Ultra-early short- and middle-latency SSEP accurately predict good and poor outcome after cardiac arrest
Maenia Scarpino, Andrea Nencioni, Pasquale Bernardo, Manuela Bonizzoli, Peiman Nazerian, Benedetta Piccardi, Riccardo Carrai, Claudio Sandroni, Antonello Grippo, Antonello Grippo, Cecilia Agostini, Pasquale Bernardo, Manuela Bonizzoli, Flavia Caniato, Riccardo Carrai, Annalisa Cassardo, Antonella Cramaro, Antonello Grippo, Maria Lombardi, Cristina Mei, Peiman Nazerian, Andrea Nencioni, Michele Ombrosi, Benedetta Piccardi, Claudio Sandroni, Claudio Sandroni, Maenia Scarpino, Marta Silvestri, Maddalena Spalletti
Abstract
BACKGROUND: Accurate prognostication following cardiac arrest (CA) is crucial for informing clinical decisions. Current guidelines do not recommend a specific time point for recording somatosensory evoked potentials (SSEPs) after CA. We evaluated the ability of ultra-early short- and middle-latency SSEPs to predict good an poor neurological outcome and compared its accuracy with that of other predictors recorded early after CA. METHODS: Prospective single-centre study. Sixty-five comatose adults underwent a multimodal prognostic assessment, including neurophysiological (SSEPs and electroencephalogram [EEG]), clinical (pupillary reflexes and myoclonus), and imaging indices (brain computed tomography [CT]) within 6 h post-CA. Serum neuron-specific enolase (NSE) was sampled 12 h post-CA. We analysed the SSEPs N20 wave amplitude and duration, and the presence of the middle-latency N70 wave. Poor outcome was defined as a Cerebral Performance Category (CPC) of 3-5 at hospital discharge. RESULTS: A bilaterally absent N20 wave predicted poor outcome with 100[89-100]% specificity and 67[48-82]% sensitivity. Adding low-amplitude (<1.2 µV), prolonged (>10 ms) N20 waves without N70 increased sensitivity to 93[79-99]% without compromising specificity. Conversely, a high-amplitude (>3 µV) N20 wave with normal duration with preserved N70 predicted good outcome with 94[79-99]% sensitivity and 100[89-100]% specificity. SSEPs outperformed all other early prognostic indices for both good and poor outcome prediction. All poor outcome patients had at least two concordant unfavourable predictors. CONCLUSIONS: Ultra-early quantitative assessment of short- and middle-latency SSEPs provides highly accurate prediction of both good and poor neurological outcomes after CA. This approach may enhance early clinical decision-making and warrants validation in larger cohorts.