Use of In-Hospital Mild Traumatic Brain Injury Symptom Checklist Within 24 Hours of Injury to Predict 3-Month Symptom
Samuel Gray1, Grace Amadon1, Nancy Temkin2
1Department of Neurosurgery, Medical College of Wisconsin, Milwaukee, Wisconsin.
Background:
Experts recommend administering a traumatic brain injury (TBI) symptom checklist as part of routine evaluation for TBI with Glasgow Coma Scale score 13-15 ("mild" TBI [mTBI]) in adult emergency departments (EDs). However, such assessment is not routine, partly due to limited guidance on interpreting symptom scores.
Objectives:
Assess the utility of Rivermead Post Concussion Symptoms Questionnaire (RPQ) scores, assessed in-hospital within 24 h of injury (day 1), in discriminating individuals with, vs. without, persistent TBI-related symptoms (persistent post-concussive symptoms [PPCS]) at 3 months post-injury, and provide interpretive guidance.
Methods:
Adults with mTBI across three Level I trauma centers completed the RPQ at day 1 and 3 months post-injury. Using binary logistic regression models and fivefold internal cross-validation, we calculated the mean area under the curve (AUC) for day 1 RPQ total score in predicting 3-month PPCS. Clinical interpretation tables were provided.
Results:
Two hundred fifty-two participants who completed a day 1 RPQ were included in the analysis. Inverse probability weighting was used to adjust for bias in attrition (n = 168 followed). The mean cross-validated AUC was 0.84 using day 1 RPQ score alone. Multivariable models, including those using previously validated sets of variables, did not outperform day 1 RPQ alone.
Conclusions:
In adults presenting to Level I trauma centers for acute mTBI, symptom burden (RPQ total score) is robustly associated with 3-month symptom outcome. The RPQ, which can be completed in about 3 min, may support recognition of mTBI symptoms in the ED and risk stratification for triage into appropriate follow-up pathways.
