Transforming Research and Clinical Knowledge in Traumatic Brain Injury
TRACK-TBI is a prospective study of patients with acute mTBI (Glasgow Coma Scale=13-15) who were enrolled ≤24 hours of injury at an emergency department of level 1 trauma centres and followed for 12 months
The effects of traumatic brain injury (TBI) are difficult to measure in longitudinal cohort studies, as disparate preinjury characteristics and injury mechanisms produce variable impairment profiles and recovery trajectories. In preparation for the Transforming Research and Clinical Knowledge in TBI (TRACK-TBI) study, which followed patients with injuries ranging from uncomplicated mild TBI to coma, we designed a multidimensional Flexible outcome Assessment Battery (FAB). The FAB relies on a decision-making algorithm that assigns participants to a Comprehensive [CAB] or Abbreviated Assessment Battery [AAB] and guides test selection across all phases of recovery. To assess feasibility of the FAB, we calculated the proportion of participants followed at 2 weeks (2w) and at 3, 6 and 12 months (3m, 6m, 12m) post-injury who completed the FAB and received valid scores. We evaluated utility of the FAB by examining differences in 6m and 12m Glasgow Outcome Scale-Extended (GOSE) scores between participant subgroups derived from the FAB-enabled versus traditional approach to outcome assessment applied at 2w. Among participants followed at 2w (N= 2,094), 3m (N=1,871), 6m (N=1,736), and 12m (N=1,607) post-injury, 95-99% received valid completion scores on the FAB, in full or in part, either in-person or by telephone. Level of function assessed by the FAB-enabled approach at 2w was associated with 6m and 12m GOSE scores (proportional odds p<0.001). These findings suggest that the participant classification methodology afforded by the FAB may enable more effective data collection to improve detection of natural history changes and TBI treatment effects 1).
The Transforming Research and Clinical Knowledge in Traumatic Brain Injury Pilot study (TRACK-TBI Pilot) enrolled patients at 3 US Level I trauma centers within 24 hours of TBI.
Patients were grouped by outcome measure concordance (good-recovery/good-satisfaction, impaired-recovery/impaired-satisfaction) and discordance (good-recovery/impaired-satisfaction, impaired-recovery/good-satisfaction). Logistic regression was utilized to determine predictors of discordance.
Functional outcome: Glasgow Outcome Scale-Extended (GOSE); SWL: Satisfaction with Life Scale (SWLS).
Of the 586 enrolled subjects, 298 had completed both outcome measures at 6-month follow-up; the correlation between GOSE and SWLS was 0.380. Patients with impaired-recovery (GOSE < 7)/impaired-satisfaction (SWLS < 20) were more likely to have mild TBI (83% vs 62%, P = .012), baseline depression (42% vs 15%, P < .0001), and 6-month depression (59% vs 21%, P < .0001) when compared with patients with impaired-recovery/good-satisfaction. Patients with good-recovery/impaired-satisfaction were more likely to have baseline depression (31% vs 13%, P < .0001) and 6-month depression (33% vs 6%, P < .0001) compared with good-recovery/good-satisfaction.
Correlation between SWL and functional outcome was not strong, and depression may modulate the association. Future research should account for functional, mental health, and patient-centered outcomes when assessing TBI recovery 2).
Mild traumatic brain injury patients (Glasgow Coma Scale score 13-15) from the prospective multicenter Transforming Research and Clinical Knowledge in Traumatic Brain Injury (TRACK-TBI) Pilot study (2010-2012) who completed the Rivermead Post Concussion Symptoms Questionnaire (RPQ) at six-months post-injury were included. Linear regression was utilized to determine the predictive value of candidate predictors for cognitive, somatic, and emotional subscales individually as well as the overall RPQ. The final prediction model was developed using Lasso shrinkage and bootstrap validation.
Cnossen et al., included 277 mTBI patients (70% male, median age 42y). No major differences in the predictive value of our set of predictors existed for the cognitive, somatic, and emotional subscales, and therefore one prediction model for the RPQ total scale was developed. Years of education, pre-injury psychiatric disorders and prior TBI were the strongest predictors of six-month post-concussive symptoms. The total set of predictors explained 21% of the variance, which decreased to 14% after bootstrap validation. Demographic and clinical variables at baseline are predictive of six-month post-concussive symptoms following mTBI, however these variables explain less than one-fifth of the total variance in outcome. Model refinement with larger datasets, more granular variables, and objective biomarkers are needed before implementation in clinical practice 3).