Sex-Dependent Cortical Volume Changes in Patients with Degenerative Cervical Myelopathy

Degenerative cervical myelopathy (DCM) is a progressive condition characterized by degeneration of osseocartilaginous structures within the cervical spine resulting in compression of the spinal cord and presentation of clinical symptoms. Compared to healthy controls (HCs), studies have shown DCM patients experience structural and functional reorganization in the brain; however, sex-dependent cortical differences in DCM patients remains largely unexplored. In the present study, we investigate the role of sex differences on the structure of the cerebral cortex in DCM and determine how structural differences may relate to clinical measures of neurological function. T1-weighted structural MRI scans were acquired in 85 symptomatic and asymptomatic patients with DCM and 90 age-matched HCs. Modified Japanese Orthopedic Association (mJOA) scores were obtained for patients. A general linear model was used to determine vertex-level significant differences in gray matter volume (GMV) between the following groups (1) male HCs and female HCs, (2) male patients and female patients, (3) male patients and male HCs, and (4) female patients and female HCs. Within patients, males exhibited larger GMV in motor, language, and vision related brain regions compared to female DCM patients. Males demonstrated a significant positive correlation between GMV and mJOA score, in which patients with worsening neurological symptoms exhibited decreasing GMV primarily across somatosensory and motor related cortical regions. Females exhibited a similar association, albeit across a broader range of cortical areas including those involved in pain processing. In sensorimotor regions, female patients consistently showed smaller GMV compared with male patients, independent of mJOA score. Results from the current study suggest strong sex-related differences in cortical volume in patients with DCM, which may reflect hormonal influence or differing compensation mechanisms.


Introduction
Degenerative cervical myelopathy (DCM) is a chronic condition involving the progressive deterioration of osseocartilaginous structures within the cervical spine resulting in compression of the spinal cord [1,2]. DCM often occurs as a consequence of age-related degeneration of the spine and is the most common spinal cord impairment in people over

Patient Population
A total of 85 patients were prospectively enrolled from 2016 to 2021 in a cross-sectional study including brain and spinal cord imaging as well as a neurological examination. Patients were recruited from an outpatient neurosurgery clinic and exhibited spinal cord compression with evidence of spinal cord deformation, mass effect, and no visible cerebrospinal fluid signal around the spinal cord at the site of maximal compression on MRI. Patients and healthy controls were excluded if they had neurological or neurocognitive impairment or significant psychiatric comorbidities. All patients signed Institutional Review Board-approved consent forms, and all analyses were performed in compliance with the Health Insurance Portability and Accountability Act (HIPAA). The patient cohort consisted of 52 males and 33 females ranging in age from 31 to 81 years with a mean age of 58.5 years for males and 58 years for females. All patients underwent brain and spinal cord imaging at UCLA. The modified Japanese Orthopedic Association (mJOA) score was used as a measure of neurological function [24]. The mJOA scoring scale ranges from 0 to 18, where lower scores represent a worse neurological impairment and an mJOA score of 18 repre-sents no impairment of neurological function. Patient demographic data is summarized in Table 1. Table 1. Cohort demographics. Age is provided in mean years ± the standard deviation, minimum and maximum years, and p-value of Wilcoxon-Mann-Whitney test between age of males and females. The modified Japanese Orthopedic Association (mJOA) score is provided in mean score ± the standard deviation, minimum and maximum scores, and p-value of Wilcoxon-Mann-Whitney test between scores of males and females. * = HCs were categorized with an mJOA score of 18 due to their healthy neurological status.

MR Imaging Acquisition
For the patient cohort, high-resolution 1 mm 3-dimensional (3D) T1-weighted structural MRIs were acquired on a 3T MR scanner (Siemens Prisma or Trio; Siemens Healthcare, Erlangen, Germany) using a 3D magnetization-prepared rapid gradient-echo (MPRAGE) sequence in either the coronal, sagittal, or axial orientation, with a repetition time (TR) of 2300 to 2500 ms, a minimum echo time (TE), an inversion time (TI) of 900 to 945 ms, a flip angle of 9 • to 15 • , FOV = 240 × 320 mm and matrix size of 240 × 320, slice thickness = 1 mm. For the HC cohort, high-resolution 3-dimensional (3D) T1-weighted structural MRIs were acquired on a 3T MR scanner using a 3D T1-weighted sequence (e.g., MPRAGE or SPGR) with a slice thickness = 1.5 mm or less with no interslice gap. All other parameters including repetition (TR) and echo (TE) time were specific to site scanner manufacturer recommendations for a T1-weighted, 3D sequence.

