Abstract
Introduction
Although cerebral edema is common following traumatic brain injury (TBI), its formation and progression are poorly understood. This is especially true for the mild TBI population, who rarely undergo magnetic resonance imaging (MRI) studies, which can pick up subtle structural details not visualized on computed tomography, in the first few days after injury. This study aimed to visually classify and quantitatively measure edema progression in relation to traumatic microbleeds (TMBs) in a cohort of primarily mild TBI patients up to 30 days after injury. Researchers hypothesized that hypointense lesions on Apparent Diffusion Coefficient (ADC) detected acutely after injury would evolve into hyperintense Fluid Attenuated Inversion Recover (FLAIR) lesions.
Methods
This study analyzed the progression of cerebral edema after acute injury using multimodal MRI to classify TMBs as potential edema-related biomarkers. ADC and FLAIR MRI were utilized for edema classification at three different timepoints: ≤48 h, ∼1 week, and 30 days after injury. Hypointense lesions on ADC (ADC+) suggested the presence of cytotoxic edema while hyperintense lesions on FLAIR (FLAIR+) suggested vasogenic edema. Signal intensity Ratio (SIR) calculations were made using ADC and FLAIR to quantitatively confirm edema progression.
Results
Our results indicated the presence of ADC+ lesions ≤48 h and ∼1 week were associated with FLAIR + lesions at ∼1 week and 30 days, respectively, suggesting some progression of cytotoxic edema to vasogenic edema over time. Ten out of 15 FLAIR + lesions at 30 days (67%) were ADC+ ≤48 h. However, ADC + lesions ≤48 h were not associated with FLAIR + lesions at 30 days; 10 out of 25 (40%) ADC + lesions ≤48 h were FLAIR + at 30 days, which could indicate that some lesions resolved or were not visualized due to associated atrophy or tissue necrosis. Quantitative analysis confirmed the visual progression of some TMB lesions from ADC + to FLAIR+. FLAIR SIRs at ∼1 week were significantly higher when lesions were ADC+ ≤48 h (1.22 [1.08–1.32] vs 1.03 [0.97–1.11], p = 0.002).
Conclusion
Awareness of how cerebral edema can evolve in proximity to TMBs acutely after injury may facilitate identification and monitoring of patients with traumatic cerebrovascular injury and assist in development of novel therapeutic strategies.
Keywords: Traumatic brain injury, Traumatic microbleeds, ADC, Cytotoxic edema, Vasogenic edema
Highlights
-
•
This study analyzed cerebral edema progression after acute injury to classify TMBs as potential edema-related biomarkers
-
•
Presence of ADC+ lesions ≤48 hours and ∼1 week were associated with FLAIR+ lesions at ∼1 week and 30 days, respectively.
-
•
Edema in proximity to acute TMBs may aid in identification of TBI patients and development of novel therapeutic strategies.
1. Introduction
Traumatic brain injury (TBI) has the highest incidence of all common neurological disorders, with 50–60 million people suffering from a TBI every year (Maas et al., 2022). Up to 90% of TBIs are classified as mild and are associated with negative outcomes for some patients, including cognitive difficulties, poorer quality of life, and delayed return to work (Carroll et al., 2004; Levy et al., 2022).
Cerebral edema typically follows TBI and is classified as ‘cytotoxic’ or ‘vasogenic,’ referring to abnormal fluid accumulation within or around cells following blood-brain barrier disruption, respectively (Marmarou, 2007; Jha et al., 2019). Cytotoxic and vasogenic edema can coexist and may be interrelated (Marmarou, 2007; Jha et al., 2019). Although cerebral edema is associated with worse clinical outcomes, the exact mechanisms of edema formation and evolution from acute injury onward are poorly understood (Jha et al., 2019). This is partly due to infrequent neuroimaging obtained within 48 h of injury, especially in mild TBI. Patients with acute head injury typically undergo a Computed Tomography (CT) scan, which may not detect subtle structural changes that Magnetic Resonance Imaging (MRI) would (Vella et al., 2017; Dhamija and Donnan, 2008; Mutch et al., 2016).
