發(fā)布時間：2018-05-04 02:01

  本文選題：重癥化 + 慢性乙型肝炎　； 參考：《遵義醫(yī)學(xué)院》2017年碩士論文

【摘要】：目的:研究重癥化慢性乙型肝炎(chronic hepatitis B,CHB)患者進(jìn)展為肝衰竭的危險因素及比較四種預(yù)測模型的應(yīng)用價值。方法:回顧性分析2011年1月至2017年2月我遵義醫(yī)學(xué)院附屬醫(yī)院感染科住院部收治的474例重癥化CHB患者臨床資料;采用SPSS19.0軟件進(jìn)行統(tǒng)計分析,找出與重癥化CHB患者進(jìn)展為ACLF相關(guān)的危險因素;分別計算MELD評分、MELD-Na評分、Gao Fangyuan等建立的預(yù)測模型(DT模型)和張緒清等建立的預(yù)測模型(SJY模型)評分分值,比較其分值大小與重癥化CHB患者中ACLF發(fā)生率的關(guān)系,用工作特征(receiver operating characteristic,ROC)曲線及曲線下面積(area under the curve,AUC)評估四種評分模型的預(yù)測價值。結(jié)果:1.共篩選出符合標(biāo)準(zhǔn)的重癥化CHB患者474例,其中進(jìn)展為ACLF 36例(ACLF組),未進(jìn)展為ACLF患者438例(non-ACLF組),通過對比兩組患者的臨床指標(biāo),發(fā)現(xiàn)年齡≥40歲、HBV DNA≥107拷貝/ml、血鈉(sodium,Na~+)、血清總膽紅素(total bilirubin,TBIL)、凝血酶原活動度(prothrombin activity,PTA)和國際標(biāo)準(zhǔn)化比值(international normalized ratio,INR)共6個因素差異有統(tǒng)計學(xué)意義(P0.05),提示重癥化CHB患者進(jìn)展為ACLF與上述因素有關(guān);2.MELD評分、MELD-Na評分、DT模型評分和SJY模型評分的分值大小與ACLF發(fā)生率均呈正相關(guān)關(guān)系,其結(jié)果有顯著統(tǒng)計學(xué)差異(P0.05);3.SJY預(yù)測模型AUC最大為0.84,其次為DT預(yù)測模型(0.79),明顯高于MELD(0.70)及MELD-Na(0.69)。SJY模型具有較高的的敏感性(0.86)和特異性(0.75)。結(jié)論:重癥化CHB患者進(jìn)展為ACLF的危險因素包括年齡≥40歲,HBV DNA≥107拷貝/ml,Na~+、TBIL、PTA和INR;MELD評分、MELD-Na評分、DT預(yù)測模型和SJY預(yù)測模型評分的分值大小與ACLF發(fā)生率均呈正相關(guān)關(guān)系;DT模型和SJY模型對重癥化CHB患者進(jìn)展為ACLF都具有較好的預(yù)測作用,優(yōu)于MELD及MELD-Na評分模型。
[Abstract]:Objective: to study the risk factors of progression to liver failure in patients with severe chronic hepatitis B (CHB) and to compare the application value of four predictive models. Methods: the clinical data of 474 patients with severe CHB were analyzed retrospectively from January 2011 to February 2017 in the Department of infection, affiliated Hospital of Zunyi Medical College, and were statistically analyzed by SPSS19.0 software. To find out the risk factors related to the progression of severe CHB patients to ACLF, to calculate the MELD score and the prediction model established by MELD-Na, Gao Fangyuan et al.) and to calculate the score value of SJY model, which were established by Zhang Xuqing, et al. The relationship between the score and the incidence of ACLF in severe CHB patients was compared. The predictive value of the four scoring models was evaluated by using the receiver operating characteristic roc curve and the area under the curve. The result is 1: 1. A total of 474 severe CHB patients who met the criteria were selected, including 36 patients with ACLF and 438 patients with non-ACLF who did not progress to ACLF. The clinical indexes of the two groups were compared. It was found that there were significant differences in six factors, including age 鈮，

本文編號：1841066