Image Processing and Analysis
Cortical segmentation and computation of GMV were performed using FreeSurfer (https://surfer.nmr.mgh.harvard.edu/fswiki, access date 1 May 2021 on the T1-weighted images described above [26]. Processed brain surfaces were smoothed with a full-width half-maximum of 10 mm, then registered to a standard space defined by the Desikan-Killiany-Tourville (DKT) atlas [27]. Whole-brain cortical volume analysis was completed using FreeSurfer. A general linear model (GLM) was used to determine vertex-level significant differences in GMV between the following groups: (1) male HCs and female HCs, (2) male patients and female patients, (3) male patients and male HCs, and (4) female patients and female HCs. To control for the influence of age on GMV, age was included as a covariate in morphometric analyses [28,29]. When comparing GMV between male patients and female patients, both age and mJOA score were included as covariates. To evaluate the association between sex, cortical volume, and neurological deficit, a GLM was used to determine vertex-level significant correlations between GMV and mJOA score in (A) male patients and HCs, and (B) female patients and HCs. Following the overlapping of significant clusters observed in the male group and significant clusters observed in the female group, we identified common cortical regions showing significant correlations between GMV and mJOA in both male and female groups. Additionally, average GMVs for each individual subject were extracted from the mutually significant clusters and corrected for subject age. In the male patient group and the female patient group, linear regression analyses were performed between age corrected average GMV and mJOA score within sensorimotor and pain related brain regions. In addition, linear regression analyses were used to identify differences in GMV and mJOA slope and intercept between male and female patients. Healthy controls were excluded in regression analyses. Regression analyses were performed using MATLAB (Release 2018a, MathWorks, Natick, MA, USA) and GraphPad Prism software (Version 7.0c GraphPad Software, San Diego, CA, USA). The vertex-wise level of significance was set at p < 0.05, with multiple comparisons correction performed by using Monte Carlo permutations with a significance level of p < 0.05.

Subject Characteristics
As summarized in Table 1, the patient cohort consisted of 52 males with a mean age of 58.5 ± 11.6 years and 33 females with a mean age of 58.0 ± 10.7 years. There was no significant difference in age between male patients and female patients (Wilcoxon-Mann-Whitney test, p = 0.8068). The mJOA scores within the cohort ranged from 9 to 18 with a mean score of 15.0 ± 2.7 for male patients and 15.6 ± 2.4 for female patients. Of the 85 total study patients, 19 had asymptomatic spinal cord compression (mJOA = 18), 38 presented with mild myelopathy (15 ≤ mJOA ≤ 17), 19 exhibited moderate myelopathy (12 ≤ mJOA ≤ 14), and 9 patients were categorized with severe myelopathy (mJOA ≤ 11). No significant difference in mJOA score was observed between male and female patients (Wilcoxon-Mann-Whitney test, p = 0.3885).
The HC cohort consisted of 53 males with a mean age of 58.7 ± 6.4 years and 37 females with a mean age of 59.8 ± 6.3 years. There was no significant difference in age between male and female HCs (Wilcoxon-Mann-Whitney test, p = 0.4076). Additionally, no significant difference in age was found between the patient cohort and the HC cohort (Mann-Whitney test, p = 0.9206). Due to lack of neurological impairment, all HC participants had an mJOA score of 18.

