Background: Though the volume-outcome relationship has been well-established in adults, low mortality rates and small sample sizes have precluded definitive demonstration in children. This study compares treatment-specific factors for children with nephroblastoma at high (HVC) versus low volume centers (LVC).
Methods: We performed a retrospective cohort study comparing patients ≤18years with unilateral nephroblastoma treated at HVCs and LVCs using the National Cancer Data Base (1998-2012). Definitions of HVCs included performing above the median, the upper two quartiles, and the highest decile of nephroblastoma resections. Outcomes included nodal sampling, margin status, time to chemotherapy and radiation, and survival. Statistical analyses included χ2, t-tests, generalized linear, and Cox regression models (p<0.05).
Results: Of 2911 patients from 210 centers, 1443 (49.6%) were treated at HVCs. There was no difference in frequency of preoperative biopsy or days to radiation (p>0.05). High volume centers were more likely to perform nodal sampling (RR 1.04, 95%CI 1.01-1.08) and had fewer days to chemotherapy (RR 0.80, 95%CI 0.69-0.93). Five-year survival was similar (HVC: 0.93, 95%CI 0.92-0.94; LVC: 0.93, 95%CI 0.91-0.94).
Conclusions: HVCs were more likely to perform nodal sampling and had fewer days to chemotherapy. There was no difference in days to radiation or survival between centers.
Level of evidence: Level II (retrospective prognosis study).
Keywords: National Cancer Data Base; Nephroblastoma; Volume-outcome; Wilms' tumor.
Copyright © 2017 Elsevier Inc. All rights reserved.