重症SARS患者预后因素的研究  

A study on prognostic factors of severe SARS: a clinical analysis of 165 cases

作　　者：邹正升[1] 杨永平[1] 陈菊梅[1] 辛绍杰[1] 张伟[1] 周先志[1] 毛远丽[1] 胡良平[2] 李保森[1] 

机构地区：[1]解放军第302医院,北京100039 [2]军事医学科学院情报所

出　　处：《解放军医学杂志》2004年第3期234-236,共3页Medical Journal of Chinese People's Liberation Army

基　　金：国家"十五"863计划重大项目 (编号 2 0 0 3AA2 0 81 0 6)资助课题

摘　　要：目的　分析影响SARS预后的单因素并建立影响SARS预后的多因素回归模型。方法　采用SPSS 11 0及SDAS软件回顾性分析 16 5例临床确诊的重症SARS患者的临床特征 ,通过对预后的单因素及Logistic多元回归分析 ,建立SARS预后回归模型。结果　①年龄与预后显著相关(r=0 5 0 6 ,P <0 0 0 1) ,≤5 0岁与 >5 0岁患者的病死率分别为 4 0 8%及 5 3 3% (P <0 0 1)。②有无基础病变与预后显著相关 (r=0 .4 5 7,P <0 0 0 1) ,它们的病死率分别为 5 4 5 %及 7 5 % (P <0 0 1)。③病情最重时淋巴细胞绝对数、PLT、尿常规改变、AST、TB、LDH、CK、CK MB、HBDH、UREA、肾功改变、呼吸频率、低氧血症程度、胸片受累肺叶数与预后相关 (r为 0 2 5 7～ 0 788,P <0 0 5 )。④SARS预后多因素回归生物数学模型与低氧血症程度与血小板数有关。多元回归模型为Py=1=es/(1+es) (S=2 4 90×低氧血症程度 - 0 0 5 0×血小板数)。若S >0 ,则Py=1大于 0 5 ,判断为死亡 ;若S<0 ,则Py=1小于 0 5 ,判断为存活。⑤该模型敏感性高 (91 6 7% ) ,特异性强 (98 33% ) ,准确性高(96 4 2 % )。结论　SARS预后模型是建立在SARS病理生理基础和生物数学之上的一种新的、简单、经济、敏感性高、特异性强和准确性高的指征 ,可有效判定SARS?Objective To analyze the single factor affecting prognosis of SARS, and to establish a model of regression analysis for multiple factors affecting the prognosis of the disease. Methods SPSS 11.0 and SDAS software packages were used to retrospectively analyze the clinical features of SARS in 165 clinically confirmed severe cases. Meanwhile, single-factor and logistic multivariate regression analyses were conducted for SARS prognosis to establish a regression model for SARS prognosis analysis. Results (1) In patients with SARS, the age was significantly correlated to prognosis (r=0.506, P<0.001). The mortality was 4.08% and 53.3% in patients under the age of 50 and those over 50, respectively. There was significant difference between these 2 groups of patients (P<0.01). (2) The existence of other diseases before SARS was significantly correlated to prognosis (r=0.457, P<0.001). The mortality was significantly higher in patients who had suffered from other diseases before SARS than in those without (54.5% vs 7.5%, P<0.01). (3) At the progressive stage of SARS, the absolute number of lymphocytes, PLT, changes in routine urine findings, AST, TB, LDH, CK, CK-MB, HBDH, UREA, changes in renal functional parameters, respiratory rate, degree of hypoxemia, and number of involved pulmonary lobes were markedly correlated to the prognosis (r=0.257～0.788, P<0.05). (4) The multiple-factor regression bio-mathematic model for SARS prognosis analysis was associated with degree of hypoxemia and number of platelets. The multivariate regression model was P y=1=е s/(1+е s) (S=2.490×degree of hypoxemia-0.050×number of platelets). In condition of S>0, then P y=1 was over 0.5, it denoted death of the patient. In condition of S<0, and P y=1 was lower than 0.5, it denoted survival of the patient. (5) The model was of high sensitivity (91.67%), specificity (98.33%) and accuracy (96.42%). Conclusion The model for analysis of SARS prognosis is a new, simple, economic, highly sensitive, specific and accurate index based on

关 键 词：严重急性呼吸综合征 预后 多因素分析 

分 类 号：R563.190.7[医药卫生—呼吸系统]