Sex-Dependent Cortical Volumetric Differences
Results from the whole-brain cortical volume analysis revealed no significant difference in GMV between males and females within the HC cohort, but significant differences in GMV between male and female within patients with DCM. We observed that male DCM patients compared to female patients ( Figure 1A, Table 2) exhibited significantly larger GMV in the caudal middle frontal, superior temporal, transverse temporal, and lingual gyrus of the left hemisphere, as well as in the precentral gyrus, insula, and lingual gyrus of the right hemisphere. Additionally, when controlling for mJOA, male DCM patients demonstrated significantly larger GMV than female patients in the bilateral lateral occipital gyri, left superior temporal gyrus, right insula, right middle temporal gyrus, and right lingual gyrus ( Figure 1B, Table 2). gual gyrus of the left hemisphere, as well as in the precentral gyrus, insula, and lingual gyrus of the right hemisphere. Additionally, when controlling for mJOA, male DCM patients demonstrated significantly larger GMV than female patients in the bilateral lateral occipital gyri, left superior temporal gyrus, right insula, right middle temporal gyrus, and right lingual gyrus ( Figure 1B, Table 2).
Male DCM patients displayed significantly larger gray matter volume (GMV) in the left parahippocampal gyrus, left paropercularis, right lateral occipital cortex, and right lingual gyrus compared with male HCs (Figure 2A, Table 3). On the contrary, female DCM patients exhibited significantly smaller GMV compared with female HCs, specifically in the left pericalcarine cortex and right lingual gyrus ( Figure 2B, Table 3). When controlling for both age and mJOA, regions with significant differences in GMV were identified in the a, left superior temporal gyrus; b, bilateral lateral occipital cortex; c, right insular cortex; d, right middle temporal gyrus; and e, right lingual gyrus. Significant clusters were determined by thresholding based on statistical significance (p < 0.05).   Male DCM patients displayed significantly larger gray matter volume (GMV) in the left parahippocampal gyrus, left paropercularis, right lateral occipital cortex, and right lingual gyrus compared with male HCs (Figure 2A, Table 3). On the contrary, female DCM patients exhibited significantly smaller GMV compared with female HCs, specifically in the left pericalcarine cortex and right lingual gyrus ( Figure 2B, Table 3).

Interaction between Cortical Volume and mJOA Scores
When examining the effect of sex on the association between GMV and mJOA score ( Figure 3, Table 4), both males ( Figure 3A) and females ( Figure 3B) demonstrated a significant positive correlation between GMV and mJOA score across multiple cortical regions. Female subjects demonstrated associations between GMV and mJOA in similar regions to male subjects, but regions in female subjects appeared to extend across a broader area of the brain perhaps suggesting more widespread cortical changes in females. Mutually significant regions with a positive correlation between GMV and mJOA common for both males and females are illustrated in Figure 4A. Within DCM patients only (excluding HCs), males and females demonstrated significant correlations between age corrected GMV and mJOA within similar regions, but the degree of change (i.e., slope of the regression line) and overall GMV (i.e., intercept of regression line) were different between males and females within the left superior frontal (p = 0.0013), right superior frontal (p = 0.0301), left paracentral (p < 0.0001), right rostral middle frontal (p < 0.0001), left precentral (p < 0.0001), and right precentral (p < 0.0001) gyri, as well as the anterior, isthmus, and posterior cingulate cortex, the insula, and the precuneus ( Figure 4B, Table 5).       Table 5. Regression analyses quantifying the association between average gray matter volume (GMV) and mJOA score for regions found significant in both sexes. LH denotes left hemisphere and RH denotes right hemisphere. The table includes the following: mutually significant anatomical region, surface area of cortical region of interest (ROI), p-value evaluating whether male and female linear fits are significantly different in slope and y-intercept, p-value evaluating whether a linear relationship occurs between average GMV and mJOA score in males, p-value evaluating whether a linear relationship occurs between average GMV and mJOA score in females, goodness of fit for males, and goodness of fit for females.  The light blue region denotes the 95% confidence interval for male patients and the pink region denotes the 95% confidence interval for female patients. Table 5. Regression analyses quantifying the association between average gray matter volume (GMV) and mJOA score for regions found significant in both sexes. LH denotes left hemisphere and RH denotes right hemisphere. The table includes the following: mutually significant anatomical region, surface area of cortical region of interest (ROI), p-value evaluating whether male and female linear fits are significantly different in slope and y-intercept, p-value evaluating whether a linear relationship occurs between average GMV and mJOA score in males, p-value evaluating whether a linear relationship occurs between average GMV and mJOA score in females, goodness of fit for males, and goodness of fit for females.

Discussion
The present study demonstrates significant sex-related differences in cortical volume in patients with degenerative cervical myelopathy. Prior to this investigation, the role of sex on brain structure in DCM remained largely understudied. Our findings may foster further investigation and understanding of the influence of sex and sex hormones on supraspinal plasticity following spinal cord injury.

