Retrospective analysis of historical, clinical, ultrasonographic, serum biochemical and haematological data in prognostic evaluation of equine liver disease

Reasons for performing study: Results of noninvasive tests of liver disease do not always correlate with the degree of hepatic disease nor outcome of the case. Objective: To investigate the prognostic value of data collected using noninvasive tests during the investigation of cases of suspected liver disease in mature horses. Hypothesis: Much of the data gathered during the investigation of suspected hepatopathy cases offers little prognostic guidance and interpretation of such data can be misleading. Methods: The results from a range of common and noninvasive diagnostic techniques applied in 116 mature horses with suspected liver disease, were assessed for their ability to predict survival within a 6 month period. Results: A significantly poorer prognosis was found in association with clinical signs suggestive of liver disease, presence of hepatic encephalopathy, ultrasonographic abnormalities, increased serum globulins, increased total bile acids (TBA), increased alkaline phosphatase (AP), increased gamma-glutamyl transferase (gammaGT), erythrocytosis, leucocytosis, low serum albumin and low serum urea. Additional, significant novel findings of interest included an association between increased plasma fibrinogen and low serum creatinine concentrations with nonsurvival in cases of liver disease, an association between raised serum concentrations of AP and gammaGT with billiary hyperplasia and also an association between hepatic fibrosis, haemosiderosis and biliary hyperplasia with ultrasonographically detected hepatic abnormalities. Conclusions: The most useful noninvasive prognostic test in cases of suspected liver disease in mature horses is the severity of clinical signs. Other data may be of some limited prognostic value. Potential relevance: Application of the findings in this study may not be directly applicable to other case populations. However, the findings should at least be considered when prognosis is based on similar criteria.