Article révisé par les pairs
Résumé : Objective To validate the complication burden index (CBI), a score quantifying 13 types of complications during status epilepticus (SE) treatment and evaluate its impact and that of its components on 6-month mortality. Methods This retrospective multicenter cohort study included adult patients with non-anoxic SE treated at Odense University Hospital (Denmark) between 2008 and 2017 and at Erasme Hospital (Belgium) between 2014 and 2017. The association of CBI and its components -obtained from clinical records- with functional outcome, in-hospital, and 6-month mortality was assessed using uni- and multivariable analyses. Long-term survival depending on CBI was compared using the log-rank test. Results The combined cohort comprised 537 patients (n = 261 from Denmark and n = 276 from Belgium), with a median CBI of 3 in both cohorts (range: 0–10) with similar, though not identical, patterns observed in both. CBI was associated with treatment response in the emergency room, SE duration, comorbidities, and etiology, but not with age. CBI independently predicted in-hospital mortality in multivariate analyses in both cohorts. In contrast, in patients discharged alive from hospital, CBI was only weakly linked to 6 month-mortality. In multivariable analyses adjusted for age, duration and etiology, cardiovascular, respiratory and renal complications were the only components independently associated with in-hospital mortality. Cardiovascular, coagulative, infectious and renal complications independently predicted all-cause mortality at six months. Conclusion CBI proves to be a valid tool for quantifying complication burden during SE. CBI and especially cardiovascular, respiratory, and renal complications are associated with in-hospital mortality. CBI was not an independent predictor of mortality after 6 months in patients discharged alive highlighting the temporal nature of complication-related mortality.

Documents en relation

DI-fusion