Efficacy and safety of local versus general anesthesia in stereotactic biopsies: a matched-pairs cohort study.

BACKGROUND: Frame-based stereotactic biopsies remain the gold standard for the diagnosis of intracerebral lesions. A major advantage is the ability to perform these procedures under local anesthesia (LA). However, there is no consensus on whether or when to use LA or general anesthesia (GA). It has been postulated that the use of LA may reduce the risk of complications. This study aims to objectify the efficacy and safety of stereotactic biopsies under LA versus GA by analyzing a prospective registry for stereotactic biopsies.

METHODS: A prospective registry for stereotactic procedures was analyzed for the activities performed between May 2009 and August 2011. Inclusion criteria were patients undergoing a stereotactic biopsy. Patients with functional stereotactic procedures, cyst or abscess evacuation, and children were excluded. Of the remaining 274 patients, 204 could be randomly matched considering sex, age, American Society of Anesthesiologists (ASA) classification, and type of hospitalization. The primary endpoint was the diagnostic yield. Secondary endpoints were safety and procedural time intervals.

RESULTS: Ninety-six percent of diagnosis was achieved in the LA group and 94% within the GA group. There was no significant difference in the overall complication rate. However, the rate of pulmonary complications was higher in the GA group (P=0.059). The rate of bleeding was significantly higher in the LA group (P=0.003) intraoperatively. However, no significant difference was found clinically or on routine postoperative computed tomography scan reports. There was a significant reduction in the time spent in the operating room within the LA group (P<0.001).

CONCLUSIONS: The type of anesthesia may not interfere with the diagnostic yield in stereotactic biopsies. Patients operated under LA are less likely to suffer pulmonary complications. Furthermore, LA could spare hospital resources in this setting.