Computed tomographic features for differentiating benign from malignant liver lesions in dogs

Area of Science:

• Veterinary Radiology
• Comparative Oncology
• Diagnostic Imaging

Background:

• Distinguishing benign from malignant liver lesions in dogs using computed tomography (CT) is challenging due to complex and subjective criteria.
• Existing CT characteristics require significant expertise, limiting their clinical applicability.

Purpose of the Study:

• To identify practical CT variables for classifying canine liver lesions as benign or malignant.
• To assess the clinical relevance of these CT variables for histopathological diagnosis.

Main Methods:

• Prospective study including dogs with liver nodules/masses undergoing CT and histopathology.
• Evaluation of 23 qualitative and quantitative CT variables using univariate and multivariate analyses.
• Analysis of signalments, CT findings, and histopathological diagnoses for 70 lesions in 57 dogs.

Main Results:

• Fifty-two malignant and 18 benign liver lesions were identified.
• Two significant CT variables for differentiation: heterogeneous postcontrast enhancement in the delayed phase (OR: 14.7) and maximal transverse diameter >4.5 cm (OR: 33.3).
• These variables achieved an area under the curve of 0.8910, indicating 88.6% accuracy in distinguishing lesion types.

Conclusions:

• Triple-phase CT features, specifically lesion size and delayed enhancement patterns, can accurately predict liver malignancy in dogs.
• These simple CT features aid in distinguishing focal liver lesion types and support clinical decision-making.