Traumatic microbleeds (TMBs) are a persistent biomarker for all severities of TBI and appear as small, punctate or linear, regions of hypointensity on susceptibility-weighted imaging (SWI) and gradient recalled echo (GRE) MRI (Rizk et al., 2020; Griffin et al., 2019). In recent research involving post-mortem brain histology, edema occurred along linear-shaped TMBs with similar orientations to venous structures, suggesting TMBs occur from direct injury to the vasculature that leads to secondary injury in the form of cytotoxic edema (Griffin et al., 2019). The mechanism of injury may be akin to that seen in stroke patients; i) vascular injury leads to a disruption in blood flow and energy substrates, ii) cytotoxic edema occurs and decreases the diffusion of water in tissue, iii) cytotoxic edema evolves into vasogenic edema with the opening of the blood brain barrier and bulk influx of water, iv) over time, vasogenic edema subsides and resulting in a) recovery, b) gliosis/infarction, and/or c) rarefication, necrosis and atrophy. The evolution of tissue necrosis or cortical atrophy makes lesion detection more challenging in the subacute window. Because TMBs are visible in acute TBI and persist over time, they are an excellent biomarker to evaluate edema formation and evolution (Griffin et al., 2019). Specific MRI sequences, including Apparent Diffusion Coefficient (ADC) and Fluid Attenuated Inversion Recovery (FLAIR) imaging may help distinguish cytotoxic edema from vasogenic edema (Hudak et al., 2014). Lower ADC values characterized by hypointense regions suggest cytotoxic edema while normalized (isointense) or increased (hyperintense) ADC regions with conspicuously positive hyperintense regions on FLAIR suggest vasogenic edema (Hudak et al., 2014; Ito et al., 1996). A recent study demonstrated hypointense ADC regions were lowest in values within 24 h of injury and later increased with corresponding hyperintense positive appearance on FLAIR, suggesting an evolution of cytotoxic to vasogenic edema (Turtzo et al., 2021).
Our study aimed to visually classify and quantitatively measure edema progression up to 30 days after injury. Our hypothesis was that hypointense lesions on ADC (ADC+), detected ≤48 h of injury, would evolve into hyperintense FLAIR (FLAIR+) lesions by 30 days, and be demonstrated by both visually conspicuous qualitative classification and quantitative measurements.
2. Methods
2.1. Study approval
The Traumatic Head Injury Neuroimaging Classification (THINC) Study (NCT01132937) protocol was reviewed and approved by the NIH Institutional Review Board (NIH Protocol ID 10-N-0122). To be enrolled in THINC, patients were required: (i) to be ≥ 18 years of age, (ii) with a suspected head injury, (iii) presenting to a Level I (Washington Hospital Center, Washington, D.C.) or Level II Trauma Center (Suburban Hospital, Bethesda, MD), and (iv) able to have MRI within ≤48 h of injury. Researchers obtained written informed consent before any research procedures were conducted, in accordance with the Declaration of Helsinki.
2.2. Patient population
Patients included in this analysis were derived from a prior THINC study which required a follow up MRI within ∼1 week of injury, and an imaging target defined as an acute or subacute lesion on DWI or evidence of hemorrhage on acute MRI and/or CT (Turtzo et al., 2021).The current population was enrolled in THINC between October 2010 and May 2017. and met the following additional inclusion criteria: i) an imaging target defined as TMB(s) seen on GRE or SWI, ii) an evaluable Diffusion Weighted Imaging (DWI) lesion proximal to a TMB, and iii) two follow-up research MRIs completed ∼1 week after injury and ∼30 days after injury (Supplemental Figure I) (Turtzo et al., 2021). Patients were excluded if they were i) missing any necessary MRI sequences or ii) had insufficient imaging data or motion artifact (Supplemental Figure I). TBI severity was stratified by Glasgow Coma Scale (GCS) upon presentation to the emergency department [mild (GCS 13–15), moderate (GCS 9–12); severe (GCS <9)] (Department of Veterans Affairs DoD, 2021).
