Predicting outcome following mild traumatic brain injury: protocol for the longitudinal, prospective, observational Concussion Recovery (CREST) cohort study

Symptoms Associated With Exercise Intolerance and Resting Heart Rate Following Mild Traumatic Brain Injury

OBJECTIVES

People may experience a myriad of symptoms after mild traumatic brain injury (mTBI), but the relationship between symptoms and objective assessments is poorly characterized. This study sought to investigate the association between symptoms, resting heart rate (HR), and exercise tolerance in individuals following mTBI, with a secondary aim to examine the relationship between symptom-based clinical profiles and recovery.

METHODS

Prospective observational study of adults aged 18 to 65 years who had sustained mTBI within the previous 7 days. Symptoms were assessed using the Post-Concussion Symptom Scale, HR was measured at rest, and exercise tolerance was assessed using the Buffalo Concussion Bike Test. Symptom burden and symptom-based clinical profiles were examined with respect to exercise tolerance and resting HR.

RESULTS

Data from 32 participants were assessed (mean age 36.5 ± 12.6 years, 41% female, 5.7 ± 1.1 days since injury). Symptom burden (number of symptoms and symptom severity) was significantly associated with exercise intolerance (P = .002 and P = .025, respectively). Physiological and vestibular-ocular clinical profile composite groups were associated with exercise tolerance (P = .001 and P = .014, respectively), with individuals who were exercise intolerant having a higher mean number of symptoms in each profile than those who were exercise tolerant. Mood-related and autonomic clinical profiles were associated with a higher resting HR (>80 bpm) (P = .048 and P = .028, respectively), suggesting altered autonomic response for participants with symptoms relating to this profile. After adjusting for age and mechanism of injury (sports- or non-sports-related), having a higher mood-related clinical profile was associated with persisting symptoms at 3 months postinjury (adjusted odds ratio = 2.08; 95% CI, 1.11-3.90; P = .013).

CONCLUSION

Symptom-based clinical profiles, in conjunction with objective measures such as resting HR and exercise tolerance, are important components of clinical care for those having sustained mTBI. These results provide preliminary support for the concept that specific symptoms are indicative of autonomic dysfunction following mTBI.

Prediction of cognitive outcome after mild traumatic brain injury from acute measures of communication within brain networks

A considerable but ill-defined proportion of patients with mild traumatic brain injury (mTBI) experience persistent cognitive sequelae; the ability to identify such individuals early can help their neurorehabilitation. Here we tested the hypothesis that acute measures of efficient communication within brain networks are associated with patients' risk for unfavorable cognitive outcome six months after mTBI. Diffusion and T1-weighted magnetic resonance imaging, alongside cognitive measures, were obtained to map connectomes both one week and six months post injury in 113 adult patients with mTBI (71 males). For task-related brain networks, communication measures (characteristic path length, global efficiency, navigation efficiency) were moderately correlated with changes in cognition. Taking into account the covariance of age and sex, more unfavorable communication within networks were associated with worse outcomes within cognitive domains frequently impacted by mTBI (episodic and working memory, verbal fluency, inductive reasoning, and processing speed). Individuals with more unfavorable outcomes had significantly longer and less efficient pathways within networks supporting verbal fluency (all t > 2.786, p < .006), highlighting the vulnerability of language to mTBI. Participants in whom a task-related network was relatively inefficient one week post injury were up to eight times more likely to have unfavorable cognitive outcome pertaining to that task. Our findings suggest that communication measures within task-related networks identify mTBI patients who are unlikely to develop persistent cognitive deficits after mTBI. Our approach and findings can help to stratify mTBI patients according to their expected need for follow-up and/or neurorehabilitation.

