Early multimodal predictors of good neurological outcome in comatose patients after cardiac arrest: a prospective single-center cohort study
摘要
Post–cardiac arrest (CA) prognostication primarily focuses on predicting poor outcome, whereas early markers of good prognosis remain less explored. We aimed to identify early predictors of favorable outcome at 3 months in comatose CA survivors.
MethodsWe prospectively enrolled adult patients with post-anoxic coma admitted to the Intensive Care Unit of a secondary-care hospital (2020-2025). Demographic, CA-related, clinical, electroencephalographic (EEG), neuron-specific enolase (NSE) and neuroimaging data were collected. Good outcome was defined as Cerebral Performance Category (CPC) 1-2 at 3 months. Logistic regression analyses identified independent predictors of good outcome. Model performance was internally validated with bootstrap resampling.
ResultsAmong 121 included patients (median age = 66; 72% male), outcome was available for 118, of whom 42.3% achieved good outcome. Patients with CPC 1–2 were younger, had predominant shockable initial rhythms and cardiac etiology of CA, higher Glasgow Coma Scale scores, lower NSE levels, and decreasing NSE trend. Peak NSE ≤ 36 ng/mL was the optimal cut-off for favorable prognosis. Early benign EEG (continuous/nearly continuous background with preserved reactivity ≤72 h) was associated with neurological recovery. In multivariable analysis, independent predictors of favorable outcome were younger age (OR 0.92, 95% CI 0.86–0.99), shockable initial rhythm (OR 28.0, 95% CI 3.87–202.1), early benign EEG (OR 10.76, 95% CI 1.95–59.4) and peak NSE ≤ 36 ng/mL (OR 43.17, 95% CI 6.01–309.9). The model showed excellent discrimination (apparent AUC 0.97, optimism-corrected AUC 0.96).
ConclusionsIn comatose post-CA patients, a model combining age, initial rhythm, early benign EEG and peak NSE independently predicted good outcome at 3 months.