Assessing the Severity of Traumatic Brain Injury—Time for a Change?

Abstract

Traumatic brain injury (TBI) has been described to be man’s most complex disease, in man’s most complex organ. Despite this vast complexity, variability, and individuality, we still classify the severity of TBI based on non-specific, often unreliable, and pathophysiologically poorly understood measures. Current classifications are primarily based on clinical evaluations, which are non-specific and poorly predictive of long-term disability. Brain imaging results have also been used, yet there are multiple ways of doing brain imaging, at different timepoints in this very dynamic injury. Severity itself is a vague concept. All prediction models based on combining variables that can be assessed during the acute phase have reached only modest predictive values for later outcome. Yet, these early labels of severity often determine how the patient is treated by the healthcare system at large. This opinion paper examines the problems and provides caveats regarding the use of current severity labels and the many practical and scientific issues that arise from doing so. The objective of this paper is to show the causes and consequences of current practice and propose a new approach based on risk classification. A new approach based on multimodal quantifiable data (including imaging and biomarkers) and risk-labels would be of benefit both for the patients and for TBI clinical research and should be a priority for international efforts in the field.

In the Aftermath of Acute Hospitalization for Traumatic Brain Injury: Factors Associated with the Direct Pathway into Specialized Rehabilitation

“…Likewise, patients with moderate or mild TBI in the acute phase may experience long-term disability. A recent proposal in assessing the severity of TBI suggests changing from severity labels to risk assessment over time [35]. Reduced functional levels at six months are reported for patients with TBI admitted to the hospital [28].…”

Evaluation of Acute Glial Fibrillary Acidic Protein and Ubiquitin C-Terminal Hydrolase-L1 Plasma Levels in Traumatic Brain Injury Patients with and without Intracranial Lesions

“…Thus, the percentage of moderate/severe TBI patients was also higher in this subcohort compared to the total cohort, where only 7% of patients were moderate/severe. Although use of GCS for severity classification has been repeatedly called into question in recent years, 32 , 33 the overlap between severity and CT findings produced a level of ambiguity in the interpretation of results. Nonetheless, we believe that these pilot data showing an increase of GFAP and UCH-L1 in the first 2 h from moderate/severe TBI contribute novel evidence and warrant further investigation of these biomarkers' kinetics and diagnostic performance in the first hours after brain injury.…”

Personalized Prediction of Postconcussive Working Memory Decline: A Feasibility Study

“…Previous studies that have aimed at predicting long-term postconcussive cognitive outcomes for mTBI have generally adopted a multivariate approach encompassing patient demographics, clinical symptoms, and neuropsychological features, as well as other factors such as health care utilization and premorbid psychiatric conditions [20,21]. However, the performance of such predictive methods can be limited due to clinical variability and complexity, as well as confounding factors such as ambiguous documentation, undeclared medication use, and other concurrent medical conditions, and assessment of morphologic information based on structural brain imaging has not demonstrated additional benefits [22,23]. In this observational study, we prospectively recruited 70 patients with mTBI and followed up their cognitive functioning with functional and neuropsychological data for 1 year.…”