Lower and higher volumes of physical exercise build up brain reserves against memory deficits triggered by a head injury in mice

Decreasing neurotrophic support and impaired mitochondrial bioenergetics are key mechanisms for long-term neurodegeneration and cognitive decline after traumatic brain injury (TBI). We hypothesize that preconditioning with lower and higher volumes of physical exercise upregulates the CREB-BDNF axis and bioenergetic capability, which might serve as neural reserves against cognitive impairment after severe TBI. Using a running wheel mounted in the home cage, mice were engaged in lower (LV, 48 h free access, and 48 h locked) and higher (HV, daily free access) exercise volumes for thirty days. Subsequently, LV and HV mice remained for additional thirty days in the home cage with the running wheel locked and were euthanized. The sedentary group had the running wheel always locked. For the same type of exercise stimulus in a given time, daily workout presents higher volume than alternate days workout. The total distance ran in the wheel was the reference parameter to confirm distinct exercise volumes. On average, LV exercise ran 27.522 m and HV exercise ran 52.076 m. Primarily, we investigate whether LV and HV protocols increase neurotrophic and bioenergetic support in the hippocampus thirty days after exercise ceased. Regardless of volume, exercise increased hippocampal pCREBSer133-CREB-proBDNF-BDNF signaling and mitochondrial coupling efficiency, excess capacity, and leak control, that may compose the neurobiological basis for neural reserves. Further, we challenge these neural reserves against secondary memory deficits triggered by a severe TBI. After thirty days of exercise LV and HV, and sedentary (SED) mice were submitted to the CCI model. Mice remained for additional thirty days in the home cage with the running wheel locked. The mortality after severe TBI was approximately 20% in LV and HV, while in the SED was 40%. Also, LV and HV exercise sustained hippocampal pCREBSer133-CREB-proBDNF-BDNF signaling, mitochondrial coupling efficiency, excess capacity, and leak control for thirty days after severe TBI. Corroborating these benefits, the mitochondrial H2O2 production linked to complexes I and II was attenuated by exercise regardless of the volume. These adaptations attenuated spatial learning and memory deficits caused by TBI. In summary, preconditioning with LV and HV exercise builds up long-lasting CREB-BDNF and bioenergetic neural reserves that preserve memory fitness after severe TBI.

Development and validation of a prehospital-stage prediction tool for traumatic brain injury: a multicentre retrospective cohort study in Korea

OBJECTIVES

Predicting diagnosis and prognosis of traumatic brain injury (TBI) at the prehospital stage is challenging; however, using comprehensive prehospital information and machine learning may improve the performance of the predictive model. We developed and tested predictive models for TBI that use machine learning algorithms using information that can be obtained in the prehospital stage.

DESIGN

This was a multicentre retrospective study.

SETTING AND PARTICIPANTS

This study was conducted at three tertiary academic emergency departments (EDs) located in an urban area of South Korea. The data from adult patients with severe trauma who were assessed by emergency medical service providers and transported to three participating hospitals between 2014 to 2018 were analysed.

RESULTS

We developed and tested five machine learning algorithms-logistic regression analyses, extreme gradient boosting, support vector machine, random forest and elastic net (EN)-to predict TBI, TBI with intracranial haemorrhage or injury (TBI-I), TBI with ED or admission result of admission or transferred (TBI with non-discharge (TBI-ND)) and TBI with ED or admission result of death (TBI-D). A total of 1169 patients were included in the final analysis, and the proportions of TBI, TBI-I, TBI-ND and TBI-D were 24.0%, 21.5%, 21.3% and 3.7%, respectively. The EN model yielded an area under receiver-operator curve of 0.799 for TBI, 0.844 for TBI-I, 0.811 for TBI-ND and 0.871 for TBI-D. The EN model also yielded the highest specificity and significant reclassification improvement. Variables related to loss of consciousness, Glasgow Coma Scale and light reflex were the three most important variables to predict all outcomes.

CONCLUSION

Our results inform the diagnosis and prognosis of TBI. Machine learning models resulted in significant performance improvement over that with logistic regression analyses, and the best performing model was EN.