Medical Student & Research Fellow Department of Neurosurgery, Renaissance School of Medicine at Stony Brook University, Stony Brook, NY, USA East Setauket, New York, United States
Disclosure(s):
Nathaniel A. Cleri: No financial relationships to disclose
Introduction: Predicting severe traumatic brain injury (sTBI) outcomes is challenging, and existing models have limited use for individual patients. We demonstrate that a posterior dominant rhythm (PDR) observed on electroencephalography (EEG) improves the predictability of outcomes after sTBI.
Methods: This retrospective study assessed all sTBI (GCS≤8) admitted adults between 2010 to 2021. Seventy-one clinical, radiographic, and EEG variables were collected. Based on the presence of a PDR within 30 days of injury, we created PDR(+) (N=77) and PDR(-) (N=154) cohorts to assess differences in presentation and four outcomes: in-hospital survival, command following at discharge, Glasgow Outcome Scale–Extended (GOS-E) at discharge and six-months. We used AutoScore, a machine-learning automatic clinical score generator that selects and assigns weights to important predictive variables to predict outcomes.
Results: At presentation, the PDR(-) cohort had a lower mean initial GCS (4.07 v. 4.70, P = 0.007) and GCS-motor (1.99 v. 2.60, P = 0.005). Comparing outcomes, the PDR(+) cohort had superior in-hospital survival (89.4% v. 64.9%, P < 0.001), command following at discharge (82.4% v. 54.5%, P < 0.001), mean discharge GOS-E (3.29 v. 2.40, P < 0.001) and six-month GOS-E (4.43 v. 3.25, P = 0.003). After demonstrating the PDR(+) cohort’s superior outcomes, AutoScore was used to identify seven variables predictive of all four outcomes—blood glucose, PDR, age, bilateral pupil reactivity, body mass index, systolic blood pressure, and GCS-motor subscore. Our model had superb discrimination for predicting in-hospital survival (AUC = 0.818), six-month GOS-E (AUC = 0.837), discharge GOS-E (AUC = 0.799), and command following at discharge (AUC = 0.788).
Conclusion : A PDR on EEG after sTBI signals favorable outcomes. Our prognostic model has strong accuracy in predicting these outcomes.
.jpg "Nathaniel A. Cleri (he/him/his) photo")