Predictive value of clinical evaluation in the follow-up of children with a brain tumor.

Background. During follow-up of children with a brain tumor, traditionally surveillance-imaging studies are done in addition to clinical evaluations, The purpose of this study was to determine the role of clinical evaluations by a multidisciplinary team for the detection of recurrent tumor. Procedure. We retrospectively assessed the predictive value of clinical evaluation, using subsequent neuroimaging as the gold standard. Ninety-eight children with a newly diagnosed primary brain tumor were included in the study. In these patients, 393 imaging studies were done, 75 because of clinical suspicion of recurrence, and 318 for routine surveillance only. Results. In 28 of these 75 imaging studies on clinical indications, a radiologic diagnosis of recurrence was made. Only 9 out of 318 routine surveillance imaging studies resulted in a nunexpected diagnosis of recurrence. Thus, the overall positive predictive value of clinical evaluation was 37%; the overall negative predictive value was 97%. The negative predictive values for specific brain tumors varied from 91% (optical glioma) to 99% (primitive neuroectodermal tumors). Conclusions. An accurate multidisciplinary clinical evaluation has a very high negative predictive value. Consequently, the added value of surveillance imaging studies is limited. The role of such studies during the follow-up of children with brain tumors should be reconsidered. Med Pediatr Oncol 2002;37: 254-257. (C) 2002 Wiley-Liss, Inc.