Ultrasonographic assessment of liver dimensions in dairy cows with different hepatic triacylglycerol content

The objective of this study was to investigate the relationship between hepatic triacylglycerol content (TAG) and hepatic size and shape assessed by transcutaneous ultrasonography in dairy cows. In a second part, the usefulness of ultrasonographic measurements pertaining to the size and shape of the liver for the diagnosis of fatty liver was evaluated. Transcutaneous ultrasonographic examination of the liver was carried out in 133 German Holstein cows (BW: 578 ± 6 kg; age: 5 ± 0.2 yr; withers height: 145 ± 0.3 cm). The size of the liver, angle of the ventral liver margin, and thickness of the liver over the portal vein and caudal vena cava were measured in the 10th and 11th intercostal spaces via ultrasonography. As the gold standard, TAG was determined enzymatically in liver biopsies. Hepatic TAG ranged from 5 to 292 mg/g of liver fresh weight (FW). Cows were allocated to 4 groups according to their hepatic TAG content (TAG ≥150 mg/g of FW, n = 32; ≥100 to 150 mg/g of FW, n = 34; ≥50 to 100 mg/g of FW, n = 38; <50 mg/g of FW, n = 29). Weak to moderate correlations (r = 0.26 to 0.49) between the ultrasonographic liver measurements and hepatic TAG indicated an increased size and angle of the liver with increasing hepatic fat accumulation. All mean liver ultrasonographic measurements revealed differences between TAG classes (P < 0.001), with the greatest values in cows with hepatic TAG concentrations ≥150 mg/g of FW. Stepwise multiple linear regression for hepatic TAG prediction (r² = 0.34, P < 0.001) and stepwise discriminant analysis used only ultrasonographic measurements obtained via the 10th intercostal space and age. Regression analysis revealed that TAG prediction widely overestimated measured TAG in the low TAG range and underestimated TAG in the high TAG range. The sensitivity and specificity based on discriminant analysis for differentiation of the TAG groups were 0.61 and 0.61 (<50 vs. ≥50 mg/g of FW), 0.67 and 0.71 (<50 vs. ≥100 mg/g of FW), 0.83 and 0.82 (<50 vs. ≥150 mg/g of FW), 0.70 and 0.77 (<100 vs. ≥100 mg/g of FW), and 0.83 and 0.85 (<100 vs. ≥150 mg/g of FW), respectively. Results revealed that with a hepatic TAG content of approximately 100 mg/g of FW, the liver increased in size and its margins became rounded. Pronounced increases in the size and thickness of the liver and in the angle of the liver occurred only in cows with very severe fatty liver (TAG ≥150 mg/g of FW). However, the determination of ultrasonographic measurements of liver size and shape appeared to be of limited diagnostic value in the recognition of fatty liver because of increased interindividual variance in dairy cows.