Skip to main content
NCBI home page
VA Author Manuscripts logoLink to VA Author Manuscripts
. Author manuscript; available in PMC: 2021 Apr 1.
Published in final edited form as: Depress Anxiety. 2019 Sep 2;36(11):1047–1057. doi: 10.1002/da.22952

White Matter Integrity and Functional Predictors of Response to Repetitive Transcranial Magnetic Stimulation for Posttraumatic Stress Disorder and Major Depression

Jennifer Barredo 1,2,*, John A Bellone 1,2, Melissa Edwards 1, Linda L Carpenter 1,3, Stephen Correia 1,2,3, Noah S Philip 1,2,3
PMCID: PMC8015421  NIHMSID: NIHMS1681184  PMID: 31475432

Abstract

Background:

Recent evidence suggests that therapeutic repetitive transcranial magnetic stimulation (TMS) is an effective treatment for pharmacoresistant posttraumatic stress disorder (PTSD) and comorbid major depressive disorder (MDD). We recently demonstrated that response to 5Hz TMS administered to dorsolateral prefrontal cortex was predicted by functional connectivity of the medial prefrontal (MPFC) and subgenual anterior cingulate (sgACC). This functionally-defined circuit is a novel target for treatment optimization research, however our limited knowledge of the structural pathways that underlie this functional predisposition is a barrier to target engagement research.

Methods:

To investigate underlying structural elements of our previous functional connectivity findings, we submitted pre-TMS diffusion-weighted imaging data from 20 patients with PTSD and MDD to anatomically constrained tract-based probabilistic tractography (FreeSurfer’s TRACULA). Averaged pathway fractional anisotropy (FA) was extracted from four frontal white matter tracts: the forceps minor, cingulum, anterior thalamic radiations, and uncinate fasciculi. Tract FA statistics were treated as explanatory variables in backward regressions testing the relationship between tract integrity and functional connectivity coefficients from MPFC and sgACC predictors of symptom improvement after TMS.

Results:

FA in the anterior thalamic radiations was consistently associated with symptom improvement in PTSD and MDD (Bonferroni-corrected p<.05).

Conclusions:

We found that structural characteristics of the ATR account for significant variance in individual-level functional predictors of post-TMS improvement. TMS optimization studies should target this circuit either in stand-alone or successive TMS stimulation protocols.

Keywords: Stress Disorders, Post-Traumatic, Depressive Disorder, Major, Transcranial Magnetic Stimulation, Diffusion Magnetic Resonance Imaging, White Matter

Introduction

Over 50% of patients with posttraumatic stress disorder (PTSD) also suffer from major depressive disorder (MDD) (Flory & Yehuda, 2015; Rytwinski, Scur, Feeny, & Youngstrom, 2013). Unfortunately, first-line treatments are less effective in these patients (Bernardy & Friedman, 2015; Chase et al., 2017; Nixon & Nearmy, 2011). Given that long-term prognoses are generally poorer for patients with co-morbid presentations (Flory & Yehuda, 2015), the generation and optimization of new treatment alternatives is a high research priority.

Repetitive transcranial magnetic stimulation (hereafter, TMS) is a promising treatment for patients with resistant symptoms. In naturalistic patient samples, 30–40% of patients receiving TMS for depression achieve clinical remission (George, Taylor, & Short, 2013). Symptom reductions are durable, lasting up to one-year post-treatment (Carpenter et al., 2012; Dunner et al., 2014). TMS is also an emerging treatment for PTSD (Koek, Roach, Athanasiou, van ‘t Wout-Frank, & Philip, 2019), including comorbid PTSD+MDD (Carpenter et al., 2018; Philip, Ridout, Albright, Sanchez, & Carpenter, 2016).

TMS involves the administration of strong, focused electromagnetic pulses generated by a coil placed on the scalp. These pulses induce changes in local neural activity (Huerta & Volpe, 2009) and propagated signals modulate activity in distal regions and functional networks (Eldaief, Halko, Buckner, & Pascual-Leone, 2011). Dorsolateral prefrontal cortex (DLPFC) is the most common region targeted by TMS as a psychiatric treatment. DLPFC is a core region of the central executive network (CEN) that plays a critical role in “top-down” regulation of other functional brain networks. Hence, the rationale for targeting DLPFC originated from the goal of increasing executive control over symptoms of psychiatric illness.

Though TMS is clearly beneficial for many patients, optimizing TMS delivery may improve outcomes for those that do not achieve remission under TMS treatment-as-usual. For example, if function in a particular circuit differentiates responders from non-responders, better targeting of the underlying anatomical circuit may benefit treatment-resistant patients. Unfortunately, little is known about the structural and functional characteristics that predispose responsiveness to TMS.

Predictors of response to TMS for depression and/or PTSD have been examined in a limited number of functional magnetic resonance imaging (fMRI) resting-state connectivity studies. Several studies have reported that stronger pre-treatment functional connectivity of the subgenual anterior cingulate cortex (sgACC) with the default mode network (DMN), a functional network involved in self-referential thought (Andrews-Hanna, Reidler, Sepulcre, Poulin, & Buckner, 2010) and rumination (Hamilton, Farmer, Fogelman, & Gotlib, 2015). Stronger pre-treatment functional connectivity of the subgenual anterior cingulate cortex (sgACC) with the DMN is characteristic of patients that experience greater post-treatment improvement, though directionality (positive vs. negative connectivity) is inconsistent across studies (Baeken et al., 2014; Liston et al., 2014; Philip et al., 2018). In PTSD+MDD, stronger amygdala-to-MPFC functional connectivity at baseline predicted superior improvement (Philip et al., 2018). Other studies have found predictive differences in pre-TMS DLPFC-striatum connectivity (Avissar et al., 2017; Kang et al., 2016), as well as between CEN and salience processing regions (Philip, Barredo, Aiken, & Carpenter, 2017).

Though the above findings suggest a number of potential novel targets for TMS, in most cases, direct anatomical connections between these functional connectivity targets are absent or sparse. As such, it is unclear which pathways an optimized protocol should target in order to induce functional states associated with optimal symptom reduction. To address this question, we investigated the relationship between white matter integrity and functional predictors identified by our earlier study of 5Hz left DLPFC TMS for PTSD+MDD (Philip et al., 2018). Pre-treatment diffusion-weighted imaging collected from a subset of our original participants (n=20) was submitted to probabilistic tractography. Our original study found evidence of predictive functional connectivity relationships between amygdala, sgACC, medial prefrontal cortex (MPFC) and DLPFC. Accordingly, we restricted our analysis to four fronto-limbic white matter pathways: the anterior thalamic radiation (ATR), forceps minor, cingulum bundle, and the uncinate fasciculus. For the selected tracts, we tested whether white matter integrity at baseline would constrain the observation of patterns of functional connectivity predictive of subsequent treatment response (Figure 1).

