (odds ratio, 2.22; 95% confidence interval, 1.28 to 3.85). Case foals were significantly more likely than control foals to have sepsis, and septic foals were significantly less likely to survive than were nonseptic foals. For case foals, other factors associated with a greater likelihood of nonsurvival were higher anion gap and higher logarithm of aspartate aminotransferase activity. When sepsis status was controlled for, the presence of high liver enzyme activities was not significantly associated with outcome.

Conclusions and Clinical Relevance—Results suggested that high liver enzyme activities were common in sick neonatal foals, especially foals with sepsis. Foals with high liver enzyme activities were more likely to be septic, and septic foals were less likely to survive than were foals without sepsis. However, high liver enzyme activities alone were not a useful negative prognostic indicator.

Abstract

Objective—To determine whether high liver enzyme activities were negatively associated with outcome in sick neonatal foals as compared with foals that did not have high liver enzyme activities.

Design—Retrospective case-control study.

Animals—147 foals < 30 days old with high γ-glutamyltransferase activity, high sorbitol dehydrogenase activity, or both (case foals) and 263 foals < 30 days old with γ-glutamyltransferase and sorbitol dehydrogenase activities within reference limits (control foals).

Procedures—Medical records were reviewed for information on signalment, physical examination findings, and results of hematologic and serum biochemical analyses performed at the time of admission to a veterinary medical teaching hospital.

Results—Case foals were significantly more likely to die or be euthanized, compared with control foals (odds ratio, 2.22; 95% confidence interval, 1.28 to 3.85). Case foals were significantly more likely than control foals to have sepsis, and septic foals were significantly less likely to survive than were nonseptic foals. For case foals, other factors associated with a greater likelihood of nonsurvival were higher anion gap and higher logarithm of aspartate aminotransferase activity. When sepsis status was controlled for, the presence of high liver enzyme activities was not significantly associated with outcome.

Conclusions and Clinical Relevance—Results suggested that high liver enzyme activities were common in sick neonatal foals, especially foals with sepsis. Foals with high liver enzyme activities were more likely to be septic, and septic foals were less likely to survive than were foals without sepsis. However, high liver enzyme activities alone were not a useful negative prognostic indicator.

Liver disease in adult horses has been well described,1–3 but to our knowledge, there is little published information about liver disease in neonatal foals beyond case reports and small case series.4–8 Reported causes of liver disease in neonatal foals include sepsis, PAS, equine herpesvirus-1 infection, neonatal isoerythrolysis, Tyzzer's disease, ascending cholangiohepatitis, toxic hepatopathy, and congenital abnormalities.4–10 With some of these diseases, including Tyzzer's disease and herpesvirus infection, the hepatic involvement is primary,5,6,8 whereas with others, hepatic abnormalities develop secondary to hypoxia, hypoperfusion, or sepsis.10 Outcome and prognostic indicators have been reported for foals with some of the liver-specific diseases6,8,11 as well as for septic and critically ill foals in general.12–17 However, the effect of hepatic involvement on prognosis in foals with these conditions has not been studied. Specifically, whether the magnitude of increase in liver enzyme activities has any relation to prognosis in neonatal foals is unknown. In addition, descriptions of histopathologic changes in foals with high liver enzyme activities are limited.4,5,7,8,18 Liver enzyme activities are commonly high in neonatal children with sepsis or perinatal asphyxia, and there is some evidence that high liver enzyme activities have an adverse effect on prognosis in these neonates.19,20

The present study was designed to determine whether liver enzyme activities were associated with outcome in sick neonatal foals. Specifically, the purpose of the study reported here was to determine whether high liver enzyme activities were negatively associated with outcome in sick neonatal foals and whether septic foals with high liver enzyme activities were less likely to survive, compared with septic foals with liver enzyme activities within reference limits. In addition, we wanted to document clinical and clinicopathologic abnormalities and outcome in sick neonatal foals with biochemical evidence of liver disease.

Materials and Methods

Case and control selection—Medical records of the William R. Pritchard VMTH were searched to identify sick foals < 30 days old that were admitted between January 1, 2000, and December 31, 2006. Foals were included if they were admitted for treatment of any illness and, at the time of admission, had a serum GGT activity > 45 U/L, a serum SDH activity > 15 U/L, or both (case group) or if they were admitted for treatment of any illness and, at the time of admission, had a serum GGT activity ≤ 45 U/L and a serum SDH activity ≤ 15 U/L (control group). Clinically normal foals brought to the VMTH for routine health checks and healthy foals hospitalized with their sick dams were excluded from the study. For the present study, GGT and SDH activities were used as markers for liver disease, because both enzymes are highly liver specific.21,22 Reference ranges for liver enzyme activities in neonatal foals are commonly higher than ranges for adult horses.23,24 Thus, VMTH reference ranges for neonatal foals developed on the basis of results for 15 healthy university-owned neonatal foals were used.

Procedures—Information obtained from medical records of foals included in the study consisted of signalment, physical examination findings, and results of hematologic and serum biochemical analyses performed at the time of admission to the VMTH. Hematologic variables were measured by use of an automated analyzer.a,b Serum biochemical data were obtained with a clinical chemistry analyzer.c Results of bacterial culture of blood samples and results of histologic examination of liver biopsy or necropsy samples were also recorded when available. Hospitalization time, survival to discharge, and clinical diagnosis (ie, the final diagnosis determined by the attending clinician) were also recorded. For foals that were euthanized, the reason for euthanasia was recorded. Foals that were euthanized for financial reasons alone were excluded from statistical analysis of mortality rate. Long-term follow-up data were obtained by telephone communication with the owner or trainer.