Cortical Volumetric Differences in HCs Are Not Sex Dependent
The current study found no statistically significant differences in GMV between HC males and females. Literature investigating sex-related differences in cortical morphometry of the healthy brain remains controversial, with some studies reporting significant sexrelated differences in GMV and others citing no significant difference [30][31][32][33]. To address these inconsistencies, Sanchis-Segura et al. examined how the number, size, and direction of sex differences in regional GMV vary depending on how total intercranial volume (TIV) is statistically controlled; and they concluded that when TIV effects are properly accounted for, sex differences in GMV are relatively small in healthy adults [34].

Sex-Dependent Cortical Volumetric Differences in Patients
When investigating volumetric differences within patients, we found male patients exhibited larger GMV in various regions compared to female patients, including motor, language, and pain related cortices. Previous studies have revealed DCM patients exhibit functional and morphological alterations within primary motor and sensorimotor cortices when compared to age-matched HCs [11,35,36]. We suspect patients experience alterations in such brain regions due to hormonal, neuroinflammatory, and neuronal compensatory differences between sexes [23]. Preclinical studies of spinal cord injury (SCI), stroke, and traumatic brain injury (TBI) have shown sex steroids, particularly 17-estradiol, estrogen, progesterone, and testosterone, can provide neuroprotective, pro-myelination, and antiinflammatory effects resulting in improved tissue sparing and motor function [12][13][14][15][16][17][18].
In humans with acute traumatic SCI, administration of progesterone and vitamin D was associated with better functional recovery and outcome [17]. Interestingly, preclinical studies have shown testosterone treatment also provides neuroprotective benefits following SCI, but in the clinical setting about 43-57% of male patients experience low levels of testosterone following SCI, and low levels of testosterone were associated with severity of injury [14,22,37,38]. Sex-dependent volumetric differences observed within DCM patients and between patients and HCs may be driven by variations in hormone levels. In the male group, DCM patients exhibited larger GMV in regions involved in memory, vision, and language. Female patients exhibited fewer alterations than male patients when compared to healthy counterparts, a possible indication of the neuroprotective effects of normal or elevated progesterone and estrogen levels.
Furthermore, significant positive associations between GMV and mJOA scores were found in both male and female groups primarily across somatosensory and motor related cortical regions. Such findings are consistent with previous reports in which cortical alterations and cerebral reorganization were correlated with neurological function, proposing a compensatory relationship between cortical alterations and symptom progression in patients with cervical spondylosis [5,10,35]. A positive association between GMV and mJOA appears to confirm that patients with worsening neurological symptoms exhibit decreasing GMV across sensorimotor related cortices. Females exhibited an association between GMV and mJOA across a broader range of brain regions compared with male patients, including regions believed to be involved in pain processing [39]. Independent of mJOA, female patients consistently showed lower GMV than males within various regions involved in sensorimotor function. Our results reflect the possible influence of sex hormones on cerebral compensatory mechanisms and disease progression between males and females with DCM. Based on these novel preliminary studies, future investigations that evaluate supraspinal microstructural and functional alterations are warranted and will provide additional insight into the role of sex hormones in DCM neural plasticity.

Limitations and Future Direction
Although our patient and healthy control cohorts were well matched in terms of age and numbers of male and female subjects, it is important to note the healthy control subjects were acquired retrospectively from an image repository. Therefore, collection of age-and gender-matched HCs with brain and spinal cord imaging and mJOA testing is warranted for validating our findings and future studies. Furthermore, collection and inclusion of additional clinical and demographic information, such as handedness, disease duration, and medical comorbidities, will contribute to analyses of cortical structure in future studies. Additionally, measurement and assessment of serum sex hormones in relation to sex and neurological function would greatly benefit our understanding of the mechanisms underlying sexual dimorphism in DCM.

Conclusions
To the best of our knowledge this is the first study to investigate sex differences in cortical volume in patients with DCM. Results suggest males with DCM exhibit significantly larger GMV compared to female DCM patients in various brain regions, and DCM patients exhibit significant sex-related differences in the association between GMV and neurological function, particularly in brain areas involved in sensorimotor function.  Data Availability Statement: Data will be made available upon request from investigators.