2.3. Image acquisition
Serial MRI studies were acquired on three scanners (1.5T, GE Healthcare, USA; 3T, Siemens, USA; or 3T Phillips, USA) at the following timepoints: ≤48 h, ∼1 week (time window per protocol was >48 h but <13 days), and ∼30 days (time window per protocol was 20–50 days) after injury. The imaging protocols at each timepoint utilized in this study were: (i) diffusion tensor imaging including DWI/ADC, (ii) T2-weighted FLAIR, (iii) GRE, and (iv) SWI as detailed previously (Department of Veterans Affairs DoD, 2021). Relevant parameters included: 240 mm FOV, 40 3.5 mm thick axial-oblique slices with 1 mm in-plane except DTI at 1.7/2.4 mm (3T/1.5T); DTI with 15-directions at b = 1000 and 4-b = 0 images; DTI voxel volume of 10/20 μL (3T/1.5T); FLAIR with TR/TE/TI ∼ 9000/120/2600 ms; GRE with TE = 12/20 ms (3T/1.5T), TR = 800 ms, FA = 30; and SWI with TE = 25/40 ms (3T/1.5T).
2.4. TMB lesion cluster qualitative analysis
MRI studies were de-identified prior to image analysis. Qualitative analysis of TMB presence and anatomical location was performed independently by three raters (LCT, LLL, ML). The same raters from the THINC study were utilized with inter-rater agreements of κ = 0.78 and 0.73 for DWI hyperintensity and ADC hypointensity respectively. GREs were primarily used in TMB detection to minimize potential false positives that could result from a more sensitive but less specific SWI image, though SWI was reviewed when the TMB presence was indeterminate on GRE. To minimize over-counting of TMBs and facilitate exploration of a variety of anatomical regions, raters grouped TMBs located in similar anatomical regions into lesion “clusters.” Clusters were defined as multiple TMBs within a few cm in diameter within the same hemisphere. These spatially defined clusters were evaluated consistently across all timepoints (Fig. 1). After independent identification of clusters was performed by the three raters, consensus was reached across all raters for the anatomical location and conspicuity of each cluster at each timepoint based on the visual presence/absence on DWI, FLAIR, and GRE. There was a limit of 3 clusters per patient, referred to simply as lesions for the remainder of the study. Additionally, at each timepoint, raters categorized each ADC lesion in proximity to a TMB lesion as hypointense (ADC+) or not hypointense (ADC-: isointense or hyperintense, as appropriate). Similarly, each FLAIR lesion proximal to a TMB lesion was categorized as hyperintense (FLAIR+) or not hyperintense (FLAIR-: isointense) at each timepoint.
Fig. 1.
Methodology approach for quantitative analysis demonstrated using serial MRI from a 21-year-old male who presented with a negative head CT to a Level I trauma center following a bicycle accident <1 h of injury with a GCS of 14. First MRI (≤48 h) was obtained approximately 19 h after injury. (A) Progression of TMB at baseline GRE (≤48 h) and 30 days after injury. Note: this patient had multiple TMB clusters: one in L frontal and another in the R frontal; however, only the progression of the L frontal cluster is shown in-depth here. (B) Progression of ADC + lesion ≤48 h and ∼1 week after injury to hyperintense lesion at 30 days after injury and corresponding ADC SI ratios. (C) Progression of FLAIR + lesion ≤48 h, ∼1 week, and 30 days after injury and corresponding FLAIR SI ratios. Abbreviations - ADC, Apparent diffusion coefficient; DWI, Diffusion-weighted imaging; FLAIR, Fluid attenuated inversion recovery; GCS, Glasgow Coma Scale; GRE, gradient recalled echo, SI, Signal Intensity; TMB, Traumatic microbleed.
ADC + lesions were considered as evidence of cytotoxic edema while FLAIR + lesions were considered evidence of vasogenic edema or, potentially, gliosis at later times from injury.