Figure 1.

Figure 1.

Open in a new tab

Visualization of methodological approach for identifying structural associates of treatment-predictive functional connectivity relationships. In Philip et al. (Philip et al., 2018), we identified several patterns of functional connectivity that were predictive of later response to treatment (panel A). To identify the underlying structural circuits associated with these functional predictors, subject-level average weighted fractional anisotropy (FA) values (Yendiki et al., 2011) for each of our pathways of interest (panel B) were treated as explanatory variables in backward stepwise regressions. One regression model was run for each functional predictors (Table 2). For each of these models, the dependent variable was composed of individual-level beta values extracted from clusters for which baseline functional connectivity predicted later meaningful clinical improvement in PTSD or depression symptoms (Philip et al., 2018).

Methods

Participants

Participants (n = 21; age=52±10.3, female=10) were a subset of larger study cohort who underwent MRI scanning prior to receiving 5Hz TMS to the left DLPFC for PTSD+MDD (delivered at 120% of motor threshold, 3000–4000 pulses, for up to 40 sessions); this subset comprised all participants with usable DTI data (defined below). Functional imaging methods were previously reported by Philip et al. (Philip et al., 2018) and clinical trial details were described by Carpenter et al (Carpenter et al., 2018). All participants completed self-reported symptom severity measures both pre- and post-treatment. PTSD symptoms were measured with the PTSD Symptom Checklist for DSM-5 (PCL-5; (Weathers et al., 2013)). Depression severity was measured with the Inventory of Depressive Symptoms Self-Report (IDS-SR; (Rush, Gullion, Basco, Jarrett, & Trivedi, 1996)). Please see Table 1 for an overview of participant demographics.

Table 1. Demographics and Clinical Measures.

Descriptive data for participants grouped according to categorical improvement following a course of TMS treatments i.e., endpoint PCL-5 below 34 (Weathers et al., 2013) and IDS-SR score below 26 (Rush et al., 2006).

All Improved(PCL<34, IDS-SR<26) Not Improved(PCL>=34, IDS-SR>=26)
n 20 9 11
Age, mean (s.d.) 52.4 (10.3) 52.9 (13.3) 52.0 (7.7)
Females, n (%) 10 (50) 3 (33) 7 (64)
PCL
 Baseline, mean (s.d.) 49.8 (12.7) 44.4 (9.1) 54.1 (13.9)
 Endpoint, mean (s.d.) 35.6 (20.0) 17.6 (10.9) 50.3 (11.6)
 % change, mean (s.d.) 28.4 (36.9) 58.4 (27.1) 3.8 (22.8)
IDS-SR
 Baseline, mean (s.d.) 45.4 (12.3) 36.8 (10.0) 52.4 (9.2)
 Endpoint, mean (s.d.) 32.1 (18.9) 13.7 (6.2) 47.1 (9.8)
 % change, mean (s.d.) 32.6 (32.1) 62.0 (15.1) 8.7 (19.4)
Medications
 Antidepressants, n (%) 15 (75) 5 (25) 10 (50)
 Benzodiazapines, n (%) 6 (30) 1 (5) 5 (25)
 Anticonvulsants, n (%) 7 (35) 3 (15) 4 (20)
 Antipsychotics, n (%) 6 (30) 4 (20) 2 (10)
 Stimulants, n (%) 6 (30) 1 (5) 5 (25)
Open in a new tab

MRI data collection

Diffusion-weighted MRI, functional MRI, and T1-weighted structural images were acquired at the Brown University MRI Research Facility on a Siemens 3T Prisma MRI scanner (Siemens Corp., Erlangen, Germany) equipped with a 32-channel head coil. For each participant, we collected a high-resolution T1-weighted anatomical image (voxel size=1.0 mm3, TR=1900 ms, TE=2.98 ms, FOV=256 mm2) and a 12-minute, 64 direction, diffusion-weighted echo-planar imaging scan (voxel size=1.8 mm3, slices=76, TR=10200 ms, TE=103.0 ms, b=1000 s/mm2, b0=12).

Diffusion-weighted MRI preprocessing and probabilistic tractography

Diffusion image preprocessing was carried out with FSL (Jenkinson, Beckmann, Behrens, Woolrich, & Smith, 2012) programs run through the TRActs Constrained by UnderLying Anatomy (TRACULA) protocol (Yendiki et al., 2011) included in Freesurfer (version 5.3.0) (Fischl, 2012). Two raters (JAB and ME) conducted visual quality assurance for all diffusion-weighted images and tractography reconstructions. One subject was excluded because of excessive motion and incomplete streamline reconstructions. The following steps were applied to all diffusion-weighted images: 1) affine registration to the first non-diffusion-weighted, or b0, image (Jenkinson, Bannister, Brady, & Smith, 2002), 2) eddy current correction (Rohde, Barnett, Basser, Marenco, & Pierpaoli, 2004), and 3) vector reorientation (Leemans & Jones, 2009). Subjects’ average b0 image was then registered to their T1-weighted anatomical images via boundary-based registration, diffusion data were then transformed to Montreal Neurological Institute (MNI) Atlas space (Greve & Fischl, 2009).

The TRACULA routine produces eighteen white matter fascicles reconstructed from each individual’s diffusion data. TRACULA employs a “ball-and-stick” diffusion model (Behrens, Berg, Jbabdi, Rushworth, & Woolrich, 2007) to estimate local fiber orientations during fiber reconstructions. Fibers are then assigned to tracts using an algorithm trained on anatomical priors (Yendiki et al., 2011). The weighted pathway averages (Yendiki et al., 2011) of white matter integrity metrics generated by TRACULA (fractional anisotropy, FA; mean diffusivity, MD; radial diffusivity, RD, axial diffusivity, AD) were then extracted from four frontal white matter pathways: the forceps minor, anterior thalamic radiations (ATR), cingulum, and uncinate fasciculi. The forceps minor and ATR were selected because these pathways provide the anatomical means for the DLPFC to interact with executive and salience regions within the prefrontal cortex, whereas the cingulum and uncinate connect limbic and default network regions to the prefrontal cortex. Though FA was our primary metric of interest, MD, RD, and AD were extracted to permit further exploration of tract characteristics.

Treatment of subject motion

Because subject motion can impact tractography accuracy (Yendiki, Koldewyn, Kakunoori, Kanwisher, & Fischl, 2014), we computed a total motion index (TMI) for each subject following Yendiki et al. (Yendiki et al., 2014). TMI was used in bivariate correlations to evaluate if it should be included in regressions as a confounding variable and was used in post-hoc testing of potential residual effects of motion.

