PASTRAMI : Patient-specific Statistics for Microstructure-augmented Connectomics

Overview

Traumatic brain injury (TBI) represent a major public health issue, accounting for approximately 1.5 million hospital admissions annually across the European Union and 160,000 cases in France. They are a significant cause of mortality and disability, with serious consequences for patients and their families. Predicting functional outcomes remains complex due to the heterogeneity of brain injuries, whether primary or secondary, as well as the mechanisms involved, particularly neuroinflammation. This difficulty is particularly pronounced for moderate to severe TBI, for which current predictive tools are still limited. The PASTRAMI project proposes to utilise diffusion magnetic resonance imaging (MRI) to identify biomarkers associated with axonal lesions in the white matter. This technique enables the analysis of brain microstructure and the reconstruction of neural connections (connectome), thereby offering innovative prospects for improving the prediction of functional recovery. This is a prospective, multicentre, interventional study involving 100 patients aged between 18 and 60 years with moderate to severe traumatic brain injury. The primary objective is to assess the prognostic value of brain microstructural measures at baseline, by correlating them with the GOSE functional score at 1 year. Secondary objectives include the analysis of mortality, functional outcome, quality of life, and early physiological factors influencing progression. Patients will undergo a standard MRI scan supplemented with additional sequences for prognostic purposes, without any change to their therapeutic management. The associated risks are minimal and relate primarily to the MRI scan. No immediate individual benefit is expected, but the results could improve our understanding of brain lesions and optimise future management. The total duration of the study is 48 months. The expected outcomes include better characterisation of lesions and improved predictive tools, contributing to more appropriate clinical decisions.