Predictive Value of CT Perfusion Parameters in Traumatic Brain Injury Patients: A Longitudinal Study

Abstract

Introduction: Traumatic Brain Injury (TBI) is the most common cause of mortality and morbidity worldwide. Moreover, cerebral contusions, Subdural Haemorrhages (SDH), Extradural Haemorrhages (EDH), and Subarachnoid Haemorrhage (SAH), which are secondary injuries that occur due to alterations in cerebral haemodynamics, remain unevaluated and unchecked by plain non contrast scans. Timely delineation of the cerebral perfusion status of the patient can assist in modifying treatment plans and could lead to a better recovery of the patient. Aim: To assess the cerebral haemodynamics of patients with TBI at admission and evaluate the predictive value of Computed Tomography (CT) perfusion. Materials and Methods: The present prospective longitudinal study was conducted in the Department of Radiodiagnosis and Neurosurgery, Jawaharlal Nehru Medical College and Hospital (JNMCH), Aligarh Muslim University (AMU), Aligarh, Uttar Pradesh, India, from September 2019 to September 2020. A total of 40 participants were included in the study using a purposive sampling method. CT Perfusion (CTP) was performed within 72 hours of head injury, followed by Glasgow Outcome at Discharge Scale (GODS) assessments at discharge and one month later. Cerebral perfusion parameters such as Cerebral Blood Flow (CBF), Cerebral Blood Volume (CBV), Mean Transit Time (MTT), and Time To Peak (TTP) were measured, and correlations were established using Spearman’s rank correlation coefficients to determine the relationship of GODS with CTP parameters. Results: The mean age of study subjects was 35.8±13.4 years, with a median (25th-75th percentile) of 32 (25-42.75). Among a total of 40 patients, 22 (55%) were males and 18 (45%) were females. Significant positive correlations were observed between GODS at discharge and CBF (mL/100 gm/min) at presentation, CBV (mL/100 gm/min) at presentation, and the number of normal/hyperemic territories, with correlation coefficients of 0.398, 0.329, and 0.504, respectively. A significant negative correlation was found between GODS at discharge and TTP (seconds) at presentation, with a correlation coefficient of -0.392. Similarly, significant positive correlations were observed between GODS at one month with CBF (mL/100 gm/min) at presentation and the number of normal/ hyperemic territories, with correlation coefficients of 0.407 and 0.827, respectively. A significant negative correlation was seen between GODS at one month and MTT (seconds) at presentation, with a correlation coefficient of -0.344. CBV (mL/100 gm/min) at presentation exhibited a sensitivity of 100%, followed by MTT (seconds) at presentation (71.43%), and CBF (mL/100 gm/min) at presentation (57.14%). Conversely, CBF (mL/100 gm/min) at presentation showed a specificity of 92.31%, followed by TTP (seconds) at presentation (92.31%), and CBV (mL/100 gm/min) at presentation (53.85%). In predicting unfavourable outcomes, MTT (seconds) at presentation had the lowest specificity of 7.69%. Conclusion: Perfusion CT provides additional information compared to Non Contrast Computed Tomography (NCCT) with insights into the pericontusional areas. Therefore, it can be a potential tool in the evaluation of TBI. Perfusion parameters have prognostic value, and thus, this modality should be studied with a larger sample size for more precise results.