A Nomogram Based on Quantitative EEG to Predict the Prognosis of Nontraumatic Coma Patients in the Neuro-intensive Care Unit

Abstract

ObjectiveWe aimed to establish a quantitative electroencephalography-based prognostic prediction model specifically tailored for nontraumatic coma patients to guide clinical work. MethodsThis retrospective study included 126 patients with nontraumatic coma admitted to the First Affiliated Hospital of Chongqing Medical University from December 2020 to December 2022. Six in-hospital deaths were excluded. The Glasgow Outcome Scale assessed the prognosis at 3 months after discharge. The least absolute shrinkage and selection operator regression analysis and stepwise regression method were applied to select the most relevant predictors. We developed a predictive model using binary logistic regression and then presented it as a nomogram. We assessed the predictive effectiveness and clinical utility of the model. ResultsAfter excluding six deaths that occurred within the hospital, a total of 120 patients were included in this study. Three predictor variables were identified, including APACHE II score [39.129 (1.4244-1074.9000)], sleep cycle [OR: 0.006 (0.0002-0.1808)], and RAV [0.068 (0.0049-0.9500)]. The prognostic prediction model showed exceptional discriminative ability, with an AUC of 0.939 (95% CI: 0.899-0.979). ConclusionA lack of sleep cycles, smaller relative alpha variants, and higher APACHE II scores were associated with a poor prognosis of nontraumatic coma patients in the neurointensive care unit at 3 months after discharge. Clinical implicationThis study presents a novel methodology for the prognostic assessment of nontraumatic coma patients and is anticipated to play a significant role in clinical practice.