Abstract Background Somatosensory evoked potentials (SSEP) are part of the multimodal approach recommended by international guidelines for neurological prognostication after cardiac arrest (CA). Objective This study aimed to evaluate the prognostic ability of SSEP for neurological outcomes after CA. Method Using a multicenter database of out‐of‐hospital CA patients in Japan, we included patients ≥ 18 years of age who underwent SSEP. The main outcome measure was poor outcome (cerebral performance category scores 3, 4, and 5) at 1 year. We compared patients' demographics, treatments, and outcomes between the SSEP N20‐negative and ‐positive groups. We also performed receiver operating characteristic analysis and calculated the area under the curve (AUC). Results Among 16,452 patients registered in the database, 66 patients were enrolled, and 40 patients lacked N20. Most variables did not differ between the two groups. Temperature control therapy was performed for 51 patients (77.2%). Of all patients without N20, the rates of poor outcomes at 7 days, 1 month, 3 months, and 1 year increased, with rates of brain death or death rising over time. In the N20 group, about 30% showed favorable outcomes. The prognostic abilities were: AUC 0.886 (95% confidence interval, 0.785–0.988), sensitivity 0.73 (95% CI, 0.60–0.83), specificity 1.00 (95% CI, 0.69–1.00), positive predictive value 1.00 (95% CI, 0.91–1.00), negative predictive value 0.40 (95% CI, 0.23–0.60), false positive rate 0.00 (95% CI, 0.00–0.31), and false negative rate 0.27 (95% CI, 0.17–0.40). Conclusion The absence of bilateral N20 in SSEP was strongly associated with poor neurological outcomes and high mortality at one year after CA; however, approximately 30% of those in the group with N20‐positive achieved favorable outcomes.
Tanaka et al. (Thu,) studied this question.