Biomarkers of long-term mortality in seizure patients treated by emergency medical services: Score development and validation

BACKGROUND: Seizures are one of the most attended neurological emergencies in the prehospital context. The categorization of the risk of these patients is a great challenge for health professionals, due to the limited information, leading to inaccurate diagnoses. The aim of this study was to explore the development and validation of a long-term mortality predictive score that considers vital signs and biomarkers in seizure patients. METHODS: A prospective, multicenter study was conducted by emergency medical services (EMS) in Spain, including five advanced life support units, 27 basic life support units and four emergency services. The sample consisted of adults who suffered prehospital seizures, in which vital signs and blood tests were recorded using point-of-care tests (POCTs) to predict long-term all cause 1 year mortality. RESULTS: The sample consisted of 198 patients, in whom 33 mortality events were recorded. Our predictive model identified age, Glasgow Coma Scale (GCS) score, hemoglobin, serum Anion Gap (SAG), international normalized ratio (INR), and Charlson comorbidity index as risk factors, revealing and AUC of the score of 0.743 (95%CI 0.624-0.862). CONCLUSION: This study has identified hemoglobin, SAG and INR as prehospital biomarkers capable of predicting long-term mortality in patients who have suffered prehospital seizures. The combination of these new biomarkers to age and GCS into a score available for EMS staff could be a practical and effective tool that improves risk stratification and patient management.