Statistical analyses

All statistical analyses were performed using SPSS Statistics 24 (IBM Corporation, Armonk, New York, USA). Prior to hypothesis testing, we assessed whether potential confounding variables of age, sex, estimated total intracranial volume (eTIV; (Buckner et al., 2004)), total motion index (TMI), and signal-to-noise ratio influenced FA in any of our pathways of interest. Confound effects for continuous variables (age, eTIV, TMI, signal-to-noise) were tested using bivariate correlations; the categorical variable of sex was tested using a one-way analysis of variance (ANOVA). All tests were not significant (ps>0.1), however we did exclude one participant for excessive motion (TMI>3.5 SD) (Yendiki et al., 2014).

Backward stepwise regressions were used to evaluate whether FA in fronto-limbic pathways accounted for significant variance in functional predictors of TMS outcome. One regression model was run for each of the functional predictors listed in Table 2. For each of these models, the dependent variable was composed of individual-level functional predictor beta values from our earlier study (Philip et al., 2018) (Figure 1). These beta values were extracted from clusters where a statistically significant association between functional connectivity at baseline and meaningful clinical improvement in PTSD or depression symptoms after TMS was previously observed (Philip et al., 2018) (i.e., functional connectivity predictors of clinical response). Cluster statistics from our earlier study (Philip et al., 2018) are available in Table 2. Subject-level average weighted FA values (Yendiki et al., 2011) for each of our pathways of interest were treated as explanatory variables in these regressions. Regression models were considered significant at p<.05, however associations between functional predictors and individual tracts were only considered significant at coefficient p-values<.013 (i.e., Bonferroni-corrected for four tracts of interest). As neighboring tracts may share variance due to nuisance factors i.e., similarity in local signal-to-noise ratios or misclassified voxels, we examined both tolerance and variance inflation factors (VIF; reciprocal of tolerance) as a check for multi-collinearity. Coefficient t-tests were only considered meaningful if VIF<3 (Hair, Anderson, Tatham, & Black, 1995), coefficient t-tests were only considered meaningful at VIF<3. We found no evidence of significant multi-collinearity in any of our regression models.

Table 2.

Seed-to-voxel functional connectivity clusters at baseline associated with symptom improvement after treatment from Philip et al. (Philip et al., 2018).

Scale Seed-to-Cluster Pair Peak MNI Coordinate Cluster
Size		 Cluster p-FDR
PCL L. Amygdala-L. MPFC −20 +54 +16 723 <.001
R. Amygdala-L. MPFC −14 +60 +10 363 <.001
R. Amygdala-VMPFC −16 +22 –8 226 <.001
R. sgACC-R. Lat. PFC 16 +2 +58 151 <.005
R. sgACC-L. Lat. PFC −20 +8 +54 143 <.005
IDSSR L. Amygdala-L. MPFC −22 +54 +16 349 <.001
L. Amygdala-L. DLPFC −26 +30 +46 60 <.001
R. Amygdala-L. MPFC −14 +62 –12 138 <.001
Open in a new tab

Abbreviations: MNI, Montreal neurological institute; FDR, false-discovery rate; L, left; R, right; MPFC, medial prefrontal cortex; PCL, PTSD checklist for DSM-5; IDSSR, inventory of depressive symptomatology, self-report; DLPFC, dorsolateral prefrontal cortex; VMPFC, ventromedial prefrontal cortex; Lat, lateral; PFC, prefrontal cortex

For significant models, we ran additional exploratory regressions substituting MD, RD, and AD for FA as the explanatory variable. FA tends to be the most sensitive measurement of general integrity among these metrics, but it cannot parse the potential source of these differences (e.g. axonal vs. membrane-related). These additional exploratory regressions were conducted to further characterize the nature of the underlying white matter.

Results

White matter associations with functional predictors - PTSD

Our previous study found that positive functional connectivity between left amygdala and left MPFC was stronger in participants exhibiting greater improvement after TMS (Philip et al., 2018). Here, backward regressions indicated that FA in bilateral ATR accounted for significant variance in left amygdala-to-left MPFC functional predictors (Adjusted R2=27.1%, F(2,19)=4.53, p<.05), though only the right ATR coefficient survived Bonferroni correction (Table 3). Exploratory post-hoc tests were not significant.

Table 3. White Matter pathways associated with imaging predictors of PTSD symptom response (PCL) to TMS.

Only statistics from regression models where tract FA accounted for significant variance in functional predictors are included. Coefficient betas, Student’s t-values, two-tailed p-values, are listed in the first three columns. P-values in the table are uncorrected; those surviving correction are denoted with asterisks. The fourth and fifth columns correspond to the categorical groups of Improved (n=9) and Not Improved (n=11), respectively. Improvement was defined as a total PCL-5 score <34 and total IDS-SR score <26. Each column contains the mean and standard error in functional connectivity between predictor regions (in units of Pearson’s r), as well as the correlation between functional connectivity and tract FA (Pearson’sr).

b t p-value Improved, FC mean (s.e.), FC-FA r Not Improved, FC mean (s.e.), FC-FA r
L. Amygdala-to-L. MPFC
 L. Anterior Thalamic Radiations −0.77 −2.88 0.022 0.18 (0.09), −0.41 0.12 (0.03), 0.08
 R. Anterior Thalamic Radiations 0.91 3.01 0.008* 0.18 (0.09), 0.48 0.12 (0.03), 0.09
R. Amygdala-to-L. MPFC
 L. Anterior Thalamic Radiations −0.87 −2.82 0.012* 0.14 (0.1), −0.26 0.06 (0.04), −0.31
 R. Anterior Thalamic Radiations 0.85 2.77 0.013* 0.14 (0.1), 0.50 0.06 (0.04), −0.26
R. sgACC-to-L. PFC
 Forceps Minor −0.6 −2.68 0.017 −0.08 (0.08), −0.54 0.06 (0.05), −0.15
 L. Anterior Thalamic Radiations 0.96 3.02 0.008* −0.08 (0.08), 0.34 0.06 (0.05), −0.04
 R. Anterior Thalamic Radiations −0.73 −2.54 0.022 −0.08 (0.08), −0.16 0.06 (0.05), −0.33
Open in a new tab
*

Significant at corrected p<.05

Abbreviations: PCL, PTSD checklist for DSM-5; TMS, repetitive transcranial magnetic stimulation; b, coefficient beta; t, coefficient Student’s t score; FA, fractional anisotropy; FC, functional connectivity; s.e., standard error; L, left; R, right; MPFC, medial prefrontal cortex; sgACC, subgenual anterior cingulate cortex; PFC, prefrontal cortex.

