Does subtotal resection ameliorate hypothalamic morbidity in pediatric craniopharyngioma? A 30-year retrospective cohort study.

Abstract

The optimal extent of resection of craniopharyngiomas to minimize the long-term risks of hypothalamic and endocrine dysfunction (obesity and panhypopituitarism) in children remains uncertain. The purpose of this study was to report long-term outcomes of pediatric patients with craniopharyngioma undergoing surgical treatment and to study rates of endocrinological and hypothalamic dysfunction in association with extent of resection. This retrospective study was performed in a cohort of children who underwent resection for craniopharyngioma at Children's of Alabama between 1990 and 2020. The primary outcome was hypothalamic dysfunction defined as a 0.5 increase in body mass index (BMI) Z-score and as a BMI > 2 SDs with or without psychiatric disturbances. Univariable analysis was performed using ANOVA, Wilcoxon rank-sum test, Pearson's chi-square test, and Fisher's exact test as appropriate. Missing data on the primary outcome were handled via multiple imputations. Relative risks were estimated using a multivariable generalized linear model with a priori variables selected using a modified Poisson regression approach with robust error variance to estimate risk ratios. The cohort includes 39 patients (24 girls and 15 boys; age range 1 month-16 years) who underwent resection of craniopharyngioma at the authors' center between 1990 and 2020. The preoperative goal of treatment was cyst decompression (CD) in 5, subtotal resection (STR) in 13, and gross-total resection (GTR) in 21 patients. The median long-term follow-up after surgery was 8.11 years (average 8.21, range 0.4-24.33 years). Univariate analysis demonstrated a statistically significant increase in hypothalamic dysfunction in patients undergoing GTR when compared to those undergoing STR or CD at 1 month postoperatively (p = 0.006) and 6-11 months postoperatively (p = 0.010), but with this difference not persisting beyond 1 year. Multivariable analysis showed patients older than 10 years at time of surgery to be the most affected and at highest risk of developing significant hypothalamic dysfunction. There was no significant difference in pituitary or neurological function between the STR/CD and GTR groups at 12-24 months or at most recent follow-up. There was no significant difference in BMI Z-scores between the STR/CD and GTR groups at 6-12 months or at most recent follow-up. Both STR and GTR of craniopharyngioma were associated with significant endocrinological sequelae after 1 year. These potential complications should be discussed with patients and their families, and postoperative protocols should include early nutritional and endocrinological interventions with endocrinologist consultation.