Foals were considered to have SIRS if they met at least 2 of the following criteria: leukocytosis (WBC count > 12,500 cells/μL), leukopenia (WBC count < 4,000 cells/μL), or high (≥ 10%) immature neutrophil fraction; hypothermia (rectal temperature < 37.2°C [99°F]) or hyperthermia (rectal temperature > 39.2°C [102.5°F]); tachycardia (heart rate ≥ 120 beats/min) or bradycardia (heart rate < 60 beats/min); and tachypnea (respiratory rate > 40 breaths/min). A SIRS score ranging from 0 to 4 was assigned on the basis of the number of criteria fulfilled. A sepsis score was calculated according to the modified sepsis scoring system25; foals were considered to have sepsis if they had a sepsis score ≥ 11.

Statistical analysis—The Mann-Whitney test was used to compare the distributions of age at the time of initial examination, hospitalization time, sepsis score, and SIRS score between the case and control groups. A non-parametric test was chosen on the basis of failure of the data to conform to a normal distribution, as determined by use of the method of Kolmogorov and Smirnov. A Fisher exact test was used to compare dichotomous factors between groups, including whether foals survived to discharge (yes vs no), results of bacterial culture of blood samples (positive vs negative), and clinical diagnosis (colic, cardiac disease, enteritis, failure of passive transfer, neonatal isoerythrolysis, umbilical infection, ocular disease, PAS, pneumonia, sepsis, septic arthritis, Tyzzer's disease, uroabdomen, or other), and to test for an association between histologic diagnosis and clinical diagnosis at necropsy. A Fisher exact test was also used to test whether survival to discharge was associated with sepsis score (≥ 11 vs < 11) or SIRS score (≥ 2 vs < 2) and, for foals with sepsis, whether survival to discharge was associated with liver enzyme activities (high vs within reference limits).

Logistic regression was used to quantify the magnitude of association between potential continuous and categorical risk factors and survival to discharge by comparing foals that either died or were euthanized because of a poor medical prognosis (nonsurvivors) with foals that were discharged alive (survivors). Foals that were euthanized for purely economic reasons were censored. Analyses were conducted for all foals and for foals with 1 or more high liver enzyme activities. Variables found to be predictive in univariate analysis were candidates for inclusion in the multivariate analysis. Final multivariate models included those variables that retained their statistical significance after controlling for other variables in the model. Interactions between potentially predictive hematologic and biochemical variables were evaluated by means of likelihood ratio tests. Models with continuous variables were evaluated for linearity in the log odds with likelihood ratio tests. All analyses were performed with standard software.d,e Values of P < 0.05 were considered significant.

Results

One hundred forty-seven foals with high GGT activity, high SDH activity, or both (case foals) met the criteria for inclusion in the study (Table 1). Median age at the time of initial examination was 2 days (range, 0 to 29 days), and median hospitalization time was 5 days (range, 0 to 71 days). Sepsis scores could not be calculated for 11 foals owing to insufficient data. Sepsis scores for the other foals ranged from 0 to 22, with 66 (49%) foals having a sepsis score ≥ 11. An SIRS score could be assigned to 137 of the foals; median SIRS score was 2 (range, 0 to 4), with 72 of the 137 (53%) foals having an SIRS score ≥ 2. Blood samples from 90 foals were submitted for bacterial culture, and results were positive for 22 of the 90 (24%) foals, including 6 foals from which 2 isolates were obtained, 1 foal from which mixed growth was obtained, and 15 foals from which a single isolate was obtained (Table 2). Sixteen of 53 (30%) foals with a sepsis score ≥ 11 had positive blood culture results, and 6 of 34 (18%) foals with a sepsis score < 11 had positive blood culture results. Of the 147 foals with high liver enzyme activities, 108 (74%) survived, 8 (5%) died, and 31 (21%) were euthanized. For 7 of the 31 foals that were euthanized, the decision for euthanasia was judged to have been made primarily for economic reasons, and these foals were excluded from mortality analyses. With these foals excluded, the survival rate was 77% (108/140).

Table 1—

Results of clinicopathologic testing performed at the time of admission for 147 sick neonatal (< 30 days old) foals with high (> 45 U/L) GGT activity, high (> 15 U/L) SDH activity, or both.

VariableMedianMinimumMaximumReference range
Hct (%)3495530–46
WBC count (cells/μL)8,80667036,6705,300–14,000
Neutrophils (cells/μL)6,6183028,9693,400–11,900
Lymphocytes (cells/μL)1,3112018,610700–2,900
Platelets (platelets/μL)210,00010,000414,000134,000–310,000
Fibrinogen (mg/dL)3001001,000100–300
Anion gap (mEq/L)197477–18
Sodium (mEq/L)138108158131–144
Potassium (mEq/L)3.92.16.53.1–5.0
Chloride (mEq/L)967110991–103
Calcium (mg/dL)11.64.317.710.8–12.8
Phosphorus (mg/dL)6.32.131.84.1–6.4
Creatinine (mg/dL)1.30.446.81.2–2.4
BUN (mg/dL)17314412–24
Glucose (mg/dL)1360392118–207
Total protein (g/dL)4.52.38.34.3–7.7
Albumin (g/dL)2.113.12.2–3.3
Globulin (g/dL)2.40.86.41.6–4.9
AST (U/L)29224,81870–213
CK (U/L)46690168,795138–508
AP (U/L)1,3321706,6961,362–3,752
GGT (U/L)49950520–45
SDH (U/L)2004952–11
Total bilirubin (mg/dL)3.10.431.81.8–3.8
Indirect bilirubin (mg/dL)30.223.31.7–3.6
Direct bilirubin (mg/dL)0.309.30.1–0.3

AP = Alkaline phosphatase. CK = Creatine kinase.