In Philip et al. (2018), stronger positive functional connectivity between right amygdala and left MPFC at baseline also distinguished participants experiencing greater post-treatment improvement. The final regression model found evidence of a significant relationship between FA in ATR FA and right amygdala-to-left MPFC functional connectivity predictors (Adjusted R2=26.5%, F(2,19)=4.43, p<.05) (Figure 2). Both the left ATR negative tract coefficient and right ATR positive coefficient were significant after multiple comparisons correction (Table 3). Post-hoc exploratory testing yielded a significant effect of RD in ATR on functional connectivity (Adjusted R2=21.7%, F(2,19)=3.63, p<.05), however, individual tract coefficients were only significant at uncorrected alpha levels (left ATR: t=2.3, p=.03; right ATR: t=−2.7, p=.02).

Figure 2.

Figure 2.

Open in a new tab

Fractional anisotropy in the anterior thalamic radiations is associated with amygdala-to-left MPFC functional connectivity predictors of PTSD treatment response to TMS. Images in panels A, D, and E are from one representative subject. Brain images are rendered in neurological orientation. A. Distribution of tract endpoints from the anterior thalamic radiations. Endpoints have been smoothed to enhance visibility. B. Left MPFC cluster where pre-treatment functional connectivity with left basolateral amygdala was predictive of responsiveness of PTSD symptoms to TMS. C. Left MPFC cluster where pre-treatment functional connectivity with right basolateral amygdala was predictive of responsiveness of PTSD symptoms to TMS. D. Sagittal view of the tract probability distribution from the right anterior thalamic radiation. E. Axial view of tract probability distributions from the left and right anterior thalamic radiations. F. Schematic summarizing the relationship between fractional anisotropy and neural correlates of symptom response. Fractional anisotropy in the left hemisphere was negatively associated with functional predictors, whereas fractional anisotropy in the right hemisphere was positively associated with PTSD symptom improvement. Abbreviations: MPFC, medial prefrontal cortex; PTSD, posttraumatic stress disorder; TMS, repetitive transcranial magnetic stimulation.

In our earlier study of 5Hz TMS for PTSD+MDD, participants experiencing greater improvement in PTSD symptoms after treatment exhibited stronger negative functional connectivity between right sgACC and left lateral PFC at baseline (Figure 3). In the current study, backward regression converged on a model in which FA in the forceps minor and bilateral ATR accounted for significant variance in this sgACC functional predictor (Adjusted R2=34.8%, F(2,19)=4.38, p=.02). Only the left ATR tract coefficient was significant at the corrected threshold (Table 3). Post-hoc tests were not significant.

Figure 3.

Figure 3.

Open in a new tab

Fractional anisotropy in the anterior thalamic radiations and forceps minor is associated with subgenual-to-lateral PFC functional connectivity predictors of PTSD treatment response to TMS. Image in panel C is from one representative subject. Brain images are rendered in neurological orientation. A-B. Lateral PFC clusters where pre-treatment functional connectivity with right subgenual ACC was predictive of responsiveness of PTSD symptoms to TMS. C. Distribution of tract endpoints (blue) from probabilistic reconstruction of the forceps minor (red). Endpoints have been smoothed to enhance visibility. D. Schematic summarizing the relationship between fractional anisotropy and neural correlates of meaningful improvement in PTSD symptoms. Fractional anisotropy in the left hemisphere was positively associated with functional predictors, whereas fractional anisotropy in the right hemisphere and forceps minor was negatively associated with functional connectivity predictors. Abbreviations: PFC, prefrontal cortex; ACC, anterior cingulate cortex; PTSD, posttraumatic stress disorder; TMS, repetitive transcranial magnetic stimulation.

In the case of the amygdala-to-VMPFC and right sgACC-to-right lateral PFC functional predictors, backward regressions did not converge on a significant model (all p-values>.1). For all functional predictors of PTSD improvement, exploratory post-hoc models using MD or AD as the explanatory variables were not significant (all ps>.1).

White matter associations with functional predictors - depression

In Philip et al. (2018), left amygdala-to-left MPFC positive functional connectivity at baseline was stronger in participants that met remission criteria at the end of treatment. Here, backward regressions converged on a model in which FA in bilateral ATR FA was marginally associated with left amygdala-to-left MPFC functional connectivity (Adjusted R2=20.0%, F(2,19)=3.38, p=.058). Tract coefficients were significant at uncorrected, but not corrected thresholds (Table 4). Exploratory post-hoc tests were not significant.

Table 4. White Matter pathways associated with imaging predictors of depression symptom response (IDSSR) to TMS.

Only statistics from regression models where tract FA accounted for significant variance in functional predictors are included. Coefficient betas, Student’s t-values, two-tailed p-values, are listed in the first three columns. P-values in the table are uncorrected; those surviving correction are denoted with asterisks. The fourth and fifth columns correspond to the categorical groups of Improved (n=9) and Not Improved (n=11), respectively. Improvement was defined as a total PCL-5 score <34 and total IDS-SR score <26. Each column contains the mean and standard error in functional connectivity between predictor regions (in units of Pearson’s r), as well as the correlation between functional connectivity and tract FA (Pearson’s r).

b t p-value Improved, FC mean (s.e.), FC-FA r Non-Improved, FC mean (s.e.), FC-FA r
L. Amygdala-to-L. MPFC
 L. Anterior Thalamic Radiations −0.69 −2.19 0.043 0.18 (0.09), −0.35 0.12 (0.03), −0.31
 R. Anterior Thalamic Radiations 0.82 2.58 0.02 0.18 (0.09), 0.33 0.12 (0.03), −0.12
R. Amygdala-to-L. MPFC
 L. Anterior Thalamic Radiations −0.85 −2.75 0.014 0.14 (0.1), −0.33 0.06 (0.04), −0.01
 R. Anterior Thalamic Radiations 0.77 2.47 0.025 0.14 (0.1), 0.58 0.06 (0.04), −0.18
Open in a new tab
*

Significant at corrected p<.05

Abbreviations: FA, fractional anisotropy; IDSSR, inventory of depressive symptomatology, self-report; TMS, repetitive transcranial magnetic stimulation; FA, fractional anisotropy; FC, functional connectivity; s.e., standard error; L, left; R, right; MPFC, medial prefrontal cortex.

