Computed tomographic liver volumetry and hepatic attenuation in dogs with histologically confirmed cirrhosis compared with dogs without liver disease

Objective: To compare CT-derived liver volume and noncontrast hepatic attenuation between dogs with histologically confirmed cirrhosis and dogs without liver disease and describe qualitative CT findings in dogs with cirrhosis. Methods: This retrospective, multi-institutional, cross-sectional study included 8 dogs with histologically confirmed cirrhosis and 121 dogs without liver disease from a previously published CT liver volume reference interval cohort. Noncontrast CT images were used for manual hepatic volumetry and standardized hepatic and splenic attenuation measurements. Liver volume was normalized to body weight. Hepatic parenchymal attenuation (HPA) and the HPA-to-splenic parenchymal attenuation (SPA) ratio were compared between groups. Qualitative CT features were recorded descriptively. Results: Dogs with cirrhosis had lower mean liver volume-to-body weight ratios than dogs without liver disease (16.5 ± 4.5 cm3/kg vs 22.8 ± 5.3 cm3/kg; mean difference, -6.3 cm3/kg; 95% CI, -10.1 to -2.5; P = .002). Mean HPA was lower in dogs with cirrhosis (47.3 ± 9.6 HU vs 58.8 ± 6.2 HU; mean difference, -11.5 HU; 95% CI, -16.1 to -6.9; P = .001). The HPA:SPA ratio was also lower in dogs with cirrhosis (0.8 ± 0.1 vs 0.9 ± 0.1; P = .03). Nodular parenchymal change, acquired portosystemic shunts, and peritoneal effusion were present in 4, 2, and 3 cirrhotic dogs, respectively. Conclusions: Dogs with histologically confirmed cirrhosis had lower weight-normalized liver volume, HPA, and HPA:SPA ratio than dogs without liver disease. Clinical Relevance: CT-derived quantitative metrics may support recognition of canine cirrhosis, but reduced liver volume alone should not be considered diagnostic.