Long-term monitoring of cardiorespiratory patterns in drug-resistant epilepsy

ObjectiveSudden unexplained death in epilepsy (SUDEP) during inpatient electroencephalography (EEG) monitoring has been a rare but potentially preventable event, with associated cardiopulmonary markers. To date, no systematic evaluation of alarm settings for a continuous pulse oximeter (SpO(2)) has been performed. In addition, evaluation of the interrelationship between the ictal and interictal states for cardiopulmonary measures has not been reported. MethodsPatients with epilepsy were monitored using video-EEG, SpO(2), and electrocardiography (ECG). Alarm thresholds were tested systematically, balancing the number of false alarms with true seizure detections. Additional cardiopulmonary patterns were explored using automated ECG analysis software. ResultsOne hundred ninety-three seizures (32 generalized) were evaluated from 45 patients (7,104 h recorded). Alarm thresholds of 80-86% SpO(2) detected 63-73% of all generalized convulsions and 20-28% of all focal seizures (81-94% of generalized and 25-36% of focal seizures when considering only evaluable data). These same thresholds resulted in 25-146 min between false alarms. The sequential probability of ictal SpO(2) revealed a potential common seizure termination pathway of desaturation. A statistical model of corrected QT intervals (QTc), heart rate (HR), and SpO(2) revealed close cardiopulmonary coupling ictally. Joint probability maps of QTc and SpO(2) demonstrated that many patients had baseline dysfunction in either cardiac, pulmonary, or both domains, and that ictally there was dissociationsome patients exhibited further dysfunction in one or both domains. SignificanceOptimal selection of continuous pulse oximetry thresholds involves a tradeoff between seizure detection accuracy and false alarm frequency. Alarming at 86% for patients that tend to have fewer false alarms and at 80% for those who have more, would likely result in a reasonable tradeoff. The cardiopulmonary findings may lead to SUDEP biomarkers and early seizure termination therapies.