2.5. TMB lesion cluster quantitative analysis
Quantitative analysis of lesion signal intensity ratios (SIR) was performed in an independent session by a fourth trained rater (JL) using the labeled TMB lesion locations from the qualitative analysis reached by consensus by the three raters. ADC, DWI, and FLAIR were midsagittally aligned, co-registered, and visually inspected for accuracy using MIPAV (NIH, CIT, v11.0.3) to allow for comparable spatial localization of lesions across sequences and timepoints. Selection of the SIR measurement location was based on the slice that the lesion was most visually conspicuous, likely reflecting the most severely ischemic region. SIRs were measured for each lesion adjacent to a TMB lesion on ADC and FLAIR at each timepoint. The SIR value was calculated by dividing the signal intensity (SI) of the lesion by its contralateral homologous region; SIR = [SI TMB lesion region/SI contralateral region]. Additionally, the images were flipped to ensure that the lesion region was placed in the comparable contralateral region for the SIR measurement. The quality of the ADC measurement for each field strength was assessed in a region of interest in normal appearing parenchyma by dividing the mean by the standard deviation of the ADC across voxels.
2.6. Statistical analysis
Descriptive and quantitative variables were reported as n (%) or median with interquartile range [IQR 25–75]. Fisher exact two-tailed tests were used to determine the predictive value of ADC + lesion at an earlier timepoint and FLAIR + lesion at a later timepoint. Positive and negative predictive values of FLAIR + lesions at 30 days to determine ADC lesions ≤48 h and FLAIR + lesions at ∼1 week were calculated. Mann-Whitney U two-tailed tests were used to compare ADC lesion classification and ADC and FLAIR SIRs. Statistical significance was defined as p < 0.05. SPSS (v28.0.0.0 (190)) was utilized.
3. Results
Twenty-eight patients, 79% male with median age 45 years [28–57], consented to THINC and were included in this analysis (Table 1, Supplemental Figure I). Twenty-two patients (79%) presented with a GCS score 13–15. There were no noted differences between the larger cohort (n = 161) and this study cohort (n = 28) as demonstrated in Table 1. Twenty-four patients (86%) were serially imaged on 3T scanners and 4 (14%) were serially scanned on 1.5T. Mean ADC values and spatial variance in the ADC were similar across field strengths measured in a control region; 0.73 (0.044) vs 0.74 (0.043) x 10−5 cm^2/s for 1.5T vs at 3T. Median times from injury to each of the three MRI scans were 22 h, 3.4 days, and 27 days, respectively. In comparison to the entire THINC study, the median times to the follow-up MRIs were 4.3 days and 31 days.
Table 1.
Patient demographics and clinical presentation of study cohort.
|
Turtzo et al., 2021 Cytotoxic Edema TBI THINC Study |
Lee et al., 2024 Current Study |
|
|---|---|---|
| (n = 161) | (n = 28) | |
| Demographics | ||
| Age (median, IQR) | 54 (33–66) | 45 (27–58) |
| Male (n, %) | 117 (73%) | 22 (79%) |
| Clinical Presentation | ||
| Trauma Level I | 85 (52%) | 15 (54%) |
| Clinical Dx Intracranial Hemorrhage | 92 (57%) | 13 (46%) |
| Clinical Dx of TBI | 114 (70%) | 22 (79%) |
| Injury Cause, n (%) | ||
