A study on data from traumatic brain injury patients from the the TARN database examined the prognostic value of various scoring and classification systems and pathologies.
Contusion and haemorrhage appeared to be less signiﬁcant predictors of outcome than the presence of brain swelling in this British dataset.
The brainstem was the most significant location of cerebral injury.
Prognostic value of various intracranial pathologies in traumatic brain injury
European Journal of Trauma and Emergency Surgery February 2012, Volume 38, Issue 1, pp 25-32
Objective Various intracranial pathologies in traumatic brain injury (TBI) can help to predict patient outcomes.
These pathologies can be categorised using the Marshall Classiﬁcation or the Abbreviated Injury Scale (AIS) dictionary or can be described through traditional descriptive terms such as subarachnoid haemorrhage (SAH), subdural haemorrhage (SDH), epidural haemorrhage (EDH) etc. The purpose of this study is to assess the prognostic value of AIS scores, the Marshall Classiﬁcation and various intracranial pathologies in TBI.
Methods A dataset of 802 TBI patients in the Trauma Audit and Research Network (TARN) database was analysed using logistic regression. First, a baseline model was constructed with age, Glasgow Coma Scale (GCS), pupillary reactivity, cause of injury and presence/absence of extracranial injury as predictors and survival at discharge as the outcome. Subsequently, AIS score, the Marshall Classiﬁcation and various intracranial pathologies such as haemorrhage, SAH or brain swelling were added in order to assess the relative predictive strength of each variable and also to assess the improvement in the performance of the model.
Results Various AIS scores or Marshal classes did not appear to signiﬁcantly affect the outcome. Among traditional descriptive terms, only brain stem injury and brain swelling signiﬁcantly inﬂuenced outcome [odds ratios for survival: 0.17 (95% conﬁdence interval [CI]; 0.08–0.40) and 0.48 (95% CI; 0.29–0.80), respectively].
Neither haemorrhage nor its subtypes, such as SAH, SDH and EDH, were signiﬁcantly associated with outcome. Adding AIS scores, the Marshall Classiﬁcation and various
intracranial pathologies to the prognostic models resulted in an almost equal increase in the predictive performance of the baseline model.
Conclusions In this relatively recent dataset, each of the brain injury classification systems enhanced equally the performance of an early mortality prediction model in traumatic brain injury patients. The significant effect of brain swelling and brain stem injury on the outcome in comparison to injuries such as SAH suggests the need to improve therapeutic approaches to patients who have sustained these injuries.