A prognostic model for pyrrolizidine alkaloids-induced hepatic sinusoidal obstruction syndrome based on sarcopenia

Sarcopenia has been shown to be associated with the prognosis of patients with liver cirrhosis, hepatocellular carcinoma, liver failure. However, the impact of sarcopenia on survival has not been investigated in patients with pyrrolizidine alkaloids (PAs)-induced hepatic sinusoidal obstruction syndrome (HSOS). In this study, we aimed to develop and internally assess a prognostic model for mortality risk stratification in patients with PAs-HSOS. A total of 156 PAs-induced HSOS patients were enrolled from January 2010 to October 2022. The relevant data of patients were collected. Univariate and multivariate Cox proportional hazards analysis were used to determine prognostic factors. Then, a prognostic model based on sarcopenia was developed and assessed. At a median follow-up interval of 45.9 months, 73.7% (115/156) patients survived. Sarcopenia was observed in 42.3% (66/156) of patients with PA-induced HSOS, 57.6% (38/66) of male patients and 42.4% (28/66) of female patients. Further study demonstrated aspartate aminotransferase (AST), urea nitrogen, thrombin time (TT) and sarcopenia (AUTS) were independent prognostic factors. The AUTS model was developed and internally validated using bootstrap resampling and showed moderate prognostic discrimination in the derivation cohort. The apparent Harrell C-index was 0.746 (bootstrap 95% CI, 0.660–0.823), and the optimism-corrected C-index after 1,000 bootstrap resamples was 0.727. A prognostic model (AST, urea nitrogen, TT and sarcopenia) based on sarcopenia showed moderate prognostic discrimination for the prognosis of patients with PAs-induced HSOS. The AUTS model showed moderate prognostic discrimination in the derivation cohort; however, its limited specificity and high false-positive rate restrict its current clinical utility. External validation is required before clinical implementation. Here, we identified sarcopenia was an independent prognostic factor of patients with PAs-HSOS; and then developed a prognostic model named AUTS (AST, urea nitrogen, TT and sarcopenia) based on sarcopenia. Further study showed AUTS showed moderate prognostic discrimination for patients with PA-induced HSOS.