论文部分内容阅读
目的分析红细胞计数、白细胞计数、血小板计数、肝功能分级与药物治疗后肝硬化食管胃静脉曲张破裂再出血的关系,建立回归方程,从而预测再出血。方法对155例肝硬化食管胃静脉曲张破裂出血患者的临床资料进行回顾性分析,并随访3年,先以单因素非条件Logistic回归分析筛选出有统计学意义的因素,然后再用多因素非条件Logistic回归分析预测再出血的因素,建立回归方程。结果单因素分析表明血小板计数、肝功能分级是再出血的相关因素,多因素分析表明肝功能分级是再出血的危险因素(P=0.008,OR值=2.645),血小板计数是保护因素(P=0.02,OR值=0.987),基于此建立的Logistic回归模型进行Hosmer和Lmeshow检验,方程的拟舍优度较好(x~2=9.861,P=0.275)。结论对于药物治疗后的肝硬化食管胃静脉曲张出血患者,血小板计数以及肝功能分级与再出血相关,以此建立的回归方程可以用来预测再出血,且与实际情况的一致性良好,对预测再出血有一定临床价值。
Objective To analyze the relationship between erythrocyte count, white blood cell count, platelet count, liver function grading and esophagogastric variceal rehemorrhage after hepatic cirrhosis and to establish regression equation to predict rebleeding. Methods The clinical data of 155 patients with cirrhosis esophagogastric rupture hemorrhage were retrospectively analyzed and followed up for 3 years. First, by univariate non-conditional logistic regression analysis, statistically significant factors were screened out and then analyzed by multivariate analysis Conditional Logistic regression analysis of predictors of rebleeding factors, the establishment of regression equations. Results Univariate analysis showed that platelet count and liver function classification were the related factors of rebleeding. Multivariate analysis showed that liver function classification was a risk factor for rebleeding (P = 0.008, OR = 2.645) and platelet count was a protective factor (P = 0.02, OR = 0.987). According to the Logistic regression model established by this method, the Hosmer and Lmeshow tests were performed. The goodness of fit of the equation is good (x ~ 2 = 9.861, P = 0.275). CONCLUSIONS: The regression equation established by this method can be used to predict the rebleeding of patients with cirrhosis of esophageal and gastric variceal bleeding, platelet count and classification of liver function after rebleeding, and the consistency with the actual situation is good. Rebleeding has some clinical value.