基于血小板参数的酒精性肝硬化预测模型构建与评估

目的 构建血小板参数用于酒精性肝炎(alcoholic hepatitis,AH)转变为酒精性肝硬化(alcoholic cirrhosis,AC)的预测模型,并评估其预测价值.方法 回顾性分析2018年1月至2020年12月台州医院肝胆外科收治的首诊AH、AC患者共计60例,其中AH组与AC组各为30例,并收集同期健...

Establishment and assessment of a prediction model for alcoholic cirrhosis based on platelet parameters

Objective To establish a predictive model for conversion from alcoholic hepatitis (AH) to alcoholic cirrhosis (AC) based on platelet parameters. Methods This study retrospectively analyzed 30 cases with AH and 30 cases with AC from Jan. 2018 to Dec. 2020 in Taizhou Hospital of Zhejiang Province. A binary Logistic regression analysis was performed to predict the disease conversion from AH to AC. Cutoff points were identified following Youden’s index of receiver operator characteristic (ROC) curve. Twenty-one patients with AH were collected for independent validation. Results The MPV level, platelet count (PLT) and macroplatelet count (P-LCC) were independent risk factors for disease conversion. The prediction model was P=EX/(1+EX), X=-3.580+1.731×PLT+2.402×MPV+2.571×P-LCC, the value for PLT ≥125×109/L is 0, <125×109/L is 1; the value for MPV ≤10.5 fL is 0, >10.5 fL is 1; the value for P-LCC≥37×109/L is 0, < 37×109/L is 1. The area under the ROC curve (AUC) was 0.957 (95%CI 0.912-0.989), the sensitivity was 87.1% and the specificity was 96.0%. The verification Results showed that the sensitivity, specificity and accuracy were 90.9%, 60.0% and 76.2%, indicating a good prediction effect. Conclusion MPV, PLT and P-LCC have predictive value in the conversion from AH to AC, which might provide reference for physicians to take advanced interfere treatment.