Brachial plexus injuries (BPI) cause permanent upper limb paralysis and serious disability in adults and children. Timely identification of the severity of nerve injury and early appreciation of the inane potential for recovery would revolutionise management. Radiology supports clinical assessment but is not an independent marker of disease severity. Surgical evaluation in severe closed nerve injuries defines the reconstructive potential. This thesis explores aspects of BPI assessment and treatment that remain unsolved. Conventional magnetic resonance imaging (MRI) and novel diffusion tensor imaging (DTI) are evaluated in adults to gain a better understanding of their current diagnostic accuracy in BPI and future value in assessing nerve health. In neonates, this thesis evaluates the preganglionic effects related to timing of injury and repair and socioeconomic factors that influence the incidence and presentation of neonates to specialist centres. These currently controversial factors are important prerequisites to designing and evaluating the optimal objective imaging modality in this age group. Data from 29 high energy trauma BPI patients were analysed. The diagnostic accuracy of 1.5T MRI for BP root avulsion was 79% (Index test MRI, Reference standard Surgery). The negative predictive value was 81% meaning there was one occult avulsion in every 5 cases. DTI data sets from 12 patients with unilateral BPI and seven matched adult controls were acquired. The test was considered positive for root avulsion when there was a visible lack of continuity between tracts in the spinal cord and the brachial plexus. The mean fractional anisotropy (FA) and mean diffusivity (MD) were calculated from a region of interest (ROI) - five 2.5mm2 pixels in the axial plane covering the lateral recess of the vertebral foramen. The overall diagnostic accuracy of DTI for determining root avulsion was 71% (95%CI 54, 85). The fractional anisotropy (FA) of avulsed roots was 10% lower than normal roots (95% [CI 7%,13%] p<0.001). The mean diffusivity (MD) of avulsed roots was 0.32x10-3mm2/s higher than normal intact roots (95%CI 0.11, 0.53; p><0.001). The T1 tracts were not clearly visualised in most BPI cases. The time course comparing survival of motoneurons in a neonatal rat BPI model, was evaluated at 2- 28 days after injury and repair. At day 10, the injury group survival of motoneurons was 22% and at 28 days only 9%of motoneurons remained. In the repair group the surviving neurons were 51% at 10 days and 20% at 28 days. The repair group had significantly reduced reactivity of macrophages and microglial cells in the C5/C6 ventral horn. In analysis (Index of Multiple Deprivation, IMD) of a 13 year, retrospective cohort series of 321 children with Obstetric Brachial Plexus injury (England), 109 (39%) were from the most deprived quintile. In Yorkshire and Humber 44% were from the most deprived quintile. No relationship was identified between severity of condition and IMD. These laboratory and clinical studies in adults, children and neonatal animals align with the real-time clinical conundrum in evaluating the injured nerve’s ability to recover to functional significance. ><0.001). The mean diffusivity (MD) of avulsed roots was 0.32x10-3mm2/s higher than normal intact roots (95%CI 0.11, 0.53;p<0.001). The T1 tracts were not clearly visualised in most BPI cases. The time course comparing survival ofmotoneurons in a neonatal rat BPI model, was evaluated at 2- 28 days after injury and repair. At day 10, the injury group survival of motoneurons was 22% and at 28 days only 9%of motoneurons remained. In the repair group the surviving neurons were 51% at 10 days and 20% at 28 days. The repair group had significantly reduced reactivity of macrophages and microglial cells in the C5/C6 ventral horn. In analysis (Index of Multiple Deprivation, IMD) of a 13 year, retrospective cohort series of 321 children with Obstetric Brachial Plexus injury (England), 109 (39%) were from the most deprived quintile. In Yorkshire and Humber 44% were from the most deprived quintile. No relationship was identified between severity of condition and IMD. These laboratory and clinical studies in adults, children and neonatal animals align with the real-time clinical conundrum in evaluating the injured nerve’s ability to recover to functional significance.