| Road Traffic Accident | 62 (39%) | 14 (50%) |
| Incidental Fall | 72 (44%) | 10 (36%) |
| Violence/Assault | 17 (10%) | 4 (14%) |
| Other | 10 (6%) | n/a |
| Injury Mechanism, n (%) | ||
| Acceleration/Deceleration | 32 (20%) | 9 (32%) |
| Direct impact- blow to head | 40 (25%) | 7 (25%) |
| Direct impact- head against object | 25 (15%) | 2 (7%) |
| Fall - ground floor | 40 (25%) | 4 (14%) |
| Fall-from height >1m | 23 (14%) | 6 (21%) |
| GCS, n (%) | ||
| 15 | 112 (70%) | 18 (64%) |
| 14 | 31 (19%) | 5 (18%) |
| ≤13 | 18 (11%) | 5 (15%) |
| LOC Reported | 100 (62%) | 19 (68%) |
| PTA Reported | 109 (68%) | 25 (89%) |
| TBI classification, n (%) | ||
| Silent | 33 (20%) | 3 (11%) |
| Mild | 99 (61%) | 19 (68%) |
| Moderate | 21 (13%) | 4 (14%) |
| Severe | 8 (5%) | 2 (7%) |
| Imaging Findings | ||
| CT – Clinical | ||
| CT positive for TBI | 95 (59%) | 15 (54%) |
| Intraparenchymal hematomaa | 38 (24%) | 4 (14%) |
| Extra-axial hemorrhage | 88 (55%) | 14 (50%) |
| Hours - Injury to CT (median, IQR) | 1.68 (1.1–6.4) | 1.5 (1.1–4.5) |
| MRI- Research | ||
| Intraparenchymal hematoma >1 cm on SWI | 49 (30%) | 8 (29%) |
| Subdural or Subarachnoid hemorrhage | 97 (60%) | 16 (57%) |
| Intraventricular hemorrhage | 16 (10%) | 3 (11%) |
| Hours - Injury to MRI ≤48 h of injury (median, IQR) |
20.5 (10–30) | 21.9 (13.1–32.1) |
| Days - Injury to <1 week MRI (median, IQR) | n/a | 3.4 (2.0–7) |
| Days - Injury to MRI within 30 daysb (median, IQR) | n/a | 27 (25–34) |
Emergency department radiologists reported contusion or intracerebral hemorrhage.
21 patients in quantitative cohort did not have 30d MRI as that was not part of the Turtzo et al., 2020 Cytotoxic Edema TBI Study inclusion criteria.
TMB lesion cluster analysis
A total of 50 TMB lesions were identified across the 28 patients; 11 (39%) had 1 lesion, 12 (43%) had 2 lesions, and 5 (18%) had 3 lesions. Despite microbleed clusters being readily apparent on MRI, only 54% of patients were CT + for TBI (Table 1). Examples of patients with CT- but MRI+ and CT+ and MRI + are shown in Supplemental Figure II.
ADC + lesions were most visually conspicuous at ≤48 h of injury (n = 25, 50%), with decreased visual appearance ∼1 week (n = 19, 38%), and were absent at 30 days (n = 0, 0%), indicating all ADC lesions were ADC-, either isointense (n = 39, 78%) or hyperintense (n = 11, 22%). An example of mixed edema evolution visualized on ADC in relation to a proximal TMB lesion is shown in Supplemental Figure III.
ADC SIRs were reduced for lesions identified as ADC + compared to ADC- lesions ≤48 h (SIR = 0.88 [0.81–0.95] vs 1.02 [0.96–1.05], p < 0.001) and 1 week (SIR = 0.96 [0.88–1.00] vs 1.03 [0.98–1.10], p < 0.001). There was no evidence of ADC + lesions at 30 days, but ADC SIRs were increased for ADC hyperintense vs isointense lesions (SIR = 1.13 [1.05–1.20] vs 1.01 [0.99–1.05], p < 0.001). These quantitative results were confirmatory of the qualitative findings (Fig. 1).
Evidence of FLAIR + lesions was present at each timepoint but was most prevalent ∼1 week (n = 30, 61%). The prevalence of FLAIR + lesions was 49%, 61% and 31% ≤ 48 h, ∼1 week, and 30 days, respectively. All FLAIR + lesions ≤48 h remained visible ∼1 week; however, 13 (∼50%) were no longer visible by 30 days.
