Related Experiment Videos
Encephalomyelopathies associated with extracerebral malignant tumors
1Institute of Brain Research, Eberhard-Karls-University of Tübingen.
Abstract:
It was the aim of this study to examine the probability of pathogenetical relations between extracerebral malignant tumors and lesions of CNS. The term paraneoplastic should be questioned. Among a running series of 2,000 brain autopsies, 456 patients (22.8%) showed such tumors, 362 of these combined with lesions in brain or spinal cord. Out of these cases, 100 had metastases, meningoses blastomatosae or leukotic hemorrhages, 218 other, non-tumoral lesions, and 44 both tumoral and non-tumoral lesions. The last-mentioned 262 cases were the target of the examination. We distinguished 6 groups: a) various well-definable impacts, diseases or genetical defects (n = 18), b) unspecific terminal resp. agonal changes (n = 49), c) vascular or circulatory disorders incl. embolizations (20 non-leukotic hemorrhages, 148 anemic infarctions or selective neuronal necroses), thromboses, angiitis or calcifications, d) infections and other inflammatory alterations (n = 37), e) metabolic and toxic lesions (9 Wernicke's disease, 12 central pontine myelinolyses, f) anomalies difficult to classify (51 cases with subacute cerebellar atrophy, diffuse leukoencephalopathy, focal spongious axonopathic lesions, myelomalacia and other). After analysing the various lesions and discussing the probable pathogenesis we grouped according to the following scheme: I) tumor-unrelated (casual coincidence) (43.2%), II) therapy-dependent (3.7%), III) agony-related (10.7%), IV) homoiogenic disorders (e.g. larynx carcinoma and Wernicke's disease) (2.0%), V) nosocomial disorders (12.9%), VI) tumor-dependent (local neighbourhood effects, primary or secondary remote effects). As the central group there remain the primary remote effects (17.3%), separable into specific functional anomalies by tumorous organ destruction, remote effects of tumor cell-born (ectopic) release of hormones or hormone-like substances (n = 2), tumor antigen-dependent immune reactions (n = 20), and pathogenetically still uncleared mechanisms (n = 51). One should apply the term paraneoplastic only for the three last-mentioned conditions.