| Objective To obtain data of postoperative infection rate, risk factors and information on pathogenic bacterias after brain surgeries (including craniotomy and CSF shunt surgery), and examine the predictive potential of the NNIS index for the postoperative infection. Finally provide guidance for mitigating postoperative complications as well as postoperative infection prevention and treatment.Methods All brain surgeries in department of Neurosurgery, the First Affiliated Hospital, Zhejiang University, dated from June2010to June2012was retrospectively examined and data were screened by specific standards. The relative information (29factors) of patients who were then included are collected and univariate and multivariate analysis was made to investigate the risk factors, and the type of pathogenic bacteria with their respective proportions were calculated.Results1783cases were examined and1476cases were included in the study with307cases excluded. The total postoperative intracranial infection rate was10.2%, with the rate for craniotomies was9.7%and for CSF shunt was14.6%. Univariate analysis revealed that the risk factors include a GCS score lower than9, surgery for aneurism, surgery for Arteriovenous Malformation (AVM), surgery for brain tumors, external ventricular shunt, external CSF drainage, a craniotomy last longer than4hours or a CSF shunt surgery last longer than2hours, subsequent surgeries, CSF leakage and a NNIS index score≥1. Multivariate logistic analysis revealed that external CSF drainage, a craniotomy last longer than4hours or a CSF shunt surgery last longer than2hours, subsequent surgeries, postoperative CSF leakage are the risk factors. Gram-positive infections account for65%, and the most common pathogenic bacteria of gram-positive infections is Staphylococcus, which constitute60%of all positive CSF culture results, while the most common pathogenic bacteria of gram-negative infections is Acinetobacter baumannii, which accounts for15%of all positive CSF culture results.Conclusion The risk factors which are concluded in this study are almost the same to what was reported both in and abroad in the recent years. According to the result, several methods could be used to effectively reduce the postoperative ICI rate, which include:avoiding the use of subdural CSF drainage as far as possible and shorten the time span of external CSF shunt, reducing the operation time of both craniotomy and CSF shunt, avoiding subsequent surgeries and suture dura mater carefully in the operation to avoid postoperative CSF leakage and if CSF leakage was diagnosed immediate measures must be taken. It is revealed that the NNIS index does not provide a convincing predictive effective for the postoperative infection, which might resulted from the discrepancies of the operation time boundary between this study and the NNIS index, and a new index for predicting postoperative ICI risk is needed in the further studies. Finally, the pathogenic bacteria that this study concluded is different from other reports, but still meet the general tendency that gram-positive infection are becoming more predominant cause for the postoperative infections. |