Ninety two percent (n = 23) of ADC + lesions ≤48 h were FLAIR+ ∼1 week while 32% (n = 8) of ADC- lesions were FLAIR+ ∼1 week (p < 0.001, Supplemental Figure IVa). Furthermore, FLAIR SIRs were increased ∼1 week when lesions were ADC+ ≤48 h (1.22 [1.08–1.32] vs 1.03 [0.97–1.11], p = 0.002). Meanwhile, 53% (n = 10) of ADC + lesions ∼1 week were FLAIR + at 30 days compared to 16% (n = 5) of ADC- lesions ∼1 week were FLAIR + at 30 days (p = 0.011, Supplemental Figure IVc). Similarly, FLAIR SIRs were increased at 30 days when ADC+ ∼1 week (1.13 [0.98–1.24] vs 0.99 [0.93–1.06], p = 0.005).
Our study findings suggest a progression of cytotoxic edema to vasogenic edema in some patients. Ten out of 15 FLAIR + lesions (67%) at 30 days were ADC+ ≤48 h.
The positive predictive values of FLAIR + lesions at 30 days were 0.67 for ADC + lesions ≤48 h and 0.93 for FLAIR + lesions ∼1 week. However, the negative predictive values of FLAIR + lesions at 30 days were 0.57 for ADC + lesions ≤48 h and 0.52 for FLAIR + lesions ∼1 week. Only 40% (n = 10) ADC + lesions ≤48 h were FLAIR + at 30 days compared to 20% (n = 5) of ADC- lesions (p = 0.22). Quantitative analysis supported this assessment as the FLAIR SIRs at 30 days were not different between ADC + vs ADC- lesion ≤48 h, (1.06 [0.97–1.1] vs 1.00 [0.93–1.06], p = 0.08).
4. Discussion
Our findings suggest ADC + lesions surrounding TMBs following acute TBI evolve into FLAIR + lesions ∼1 week, and either partially resolve, or progress to conspicuously evident gliosis at 30 days. The quantitative results confirmed our qualitative assessment as FLAIR SIRs were significantly different ∼1 week and at 30 days when there was evidence of ADC + lesions at either ≤48 h or ∼1 week of injury, respectively. Additionally, ADC SIRs were different amongst those lesions categorized as ADC + vs ADC- lesions ≤48 h and ∼1 week. However, ADC + lesions ≤48 h were not associated with FLAIR + lesions at 30 days.
This edema progression is consistent with previous research that found hypointense ADC lesions immediately after injury gradually increased in signal intensity (Turtzo et al., 2021). The close proximity of ADC + lesions surrounding TMBs ≤48 h was also consistent with findings suggesting TMBs are due to vascular injury (Griffin et al., 2019). Researchers often conceptualize the pathophysiology of cytotoxic and vasogenic edema separately (Michinaga and Koyama, 2015). However, these results add to research suggesting the two are associated over time including recent studies in humans and mice that utilized free water diffusion MRI to distinguish cytotoxic versus vasogenic edema after TBI (Marmarou, 2007; Jha et al., 2019; Eisenberg et al., 2020; Hu et al., 2023). We demonstrated here that the temporal evolution from cytotoxic to vasogenic edema near hemorrhage may also occur in TBI patients with GCS≥13–15.
Whether these FLAIR + lesions evolve over time into chronic gliosis and/or neuronal atrophy requires further study. Loss of white matter integrity following brain injury is associated with demyelination and axonal degeneration in rats (Budde et al., 2011). Animal models have demonstrated gliosis in mice months after repeated mild traumatic injuries (Mannix et al., 2014). Further understanding the mechanisms of edema evolution is relevant for researchers and clinicians.
Unexpectedly, ADC + lesions ≤48 h after injury were not associated with FLAIR + lesions at 30 days. This may be due to the mixed presence of vasogenic and cytotoxic edema around microbleeds that were not detectable by 30 days. A negative FLAIR at 30 days may be secondary to edema-related atrophy which could imply cytotoxic edema is not observed by clinicians in mild TBI patients unless MRI is acquired acutely after injury.
