Logistic regression analysis of risk factors predicting postoperative intracerebral re-hemorrhage of patients with traumatic intracerebral hemorrhage

Abstract

Objective To investigate the risk factors of postoperative intracerebral re-hemorrhage of patients with traumatic intracerebral hemorrhage by analyzing the clinic factors through case control study.Methods Clinical data of 237 patients with traumatic intracerebral hemorrhage,admitted to our hospital from January 2007 to September,were collected; these patients were divided into postoperative intracerebral re-hemorrhage group (n=84) and non-postoperative intracerebral re-hemorrhage group (n=153) according to the re-hemorrhage; the 11 independent risk factors were proved by univariate and multivariate binary logistic regression analysis,of which predicting value was evaluated by receiver operating characteristic (ROC) curve.Results Univariate binary logistic regression analysis indicated that 5 factors as the time from injure to operation,volume of intracranial hematoma (VOL),thrombin time (TT),prealbumin (PAB) and blood sugar were significantly different between patients with postoperative intracerebral re-hemorrhage and patients without postoperative intracerebral re-hemorrhage (P＜0.05).Multivariate binary logistic regression analysis indicated that PAB (P=0.016,OR=0.994,95％CI:0.988-0.999),TT (P=0.044,OR=1.114,95％CI:1.003-1.238) and VOL (P=0.003,OR=1.020,95％CI:1.007-1.033) were the independent risk factors of postoperative intracerebral re-hemorrhage of patients with traumatic intracerebral hemorrhage,and ROC curve indicated that all these three factors had predictive value.Conclusion Patients with low PAB,prolonged TT or large VOL have higher risk of postoperative intracerebral re-hemorrhage; more attention should be paid on these patients to prevent intracerebral hemorrhage recurrence. Key words: Traumatic intracerebral hemorrhage; Postoperative intracerebral re-hemorrhage; Risk factor