The control group consisted of 263 sick neonatal foals with GGT and SDH activities within reference limits. Median age at the time of initial examination was 2 days (range, 0 to 29 days), and median hospitalization time was 4 days (range, 0 to 58 days). Sepsis scores could not be calculated for 14 foals owing to insufficient data. Range of sepsis scores for the other foals was 0 to 21, with 64 (26%) foals having a sepsis score ≥ 11. An SIRS score could be assigned to 248 foals; median SIRS score was 1 (range, 0 to 4), with 114 of the 248 (46%) foals having an SIRS score ≥ 2. Blood samples from 130 foals were submitted for bacterial culture, and results were positive for 22 (17%; Table 2), including 2 foals from which 2 isolates were obtained and 20 foals from which a single isolate was obtained. Of the 263 foals in the control group, 225 (86%) survived, 4 (2%) died, and 34 (13%) were euthanized. For 8 of the 34 foals that were euthanized, the decision for euthanasia was judged to have been made primarily for economic reasons, and these foals were excluded from mortality analyses. With these foals excluded, the survival rate was 88% (225/255).

Table 2—

Results of bacterial culture of blood samples from sick neonatal foals with high GGT activity, high SDH activity, or both (case foals) or with GGT and SDH activities within reference limits (control foals).

Bacterial speciesCase foals 
Actinobacillus spp64
Escherichia coli78
Enterococcus spp21
Corynebacterium spp31
Streptococcus spp04
Klebsiella spp20
Staphylococcus spp02
Pasteurella spp10
Aeromonas spp10
Bacillus spp10
Micrococcus spp10
Clostridium spp10
Pantoea spp10
Unclassified species22
Moraxella spp01
Morganella spp01

Data are given as number of foals; results are given only for foals with positive blood culture results. Blood culture results were positive for 22 of 90 (24%) case foals, including 6 foals from which 2 isolates were obtained, 1 foal from which mixed growth was obtained (listed as unclassified species), and 15 foals from which a single isolate was obtained, and for 22 of 130 (17%) control foals, including 2 foals from which 2 isolates were obtained and 20 foals from which a single isolate was obtained.

When foals euthanized for financial reasons were excluded, foals with high liver enzyme activities were significantly more likely to die or be euthanized, compared with foals in the control group (23% vs 12%; OR, 2.22; 95% CI, 1.28 to 3.85; P = 0.006). Foals in the case group were hospitalized for significantly (P = 0.014) longer than were foals in the control group, but SIRS scores (P = 0.078) and percentages of foals with positive blood culture results (P = 0.176) were not significantly different between groups. The percentage of foals with a clinical diagnosis of sepsis was significantly higher for the case group than for the control group (OR, 5.77; 95% CI, 3.10 to 10.76; P < 0.001), and percentage of foals with a clinical diagnosis of colic (OR, 2.24; 95% CI, 1.08 to 4.66; P = 0.03) and percentage of foals with a clinical diagnosis of PAS (OR, 1.80; 95% CI, 0.99 to 3.25; P = 0.05) were higher in the control group than the case group (Figure 1).

Figure 1—
Figure 1—

Clinical diagnoses for 410 sick neonatal (< 30 days old) foals with high (> 45 U/L) GGT activity, high (> 15 U/L) SDH activity, or both (n = 147; white bars) or with GGT and SDH activities within reference limits (263; black bars). *Percentage was significantly (P < 0.05) different between groups.

Citation: Journal of the American Veterinary Medical Association 239, 5; 10.2460/javma.239.5.661

For foals with high liver enzyme activities, foals with a sepsis score ≥ 11 were significantly less likely to survive than were foals with a sepsis score < 11 (58% vs 96% survival rate, respectively; OR, 15.65; 95% CI, 4.43 to 55.33; P < 0.001), and foals with an SIRS score ≥ 2 were significantly less likely to survive than were foals with an SIRS score < 2 (64% vs 88% survival rate, respectively; OR, 3.91; 95% CI, 1.60 to 9.55; P = 0.002).

When the case and control groups were combined, there were 123 foals with a sepsis score ≥ 11, of which 75 (61%) survived, and 248 foals with a sepsis score < 11, of which 237 (96%) survived (with foals euthanized for financial reasons excluded). Foals with a sepsis score < 11 were significantly more likely to survive than were foals with a score ≥ 11 (OR, 13.79; 95% CI, 6.81 to 27.91; P < 0.001; with foals euthanized for financial reasons excluded). Septic foals were significantly (OR, 2.73; 95% CI, 1.76 to 4.23; P < 0.001) more likely to have high liver enzyme activities (66/130 [51%]) than were nonseptic foals (70/255 [27%]; all foals included).

A total of 372 foals had sufficient data for calculation of an SIRS score. Of these, 175 had an SIRS score ≥ 2, of which 129 (74%) survived, and 197 had an SIRS score < 2, of which 182 (92%) survived. Foals with an SIRS score < 2 were significantly more likely to survive than were foals with a score ≥ 2 (OR, 4.33; 95% CI, 2.32 to 8.08; P < 0.001). Eighty-eight of the 175 (50%) foals with an SIRS score ≥ 2 also had a sepsis score ≥ 11, and 85 of the 127 (67%) foals with a sepsis score ≥ 11 also had an SIRS score ≥ 2.

