重症急性胆管炎的死亡风险预测  

Prediction of survival or death of patients with acuteobstructiVe suppurative cholangitis

作　　者：屠长林 方平[1] 

机构地区：[1]上海医科大学附属金山医院外科教研室,上海医科大学附属金山医院计算机室

出　　处：《中国危重病急救医学》1994年第2期96-99,共4页Chinese Critical Care Medicine

摘　　要：根据重症急性胆管炎（急性梗阻性化脓性胆管炎ＡＯＳＣ）患者入院后治疗前的基本生理指标（临床变量），应用逐步判别分析，检出其中最能影响预后的变量组合，即：心率、呼吸次数、意识状况和白细胞计数，并定量地赋予各变量以不同的系数，建立判别函数，从而对ＡＯＳＣ患者进行死亡风险预测。以该院１９７６～１９８５年８１例经确诊为ＡＯＳＣ的病史资料作为训导病史，以之衍生出判别函数后，其回代检验结果为：敏感度６５．０％，特异度８５．２％，正确率８０．３％。再将该院１９８６～１９９１年连续收治的７７例ＡＯＳＣ患者的资料作为考核病史，前瞻性地验证上述判别函数的结果是：敏感度８５．７％，特异度９０．０％，正确率为８９．６％。研究结果说明：（１）ＡＯＳＣ的死亡风险可以预测。（２）用以建立判别函数的临床变量可以是患者入院后治疗前的最基本生理指标。（３）通过预测，可以在临床上较早地发现ＡＯＳＣ中的高危患者，以便尽早组织抢救。Severe physiological abnormalities are seeninevitably in every patient with acute critical illnesssuch as acute obstructive suppurative cholangitis,and there is evidence that the risk of death of thesepatients is closely correlated to the extent of physio-logical deteriorations in the body,and thus,predic-table,in order to investigate the relationship,with theaid of IMB/PC computer and using program of stepwise discriminant analysis of SPSS/PC+,theauthors conducted a trial to develop and validate a model for the prediction of survival or death of 158cases of AOSC.After careful review of clinical re-cords of 81 patients with AOSC admitted from 1976through l985, the authors found that differnces ex-isted in certain pre-treatment clinical variables be-tween survlvors and non-survlvors,and the mostSigificant clinical variables that influenced the riskof death were selected by stepwise discriminant ana-lysis.the method also allowed estimation of discriminant functions and thus made possible topredict every patient the risk of death with anaccruacy of 80.3%,A prospective validation study on77 concecutive patients admitted January 1986through 1991 revealed the prediction was correct in89.6% of the patients ,a result that was very encour-aging.Thus the authors are confident that the predic-tion is feasible and the information may becomeclincally useful for individual patients.

关 键 词：胆管炎 急性 梗阻性 死亡 预测 

分 类 号：R657.45[医药卫生—外科学]