Decreased grey matter volume in mTBI patients with post- traumatic headache compared to headache-free mTBI patients and healthy controls: a longitudinal MRI study

DOI: https://doi.org/10.1007/s11682- 019- 00095- 7

Abstract-Summary Traumatic brain injury (TBI) occurs in 1.7 million people annually and many patients go on to develop persistent disorders including post-traumatic head- ache (PTH).

930

5 Future Directions

We aimed to identify changes in cerebral grey matter volume (GMV) associated

with PTH in mild TBI patients.

Fifty mTBI patients (31 Non-PTH; 19 PTH) underwent MRI scans: within 10

days post-injury, 1 month, 6 months and 18 months.

Compared to non-PTH, PTH patients had decreased GMV across two large clus- ters described as the right anterior-parietal (p = 0.012) and left temporal-opercular (p = 0.027).

Compared to healthy controls non-PTH patients had decreased GMV in the left thalamus (p = 0.047); PTH patients had decreased GMV in several extensive clus- ters: left temporal-opercular (p  =  0.003), temporal-parietal (p  =  0.041), superior frontal gyrus (p = 0.008) and right middle frontal/superior frontal gyrus (0.004) and anterior-parietal (p = 0.003).

Patients with these changes should be monitored for chronic PTH.

Introduction The prevalence of PTH has been reported as high as 79% at 3 months and 65% at 12 months following mTBI (Lucas and others [34]).

Obermann and others [42] reported that PTH in patients post-whiplash was asso- ciated with significantly decreased GMV in the anterior cingulate cortex and dorso- lateral prefrontal cortex (DLPFC) at 2 weeks and 3 months post-injury; however, these changes completely resolved along with the headache after a year.

Given the difficulty in assessing these predictors and changes associated with PTH in TBI, we assessed patients both with and without headache and explored GMV changes that are unique to mTBI and those specific to patients with headache post-mTBI.

We aimed to compare changes in GMV in mTBI patients with and without PTH

and healthy controls over four time points occurring up to 18 months after injury.

Materials and Methods Patient scans were acquired in both axial and sagittal sequences over time; the sagit- tal sequence which was conducted on the same scanner, was introduced later in the study during data acquisition as it produced better quality scans.

This resulted in the exclusion of 25 scans from 19 patients across all four time

points in the non-PTH patients and 10 scans from 8 patients in the PTH group.

We first compared how GMV changed over time between the two mTBI patient

groups (those with and without PTH).

We examined change in GMV over time in all mTBI patients compared to

healthy controls.

Healthy controls were only scanned at two time points, this analysis only com-

pared patients to healthy controls at visit 1 and visit 3.

Differences between patient groups at each time point and patient groups and

controls were assessed using 2-sample T-tests and Wilcoxon tests.

Results In the left temporal-opercular cluster the difference between the two groups is sig- nificant only at visit 1 (p = 0.022), while in the right anterior-parietal cluster differ- ences were most evident at visits 2 and 4.

5.1

Post-concussion Syndrome

931

This change in GMV was significant in PTH in both clusters (p  <  0.0001) as indicated by the interaction p value whereas in non-PTH the change over time was relatively constant and not significant.

Thalamic GMV was consistent in controls over time (p = 0.27) but is initially

higher in the non-PTH group and decreases significantly over time (p < 0.0001).

PTH and controls start with comparable GMV in the left temporal-parietal and SFG clusters, while controls increase over time, PTH patients decrease between visits 1 and 3 (p < 0.0001) (plots not shown).

In the left temporal-opercular cluster, right anterior-parietal, and MFG/SFG clus- ters, PTH initially had larger volumes which decrease over time compared to con- trols (who have an increase in volume).

Discussion We found that mTBI patients with evidence of PTH approximately 18 months post- injury had decreased GMV in two main regions of the brain compared to mTBI patients who do not develop headache.

The observed decrease in PTH patients in the R MFG/SFG may be attributed to patients having initially higher GMV at visit 1 which declined to comparable levels to controls at visit 3.

Non-PTH patients compared to controls had decreased thalamic GMV over time. Thalamic GMV loss post-TBI up to 1 year after injury has been reported, and this trajectory differs from healthy controls (Eierud and others [189]; Lucas and others [1]), consistent with our findings in non-PTH patients.

Not only do we report decreased GMV over time when compared to healthy controls and non-PTH patients, but our results show that these differences are pres- ent in PTH patients as early as 1 week post-TBI.

We found brain regions in PTH with decreased GMV compared to both healthy

controls and non-PTH patients.

Acknowledgement A machine generated summary based on the work of Burrowes, Shana A. B.; Rhodes, Chandler Sours; Meeker, Timothy J.; Greenspan, Joel D.; Gullapalli, Rao P.; Seminowicz, David A. 2019 in Brain Imaging and Behavior.

Altered static and dynamic functional network connectivity in post-traumatic headache