Temporo-spatial correlations between scalp and centromedian thalamic EEG activities of stage II slow wave sleep in patients with generalized seizures of the cryptogenic Lennox-Gastaut syndrome.

Objectives: Temporo-spatial correlations between scalp and centromedian thalamic (CM) normal and abnormal electroencephalographic (EEG) activities of stage II slow wave sleep (SWS II) were investigated in 5 patients with cryptogenic Lennox-Gastaut syndrome (CLGS).

Methods: In each patient, 8h/all-night sleep studies were performed with routine methods; and a total of 1233 normal and 206 abnormal individual activities, spontaneously occurring during 200 epochs of early and late SWS II, were analyzed. Normal activities included scalp-CM K-complexes (KC-CMKC), vertex waves (VW-CMVW), and sleep spindles (SS-CMSS). Abnormal activities included: thalamo-cortical spikes (TCS-CMTCS), and epileptic (EPKC-CMEPKC) and W K-complexes (WKC-CMWKC).

Results: (1) All abnormal and normal spontaneous SWS II activities occurred associated in scalp and CM regions except the SS. Associated spindles were significantly larger (P<0.01) than dissociated ones, this occurring during both early and late SWS II. (2) The peak of VW significantly anticipated (P<0.02) that of its CM counterpart (CM-VW), while the peak of CMTCS anticipated that of its scalp counterpart. The onset of CMSS significantly anticipated (P=0.02) that of its scalp counterpart (SS). The behavior of VW-CMVW and TCS-CMTCS of the abnormal KC was similar to those of the normal complexes, while the onset of abnormal spindles was simultaneous in scalp and CM regions. Scalp VW, CTS, and SS attained maximal amplitude at the parietal region bilaterally with decreasing amplitude gradients to other scalp regions, while CMVW, CMTCS, and CMSS attained maximal amplitude in all thalamo-mesencephalic regions of CM. (3) Normal spindles significantly reduced (P<0.02) the amplitude of the positive CM, CMVW, and scalp TCS counterparts of the negative scalp VW and CM (CMTCS), respectively, while abnormal spindles reduced the amplitudes (P<0.01) of both negative VW and CMTCS and positive counterparts.

Conclusion: These data suggest the following: (1) that all SWS II activities, including SS, are mediated by common thalamo-cortical systems; (2) that VW originate from the parietal scalp and normal spindles and TCS from the CM regions bilaterally while abnormal spindles originate either from widespread cortical and CM regions or from a site outside the thalamo-cortical systems, and (3) that the functional role of SS is to inhibit non-specific thalamo-cortical systems for sleep preservation.