A medical history rather than neuropsychological scores, biomarkers or imaging is most likely to help predict which patients with mild traumatic brain injury (TBI) will develop post-concussion syndrome (PCS).
Speaking at the Australasian Neuroscience Society 2018 conference in Brisbane, Professor Melinda Fitzgerald said a small pilot study in WA had found a history of concussion increased the risk of PCS almost five times (OR=4.86).
Similarly, a history of a psychological disorder also substantially increased the risk of PCS (OR=4.36) and anxiety to a lesser extent (R=1.37).
For every one-point increase on the Depression, Anxiety and Stress Scale (DASS-21), the odds of PCS increased by 37.2%, she said.
Professor Fitzgerald, who holds a joint appointment in neurotrauma with the Perron Institute and the Curtin Health Innovation Research Institute at Curtin University, said there was currently a lack of predictive measures that could be used to direct clinical care.
“We’re very interested in being able to predict outcome because it’s what is necessary in order to triage people for appropriate treatment.”
She told the limbic New Zealand data was showing that patients directed swiftly towards appropriate care (within about four days) specific to their symptoms had substantially improved outcomes.
The WA study involved 62 patients recruited from the emergency department at the Royal Perth Hospital. Almost half (47%) had been injured in a fall while assaults, motor vehicle accidents and sports injuries were also involved.
Only 36 presented for follow-up one month after their TBI and five had PCS.
The study found the Repeatable Battery for the Assessment of Neuropsychological Status (RBANS) total score and scores for immediate memory, attention and delayed memory were lower at baseline in patients who subsequently developed PCS compared to those who recovered normally.
And a trail-making task took longer in patients who developed PCS however the findings were not strong enough to be predictive.
Professor Fitzgerald said blood biomarkers, requiring a breach of the blood-brain barrier, were the ‘holy grail’ for predicting poor outcome following TBI.
The study found plasma concentrations of glial fibrillary acidic protein (GFAP) were increased in patients with TBI compared to age and sex matched controls.
However there was no difference between patients who recovered well and those with prolonged symptoms.
MRI was conducted in a subset of patients and controls but the numbers were small and there was little indication it was predictive of TBI outcome at this stage.
Professor Fitzgerald said no one measure was likely to be predictive of poor outcomes in the 180,000 concussions reported each year in Australia.
“My goal would be to set up concussion clinics so we can actually do interventions and test those interventions. We need a nationwide approach to traumatic brain injury of all severities, certainly in terms of concussion because people slip through the net.”
People who did not necessarily present to hospital but saw their doctor, a school nurse, or their sports first aid officer should also be included in a system-wide approach to improve outcomes for adults and children with TBI of any severity.
Professor Fitzgerald said symptom-based treatments for TBI included physiotherapy for vestibular disturbances or headache, especially if it was associated with muscular problems such as whiplash, and cognitive behavioural therapy for neuropsychological disturbances and particularly sleep disturbances.
“The cognitive disturbances – that fuzzy feeling – that is less easy to treat. There are some interventions to improve cognition but I think that is the probably the exciting area as a researcher with a background in cell biology – that’s where the growth is going to be. That’s what we need. We need better treatments to actually protect neurons and protect the supporting cells in the brain.”
She remained positive that larger scale studies would reveal the utility of ‘exciting new blood biomarkers in the pipeline’ or be able to harness machine learning to improve the predictive capability of MRI.