Wiring Recovery: A Systematic Review of Connectomics in Prognostication of Severe Traumatic Brain Injury

Introduction: Traumatic brain injury (TBI) is a significant public health concern, leading to substantial morbidity and mortality globally. Prognostication in severe TBI is a complex process that guides clinical decision-making and family counseling. Recent advancements in connectomics, the study of neural connections within the brain, have provided new insights into the brain's ability to reorganize and recover after severe injuries. This systematic review aims to synthesize current evidence on the role of connectomics in the prognostication of outcomes following severe TBI.

Methods: Following PRISMA guidelines, we systematically searched PubMed, Ovid Embase, Web of Science, and the Cochrane databases from 2000 to May 2021, focusing on studies that employed connectomic methods to prognosticate outcomes in severe traumatic brain injury. Inclusion and exclusion criteria were stringently applied, and selected studies underwent rigorous data extraction and quality assessment, with findings synthesized according to the type of connectomic analysis utilized.

Results: Twenty studies met our inclusion criteria, encompassing patient cohorts ranging from 13 to 118 individuals. Resting-state functional MRI (rsfMRI) was employed in 12 studies, while diffusion tensor imaging (DTI) was utilized in 16, with 7 specifically investigating patients with diffuse axonal injury. The average interval from injury to neuroimaging spanned 18 days, ranging from one to six weeks. The Glasgow Outcome Scale Extended, Coma Recovery Scale-Revised, and assessments of specific neurological functions were the most commonly reported outcome measures. Fractional Anisotropy emerged as the predominant scalar metric of interest, and the default mode network was the most frequently studied brain network. Reported predictive models using statistical methods including machine-learning demonstrated a wide accuracy range of 0.46 to 0.97.

Conclusion : This systematic review demonstrates the potential of connectomics, using non-invasive neuroimaging, for predicting outcomes in severe TBI cases. However, the wide range of predictive accuracies and analyzed regions underscores the necessity for standardized connectomic research methodologies. Further research with standardized protocols is required to consolidate the role of connectomics in TBI prognosis.