Abstract □ 81

Introduction: This study was aimed to establish the distribution and morphology of microglia and astroglia in arousal centers. Arousal is that group of behavioural changes that occur when a person awakens from sleep or makes the transition to a state of readiness. These centers are located adjacent to the third ventricle in dorsomedial and reticular thalamic nuclei and in paraventricular, supraoptic and posterior hypothalamic nuclei. As SID is associated with sleep it is postulated that disturbance of the arousal system is involved in the pathogenesis of SID. Microglia are an early and sensitive indicator of irritation on the cellular level and they react to extracellular stimuli. Astroglial response is starting later and remains longer. Material and methods: For investigation 36 SID and 13 control cases of infants without known neurological disease were selected. All infants had died during the first year of life; 30 were males, and 19 were females. Prior to histological processing the brain tissue was fixed in formalin for 2 months up to 2 years. Representative specimens of diencephalon were embedded in paraffin. Paraffin sections were investigated by routine stainings and by application of monoclonal antibodies to characterize microglia: KiM1P (unknown antigen; kindly provided by Prof. Parwaresch, Kiel), CR3/43 (MHC II; HLA-DR, DAKO), and KP1 (macrophage membranes; DAKO). Polyclonal antibody to glial fibrillary acidic protein (GFAP; DAKO) was used to visualize the cytoskeleton of astrocytes. Lectin histochemistry with Ricinus communis agglutinin 120 (RCA I; Sigma) was applied as a specific microglial marker.

Results: Microglial cells were more frequently distributed in the white matter than in the grey matter. They were mainly situated in subependymal myelinated areas. A selective accumulation and/or activation of microglia in arousal-related nuclei in thalamus and hypothalamus could not be ascertained. The highest intensity of GFAP expression was observed in the marginal/periventricular and perivascular areas. A gliosis in arousal centers of diencephalon could not be demonstrated by immunohistochemistry. In 11 cases (7 SID, 4 controls) erythrodiapedesis in periventricular sites was found. All of them had a spatial relationship to either venous or arterial vessels with intact walls. The haemorrhages were small in diameter and restricted to the perivascular space. An astroglial or microglial reaction had not been induced.

Conclusions: There was no significant difference in histopathological findings in the investigated brains of SID and controls. The lack of glial reaction does not support the hypothesis of disturbed arousal causing SID. The high frequency of erythrodiapedetic haemorrhages in both groups indicates a non-specific agonal event.