Med Pediatr Oncol 2002;39:566–572 On the Origin of EEG-Slowing and Encephalopathy During Induction Treatment of Acute Lymphoblastic Leukemia Rudolf Korinthenberg, MD, 1 * Bernadette Scheuring, 1 Joachim Boos, MD, 2 and Charlotte Niemeyer, MD 3 INTRODUCTION Neurological complications frequently occur during the treatment of childhood acute lymphoblastic leukemia (ALL). A total of 25%–50% of patients can be expected to exhibit abnormal neurological signs due to various etiologies. Gross and fine motor dysfunction, as well as neuropsychological deficits, are encountered in a signi- ficant number of children years after the termination of treatment [1,2]. Beyond these clinical abnormalities, reversible EEG changes have been reported in a high proportion of patients, especially during induction and re-induction treatment. These can be due to hematological or infec- tious complications, though in the largest number of children, toxicity of treatment was suspected to be causative [3–5]. We carried out prospective sequential EEG and CSF- amino acid investigations with two aims: (i) to draw conclusions on the etiology of the EEG changes from their exact timing during the treatment course and (ii) to investigate the possible role of treatment-induced CSF amino acid changes in neurological dysfunction. PATIENTS AND METHODS All children with standard-risk (SR) and medium-risk (MR) ALL diagnosed between October 1993 and May 1996 were eligible to participate in this study. Four had to be excluded due to a lack of parental consent or insufficient CSF samples. The data of 29 patients (17 boys and 12 girls) were available for analysis. Their mean age was 6.1 years (SD 3.6 years, range 1.5–15 years). The duration of specific or unspecific symptoms prior to the diagnosis of ALL ranged from 1 to 24 weeks (mean 8.4 weeks, SD 7.5). Background. Neurological complications and EEG slowing frequently occur in children undergoing induction treatment for acute lym- phoblastic leukemia (ALL). Disease-related fac- tors and treatment-related toxicity are believed to play causative roles. We wanted to elucidate the etiology further by serial EEG examinations and parallel CSF amino acid analyses. Proce- dure. Twenty-nine children participated in the study. EEG examinations with quantitative computerized analysis were scheduled on day 1, 10, 29, and 59 of protocol I of BFM-ALL 90 and 95 Study Protocols. CSF analysis for amino acids was carried out on day 1, 15, 29, 45, and 59. Results. A total of 21 of 25 available EEGs showed slight-to-moderate slowing already at diagnosis. The abundance of slow waves was significantly correlated to the white blood count and the CSF glutamine concentration. The EEGs significantly worsened during the first 10 days of treatment with prednisone, VCR, daunorubicin, and intrathecal methotrexate. The following treatment including asparaginase (ASP) gave rise to depletion of CSF from asparagine and a rise of aspartate; glutamine, and glutamate did not follow this pattern. The EEGs remained abnormal, but did not worsen further; the CSF amino acid changes were not related to the EEG. During the subsequent consolidation treatment, the EEGs normalized despite ad- ministration of cyclophosphamide, cytara bine, intrathecal methotrexate, and mercaptopurin. Conclusions. The greater part of EEG changes observed in the early treatment of ALL is due to disease-related factors. Treatment with predni- sone, vincristine, and to a much lesser degree asparaginase aggravates the pre-existing ence- phalopathy. Depletion of CSF from asparagine does not give rise to additional changes. In the second month, the EEG normalizes despite on- going treatment with different cytotoxic drugs. Med Pediatr Oncol 2002;39:566–572. ß 2002 Wiley-Liss, Inc. Key words: acute lymphoblastic leukemia; childhood cancer; neurotoxicity; electro- encephalography; cytostatic drugs —————— 1 Department of Neuropediatrics and Muscular Disorders, Pediatric University Hospital Freiburg, Germany 2 Department of Oncology, Pediatric University Hospital, Mu ¨nster, Germany 3 Department of Pediatrics, Pediatric University Hospital Freiburg, Germany *Correspondence to: Rudolf Korinthenberg, Klinik II: Neuro- pa ¨diatrie und Muskelerkrankungen, Zentrum fu ¨r Kinderheilkunde und Jugendmedizin, Mathildenstr, D-79106 Freiburg, Germany. E-mail: rudokori@kikli.ukl.uni-freiburg.de Received 30 May 2001; Accepted 7 May 2002 ß 2002 Wiley-Liss, Inc. DOI 10.1002/mpo.10189