Risk model of seizure cluster or status epilepticus and intervention in the emergency department

Abstract

Objectives

To identify possible predictors of seizure cluster or status epilepticus (SE) and to evaluate whether these patients receive greater interventions in emergency departments.

Methodology

We conducted a secondary analysis of the ACESUR Registry, a multipurpose, observational, prospective, multicentre registry of adult patients with seizures from 18 emergency departments. Clinical and care-related variables were collected. We identified risk factors and risk models for seizure cluster or SE and assessed the effect of interventions by prehospital emergency services and the hospital emergency department.

Results

We identified a total of 186 (28%) patients from the ACESUR registry with seizure cluster (126 [19%]) or SE (60 [9%]); the remaining 478 patients (72%) had isolated seizures. The risk model for seizure cluster or SE in the emergency department included Charlson Comorbidity Index scores ≥ 3 (OR: 1.60; 95% CI, 1.05–2.46; P = .030), ≥ 2 habitual antiepileptic drugs (OR: 2.29; 95% CI, 1.49–3.51; P < .001), and focal seizures (OR: 1.56; 95% CI, 1.05–2.32; P = .027). The area under the curve of the model was 0.735 (95% CI, 0.693–0.777; P = .021). Patients with seizure cluster and SE received more aggressive interventions both by prehospital emergency services (OR: 2.89; 95% CI, 1.91–4.36; P < .001) and at the emergency department (OR: 4.41; 95% CI, 2.69–7.22; P < .001).

Conclusions

This risk model may be of prognostic value in identifying adult patients at risk of presenting seizure cluster or SE in the emergency department. In our sample, these patients received more aggressive treatment than adult patients with isolated seizures before arriving at hospital, and even more so in the emergency department.