The TRACK-TBI cohort study revealed that patients with traumatic brain injury and a GCS score of 3 to 12 had a significant increase in the accuracy of prognostic accuracy of IMPACT models.
Frederick K. Korley, MD, PhD
In a recent observational cohort study, Transforming Research and Clinical Knowledge in Traumatic Brain Injury (TRACK-TBI; NCT02119182), a significant increase in the prognostic accuracy of the International Mission for Prognosis and Analysis of Clinical Trials (IMPACT) models was displayed for patients with traumatic brain injury and a Glasgow Coma Scale (GCS score) of 3 to 12.1
From the participants with brain injury (n = 1696) at baseline and 6 months, 7.1% died (n = 120), 13.9% had an unfavorable outcome (n= 235), 66.9% had an incomplete recovery (n = 1135), and 33.1% had recovered fully (n = 561).1 The corresponding area under the curve (ACU) for ubiquitin C-terminal hydrolase L1 (UCH-L1) were 0.89 (95% CI, 0.86-0.92) for predicting death, 0.86 (95% CI, 0.84-0.89) for unfavorable outcome, and 0.61 (95% CI, 0.59-0.64) for incomplete recovery at 6 months.
Frederick K. Korley, MD, PhD, associate professor of Emergency Medicine at the University of Michigan, colleagues noted that the “day-of-injury GFAP and UCH-L1 plasma concentrations have good to excellent prognostic value for predicting death and unfavorable outcome, but not for predicting incomplete recovery at 6 months. These biomarkers contribute the most prognostic information for participants presenting with a GCS score of 3 [to] 12.”
The study enrolled 2552 patients from the Transforming Research and Clinical Knowledge in TRACK-TBI observational cohort study from Feb 26, 2014, to Aug 8, 2018.TRACK-TBI included patients who were 17 years and older and were evaluated for TBI at 18 US level 1 trauma centers. The analysis included participants who had day-of-injury plasma samples for measurement of GFAP and UCH-L1 and completed 6-month assessments with the Glasgow Outcomes Scale-Extended (GOSE-TBI) because of traumatic brain injury.
Other findings showed the AUC of GFAP for predicting death at 6 months in all patients was 0.87 (95% CI, 0.83-0.91), for unfavorable outcome was 0.86 (95% CI, 0.83-0.89), and for incomplete recovery was 0.62 (95% CI, 0.59-0.64). AUCs were higher for participants with traumatic brain injury and GCS score of 3–12 than for those with GCS score of 13-15. Among participants with GCS score of 3 to 12 (n = 353), added with the GFAP and/or UCH-L1 to each of the 2 IMPACT in traumatic brain injury models significantly increased AUCs for predicting death (range, 0.90-0.94) and unfavorable outcome (range, 0.83-0.89).
The study reported 3 key findings. One of the findings was the day-of-injury GFAP and UCH-L1 concentrations predicted an unfavorable outcome and death at 6 months was good to excellent in discriminative ability. The second finding was that the day-of-injury GFAP and UCH-L1 concentrations increased the prognostic value of the IMPACT models for predicting outcomes in patients with traumatic brain injury with a GCS score of 3 to 12. Lastly, as the researchers examined the biomarkers, there were no differences between the AUC of GFAP and UCH-L1 combined in comparison with the AUC of either biomarker alone.
The study results showed patients with traumatic brain injury and a GCS score of 3 to 12, adding day-of-injury GFAP and UCH-L1 concentrations to the IMPACT models significantly increased the prognostic accuracy of these models. Korley et al noted that, “Improving the prognostic value of the IMPACT model by adding GFAP and UCH-L1 values could improve the efficiency of clinical trials, clinicians can use this tool to provide a more accurate prognosis to patients and their family members.”
For the study’s limitations, the researchers could not determine whether GFAP and UCH-L1 had prognostic value for predicting other outcomes, such as mental health and cognition. In addition, the population comprised of adult patients who might have been medically healthier than the general population of patients with traumatic brain injuries. Also, the findings might not be applicable to pediatric patients or to patients outside the level 1 trauma setting. Furthermore, at the follow-up of 6 months, several patients were lost, and the sampling times of blood draws could have influenced the prognostic results of the biomarkers.
Korley et al wrote, “Day-of-injury GFAP and UCH-L1 concentrations have good to excellent prognostic value for predicting death and unfavorable outcome at 6 months, but not for incomplete recovery,” adding that future research is needed to investigate the reproducibility of the study’s findings before conducting a broader clinical adoption.