Abstract
It is a pleasure to present a terrific line up of articles in this issue of Journal of Neurotrauma. I would like to draw special attention to the article by Dr. Le Sage and colleagues entitled “PoCS Rule: Derivation and Validation of a Clinical Decision Rule for Early Prediction of Persistent Symptoms after a Mild Traumatic Brain Injury.” In their article, the authors, representing a consortium of seven Canadian Emergency Department researchers, present their derivation and independent validation of a prognostic clinical decision rule to predict which “mild” TBI/concussion patients are likely to continue to have persistent symptoms 90 days after injury. I would like to point out several remarkable features of the article: The decision rule was developed using a relatively large sample: 471 patients. The decision rule was independently validated in an additional 612 patients. To my knowledge, this is the first early prediction of post-concussive symptoms rule that has been independently validated in a well-powered cohort. Follow-up rates were impressive: Of a total of 1339 patients enrolled, data from 1083 (81%) were included in the analyses. Missing data rates were relatively modest, and state-of-the-art approaches to imputation of missing data were used. Patients were consulted in the design phase and participated in the definition of the primary outcomes of interest.
To summarize several of the key takeaway lessons from this article:
Approximately 16% of patients had persistent post-concussive symptoms at 90 days based on structured interviews conducted by trained staff. This rate is somewhat lower than in other studies, likely due to the relatively strict definition of persistent post-concussive symptoms: “The presence of spontaneously described, severe, persistent symptoms that have an impact on patient's life.”
The predictors of persistent post-concussive symptoms on presentation in the Emergency Department included age (highest risk for 35-44 year olds), female sex, prior multiple TBIs, prior TBI within the past year, history of mental health disorder, headache in the ED, cervical sprain, and hemorrhage on CT.
For those in an “intermediate” risk category based on presentation in the Emergency Department, the authors found that 7 day follow-up characteristics predicting persistent post-concussive symptoms at 90 days included headaches, sleep disturbance, fatigue, and a total score on the Rivermead Post-Concussion Questionnaire of 21 or higher.
Taken together, the decision rule based on initial presentation and 7 day follow-up had approximately 90% sensitivity, approximately 50% specificity, and over 95% negative predictive value. The negative predictive value is especially important for triaging patients who are unlikely to require extensive care following concussion.
Blood levels of the biomarkers S100ß, NSE, GFAP, and C-Tau measured by ELISA were not associated with persistent post-concussive symptoms and not retained in any of the final clinical decision rules. It remains to be determined whether other blood biomarkers or additional objective assessment approaches would add prognostically useful information when combined with this clinical decision rule.
Alcohol and drug consumption, mechanism of injury, multiple body injuries, level of education, loss of consciousness, and post-traumatic amnesia were not associated with persistent post-concussive symptoms. These are surprising findings, but the results seem solid.
Litigation was not considered and will need to be addressed in future studies.
Why do decision rules like this one matter? As the authors point out, a decision rule could be used to make appropriate referrals for concussion care and plan post-discharge resources. This is a critically important goal. However, this study is probably not the last word: The positive predictive value of the decision rule was still relatively modest (∼30%) and there is clearly room for improvement. Perhaps additional/alternative blood biomarkers, quantitative assessments of eye movements, quantitative pupillometry, EEG, advanced imaging, or other point-of-care modalities will take us even further.
From my perspective, the most exciting direction for future research will involve the use of this or other future decision rules to stratify concussion patients for clinical trials of candidate therapeutics. It is easy to imagine trials of treatments aimed at alleviating cervical strain, post-traumatic headaches, post-traumatic sleep disturbance, fatigue, or a combination of these using currently available treatments and emerging/experimental modalities. Decision rules such as the one developed and validated by Le Sage and colleagues will make it easier to justify and plan such trials. Together, we are moving closer to a mature scientific foundation for our field in which well-validated metrics and outcomes serve as anchors for evidence-based care.