Limitations of this analysis included the small sample size which did not accommodate multivariable regression analyses including age and sex comparisons. There were also potential inconsistencies in co-registration that may have impacted the anatomical specificity of the quantitative analysis. Approximately 14% of patients were imaged on 1.5T rather than 3T. The variance of the ADC measurements were similar across field strengths; however the volume of the imaging voxel at 1.5T was about twice that at 3T, raising some concern for some bias given the size of the focal lesions. Another challenge for comparison of our study with other studies is the evolving definition of TBI classifications, which has made comparisons difficult and is still being addressed by research workshops including the upcoming National Institute of Neurological Disorders and Stroke (NINDS) Traumatic Brain Injury (TBI) Classification and Nomenclature Workshop (Natcher Conference Center, NIH Campus, Bethesda, MD, January 22–23, 2024). Despite the small sample size, the quantitative results confirmed the findings from the visually conspicuous qualitative reads, validating the study methodology.
5. Conclusions
This study evaluated cerebral edema progression up to 30 days following TBI. Our results demonstrated the majority of TMBs with corresponding ADC + lesions ≤48 h appeared FLAIR-at 30 days. However, ADC + lesions ≤48 h and ∼1 week were associated with FLAIR + lesions ∼1 week and at 30 days after injury, respectively. This indicates there was an evolution of cytotoxic to vasogenic edema while raising questions as to the exact timing of that progression. Future studies may include further exploration and quantification of injury related atrophy beyond 30 days.
Statement of ethics
The THINC study protocol was reviewed and approved by the NIH Institutional Review Board (NIH Protocol ID 10-N-0122). Researchers obtained written informed consent from patients or their legally authorized representative prior to participation in the study, in accordance with the Declaration of Helsinki.
Funding sources
Financial support for this work was provided by the Intramural Research Program of the National Institute of Neurological Disorders and Stroke, National Institutes of Health, United States, and the Center for Neuroscience and Regenerative Medicine, Department of Defense, United States. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health or of the Department of Defense.
CRediT authorship contribution statement
Jacquie Lee: Writing – review & editing, Writing – original draft, Methodology, Formal analysis. Emily Baniewicz: Writing – review & editing, Formal analysis. Nicole L. Peterkin: Writing – review & editing, Data curation. Danielle Greenman: Writing – review & editing, Methodology, Data curation. Allison D. Griffin: Writing – review & editing, Data curation. Neekita Jikaria: Writing – review & editing, Methodology, Data curation. L. Christine Turtzo: Writing – review & editing, Conceptualization. Marie Luby: Writing – review & editing, Methodology, Conceptualization. Lawrence L. Latour: Writing – review & editing, Resources, Methodology, Conceptualization.
Declaration of competing interest
The authors have no conflicts of interest to declare.
Acknowledgement
We would like to thank our patients and their families, without whom this research would not have been possible. We also greatly appreciate the clinicians and administrators who supported the THINC study.
Footnotes
Supplementary data to this article can be found online at https://doi.org/10.1016/j.ynirp.2024.100199.
Contributor Information
Marie Luby, Email: lubym@ninds.nih.gov.
Lawrence L. Latour, Email: latourl@ninds.nih.gov.
Appendix A. Supplementary data
The following are the Supplementary data to this article.
figs1.
figs2.
figs3.
figs4.
Data availability
Data will be made available on request.
