Monitoring of therapy-related late effects after acute lymphoblastic leukaemia (ALL) therapy in childhood has become an increasingly important field in posttherapeutic patient surveillance. The usefulness of neurophysiological investigations (e.g. EEG, evoked potentials (EP)) as part of these attempts is controversial. The present report focuses on this problem and the question whether and to what extent routinely performed EEG recordings and visual evoked potentials (VEP) were correlated with further measures of CNS integrity. EEGs and VEPs were recorded in 163 asymptomatic long-term survivors of ALL in childhood during a large retrospective multicentre study evaluating CNS late sequelae following antileukaemic therapy. Fifty-two ALL long-term survivors (4.5-10.6 years after end of therapy, median: 8.8 years), who had been treated according to BFM-81 SR-A (n = 30) or SR-B (n = 22) were selected for this analysis focusing on therapy-related CNS late effects. Therapy protocols differed with regard to the mode for CNS prophylaxis: SR-A, cranial irradiation with intrathecal methotrexate; SR-B, intrathecal and iv methotrexate. Neurophysiological findings were correlated with illness- and treatment-related parameters, as well as with data on the morphological, neurological and psychological status of the CNS. At the time of follow-up neurophysiological measures were abnormal in 28/52 cases (53.8%). Neither illness- nor therapy-specific differences in CNS prophylaxis showed any relationship to EEG/VEP outcome any relationship to EEG/VEP outcome in this reduced group of the whole study population. Children with EEG/VEP abnormalities showed a significantly higher incidence of structural CNS disturbances compared to those with inconspicuous neurophysiological recordings (60.9% vs 31.8%). However, in this special subject group there was no specific neurophysiological finding for a specific morphological substrate, neurological or psychological deficiency and vice versa.
Conclusion: Routinely performed EEG/VEP investigations are not very helpful measures to predict the presence or degree of behavioural deficiencies, neurological disturbances, or morphological CNS abnormalities. Patients who received cranial irradiation or systemic methotrexate applications showed the same incidence of neurophysiological disturbances without evidence for specific neurotoxic correlates.