In our earlier study, positive functional connectivity between right amygdala and left MPFC at baseline was observed in participants that achieved remission at the end of treatment. The final regression model indicated that bilateral ATR FA also accounted for significant variance in this functional predictor (Adjusted R2=23.6%, F(2,19)=3.93, p<.05). Tract coefficients were significant at uncorrected thresholds. Post-hoc tests revealed a significant relationship between RD and functional predictors (Adjusted R2=25.5%, F(2,19)=4.25, p<.05). This relationship was negative and significant at corrected levels in right ATR (t=−2.89, p=.01), but the positive relationship in left ATR did not survive correction (t=2.50, p=.02). Post-hoc tests using MD or AD as explanatory variables were not significant (all ps>.1).

The backward regression examining FA relationships to amygdala-to-DLPFC functional predictors did not converge on a significant model (p>.1).

Discussion

To our knowledge, this is the first study to investigate the relationship between functional predictors of response to TMS for PTSD+MDD and white matter integrity. TMS is a promising treatment for patients with treatment-resistant PTSD+MDD. TMS has demonstrated effects on specific functional circuits, and the structural integrity of those circuits may place limitations on symptom response. Here, we identified structural correlates of previously-defined functional connectivity predictors. This structurally and functionally-defined circuit is a robust, novel target for future treatment optimization research.

In this study, we assessed the relationship between FA in four fronto-limbic white matter pathways and functional connectivity predictors identified in our previous published work (Philip et al., 2018). Our regression models found a significant association between variance in ATR integrity and amygdala-to-MPFC functional connectivity relationships predictive of later symptom response. Importantly, this association was significant for functional predictors of both PTSD and depression improvement.

The ATR projects throughout frontal cortex (Morecraft, Geula, & Mesulam, 1992; Romanski, Giguere, Bates, & Goldman-Rakic, 1997; Shah, Jhawar, & Goel, 2012) and is part of the cortico-striatal-thalamo-cortical (CSTS) system of white matter projections facilitating information transfer between distributed brain regions, including amygdala and prefrontal cortex (G. E. Alexander, DeLong, & Strick, 1986; Haber, 2016). Low FA in the ATR was associated with symptom severity across emotional disorders in a recent meta-analysis (Jenkins et al., 2016). Moreover, ATR terminations in orbitofrontal cortex (OFC) and MPFC are substantial (Goldman-Rakic & Porrino, 1985; O’Muircheartaigh, Keller, Barker, & Richardson, 2015) and aberrant MPFC/OFC function is a core feature of many psychiatric and emotional disorders (Drevets, 2007; Fettes, Schulze, & Downar, 2017; Jenkins et al., 2016). CSTC and MPFC/OFC dynamics may also play a central role in the development of psychopathology (Fettes et al., 2017; Goodkind et al., 2015). Our results linking the ATR, a component of the CSTC, to a MPFC functional predictor underscores its role in psychopathology and recommends this circuit should be a focus TMS protocol optimization research.

Though the role of thalamus and ATR in PTSD and depression treatment response is understudied, our results and those of several recent studies, suggest they deserve additional consideration. For example, FA in the neighboring stria terminalis has been shown to predict antidepressant response (Korgaonkar, Williams, Song, Usherwood, & Grieve, 2014). A small sample (n=13) fMRI study found that activation during response inhibition in thalamus and arousal regions predicted poor response to cognitive behavioral therapy for PTSD (Falconer, Allen, Felmingham, Williams, & Bryant, 2013). Baseline levels of cortico-thalamic, as well as cortico-striatal functional connectivity have also been shown to predict response to dorsomedial prefrontal targeted TMS for depression (Salomons et al., 2014). Finally, negative correlations between thalamic cerebral blood flow and treatment outcomes have been observed with positron emission tomography (Mottaghy et al., 2002), and single-photon emission computerized tomography has found evidence of thalamic hypoperfusion in nonresponders to antidepressant therapy (Richieri et al., 2011).

The observed left-right asymmetry in FA associations was unanticipated. Generally, right ATR FA and functional connectivity predictors were correlated positively, whereas correlations were negative in the case of left ATR. We speculate that this laterality may reflect some of the neurocognitive characteristics of PTSD (Vasterling & Arditte Hall, 2018).

We also observed negative correlations between forceps minor FA and functional connectivity predictors of PTSD symptom improvement in DLPFC, although observed relationships were only marginally significant after correction for multiple comparisons. FA throughout frontal white matter tracts including the forceps minor is reported to be lower in individuals with chronic or treatment-resistant depression, compared to those more responsive to treatment (de Diego-Adelino et al., 2014). Low FA in parts of the forceps minor and ATR is also present in PTSD (Olson et al., 2017), underscoring the transdiagnostic impact of this region.

Of note, significant relationships between FA in the cingulum and functional connectivity predictors were not observed in this study, despite its prior association with response to antidepressant medication in MDD and exposure-based psychotherapy in PTSD (Kennis et al., 2015; Korgaonkar et al., 2014). Uncinate fasciculus integrity was also not associated with our functional predictors. This was unexpected given its prior implication in PTSD (Admon et al., 2013; Koch et al., 2017). While negative results should be interpreted with caution, tract associations with responses to specific therapies (e.g. TMS vs. antidepressant) supports the use of structural imaging as a component of decision algorithms to deliver individualized treatment (Korgaonkar et al., 2014).

For completeness, we also note that several of our exploratory regression models examining radial diffusivity (RD) were also significant. These analyses were conducted because though FA is highly sensitive to general, non-specific differences in microstructure, it yields little information about the potential origin of differences in tract integrity (e.g. myelination, axonal injury, cellularity) (A. L. Alexander, Lee, Lazar, & Field, 2007). Though we cannot draw strong conclusions from our exploratory analyses, our observation that RD, but not AD or MD, was associated with functional connectivity predictors, suggests that differences in anisotropy may stem from myelination (Song et al., 2002).

Limitations

Limitations include those inherent to studies of modest sample size, cross-sectional neuroimaging, and probabilistic tractography. For instance, probability estimates are less accurate in areas rich in crossing fibers and in voxels at the interface of gray and white matter due to partial-volume effects. While the estimation of two principle components that FSL’s “bedpostx” algorithm allows is an improvement on “winner-take-all” single tensor models, ambiguous estimates in these regions remains persistent. We also acknowledge that while 5Hz TMS to DLPFC has been recently studied in this patient population, other frequencies and targets are in clinical use and it remains unknown whether our results will generalize to other approaches. Furthermore, because participants had comorbid diagnoses, we cannot completely disambiguate whether observed results are truly unique to categorical diagnoses. Yet, the convergence of findings underlying both symptom domains provides insight about shared neuroanatomical pathways associated with the predisposition for clinical improvement. Our results suggest that the ATR relationship to amygdala-to-MPFC functional predictors is transdiagnostic, however we acknowledge that in the case of depression, this association was marginally significant. We also note that in the absence of a sham control group, we are unable to separate predictors of TMS response from potential placebo effects. We also acknowledge that while the restriction of our analysis to four paths of interest was reasonable given our a priori hypotheses, limiting our analytic scope may also ignore other important polysynaptic relationships.