When the case and control groups were combined, results of blood culture were positive for 29 of 96 (30%) foals with a sepsis score ≥ 11, but for only 15 of 114 (13%) foals with a sepsis score < 11.

Twenty-six of 62 (42%) septic foals with high liver enzyme activities did not survive, compared with 22 of 61 (36%) septic foals with liver enzyme activities within reference limits; these proportions were not significantly (P = 0.58) different. For the control group, 22 of 108 (20%) foals with an SIRS score ≥ 2 did not survive, whereas for the case group, 24 of 67 (36%) foals with an SIRS score ≥ 2 did not survive. These proportions were significantly (P = 0.03) different.

Several factors were identified in univariate analyses to be associated with outcome (Table 3). When controlling for sepsis status, factors significantly associated with nonsurvival (died or euthanized) in univariate analyses included a lower WBC count, lower neutrophil count, lower albumin concentration, higher anion gap, higher BUN concentration, higher AST activity, and higher phosphorus concentration. In the final multivariate model, sepsis score ≥ 11, higher anion gap, and higher logarithm of AST activity were significantly associated with nonsurvival (Table 4).

Table 3—

Results of univariate analysis of clinical and clinicopathologic variables associated with discharge status (died or euthanized vs discharged alive) for 147 sick neonatal foals with high GGT activity, high SDH activity, or both.

 Median value 
VariableSurviving foalsNonsurviving foalsP value
Days hospitalized6.52<0.001
Temperature (°C)38.037.30.04
WBC count (cells/μL)10,4515,984<0.001
Band neutrophils (cells/μL)255475<0.001
Neutrophils (cells/μL)8,1563,970<0.001
Lymphocytes (cells/μL)1,5651,2750.014
Anion gap1826<0.001
Phosphorus (mg/dL)6.26.650.036
Creatinine (mg/dL)1.32.30.003
BUN (mg/dL)1630<0.001
Glucose (mg/dL)14353<0.001
Albumin (g/dL)2.12.00.002
Total protein (g/dL)6.26.70.03
AST (U/L)269430<0.001
CK (U/L)4111,372<0.001
Total bilirubin (mg/dL)2.94.850.003
Direct bilirubin (mg/dL)0.30.40.004
Lactate (mmol/L)3.699.65<0.001

Lactate concentration was only measured in 43 surviving and 22 nonsurviving foals.

See Table 1 for remainder of key.

Table 4—

Results of multivariate logistic regression analysis of clinical and clinicopathologic variables associated with discharge status for 147 sick neonatal foals with high GGT activity, high SDH activity, or both.

VariableOR  
Sepsis* (yes vs no)10.892.56–46.260.001
Anion gap (for every 1 mEq/L increase)1.151.06–1.25<0.001
Log AST (for every 1-U increase)2.711.30–5.660.008

Modified sepsis score ≥11.

Logistic regression was also used to test whether sepsis status or high liver enzyme activities were associated with survival to discharge for all foals combined. Sepsis was again significantly (P < 0.001) associated with nonsurvival, but high liver enzyme activities was not independently associated with nonsurvival (P = 0.56).

Long-term follow-up information was available for 59 foals with high liver enzyme activities. Of these foals, 49 were alive at 1 to 6 years of age. Reasons for death or euthanasia of the remaining 10 horses included lameness or fracture (n = 4), oleander intoxication (1), head injury (1), and small intestinal perforation (1). Three foals were euthanized or died within 2 months after discharge for reasons related to their primary condition, although not specifically related to liver disease, including persistent dysphagia (n = 1), persistent upper respiratory tract obstruction (1), and pyelonephritis (1). In the 49 horses that were alive at the time of long-term follow-up, there were no reports of clinically apparent liver disease; 48 of these horses were reported to be healthy, and 1 reportedly had poor performance of undetermined etiology.

Thirty-six foals with high liver enzyme activities underwent necropsy. In 10 foals (3 that were premature, 3 with sepsis, 2 with septic arthritis, and 1 with colic), no specific comments were made in the necropsy report about the histologic or gross appearance of the liver. Histologic abnormalities identified during examination of liver specimens obtained at necropsy included neutrophilic hepatitis (7 foals with sepsis, 2 with Tyzzer's disease, and 1 with equine herpesvirus-1 infection), congestion (4 foals with sepsis, 3 with a congenital cardiac defect, and 1 with pneumonia), necrosis (3 foals with sepsis), and other abnormalities, including hemorrhage, hypoplasia, bile stasis, abscess, and lipidosis (1 each in 1 foal with sepsis; total of 5 foals) and telangiectasia (1 foal that was premature).

One additional foal that did not undergo necropsy underwent hepatic biopsy; histologic examination of the biopsy specimen revealed severe, acute, multifocal necrotizing hepatitis consistent with Tyzzer's disease. A significant association was not found between clinical and histopathologic diagnoses.

Discussion

Results of the present study suggested that high liver enzyme activities were common in neonatal foals hospitalized because of illness, especially in foals with sepsis (ie, foals with a sepsis score ≥ 11). Of the 410 neonatal (< 30 days old) foals hospitalized during the 7-year study period, 147 (36%) had high liver enzyme (GGT, SDH, or both) activities, and septic foals were significantly more likely to have high liver enzyme activities (66/130 [51%]) than were nonseptic foals (70/255 [27%]). In addition, foals with sepsis were significantly less likely to survive than were foals without sepsis. However, high liver enzyme activities alone were not a useful negative prognostic indicator.

