Between-hospital variation in biopsy indication for patients with newly diagnosed glioblastoma in the Dutch Quality Registry for Neurosurgery

Abstract

Purpose: This registry-based study aims to examine the variation in biopsy and resection indications for glioblastoma patients across Dutch hospitals and to identify patient- and hospital-related factors associated with the surgical treatment. Methods: Data from all 7443 adults with first-time glioblastoma surgery at 12 hospitals were obtained from the prospective population-based Quality Registry Neurosurgery in the Netherlands between 2011 and 2021. Patients were stratified by either biopsy or resection. We analyzed variation in American Association of Anesthesiologist (ASA) classification, Karnofsky Performance Score (KPS), gender and age distribution between the different centers. Between-hospital variation in biopsy percentage was analyzed using a funnel plot. Logistic regression was used to identify associated patient- and hospital-related factors. Results: In total, 32% of the newly diagnosed glioblastoma patients underwent a biopsy, with wide variations between the different centers (23–56%). Patients-related variables such as higher age or ASA classification and lower KPS were significantly associated with the indication for biopsy. After correction for these factors, between-hospital variation persisted, with two institutes performing more biopsies than expected and one less than expected. Median overall survival was 12.5 months (95% CI 12.2–12.9) in the resection group and 5.6 months (95% CI 5.1–6) in the biopsy group, with wide variations between the different centers. Conclusion: A substantial between-hospital variation in biopsy percentages was found. Patient factors (age, ASA classification and KPS) but also hospital factors (such as academic setting) impact surgical decisions. Variation persisted also after correction for potential confounders, indicating that other factors play a role in decision-making.