Conclusions

To our knowledge, this is the first study to examine the underlying structural components of a functional predictor of response to 5Hz TMS for PTSD+MDD. Using diffusion imaging and probabilistic tractography, we found that structural characteristics of the ATR account for significant variance in individual-level functional connectivity relationships that are predictive of post-treatment improvement. Importantly, this finding was consistent across disorders examined. Future treatment optimization studies should investigate whether targeting this circuit either in stand-alone or successive TMS stimulation protocols can enhance therapeutic outcomes.

Data Availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Acknowledgements

This study was funded in part by an investigator-initiated grant from Neuronetics, Inc. to Butler Hospital (Drs. Carpenter and Philip). Dr. Philip has also received grant support from Neosync and Janssen through clinical trial contracts, and has been an unpaid scientific advisory board member for Neuronetics. Dr. Carpenter is a consultant for Magstim, Feelmore Labs, Nexstim, and Janssen. Other authors report no biomedical conflicts of interest.

This study was supported by U.S. Veterans Affairs grants I01 RX002450 (NSP), IK2 CX001824 (JB), and the VA RR&D Center for Neurorestoration and Neurotechnology at the Providence VA Medical Center. The funders had no role in the conduct of the study, manuscript preparation or the decision to submit for publication. The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the Department of Veterans Affairs.

Conflicts of Interest

Dr. Barredo reports grants from Veterans Affairs Clinical Science Research & Development (1IK2CX001824-01A1) and Veterans Affairs Rehabilitation Research & Development (1I50RX002864-01) during the conduct of the study. Dr. Bellone has nothing to disclose. Dr. Edwards has nothing to disclose. Dr. Correia has nothing to disclose. Dr. Carpenter reports an investigator-initiated grant from Neuronetics supporting the conduct of the study; personal fees for consulting from Magstim, research equipment support from Nexstim, grants from Feelmore Labs, grants and personal fees for consulting from Janssen, grants from Neosync, and research equipment from Neuronetics, outside the submitted work. This study was funded in part by an investigator-initiated grant from Neuronetics, Inc. to Butler Hospital (Drs. Carpenter and Philip). Dr. Philip has also received grant support from Neosync and Janssen through clinical trial contracts, and has been an unpaid scientific advisory board member for Neuronetics.