Primary hepatic diseases such as Clostridium piliformis infection were rare in this study, and high liver enzyme activities were most commonly detected secondary to other diseases, such as sepsis, PAS, and enteritis. In foals with these conditions, high liver enzyme activities are most likely a result of hepatic injury associated with hematogenous spread of bacteria or endotoxin, perfusion abnormalities, or hypoxic injury.

High liver enzyme activities are common in premature human infants with sepsis.26 Fibrin deposits have been detected in the hepatic sinusoids of septic foals at postmortem examination and are attributed to disseminated intravascular coagulation secondary to activation of the inflammatory and coagulation cascades.27 Direct hematogenous spread of bacteria is also possible,21 and for 2 foals euthanized in the present study, bacterial culture of liver specimens obtained at necropsy yielded pure cultures of Actinobacillus equuli. Sepsis has also been associated with cholestasis secondary to endotoxin-induced inhibition of bile salt uptake, which could possibly cause an increase in GGT.26 Histopathologic diagnoses for foals with sepsis in the present study included hepatitis, congestion, and necrosis, consistent with injury resulting from perfusion abnormalities, bacteremia, or systemic inflammation.

Foals with sepsis in the present study were significantly less likely to survive, compared with foals without sepsis, which is consistent with findings of a previous study.28 Sepsis has been reported to be the leading cause of death in neonatal foals < 7 days of age.29 Only 61% (75/123) of foals with sepsis in the present study survived, compared with 96% (237/248) of foals without sepsis. Systemic inflammatory response syndrome has previously been identified as a risk factor for death in foals,12 and our study supported this finding, with 74% (129/175) of foals with SIRS surviving, compared with 92% (182/197) of foals without SIRS.

The high percentage (66/136 [49%]) of foals in the present study with high liver enzyme activities that also had sepsis illustrates that hepatic injury is common during sepsis. Interestingly, septic foals with high liver enzyme activities were no less likely to survive than were septic foals with liver enzyme activities within reference limits. However, foals with SIRS and high liver enzyme activities were less likely to survive than were foals with SIRS and liver enzyme activities within reference limits.

Result of the multivariate analysis in the present study indicated that the presence of high liver enzyme activities alone, independent of sepsis, was not prognostic of a poor outcome. In addition, we did not detect an association between the magnitude of increase in serum GGT or SDH activity and outcome, and some foals with GGT activity > 10 times the upper reference limit survived. In adult horses with liver disease, a high serum GGT activity is negatively associated with survival.1,2 The likely reasons for this difference in results from previous studies include the fact that our study evaluated neonatal foals with presumably acute liver disease, whereas the previous studies primarily included horses with chronic liver disease.

In the present study, only a few factors were found to be significantly associated with outcome when multivariate logistic regression was performed. Other than sepsis, the only factors associated with nonsurvival were higher anion gap and higher logarithm of AST activity. Higher blood lactate concentration and high anion gap have consistently been shown to be negative prognostic indicators in neonatal foals.12,30–32 Anion gap can be used to approximate blood lactate concentration,22 although direct measurement is more accurate.12 For case foals in the present study, blood lactate concentration was significantly higher among foals that did not survive, compared with foals that survived to discharged; however, data for too few foals were available for this factor to be included in the multivariate analysis. Therefore, anion gap was used as a proxy. Lactate concentration is an indicator of systemic perfusion abnormalities and is typically high in foals with sepsis.12 Lactate is metabolized by the liver, and hepatic disease can lead to decreased clearance.

An increase in the logarithm of serum AST activity was also predictive of nonsurvival in the present study. Aspartate aminotransferase is not liver specific and can be derived from muscle tissue and erythrocytes as well as hepatocytes. It was not possible to identify the origin of the AST in the foals of our study. However, there was a lack of correlation between AST activity and activities of other hepatic-derived enzymes, including alkaline phosphatase, GGT, and SDH (data not shown), suggesting that AST may not have been of hepatic origin. Creatine kinase activity was higher in foals that did not survive (median, 1,372 U/L) than in foals that did survive (411 U/L), suggesting that AST may have been of muscular origin, but a significant difference in creatine kinase activity between groups was not identified after multivariate analysis. Irrespective of its origin, an increase in AST activity is suggestive of organ injury, likely from perfusion abnormalities or systemic inflammation.

The fact that among foals with high liver enzyme activities, BUN and phosphorus concentrations were higher in nonsurviors than in survivors suggested that multiorgan injury played a role in outcome. These variables, although not specific, are suggestive of impaired renal function, and azotemia has been reported to be negatively associated with survival in a previous study.33 White blood cell and neutrophil counts were significantly lower in nonsurvivors than in survivors, consistent with the finding that foals with the most severe systemic inflammation or sepsis were less likely to survive.

The long-term prognosis for foals in the present study that survived to discharge was good. Forty-nine of 59 (83%) foals for which follow-up information was obtained were alive, and none had been euthanized because of hepatic disease. All of the surviving horses were reported to be healthy with the exception of one with nonspecific poor performance.

To our knowledge, histologic findings have not previously been reported for foals with high liver enzyme activities, except in individual case reports describing foals with specific diseases such as Tyzzer's disease. Neutrophilic hepatitis was the most common histopathologic diagnosis, especially in foals with sepsis. Other diagnoses included hepatic congestion and necrosis, which occurred in foals with a variety of clinical diagnoses. These histologic findings coupled with the clinical diagnoses of sepsis and PAS suggest that hepatic disease was associated with perfusion abnormalities, ischemic injury, or systemic inflammation associated with the primary disease process.

