The assessment of liver fibrosis is of major importance for treatment and prognosis of canine chronic liver diseases. A panel of blood scores are currently used in human medicine and combine several biochemical parameters in proprietary algorithms. These tests help the clinician to stage and monitor liver diseases, thereby decreasing the need for liver biopsy.

We already built a blood score using a training set of 58 dogs. This score uses alanine aminotransferase, alkaline phosphatase, total bilirubin, potassium, and gamma glutamyl transferase. This blood score had a sensitivity and specificity of 81% and 68%, respectively (AUROC [95% CI] = 0,80 [0,67–0,92]).

The aim of the present study was to validate the blood score performance in a new population.

Following ethical committee approval and informed owner consent, client-owned dogs ≥2 years old that underwent liver biopsy to investigate abnormal liver enzyme activity were included. Exclusion criteria were: focal liver lesion and/or extrahepatic cholestasis on abdominal ultrasound, previous medications or extrahepatic comorbid diseases that could have influenced liver enzymes activity, or diagnosis of neoplasia on histologic examination.

Fasting blood samples were collected on the day of liver biopsy. Liver fibrosis was evaluated according to the criteria of the World Small Animal Veterinary Association guidelines on canine liver pathology. Dogs were classified as follows: no or only portal fibrosis (group 1) and bridging fibrosis or cirrhosis (group 2).

Thirty-eight dogs were recruited (53% female, 47% male): 20 in group 1 and 18 in group 2.

In this validation population, the blood score discriminates dogs with significant fibrosis, with an AUROC curve of 0,83 [0,69–0,96] (sensitivity 83%, specificity 70%) versus 0,68 [0,50–0,86] for ALT and 0,59 [0,41–0,78] for ALKP.

The validation protocol demonstrated similar performances between the training and the validation population. The blood score developed in this study was designed to help clinicians to screen for dogs with liver fibrosis in order to guide further investigations. This test provides accurate and reliable results in a fast, simple and cost-effective manner. An external validation protocol is needed for further assessment of the robustness of the test.

Disclosures

Disclosures to report
The authors declare a potential conflict of interest with the company Echosens that developed the blood score described in this study. Echosens supports part of the residency funding of the presenting author.