Abstracts

RATIONALE: Tachycardia occurs in the majority of temporal lobe seizures and appears to be independent of clinical signs and symptoms. Heart rate elevation during seizures has been used in computerized seizure detection during VEEG monitoring at our institution. The aim of this study was to determine the factors associated with lesser degrees of heart rate increase in temporal lobe seizures. METHODS: 95 temporal lobe seizures in 38 patients were recorded during VEEG. Seizures with heart rate increase not exceeding our computerized seizure detection threshold of heart rate greater than mean and three standard deviations for 15 seconds were considered to have lesser degrees of heart rate change and were analyzed. Seizure duration, patient activity, side of seizure focus and heart rate change from baseline during seizures was assessed in these patients. RESULTS: Three seizures were excluded due to artefact affecting analysis. Five of 92 (5%) seizures were not detected by heart rate seizure detection method. Two of five patients met the threshold but not the duration required for seizure detection. Heart rate only exceeded two standard deviations above mean in one seizure of a patient who did meet detection parameters in another seizure. Two patients were not observed to have any significant change in heart rate. Left temporal onset occurred in four of the five(80%)seizures. Seizure duration and clinical signs in these seizures did not appear to be significantly different from seizures with tachycardia. CONCLUSIONS: Heart rate changes are significantly increased above baseline in the vast majority of temporal lobe seizures, even if ictal tachycardia as defined in studies looking at its incidence in partial seizures does not occur before or during the seizure. Left temporal onset seizures may have relatively less tachycardia and may be one of the factors affecting the sensitivity of this seizure detection algorithm when applied to all patients. Adjustments of the heart rate detection parameters for individual patients may increase the clinical usefulness of this algorithm without loss of specificity. Supported by Developmental Technology Award #B-01-96, NCBH, Winston-Salem