The modified sepsis scoring system used in the present study has been reported to have a sensitivity of 93% and a specificity of 88% for identifying sepsis.34 A more recent study35 questioned the accuracy of the modified sepsis score. However, restricting the diagnosis of sepsis to those foals with positive blood culture results leads to underdiagnosis of sepsis. In an in vitro study36 that used conventional blood culture methods, only 31% of samples yielded a positive culture result. Thus, despite its limitations, we considered the modified sepsis score to be the most objective way to detect foals at high risk for sepsis. A small percentage of foals in this study with positive blood culture results had a sepsis score < 11, which highlights the concern that there may be limitations to the use of the sepsis score as a sole diagnostic criterion. However, false-positive bacterial culture results as a result of contamination during sample collection are also possible.

ABBREVIATIONS

AST

Aspartate aminotransferase

CI

Confidence interval

GGT

γ-Glutamyltransferase

OR

Odds ratio

PAS

Perinatal asphyxia syndrome

SDH

Sorbitol dehydrogenase

SIRS

Systemic inflammatory response syndrome

VMTH

Veterinary Medical Teaching Hospital

a.

Advia 120 Hematology Analyzer, Siemans Healthcare Diagnostics, Tarrytown, NY.

b.

Serano-Baker Hematology Analyzer, Serano-Baker Co, Allentown, Pa.

c.

Hitachi 717, Roche Diagnostics, Indianapolis, Ind.

d.

GraphPad InStat 3, GraphPad Software Inc, San Diego, Calif.

e.

Stata 10, StataCorp, College Station, Tex.

References

  • 1. Durham AE, Newton JR, Smith KC, et al. Retrospective analysis of historical, clinical, ultrasonographic, serum biochemical and haematological data in prognostic evaluation of equine liver disease. Equine Vet J 2003; 35: 542547.

    • Search Google Scholar
    • Export Citation
  • 2. McGorum BC, Murphy D, Love S, et al. Clinicopathological features of equine primary hepatic disease: a review of 50 cases. Vet Rec 1999; 145: 134139.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 3. Smith MR, Stevens KB, Durham AE, et al. Equine hepatic disease: the effect of patient- and case-specific variables on risk and prognosis. Equine Vet J 2003; 35: 549552.

    • Search Google Scholar
    • Export Citation
  • 4. Acland HM, Mann PC, Robertson JL, et al. Toxic hepatopathy in neonatal foals. Vet Pathol 1984; 21: 39.

  • 5. Bastianello SS, Nesbit JW. The pathology of a case of biliary atresia in a foal. J S Afr Vet Assoc 1986; 57: 117120.

  • 6. Borchers A, Magdesian KG, Halland S, et al. Successful treatment and polymerase chain reaction (PCR) confirmation of Tyzzer's disease in a foal and clinical and pathologic characteristics of 6 additional foals (1986–2005). J Vet Intern Med 2006; 20: 12121218.

    • Search Google Scholar
    • Export Citation
  • 7. Haechler S, Van den Ingh TS, Rogivue C, et al. Congenital hepatic fibrosis and cystic bile duct formation in Swiss Freiberger horses. Vet Pathol 2000; 37: 669671.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 8. Murray MJ, del Piero F, Jeffrey SC, et al. Neonatal equine herpesvirus type 1 infection on a thoroughbred breeding farm. J Vet Intern Med 1998; 12: 3641.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 9. Hess-Dudan F. Four possible causes of hepatic failure and or icterus in the newborn foal. Equine Vet Educ 1994; 6: 310315.

  • 10. Magdesian KG. Liver failure in the foal. In: Paradis MR, ed. Equine neonatal medicine: a case-based approach. Philadelphia: Elsevier Saunders, 2006; 221229.

    • Search Google Scholar
    • Export Citation
  • 11. Boyle AG, Magdesian KG, Ruby RE. Neonatal isoerythrolysis in horse foals and a mule foal: 18 cases (1988–2003). J Am Vet Med Assoc 2005; 227: 12761283.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 12. Corley KT, Donaldson LL, Furr MO. Arterial lactate concentration, hospital survival, sepsis and SIRS in critically ill neonatal foals. Equine Vet J 2005; 37: 5359.

    • Search Google Scholar
    • Export Citation
  • 13. Furr M, Tinker MK, Edens L. Prognosis for neonatal foals in an intensive care unit. J Vet Intern Med 1997; 11: 183188.

  • 14. Gayle JM, Cohen ND, Chaffin MK. Factors associated with survival in septicemic foals: 65 cases (1988–1995). J Vet Intern Med 1998; 12: 140146.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 15. Koterba AM, Brewer BD, Tarplee FA. Clinical and clinicopathological characteristics of the septicaemic neonatal foal: review of 38 cases. Equine Vet J 1984; 16: 376382.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 16. Peek SF, Darien BJ, Semrad SD, et al. A prospective study of neonatal septicemia and factors influencing survival, in Proceedings. 50th Annu Conv Am Assoc Equine Pract 2004;60–63.

    • Search Google Scholar
    • Export Citation
  • 17. Peek SF, Semrad S, McGuirk SM, et al. Prognostic value of clini-copathologic variables obtained at admission and effect of antiendotoxin plasma on survival in septic and critically ill foals. J Vet Intern Med 2006; 20: 569574.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 18. Pulley LT, Shively JN. Tyzzer's disease in a foal. Light- and electron-microscopic observations. Vet Pathol 1974; 11: 203211.

