Abstracts

Rationale: Hemodynamic changes in the brain are often used as surrogates for epileptic neuronal activity in both the laboratory and the clinic (e.g. intrinsic signal, fMRI and SPECT), in spite of the fact that the precise spatiotemporal relationship between perfusion, oxygenation and the neuronal dynamic is not well understood. What is missing is a technique for simultaneously measuring neuronal activity and perfusion in a large area of cortex with high spatial and temporal resolution. We developed a simultaneous recording method to image ictal activity with voltage sensitive dyes and the blood volume changes with intrinsic optical imaging. Methods: The cortex was stained with the new blue dye RH-1692. 1.5μl of 4-aminopyridine (4-AP) was injected into the superficial layer of the neocortex to induce seizures and interictal spikes. A second glass electrode (2-4 MΩ) filled with 0.9% saline was positioned < 1 mm from the 4-AP injection site to record the local field potential (LFP). Two illumination sources were employed for VSDI (630±20nm, for new blue dye RH-1692) and ORIS (546±20nm, reflecting blood volume change) respectively. A 570nm dichroic mirror was used to separate VSDI signal (>665nm) and ORIS signal (546 nm). VSD signal and ORIS signal were simultaneously recorded with two cameras. The VSD data was recorded with the MACAM ULTIMA fast camera system with 1 ms temporal resolution and 50 μm spatial resolution. Images were taken for 20 seconds each session. The intrinsic optical was recorded with the video camera (Imager 3001, Optical Imaging). Results: The ictal event initiated with low-amplitude high frequency activity (initiation), and evolved into regular spike-wave activity with the frequency of 10-20 Hz (evolution) and terminated with intermittent poly-spikes (termination). During initiation and evolution, the propagation of epileptiform activity followed a similar pathway. However, during the termination, different spikes followed different pathways, indicating the desynchronization of the neuronal network. The intrinsic signal, on the other hand, showed a slow propagation, which does not accurately reflect the spatial-temporal dynamic of the underlying neuronal activity. Conclusions: Our data indicated that each spike of a seizure consists of an inhomogeneous multidirectional propagating wave that involves a large area of cortex. Maps of seizures based on perfusion do not clearly reflect the complexity of the underlying electrophysiological activity.