Introduction: Traumatic brain injuries (TBIs) are a significant cause of disability and mortality in the United States, impacting an estimated 1.7 million people annually and contributing to one third of all injury related deaths. Traumatic intracerebral hemorrhages (tICH), a subtype of TBIs that occurs in 13-35% of cases, can lead to severe long-term neurological complications. The blossoming (i.e. expanding) of a contusion, also known as a hemorrhagic progression of a contusion (HPC), is a form of delayed brain injury that occurs in 38-63% of tICH patients and is a strong predictor of mortality. HPCs can insidiously exacerbate the severity of a TBI and worsen an already poor patient prognosis. Though HPCs have a significant impact on neurotrauma patients, this phenomenon is incompletely characterized and inconsistently defined in the literature. Using 3D imaging technology, the aim of our study is threefold: 1) to characterize the phenomenon of blossoming contusions including time course, composition, and evolution; 2) to identify risk factors for contusion expansion; and 3) to determine if contusion characteristics can predict long-term clinical outcomes.

Methods:  From a database of 2187 patients who were admitted to UCMC with a traumatic brain injury from 2017 to present, participants ages 18 and older were included in this retrospective study if they presented with an intraparenchymal hemorrhage/contusion on their baseline CT scan, had a moderate to severe head injury (GCS < 13), and at least one follow-up CT scan within 24 hours of baseline CT. Exclusion criteria included a penetrating form of injury. An extensive chart review was conducted for each patient to collect comprehensive neurotrauma data. Radiology reports from all non-contrast head CT scans within 7 days of hospital admission were analyzed for indications of contusion location, expansion, and composition. Following, approximately 250 baseline and follow-up non-contrast head CTs for patients with identified contusion expansions have been requested from the UC Radiology Department for further analysis. Using two online 3D imaging analysis platforms, ITK-SNAP and Mango, CTs for each patient demonstrating an HPC will be quantified and analyzed over time to identify trends and to further characterize this phenomenon. A multivariate analysis will be conducted to identify predictors of contusion progression and clinical outcomes.

Results: 218 patients met inclusion and exclusion criteria and were included in this study. 79% of the study population was male and 21% was female, with an average age 49.8 at time of injury. 55% of patients had contusion expansion reported on CT radiology reports (n=120). Criteria for classifying an HPC (i.e. percent expansion of contusion size over a specific time interval) will be determined after 3D analysis is complete, and patients with a contusion expansion reported on radiology reports will be reassessed using the new criteria and nomenclature. Variables identified in the literature as risk factors for HPCs will be included in a multivariate statistical analysis to predict contusion expansion and clinical outcomes.

Conclusion: This project is one of the first studies to date to use 3D imaging technology to investigate blossoming contusions. Data collection is ongoing, and 3D image analysis will continue through 2020.