  • 19. Shamir R, Maayan-Metzger A, Bujanover Y, et al. Liver enzyme abnormalities in gram-negative bacteremia of premature infants. Pediatr Infect Dis J 2000; 19: 495498.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 20. Tarcan A, Tiker F, Guvenir H, et al. Hepatic involvement in perinatal asphyxia. J Matern Fetal Neonatal Med 2007; 20: 407410.

  • 21. Koterba AM, Drummond WH, Kosch PC. Neonatal hyperbilirubinemia. In: Koterba AM, ed. Equine clinical neonatology. Philadelphia: Lea & Febiger, 1990; 589601.

    • Search Google Scholar
    • Export Citation
  • 22. Carlson GP. Clinical chemistry rests. In: Smith BP, ed. Large animal internal medicine. 3rd ed. St Louis: Mosby, 2002; 389412.

  • 23. Bauer JE, Asquith RL, Kivipelto J. Serum biochemical indicators of liver function in neonatal foals. Am J Vet Res 1989; 50: 20372041.

  • 24. Patterson WH, Brown CM. Increase of serum γ-glutamyltransferase in neonatal Standardbred foals. Am J Vet Res 1986; 47: 24612463.

  • 25. Brewer B. Neonatal foal evaluation: sepsis and survival scoring in private practice, in Proceedings. 33rd Annu Conf Am Assoc Equine Pract 1987;812–822.

    • Search Google Scholar
    • Export Citation
  • 26. Moseley RH. Sepsis and cholestasis. Clin Liver Dis 2004; 8: 8394.

  • 27. Cotovio M, Monreal L, Armengou L, et al. Fibrin deposits and organ failure in newborn foals with severe septicemia. J Vet Intern Med 2008; 22: 14031410.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 28. Castagnetti C, Veronesi MC. Prognostic factors in the sick neonatal foal. Vet Res Commun 2008; 32: S87S91.

  • 29. Cohen ND. Causes of and farm management factors associated with disease and death in foals. J Am Vet Med Assoc 1994; 204: 16441651.

    • Search Google Scholar
    • Export Citation
  • 30. Henderson ISF, Franklin RP, Wilkins PA, et al. Association of hyperlactatemia with age, diagnosis, and survival in equine neonates. J Vet Emerg Crit Care 2008; 18: 496502.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 31. Wotman K, Wilkins PA, Palmer JE, et al. Association of blood lactate concentration and outcome in foals. J Vet Intern Med 2009; 23: 598605.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 32. Saulez MN, Gummow B, Slovis NM, et al. Admission clinicopathological data, length of stay, cost and mortality in an equine neonatal intensive care unit. J S Afr Vet Assoc 2007; 78: 153157.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 33. Rohrbach BW, Buchanan BR, Drake JM, et al. Use of a multi-variable model to estimate the probability of discharge in hospitalized foals that are 7 days of age or less. J Am Vet Med Assoc 2006; 228: 17481756.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 34. Brewer BD, Koterba AM. Development of a scoring system for the early diagnosis of equine neonatal sepsis. Equine Vet J 1988; 20: 1822.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 35. Corley KT, Furr MF. Evaluation of a score designed to predict sepsis in foals. J Vet Emerg Crit Care 2003; 13: 149155.

  • 36. Lorenzo-Figueras M, Pusterla N, Byrne BA, et al. In vitro evaluation of three bacterial culture systems for the recovery of Esch-erichia coli from equine blood. Am J Vet Res 2006; 67: 20252029.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Figure 1—

    Clinical diagnoses for 410 sick neonatal (< 30 days old) foals with high (> 45 U/L) GGT activity, high (> 15 U/L) SDH activity, or both (n = 147; white bars) or with GGT and SDH activities within reference limits (263; black bars). *Percentage was significantly (P < 0.05) different between groups.

  • 1. Durham AE, Newton JR, Smith KC, et al. Retrospective analysis of historical, clinical, ultrasonographic, serum biochemical and haematological data in prognostic evaluation of equine liver disease. Equine Vet J 2003; 35: 542547.

    • Search Google Scholar
    • Export Citation
  • 2. McGorum BC, Murphy D, Love S, et al. Clinicopathological features of equine primary hepatic disease: a review of 50 cases. Vet Rec 1999; 145: 134139.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 3. Smith MR, Stevens KB, Durham AE, et al. Equine hepatic disease: the effect of patient- and case-specific variables on risk and prognosis. Equine Vet J 2003; 35: 549552.

    • Search Google Scholar
    • Export Citation
  • 4. Acland HM, Mann PC, Robertson JL, et al. Toxic hepatopathy in neonatal foals. Vet Pathol 1984; 21: 39.

  • 5. Bastianello SS, Nesbit JW. The pathology of a case of biliary atresia in a foal. J S Afr Vet Assoc 1986; 57: 117120.

  • 6. Borchers A, Magdesian KG, Halland S, et al. Successful treatment and polymerase chain reaction (PCR) confirmation of Tyzzer's disease in a foal and clinical and pathologic characteristics of 6 additional foals (1986–2005). J Vet Intern Med 2006; 20: 12121218.

    • Search Google Scholar
    • Export Citation
  • 7. Haechler S, Van den Ingh TS, Rogivue C, et al. Congenital hepatic fibrosis and cystic bile duct formation in Swiss Freiberger horses. Vet Pathol 2000; 37: 669671.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 8. Murray MJ, del Piero F, Jeffrey SC, et al. Neonatal equine herpesvirus type 1 infection on a thoroughbred breeding farm. J Vet Intern Med 1998; 12: 3641.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 9. Hess-Dudan F. Four possible causes of hepatic failure and or icterus in the newborn foal. Equine Vet Educ 1994; 6: 310315.

