Predictive factors for liver abscess liquefaction degree based on clinical, laboratory, and computed tomography data.

Background: Effective management of liver abscess depends on timely drainage, which is influenced by the liquefaction degree. Identifying predictive factors is crucial for guiding clinical decisions.

Aim: To investigate the predictive factors of liver abscess liquefaction and develop a predictive model to guide optimal timing of percutaneous drainage.

Methods: This retrospective study included 110 patients with pyogenic liver abscesses who underwent percutaneous catheter drainage. Patients were divided into a poor liquefaction group (n = 28) and a well liquefaction group (n = 82) based on the ratio of postoperative 24-hour drainage volume to abscess volume, using a cutoff value of 0.3. Clinical characteristics, laboratory indicators, and computed tomography imaging features were compared. A predictive model was constructed using logistic regression and evaluated using receiver operating characteristic curves and five-fold cross-validation.

Results: Independent predictive factors for good liquefaction included the absence of diabetes [odds ratio (OR) = 0.339, P = 0.044], absence of pneumonia (OR = 0.218, P = 0.013), left-lobe abscess location (OR = 4.293, P = 0.041), cystic features (OR = 5.104, P = 0.025), and elevated preoperative serum alanine aminotransferase (ALT) levels (OR = 1.013, P = 0.041). The logistic regression model based on these factors demonstrated an area under the curve of 0.814, with a sensitivity of 90.24% and specificity of 67.86%. Five-fold cross-validation yielded an average accuracy of 83.61% and a kappa coefficient of 0.5209.

Conclusion: Pneumonia, diabetes, abscess location, abscess composition, and preoperative serum ALT levels are significant predictors of liver abscess liquefaction. The model can guide clinical decision-making.