Abstracts

Rationale: Epilepsy is one of the most prevalent neurological diseases in humans and the most common in dogs. EEG is necessary for definitively diagnosing epileptic seizures. For humans, the 10-20 electrode placement grid is the internationally accepted standard, with each named electrode location overlying known cortical sites. Veterinary neurology must eliminate other diagnoses due to a lack of protocol standardization for canine EEG. Clarifying the variation of canine skull morphology and comparing neuroanatomy between species permits craniocerebral topographical mapping, thus noninvasively standardizing the placement of EEG electrodes in dogs. A standardized EEG system will contribute to a translational model of spontaneous epilepsy. Methods: Part I, skull dimensions across breeds – 100 CT studies retrieved from the Ontario Veterinary College (OVC) medical imaging database. Exclusion criteria include age, pathology and image quality. Using medical imaging processing software (Osirix/Horos), skull shape is categorized by skull index (skull width L-R zygion : length inion-prosthion) into three groups: dolicho-, mesati- and brachycephalic. Calvarial size is measured by cranial index (cavity width L-R euryon : length inion-nasion). Part II, comparative cortical anatomy – literature review comparing topography of the four cortical lobes based on location and function using sulci and gyri. Part III, canine electrode placement grid – proposal based on findings from Parts I & II. Results: Domestic dogs display skull and cranial index values along a spectrum rather than within distinct categorical groups. There is more variation in skull than cranial index across breeds (cranial length 5.44-11.40 cm; cranial width 4.22-6.17 cm; skull length 9.03-28.30cm; skull width 6.06-15.00 cm) suggesting a standardized electrode placement system based on cortical topography. This is supported by comparative neuroanatomical features: the occipital lobe and visual areas are the most complementary, temporal lobe association areas are similarly located in ventral-caudal regions of the temporal lobe, auditory cortex areas involve anterior temporal regions in both, the primary somatosensory area occupies the parietal lobe in both, motor regions of the frontal lobe in humans extend to areas of the gyrus sigmoideus in dogs. Based on these cortical correlations, an electrode placement grid is proposed for dogs using analogous nomenclature to the human 10-20 system. Variations in skull morphology finally determine palpable landmarks to anchor the grid. Conclusions: This study addresses animal model validity criteria for a proposed EEG electrode placement grid. Next stages involve validation with advanced medical imaging using regression models of predicted location values compared to actual locations indicated by fiducial imaging markers. Establishing dog EEG as a translational model potentiates meaningful progress in epilepsy and EEG research. Funding: Andrea Leger Dunbar Summer Research Assistantship, University of Guelph GEES, OVC MSc Scholarship, OVC RaPPID and OVC Pet Trust.