  • 10. Magdesian KG. Liver failure in the foal. In: Paradis MR, ed. Equine neonatal medicine: a case-based approach. Philadelphia: Elsevier Saunders, 2006; 221229.

    • Search Google Scholar
    • Export Citation
  • 11. Boyle AG, Magdesian KG, Ruby RE. Neonatal isoerythrolysis in horse foals and a mule foal: 18 cases (1988–2003). J Am Vet Med Assoc 2005; 227: 12761283.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 12. Corley KT, Donaldson LL, Furr MO. Arterial lactate concentration, hospital survival, sepsis and SIRS in critically ill neonatal foals. Equine Vet J 2005; 37: 5359.

    • Search Google Scholar
    • Export Citation
  • 13. Furr M, Tinker MK, Edens L. Prognosis for neonatal foals in an intensive care unit. J Vet Intern Med 1997; 11: 183188.

  • 14. Gayle JM, Cohen ND, Chaffin MK. Factors associated with survival in septicemic foals: 65 cases (1988–1995). J Vet Intern Med 1998; 12: 140146.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 15. Koterba AM, Brewer BD, Tarplee FA. Clinical and clinicopathological characteristics of the septicaemic neonatal foal: review of 38 cases. Equine Vet J 1984; 16: 376382.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 16. Peek SF, Darien BJ, Semrad SD, et al. A prospective study of neonatal septicemia and factors influencing survival, in Proceedings. 50th Annu Conv Am Assoc Equine Pract 2004;60–63.

    • Search Google Scholar
    • Export Citation
  • 17. Peek SF, Semrad S, McGuirk SM, et al. Prognostic value of clini-copathologic variables obtained at admission and effect of antiendotoxin plasma on survival in septic and critically ill foals. J Vet Intern Med 2006; 20: 569574.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 18. Pulley LT, Shively JN. Tyzzer's disease in a foal. Light- and electron-microscopic observations. Vet Pathol 1974; 11: 203211.

  • 19. Shamir R, Maayan-Metzger A, Bujanover Y, et al. Liver enzyme abnormalities in gram-negative bacteremia of premature infants. Pediatr Infect Dis J 2000; 19: 495498.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 20. Tarcan A, Tiker F, Guvenir H, et al. Hepatic involvement in perinatal asphyxia. J Matern Fetal Neonatal Med 2007; 20: 407410.

  • 21. Koterba AM, Drummond WH, Kosch PC. Neonatal hyperbilirubinemia. In: Koterba AM, ed. Equine clinical neonatology. Philadelphia: Lea & Febiger, 1990; 589601.

    • Search Google Scholar
    • Export Citation
  • 22. Carlson GP. Clinical chemistry rests. In: Smith BP, ed. Large animal internal medicine. 3rd ed. St Louis: Mosby, 2002; 389412.

  • 23. Bauer JE, Asquith RL, Kivipelto J. Serum biochemical indicators of liver function in neonatal foals. Am J Vet Res 1989; 50: 20372041.

  • 24. Patterson WH, Brown CM. Increase of serum γ-glutamyltransferase in neonatal Standardbred foals. Am J Vet Res 1986; 47: 24612463.

  • 25. Brewer B. Neonatal foal evaluation: sepsis and survival scoring in private practice, in Proceedings. 33rd Annu Conf Am Assoc Equine Pract 1987;812–822.

    • Search Google Scholar
    • Export Citation
  • 26. Moseley RH. Sepsis and cholestasis. Clin Liver Dis 2004; 8: 8394.

  • 27. Cotovio M, Monreal L, Armengou L, et al. Fibrin deposits and organ failure in newborn foals with severe septicemia. J Vet Intern Med 2008; 22: 14031410.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 28. Castagnetti C, Veronesi MC. Prognostic factors in the sick neonatal foal. Vet Res Commun 2008; 32: S87S91.

  • 29. Cohen ND. Causes of and farm management factors associated with disease and death in foals. J Am Vet Med Assoc 1994; 204: 16441651.

    • Search Google Scholar
    • Export Citation
  • 30. Henderson ISF, Franklin RP, Wilkins PA, et al. Association of hyperlactatemia with age, diagnosis, and survival in equine neonates. J Vet Emerg Crit Care 2008; 18: 496502.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 31. Wotman K, Wilkins PA, Palmer JE, et al. Association of blood lactate concentration and outcome in foals. J Vet Intern Med 2009; 23: 598605.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 32. Saulez MN, Gummow B, Slovis NM, et al. Admission clinicopathological data, length of stay, cost and mortality in an equine neonatal intensive care unit. J S Afr Vet Assoc 2007; 78: 153157.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 33. Rohrbach BW, Buchanan BR, Drake JM, et al. Use of a multi-variable model to estimate the probability of discharge in hospitalized foals that are 7 days of age or less. J Am Vet Med Assoc 2006; 228: 17481756.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 34. Brewer BD, Koterba AM. Development of a scoring system for the early diagnosis of equine neonatal sepsis. Equine Vet J 1988; 20: 1822.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 35. Corley KT, Furr MF. Evaluation of a score designed to predict sepsis in foals. J Vet Emerg Crit Care 2003; 13: 149155.

  • 36. Lorenzo-Figueras M, Pusterla N, Byrne BA, et al. In vitro evaluation of three bacterial culture systems for the recovery of Esch-erichia coli from equine blood. Am J Vet Res 2006; 67: 20252029.

    • Crossref
    • Search Google Scholar
    • Export Citation

Advertisement