| Presenter: Nicholas Berlin, B.S., University of Colorado School of Medicine; Yale University School of Public Health
Authors: Nicholas Berlin (1), Brooke French (2), Frederic Deleyiannis (3), Austin Badeau (4)
(1) University of Colorado School of Medicine, Yale University School of Public Health, New Haven, CT, (2) Anschutz Medical Campus, Denver, CO, (3) Children's Hospital of Colorado, Aurora, CO, (4) University of Colorado School of Medicine, Denver, CO
Nicholas Berlin, B.S.
University of Colorado School of Medicine, Yale University School of Public Health
The efficacy of autologous bone-assisted cranioplasty for children who have undergone decompressive craniectomy has been disputed in recent studies. Development of symptomatic resorption has been reported at 50%, with most patients undergoing a revision cranioplasty. Risk factors for resorption in children are not well established. Previous studies identified age <=2.5 years, permanent ventriculoperitoneal (VP) shunt, comminuted skull fracture, size of defect and interval >6 weeks to be associated with resorption. The purpose of this study is to evaluate potential risk factors for resorption in children for future studies of efficacy and preoperative risk stratification.
All pediatric patients who underwent decompressive craniectomy and autologous bone-assisted cranioplasty from 2006 to 2013 were identified at our institution. A retrospective cohort study was conducted to evaluate risk factors for bone flap resorption. The primary risk factors evaluated were age <=2.5 years, permanent VP shunt, comminuted fracture (>=3 pieces), removal of >=2 cranial bones, interval >6 weeks, size of defect, cranioplasty surgeon, and cranioplasty surgical site infection. Multivariate analyses were performed to identify independent predictors of outcomes.
Fifty-two patients met eligibility criteria with 20 patients (38.5%) developing resorption. Of the entire cohort, 57.7% were male, 21.2% were <=2.5 years old, 63.5% suffered traumatic brain injury, 13.5% required a permanent VP shunt, 42.3% had a comminuted skull fracture, 64% had removal of >=2 cranial bones, and 73.1% underwent craniectomy to cranioplasty interval >6 weeks. The mean time to resorption was 7 months and mean follow-up time was 19.5 months. Among patients with resorption, 11 (55%) underwent revision with 4 (36.3%) undergoing an additional revision procedure. Multivariate analyses demonstrated age <=2.5 years (adjusted OR 17.0) and comminuted skull fracture (adjusted OR 12.1) as independent predictors of resorption.
Although the rates of bone resorption and accompanying revision in this cohort are less than previously reported, these complications remain important considerations from a standpoint of efficacy and safety. Age <=2.5 years and comminuted skull fracture were independent predictors of resorption, thus warranting further investigation.