References
- Budde M.D., Janes L., Gold E., Turtzo L.C., Frank J.A. The contribution of gliosis to diffusion tensor anisotropy and tractography following traumatic brain injury: validation in the rat using Fourier analysis of stained tissue sections. Brain. 2011;134:2248–2260. doi: 10.1093/brain/awr161. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Carroll L.J., Cassidy J.D., Peloso P.M., et al. Prognosis for mild traumatic brain injury: results of the WHO collaborating centre task force on mild traumatic brain injury. J. Rehabil. Med. 2004:84–105. doi: 10.1080/16501960410023859. [DOI] [PubMed] [Google Scholar]
- Department of Veterans Affairs DoD . 2021. VA/DoD Clincial Practice Guideline for the Management and Rehabilitation of Post-Acute Mild Traumatic Brain Injury. [Google Scholar]
- Dhamija R.K., Donnan G.A. The role of neuroimaging in acute stroke. Ann. Indian Acad. Neurol. 2008;11:S12–S23. [PMC free article] [PubMed] [Google Scholar]
- Eisenberg H.M., Shenton M.E., Pasternak O., et al. Magnetic resonance imaging pilot study of intravenous glyburide in traumatic brain injury. J. Neurotrauma. 2020;37:185–193. doi: 10.1089/neu.2019.6538. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Griffin A.D., Turtzo L.C., Parikh G.Y., et al. Traumatic microbleeds suggest vascular injury and predict disability in traumatic brain injury. Brain. 2019;142:3550–3564. doi: 10.1093/brain/awz290. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hu S., Exner C., Sienel R.I., et al. Characterization of vasogenic and cytotoxic brain edema formation after experimental TBI by free water diffusion MRI. J Neurotrauma. 2023;41:393–406. doi: 10.1089/neu.2023.0222. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hudak A.M., Peng L., Marquez de la Plata C., et al. Cytotoxic and vasogenic cerebral oedema in traumatic brain injury: assessment with FLAIR and DWI imaging. Brain Inj. 2014;28:1602–1609. doi: 10.3109/02699052.2014.936039. [DOI] [PubMed] [Google Scholar]
- Ito J., Marmarou A., Barzo P., Fatouros P., Corwin F. Characterization of edema by diffusion-weighted imaging in experimental traumatic brain injury. J. Neurosurg. 1996;84:97–103. doi: 10.3171/jns.1996.84.1.0097. [DOI] [PubMed] [Google Scholar]
- Jha R.M., Kochanek P.M., Simard J.M. Pathophysiology and treatment of cerebral edema in traumatic brain injury. Neuropharmacology. 2019;145:230–246. doi: 10.1016/j.neuropharm.2018.08.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Levy A.M., Saling M.M., Anderson J.F.I. Frequency and extent of cognitive complaint following adult civilian mild traumatic brain injury: a systematic review and meta-analysis. Brain Impair. 2022:1–24. doi: 10.1017/BrImp.2022.19. [DOI] [PubMed] [Google Scholar]
- Maas A.I.R., Menon D.K., Manley G.T., et al. Traumatic brain injury: progress and challenges in prevention, clinical care, and research. Lancet Neurol. 2022;21:1004–1060. doi: 10.1016/S1474-4422(22)00309-X. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Mannix R., Berglass J., Berkner J., et al. Chronic gliosis and behavioral deficits in mice following repetitive mild traumatic brain injury. J. Neurosurg. 2014;121:1342–1350. doi: 10.3171/2014.7.JNS14272. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Marmarou A. A review of progress in understanding the pathophysiology and treatment of brain edema. Neurosurg. Focus. 2007;22:E1. doi: 10.3171/foc.2007.22.5.2. [DOI] [PubMed] [Google Scholar]
- Michinaga S., Koyama Y. Pathogenesis of brain edema and investigation into anti-edema drugs. Int. J. Mol. Sci. 2015;16:9949–9975. doi: 10.3390/ijms16059949. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Mutch C.A., Talbott J.F., Gean A. Imaging evaluation of acute traumatic brain injury. Neurosurg. Clin. 2016;27:409–439. doi: 10.1016/j.nec.2016.05.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Rizk T., Turtzo L.C., Cota M., et al. Traumatic microbleeds persist for up to five years following traumatic brain injury despite resolution of other acute findings on MRI. Brain Inj. 2020;34:773–781. doi: 10.1080/02699052.2020.1725835. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Turtzo L.C., Luby M., Jikaria N., et al. Cytotoxic edema associated with hemorrhage predicts poor outcome after traumatic brain injury. J. Neurotrauma. 2021;38:3107–3118. doi: 10.1089/neu.2021.0037. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Vella M.A., Crandall M.L., Patel M.B. Acute management of traumatic brain injury. Surg. Clin. 2017;97:1015–1030. doi: 10.1016/j.suc.2017.06.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data will be made available on request.