To investigate the influencing factors and to construct a prediction model for refractory hypersplenism after liver transplantation for benign end-stage liver diseases.

Clinical data of 97 patients with benign end-stage liver diseases complicated with hypersplenism who underwent liver transplantation in Beijing Chaoyang Hospital, Capital Medical University from January 2013 to December 2023 were analyzed retrospectively. The informed consents of all patients were obtained and the local ethical committee approval was received. Among them, 70 patients were male and 27 female, aged (52±9) years on average. Before operation, 7 patients were classified with Child-Pugh A, 30 cases of Child-Pugh B and 60 cases of Child-Pugh C. 63 patients underwent modified piggyback liver transplantation and 34 cases of classic orthotopic liver transplantation. According to the recovery of hypersplenism at postoperative 6 months, all patients were divided into the remission group and refractory hypersplenism group. The risk factors of refractory hypersplenism after liver transplantation were analyzed by t-test and rank-sum test. The independent risk factors for refractory hypersplenism were identified by Logistic multivariate regression analysis. Based on the independent risk factors, the prediction model was constructed. The prediction efficiency of these independent risk factors was determined by the ROC curve.

Univariate analysis showed that preoperative Plt (Z=3.201), TB (Z=2.036), DB (Z=2.085), bipolar spleen diameter (t=3.291), thick spleen diameter (t=4.441), long spleen diameter (Z=3.309) and splenic artery diameter (t=3.634), splenic vein diameter (Z=3.494) and portal vein diameter (t=3.379) were associated with the incidence of refractory hypersplenism (all P<0.05).Lower preoperative Plt (OR=0.971, 95%CI: 0.945-0.997), longer preoperative thick spleen diameter (OR=1.080, 95%CI: 1.014-1.151) and preoperative splenic artery diameter (OR=1.472, 95%CI: 1.010-2.145) were the independent risk factors for patients with postoperative refractory hypersplenism (all P<0.05). Based on preoperative Plt, preoperative thick spleen diameter and preoperative splenic artery diameter, a prediction model was constructed. The prediction model yielded a high prediction efficiency for patients with postoperative refractory hypersplenism (ROC AUC=0.888, 95%CI: 0.823-0.953, P<0.001), with a sensitivity of 0.771 and a specificity of 0.855.

For benign end-stage liver disease patients with refractory hypersplenism after liver transplantation, the recovery of spleen function can be predicted by evaluating preoperative Plt, thick spleen diameter and splenic artery diameter.