Related Experiment Video
Updated: Aug 6, 2026
Comprehensive Endovascular and Open Surgical Management of Cerebral Arteriovenous Malformations
Published on: October 20, 2017
Post-operative events following elective craniotomy for tumor in children
C Stewart Nichols1, Emal Lesha2, Delaney Graham3
1College of Medicine, University of Tennessee Health Science Center, 910 Madison Avenue, Memphis, TN, 38163, USA. cnicho44@uthsc.edu.
Introduction:
Craniotomy for tumor resection in the pediatric population can result in many potential postoperative events (POEs). POEs, in turn, are a function of many variables, such as tumor type, size, location, surgeon experience and goal(s) of surgery. The current literature is limited to either specific POEs or complications associated with specific tumor pathologies.
Objectives:
The primary purpose of this study was to undertake a holistic approach and identify those variables predictive of all types of POEs and only surgical ones, following elective craniotomy for tumor in a diverse pediatric population.
Methods:
All elective craniotomies for tumor resection performed from 2010 to 2022 were included, excluding patients > 21 years of age. Demographic, clinical, and procedural covariates for each encounter (i.e., operation) were collected. POE was defined as a postoperative incident (medical or surgical, expected or unexpected) that necessitated further diagnostic testing, evaluation, or intervention within 90 days of surgery. Bivariate and multivariate analysis were performed; backward model selection was done to yield a final multivariate model retaining only variables with p < 0.05. Adjusted odds ratios (ORs) and 95% confidence intervals (CIs) were reported.
Results:
A total of 1,276 patients underwent 1,497 elective craniotomies for tumor resection with a median age of 9.45 years at index operation. In 535 (36%) encounters, there was at least one POE: 329 (61%) had surgical only, 109 (20%) had medical only, and 97 (18%) had both surgical and medical POEs. Multivariate analysis found several predictors of all types of POEs: length of stay (LOS) > 7 days, length of ICU stay (ICUS), attending surgeon, surgical time, and prior craniotomy. All of these, in addition to previously treated hydrocephalus and tumor type, were predictors of surgical-only POEs. All identified covariates were promoters of POE except for tumor type and redo craniotomy, which were protective.
Conclusion:
Analysis of our large pediatric tumor database identified several key statistical drivers of all POEs and surgical-only POEs following elective tumor resection, of which LOS > 7 days was the strongest.