References

  1. Admon R, Leykin D, Lubin G, Engert V, Andrews J, Pruessner J, & Hendler T (2013). Stress-induced reduction in hippocampal volume and connectivity with the ventromedial prefrontal cortex are related to maladaptive responses to stressful military service. Hum Brain Mapp, 34(11), 2808–2816. doi: 10.1002/hbm.22100 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Alexander AL, Lee JE, Lazar M, & Field AS (2007). Diffusion tensor imaging of the brain. Neurotherapeutics, 4(3), 316–329. doi: 10.1016/j.nurt.2007.05.011 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Alexander GE, DeLong MR, & Strick PL (1986). Parallel organization of functionally segregated circuits linking basal ganglia and cortex. Annual review of neuroscience, 9(1), 357–381. [DOI] [PubMed] [Google Scholar]
  4. Andrews-Hanna JR, Reidler JS, Sepulcre J, Poulin R, & Buckner RL (2010). Functional-anatomic fractionation of the brain’s default network. Neuron, 65(4), 550–562. doi: 10.1016/j.neuron.2010.02.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Avissar M, Powell F, Ilieva I, Respino M, Gunning FM, Liston C, & Dubin MJ (2017). Functional connectivity of the left DLPFC to striatum predicts treatment response of depression to TMS. Brain Stimul, 10(5), 919–925. doi: 10.1016/j.brs.2017.07.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Baeken C, Marinazzo D, Wu GR, Van Schuerbeek P, De Mey J, Marchetti I, … De Raedt R (2014). Accelerated HF-rTMS in treatment-resistant unipolar depression: Insights from subgenual anterior cingulate functional connectivity. World J Biol Psychiatry, 15(4), 286–297. doi: 10.3109/15622975.2013.872295 [DOI] [PubMed] [Google Scholar]
  7. Behrens TE, Berg HJ, Jbabdi S, Rushworth MF, & Woolrich MW (2007). Probabilistic diffusion tractography with multiple fibre orientations: What can we gain? Neuroimage, 34(1), 144–155. doi: 10.1016/j.neuroimage.2006.09.018 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Bernardy NC, & Friedman MJ (2015). Psychopharmacological strategies in the management of posttraumatic stress disorder (PTSD): what have we learned? Curr Psychiatry Rep, 17(4), 564. doi: 10.1007/s11920-015-0564-2 [DOI] [PubMed] [Google Scholar]
  9. Buckner RL, Head D, Parker J, Fotenos AF, Marcus D, Morris JC, & Snyder AZ (2004). A unified approach for morphometric and functional data analysis in young, old, and demented adults using automated atlas-based head size normalization: reliability and validation against manual measurement of total intracranial volume. Neuroimage, 23(2), 724–738. doi: 10.1016/j.neuroimage.2004.06.018 [DOI] [PubMed] [Google Scholar]
  10. Carpenter LL, Conelea C, Tyrka AR, Welch ES, Greenberg BD, Price LH, … Philip NS (2018). 5Hz Repetitive transcranial magnetic stimulation for posttraumatic stress disorder comorbid with major depressive disorder. J Affect Disord, 235, 414–420. doi: 10.1016/j.jad.2018.04.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Carpenter LL, Janicak PG, Aaronson ST, Boyadjis T, Brock DG, Cook IA, … Demitrack MA (2012). Transcranial magnetic stimulation (TMS) for major depression: a multisite, naturalistic, observational study of acute treatment outcomes in clinical practice. Depress Anxiety, 29(7), 587–596. doi: 10.1002/da.21969 [DOI] [PubMed] [Google Scholar]
  12. Chase HW, Fournier JC, Bertocci MA, Greenberg T, Aslam H, Stiffler R, … Phillips ML (2017). A pathway linking reward circuitry, impulsive sensation-seeking and risky decision-making in young adults: identifying neural markers for new interventions. Translational psychiatry, 7(4), e1096. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. de Diego-Adelino J, Pires P, Gomez-Anson B, Serra-Blasco M, Vives-Gilabert Y, Puigdemont D, … Portella M (2014). Microstructural white-matter abnormalities associated with treatment resistance, severity and duration of illness in major depression. Psychological medicine, 44(6), 1171–1182. [DOI] [PubMed] [Google Scholar]
  14. Drevets WC (2007). Orbitofrontal cortex function and structure in depression. Ann N Y Acad Sci, 1121, 499–527. doi: 10.1196/annals.1401.029 [DOI] [PubMed] [Google Scholar]
  15. Dunner DL, Aaronson ST, Sackeim HA, Janicak PG, Carpenter LL, Boyadjis T, … Demitrack MA (2014). A multisite, naturalistic, observational study of transcranial magnetic stimulation for patients with pharmacoresistant major depressive disorder: durability of benefit over a 1-year follow-up period. J Clin Psychiatry, 75(12), 1394–1401. doi: 10.4088/JCP.13m08977 [DOI] [PubMed] [Google Scholar]
  16. Eldaief MC, Halko MA, Buckner RL, & Pascual-Leone A (2011). Transcranial magnetic stimulation modulates the brain’s intrinsic activity in a frequency-dependent manner. Proc Natl Acad Sci U S A, 108(52), 21229–21234. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Falconer E, Allen A, Felmingham KL, Williams LM, & Bryant RA (2013). Inhibitory neural activity predicts response to cognitive-behavioral therapy for posttraumatic stress disorder. J Clin Psychiatry, 74(9), 895–901. doi: 10.4088/JCP.12m08020 [DOI] [PubMed] [Google Scholar]
  18. Fettes P, Schulze L, & Downar J (2017). Cortico-Striatal-Thalamic Loop Circuits of the Orbitofrontal Cortex: Promising Therapeutic Targets in Psychiatric Illness. Frontiers in Systems Neuroscience, 11, 25. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Fischl B (2012). FreeSurfer. Neuroimage, 62(2), 774–781. doi: 10.1016/j.neuroimage.2012.01.021 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Flory JD, & Yehuda R (2015). Comorbidity between post-traumatic stress disorder and major depressive disorder: alternative explanations and treatment considerations. Dialogues Clin Neurosci, 17(2), 141–150. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. George MS, Taylor JJ, & Short EB (2013). The expanding evidence base for rTMS treatment of depression. Curr Opin Psychiatry, 26(1), 13–18. doi: 10.1097/YCO.0b013e32835ab46d [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Goldman-Rakic PS, & Porrino LJ (1985). The primate mediodorsal (MD) nucleus and its projection to the frontal lobe. J Comp Neurol, 242(4), 535–560. doi: 10.1002/cne.902420406 [DOI] [PubMed] [Google Scholar]
  23. Goodkind M, Eickhoff SB, Oathes DJ, Jiang Y, Chang A, Jones-Hagata LB, … Etkin A (2015). Identification of a common neurobiological substrate for mental illness. JAMA Psychiatry, 72(4), 305–315. doi: 10.1001/jamapsychiatry.2014.2206 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Greve DN, & Fischl B (2009). Accurate and robust brain image alignment using boundary-based registration. Neuroimage, 48(1), 63–72. doi: 10.1016/j.neuroimage.2009.06.060 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Haber SN (2016). Corticostriatal circuitry. Dialogues Clin Neurosci, 18(1), 7–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Hair JFJ, Anderson RE, Tatham RL, & Black WC (1995). Multivariate Data Analysis (3 ed.). New York: Macmillan. [Google Scholar]
  27. Hamilton JP, Farmer M, Fogelman P, & Gotlib IH (2015). Depressive Rumination, the Default-Mode Network, and the Dark Matter of Clinical Neuroscience. Biol Psychiatry, 78(4), 224–230. doi: 10.1016/j.biopsych.2015.02.020 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Huerta PT, & Volpe BT (2009). Transcranial magnetic stimulation, synaptic plasticity and network oscillations. J Neuroeng Rehabil, 6, 7. doi: 10.1186/1743-0003-6-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Jenkins LM, Barba A, Campbell M, Lamar M, Shankman SA, Leow AD, … Langenecker SA (2016). Shared white matter alterations across emotional disorders: A voxel-based meta-analysis of fractional anisotropy. Neuroimage Clin, 12, 1022–1034. doi: 10.1016/j.nicl.2016.09.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Jenkinson M, Bannister P, Brady M, & Smith S (2002). Improved optimization for the robust and accurate linear registration and motion correction of brain images. Neuroimage, 17(2), 825–841. [DOI] [PubMed] [Google Scholar]
