The Cox Regression Analysis Of Predictive Factors Of Hepatorenal Syndrome

Objectives Hepatorenal syndrome is a severe complication of end-stage liver diseases. Its mortality rate is very high. The prognosis of hepatorenal syndrome is dismal. But there was still no effective treatment to it. So it is most important to prevent the occurrence and ongoing of hepatorenal syndrome in clinical working. It was supposed that if the predictors of hepatorenal syndrome were clear, it might be prevented properly. So the aim of this study was to investigate the predictive factors and the best prognostic parameter of hepatorenal syndrome.Methods A retrospective cohort study was carried out. 94 consecutive patients with cirrhosis or other acute or chronic advanced liver diseases who underwent hepatorenal syndrome hospitalized in the two affiliated hospitals from Jan 1983 to Dec 2003 were enrolled for multivariate Cox regression analysis. 35 possible predictive factors including clinical and physiochemical features contributing to the survival of these patients were analyzed. All the data were collected with the date of diagnosis being the start-point. Bivariate correlate analysis, chi-square test and independent-t test were carried out as well.Results On multivariate Cox regression, only Child-Pugh score at the time of diagnosis was the only independent risk factor of hepatorenal syndrome, relative risk was 1.823, 95%CI(1.646, 1.820), The hazard function was h(t)=[h0(t)]e1.823x. The other analyses verified the result respectively.Conclusion Child-Pugh score at the time of diagnosis is the best parameter for predicting the clinical outcome of hepatorenal syndrome. To monitor the patients' Child-Pugh score dynamically may provide guidance to their treatments and to judge their prognoses. To lower their Child-Pugh score may improve the prognoses of those patients with hepatorenal syndrome.