Diagnostic delay in children with central nervous system tumors and the need to improve education

To the Editor,

We read with interest the recent study by Patel et al. reporting delayed diagnoses in children with central nervous system (CNS) tumors, in which they reviewed a single-institution experience at Akron Children’s Hospital in Ohio [1]. The authors described a scarcity of studies on this topic in the US and cited only 1 study from Nationwide Children’s Hospital, also located in Ohio [2], but did not cite our study published a few years earlier on delayed diagnoses in children with low-grade glioma (LGG) [3].

We commend the authors for their work addressing this important topic that should be indeed further studied in the US. Although this study, like many others, did not show an effect of prediagnostic symptom interval (PSI) length on survival or long-term morbidity, the authors stated that shorter PSIs may decrease morbidity and cited a study from Japan suggesting that shorter PSIs yield better functional outcomes [4]. We believe it is important to address this conclusion because most childhood CNS tumors are associated with good long-term survival. Therefore, it is difficult to show that delayed diagnosis of CNS tumors affects overall survival. For this reason, measuring the effect of delayed diagnoses on patient functionality and morbidity should be emphasized equally as overall survival. Unlike many high-grade tumors which are widespread and have significant morbidity and decrease overall survival due to delayed diagnosis, LGG are not exactly similar. Moreover, future studies should investigate the effect of PSI length on distinct tumor types separately because it is illogical to group disparate CNS tumor types in studies of PSI. Pooling of distinct CNS tumor types with distinct pathologies does not provide informative data on the magnitude of the effect of delayed diagnosis and may even confound the findings of such studies by showing that longer PSIs are associated with better prognoses [3]. Because our study was specifically limited to pediatric LGG, we were able to show that longer PSIs are associated with increased risk of progressive disease and seizures, thereby exacerbating long-term treatment toxicities, disease morbidity, and impaired functionality.

We agree with Patel et al. that improved education on this health care challenge is needed, but we do not believe that the HeadSmart awareness campaign on the presentation of childhood brain tumors in the UK is a panacea for every country [5]. The US is much larger than the UK, and the health care systems are markedly different, especially regarding the universality of health care insurance and medical training structures. Moreover, education should be further emphasized in residency programs because delayed diagnosis occurs at different stages in various disciplines, such as psychiatry, ophthalmology, and endocrinology, which can widely differ from that in general pediatrics or family medicine. Furthermore, we believe that nursing programs/schools should also be targeted to improve the education of childhood brain tumor presentation. Indeed, we proposed in our study the acronym “LOW OR PAY” (i.e., think low-grade glioma or patient will pay the price) as an educational tool to help students, trainees, and health care providers consider the different presentations of brain tumors in children.

Most childhood CNS neoplasms are low-grade tumors, such as LGG, craniopharyngioma, choroid plexus papilloma, and myxopapillary ependymoma. For such tumors, symptoms can be atypical, wax and wane, and last for many years before being diagnosed. The triad of headache, nausea, and vomiting does not represent the typical presentation for brain tumors. Although the presence of this triad is important for suspecting aggressive brain tumors, a lack of this triad does not exclude CNS tumors, especially in children, and should be disseminated accordingly in our educational efforts. Indeed, Patel et al. showed that this triad was present in only 29.4% of the cases they studied. Although education is a key element for solving this problem, root cause analyses should be performed first for each health care system, and subsequent educational campaigns tailored accordingly.