A combination of liver stiffness and international normalized ratio is an ideal prognostic predictor of portosystemic shunt occlusion in patients with portal hypertension

Background We previously reported liver stiffness (LS) as a prognostic predictor of portosystemic shunt (PSS) occlusion. This study aims to reinvestigate the predictive factors of the model for end-stage liver disease-sodium (MELD-Na) score amelioration following balloon-occluded retrograde transvenous obliteration (BRTO) and to evaluate the postoperative prognoses of patients with portal hypertension by using newly identified factors. Methods Seventy-five patients who underwent BRTO between 2008 and 2021 were retrospectively enrolled. The MELD-Na scores were calculated preoperatively and one month postoperatively. We monitored long-term outcomes and analyzed postoperative survival. Results At one month postoperatively, the MELD-Na score decreased in 46 (61.3%) patients. Univariate analyses revealed a significant association of the score amelioration with nine factors, including lower LS levels and a higher international normalized ratio (INR). A multivariate logistic regression analysis with receiver operating characteristic curve analyses identified preoperative LS levels and INR as significant independent predictors of the postoperative MELD-Na score amelioration, with optimal cutoffs of 28.1 kPa and 1.06, respectively. The combination of LS < 28.1 kPa and INR ≥ 1.06 showed a sensitivity and specificity of 84.8% and 75.9% for the prediction of the score amelioration, respectively. For the propensity score model, we matched 24 patients with similar age, sex, MELD-Na score, and concomitant hepatocellular carcinoma. Kaplan–Meier analysis determined significantly higher cumulative survival rates in patients with LS < 28.1 kPa and INR ≥ 1.06 than in other populations. Conclusions A combination of LS and INR can predict the MELD-Na score amelioration and prognosis improvement following PSS occlusion.