The SEDS score: A daily severity score for monitoring patients with refractory SE in the ICU

Abstract

Introduction

Refractory status epilepticus (RSE) usually requires admission in an intensive care unit (ICU). Available specific prognostic scores for SE do not account for refractoriness, life support dependency, or related comorbidities, and usually fail to accurately predict outcome in RSE. Our goal was to develop a daily severity score, like existing scores used for critical illness, to follow patients' trajectory in the ICU and predict their outcome at discharge.

Methods

Retrospective observational study of two independent prospectively identified cohorts (derivation and validations cohorts) of RSE episodes admitted in the ICU of two tertiary-care centers, between January 2015 and May 2023. The sequential organ failure assessment (SOFA) score, additional neurological, EEG, treatment-related and biological variables were collected and compared between poor and good outcome groups (modified Rankin Scale (mRS) ≥ 4 or <4), on admission and during the first week of ICU stay, using Mann-Whitney tests and generalized linear mixed models. Based on significant univariate variables, we derived the SE daily severity (SEDS) score and measured its predictive performance for functional outcome using the area under the receiver operating characteristic curve (AUC).

Results

54 episodes and 71 episodes were included in derivation and validation cohorts, respectively. In the derivation cohort, no difference was found at ICU admission. The course of the SOFA score over the first week differed (p < 0.001) between groups, especially their cardiovascular and respiratory components, C-reactive protein (p < 0.001), SE-related and EEG variables. The SEDS score, which includes cardiovascular and respiratory components of SOFA score, CRP levels, GCS, EEG patterns and the use of continuous intravenous anesthetic drugs, had better AUCs than each components considered separately, ranging from 0.70 to 0.83 in the derivation cohort from 0.72 to 0.84 in the validation cohort.

Conclusion

In the absence of differences in existing scores on ICU admission, the SEDS score achieved an AUC >0.70 throughout the first week of ICU stay, indicating good performance to predict poor outcome at hospital discharge.