Importance: While research has focused on veterans after deployment, fewer studies have examined longitudinal outcomes associated with neurodegeneration. Questions remain about how these trajectories may connect and inform overall risk for later-life neurodegeneration and possible dementia.
Objective: To analyze acute, 1-year, 5-year, and 10-year longitudinal outcomes and magnetic resonance imaging (MRI) data from US military personnel to understand profiles of neurodegenerative resilience and risk.
Design, setting, and participants: This prospective longitudinal, cohort study included data from 4 combat-deployed groups: 2 control groups without head injury (36 individuals with blast exposure and 94 individuals without) and 2 groups with combat-related mild traumatic brain injury (TBI) from blast (137 individuals) or blunt trauma (21 individuals). Enrollment occurred from 2008 to 2013 in Afghanistan or following medical evacuation to Landstuhl, Germany, with 1-year, 5-year, and 10-year follow-ups completed in the US. Data were analyzed from October 2024 to March 2025.
Exposures: All participants experienced combat deployment and some participants experienced mild TBI.
Main outcomes and measures: Clinical outcomes included 34 measures of neurobehavioral, psychological, and cognitive function completed at 1-year, 5-year, and 10-year follow-ups and high-resolution structural MRI completed at baseline and at the 3 follow-up evaluations. Ten-year post-deployment clinical outcome subtypes were derived from the 34 measures using k-means clustering.
Results: A total of 288 service members (mean [SD] age at final follow-up, 41.5 [8.1] years; 263 [91%] male) were assessed. Clustering optimization identified 4 clusters: 1 multidomain resilient to cognitive dysfunction, neurobehavioral and psychological symptoms (124 individuals); 1 resilient to cognitive dysfunction but with elevated neurobehavioral and psychological symptoms (122 individuals); 1 with severe cognitive dysfunction and moderately elevated neurobehavioral and psychological symptoms (14 individuals); and 1 with moderate cognitive dysfunction and severe neurobehavioral and psychological symptoms (28 individuals). Participants in the resilient group were largely nonblast controls, but 45 individuals (36%) had experienced blast mild TBI. Comparing the clusters over time revealed they were significantly different. Significant brain volume differences by cluster were observed comparing them at each time point. The cluster with severe cognitive dysfunction and moderately elevated neurobehavioral and psychological symptoms had the lowest brain volume, particularly in the cerebellum and brainstem.
Conclusions and relevance: In this cohort study of service members, clinical outcome clustering revealed key postdeployment subtypes that evolved over time. Cerebellar and brainstem volumes were significantly different by cluster. Regional differences were evident years before when stratified by these 10-year clusters. These findings underscore the need to appreciate not only the heterogeneity of brain injury but the heterogeneity of these mild TBI outcomes.