Rational use of systematic postoperative CT scans after neurosurgical craniotomy.

INTRODUCTION: The aim of this retrospective study was to evaluate the relevance of a systematic postoperative CT scan after neurosurgical craniotomy and to identify predictive factors of complications. MATERIAL AND METHODS: This retrospective analysis included all the patients at our institution who benefited from a cerebral postoperative CT scan within 24 hours post-craniotomy. Patient characteristics and neuroimaging abnormalities were recorded. Predictive factors were identified using a recursive partitioning analysis. RESULTS: A total of 633 patients were included. These 17.9% of patients suffered postoperative complications and 7.4% of them required a new surgery. The decision for reoperation was based on the neurological deterioration and the CT scan, but never on the CT scan alone. The mortality rate was 1.1%. The risk to be reoperated was correlated to the occurrence of a new postoperative neurological deficit (P < 0.001, HR = 4.60) and in situ hemorrhage (P < 0.001, HR = 4.19). The risk of postoperative hematoma was correlated to the supratentorial location versus infratentorial (P = 0.027, HR = 2.50). With clinical factors, such as location and etiology of the lesion, schedule type of surgery, and the age of patients, we proposed six classes with the risk to present with hemorrhage or midline shift on postoperative CT scans. CONCLUSIONS: The post-craniotomy CT scan did not impact patient management as an independent decisional tool. We identified several variables associated with the risk of clinical modification that can impact the decision to reoperate and allow establishment of a risk score. This score could be an interesting tool in order to reduce the systematic use of CT scans in the post-surgical period but has to be validated in a prospective study.