Long time to diagnosis of medulloblastoma in children is not associated with decreased survival or with worse neurological outcome.

Jean-Francois Brasme,Stéphanie Puget,Brigitte Lacour,Olivier Delalande,Francois Doz,Jacques Grill,Stephan Gaillard,Martin Chalumeau,Dominique Valteau-Couanet,Nozar Aghakhani

doi:10.1371/journal.pone.0033415

Abstract

BackgroundThe long time to diagnosis of medulloblastoma, one of the most frequent brain tumors in children, is the source of painful remorse and sometimes lawsuits. We analyzed its consequences for tumor stage, survival, and sequelae.Patients and MethodsThis retrospective population-based cohort study included all cases of pediatric medulloblastoma from a region of France between 1990 and 2005. We collected the demographic, clinical, and tumor data and analyzed the relations between the interval from symptom onset until diagnosis, initial disease stage, survival, and neuropsychological and neurological outcome.ResultsThe median interval from symptom onset until diagnosis for the 166 cases was 65 days (interquartile range 31–121, range 3–457). A long interval (defined as longer than the median) was associated with a lower frequency of metastasis in the univariate and multivariate analyses and with a larger tumor volume, desmoplastic histology, and longer survival in the univariate analysis, but not after adjustment for confounding factors. The time to diagnosis was significantly associated with IQ score among survivors. No significant relation was found between the time to diagnosis and neurological disability. In the 62 patients with metastases, a long prediagnosis interval was associated with a higher T stage, infiltration of the fourth ventricle floor, and incomplete surgical resection; it nonetheless did not influence survival significantly in this subgroup.ConclusionsWe found complex and often inverse relations between time to diagnosis of medulloblastoma in children and initial severity factors, survival, and neuropsychological and neurological outcome. This interval appears due more to the nature of the tumor and its progression than to parental or medical factors. These conclusions should be taken into account in the information provided to parents and in expert assessments produced for malpractice claims.

Highlights

Brain tumors are the leading cause of solid cancers in children [1]
In the 62 patients with metastases, a long prediagnosis interval was associated with a higher T stage, infiltration of the fourth ventricle floor, and incomplete surgical resection; it did not influence survival significantly in this subgroup
We found complex and often inverse relations between time to diagnosis of medulloblastoma in children and initial severity factors, survival, and neuropsychological and neurological outcome

Summary

Introduction

Brain tumors are the leading cause of solid cancers in children [1]. Medulloblastoma, one of the most common types [1], has a 10-year survival rate of 50% [2,3,4,5], and many survivors have neurological and cognitive sequelae [6,7]. Several factors limit the usefulness of these results: the limited numbers (,100) of pediatric patients [13,17], pooled analyses for pediatric and adult patients [13], single-center recruitment subject to selection bias [13], incomplete initial disease staging for some patients that increases the likelihood of classification bias [13], a study period partially preceding the availability of CT and MRI [13], and a lack of multivariate analyses despite the presence of potential confounders [13,17] None of these studies analyzed the relation between time to diagnosis and either local tumor stage, complete surgical resection or neurological and cognitive sequelae. We analyzed its consequences for tumor stage, survival, and sequelae

Objectives

Methods

Results

Conclusion