  31. Jenkinson M, Beckmann CF, Behrens TE, Woolrich MW, & Smith SM (2012). Fsl. Neuroimage, 62(2), 782–790. doi: 10.1016/j.neuroimage.2011.09.015 [DOI] [PubMed] [Google Scholar]
  32. Kang JI, Lee H, Jhung K, Kim KR, An SK, Yoon K-J, … Lee E (2016). Frontostriatal connectivity changes in major depressive disorder after repetitive transcranial magnetic stimulation: a randomized sham-controlled study. Journal of clinical psychiatry, 77(9), e1137–e1143. [DOI] [PubMed] [Google Scholar]
  33. Kennis M, van Rooij SJ, Tromp do PM, Fox AS, Rademaker AR, Kahn RS, … Geuze E (2015). Treatment Outcome-Related White Matter Differences in Veterans with Posttraumatic Stress Disorder. Neuropsychopharmacology, 40(10), 2434–2442. doi: 10.1038/npp.2015.94 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Koch SBJ, van Zuiden M, Nawijn L, Frijling JL, Veltman DJ, & Olff M (2017). Decreased uncinate fasciculus tract integrity in male and female patients with PTSD: a diffusion tensor imaging study. J Psychiatry Neurosci, 42(5), 331–342. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Koek RJ, Roach J, Athanasiou N, van ‘t Wout-Frank M, & Philip NS (2019). Neuromodulatory treatments for post-traumatic stress disorder (PTSD). Prog Neuropsychopharmacol Biol Psychiatry, 92, 148–160. doi: 10.1016/j.pnpbp.2019.01.004 [DOI] [PubMed] [Google Scholar]
  36. Korgaonkar MS, Williams LM, Song YJ, Usherwood T, & Grieve SM (2014). Diffusion tensor imaging predictors of treatment outcomes in major depressive disorder. Br J Psychiatry, 205(4), 321–328. doi: 10.1192/bjp.bp.113.140376 [DOI] [PubMed] [Google Scholar]
  37. Leemans A, & Jones DK (2009). The B-matrix must be rotated when correcting for subject motion in DTI data. Magn Reson Med, 61(6), 1336–1349. doi: 10.1002/mrm.21890 [DOI] [PubMed] [Google Scholar]
  38. Liston C, Chen AC, Zebley BD, Drysdale AT, Gordon R, Leuchter B, … Dubin MJ (2014). Default mode network mechanisms of transcranial magnetic stimulation in depression. Biol Psychiatry, 76(7), 517–526. doi: 10.1016/j.biopsych.2014.01.023 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Morecraft RJ, Geula C, & Mesulam MM (1992). Cytoarchitecture and neural afferents of orbitofrontal cortex in the brain of the monkey. J Comp Neurol, 323(3), 341–358. doi: 10.1002/cne.903230304 [DOI] [PubMed] [Google Scholar]
  40. Mottaghy FM, Keller CE, Gangitano M, Ly J, Thall M, Parker JA, & Pascual-Leone A (2002). Correlation of cerebral blood flow and treatment effects of repetitive transcranial magnetic stimulation in depressed patients. Psychiatry Res, 115(1–2), 1–14. [DOI] [PubMed] [Google Scholar]
  41. Nixon RD, & Nearmy DM (2011). Treatment of comorbid posttraumatic stress disorder and major depressive disorder: a pilot study. J Trauma Stress, 24(4), 451–455. doi: 10.1002/jts.20654 [DOI] [PubMed] [Google Scholar]
  42. O’Muircheartaigh J, Keller SS, Barker GJ, & Richardson MP (2015). White Matter Connectivity of the Thalamus Delineates the Functional Architecture of Competing Thalamocortical Systems. Cereb Cortex, 25(11), 4477–4489. doi: 10.1093/cercor/bhv063 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Olson EA, Cui J, Fukunaga R, Nickerson LD, Rauch SL, & Rosso IM (2017). Disruption of white matter structural integrity and connectivity in posttraumatic stress disorder: A TBSS and tractography study. Depress Anxiety, 34(5), 437–445. doi: 10.1002/da.22615 [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Philip NS, Barredo J, Aiken E, & Carpenter LL (2017). Neuroimaging Mechanisms of Therapeutic Transcranial Magnetic Stimulation for Major Depressive Disorder. Biological Psychiatry: Cognitive Neuroscience and Neuroimaging. doi: 10.1016/j.bpsc.2017.10.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Philip NS, Barredo J, van ‘t Wout-Frank M, Tyrka AR, Price LH, & Carpenter LL (2018). Network Mechanisms of Clinical Response to Transcranial Magnetic Stimulation in Posttraumatic Stress Disorder and Major Depressive Disorder. Biol Psychiatry, 83(3), 263–272. doi: 10.1016/j.biopsych.2017.07.021 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Philip NS, Ridout SJ, Albright SE, Sanchez G, & Carpenter LL (2016). 5-Hz Transcranial Magnetic Stimulation for Comorbid Posttraumatic Stress Disorder and Major Depression. J Trauma Stress, 29(1), 93–96. doi: 10.1002/jts.22065 [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Richieri R, Boyer L, Farisse J, Colavolpe C, Mundler O, Lancon C, & Guedj E (2011). Predictive value of brain perfusion SPECT for rTMS response in pharmacoresistant depression. Eur J Nucl Med Mol Imaging, 38(9), 1715–1722. doi: 10.1007/s00259-011-1850-9 [DOI] [PubMed] [Google Scholar]
  48. Rohde GK, Barnett AS, Basser PJ, Marenco S, & Pierpaoli C (2004). Comprehensive approach for correction of motion and distortion in diffusion-weighted MRI. Magn Reson Med, 51(1), 103–114. doi: 10.1002/mrm.10677 [DOI] [PubMed] [Google Scholar]
  49. Romanski LM, Giguere M, Bates JF, & Goldman-Rakic PS (1997). Topographic organization of medial pulvinar connections with the prefrontal cortex in the rhesus monkey. J Comp Neurol, 379(3), 313–332. [PubMed] [Google Scholar]
  50. Rush AJ, Gullion CM, Basco MR, Jarrett RB, & Trivedi MH (1996). The inventory of depressive symptomatology (IDS): Psychometric properties. Psychol Med, 26, 477–486. [DOI] [PubMed] [Google Scholar]
  51. Rytwinski NK, Scur MD, Feeny NC, & Youngstrom EA (2013). The co-occurrence of major depressive disorder among individuals with posttraumatic stress disorder: a meta-analysis. J Trauma Stress, 26(3), 299–309. doi: 10.1002/jts.21814 [DOI] [PubMed] [Google Scholar]
  52. Salomons TV, Dunlop K, Kennedy SH, Flint A, Geraci J, Giacobbe P, & Downar J (2014). Resting-state cortico-thalamic-striatal connectivity predicts response to dorsomedial prefrontal rTMS in major depressive disorder. Neuropsychopharmacology, 39(2), 488–498. doi: 10.1038/npp.2013.222 [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Shah A, Jhawar SS, & Goel A (2012). Analysis of the anatomy of the Papez circuit and adjoining limbic system by fiber dissection techniques. J Clin Neurosci, 19(2), 289–298. doi: 10.1016/j.jocn.2011.04.039 [DOI] [PubMed] [Google Scholar]
  54. Song SK, Sun SW, Ramsbottom MJ, Chang C, Russell J, & Cross AH (2002). Dysmyelination revealed through MRI as increased radial (but unchanged axial) diffusion of water. Neuroimage, 17(3), 1429–1436. [DOI] [PubMed] [Google Scholar]
  55. Vasterling JJ, & Arditte Hall KA (2018). Neurocognitive and Information Processing Biases in Posttraumatic Stress Disorder. Curr Psychiatry Rep, 20(11), 99. doi: 10.1007/s11920-018-0964-1 [DOI] [PubMed] [Google Scholar]
  56. Weathers F, Litz B, Keane T, Palmieri P, Marx B, & Schnurr P (2013). The PTSD Checklist for DSM-5 (PCL-5). Available at: https://www.ptsd.va.gov/professional/assessment/adult-sr/ptsd-checklist.asp. Accessed October 15, 2014.
  57. Yendiki A, Koldewyn K, Kakunoori S, Kanwisher N, & Fischl B (2014). Spurious group differences due to head motion in a diffusion MRI study. Neuroimage, 88, 79–90. doi: 10.1016/j.neuroimage.2013.11.027 [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Yendiki A, Panneck P, Srinivasan P, Stevens A, Zollei L, Augustinack J, … Fischl B (2011). Automated probabilistic reconstruction of white-matter pathways in health and disease using an atlas of the underlying anatomy. Front Neuroinform, 5, 23. doi: 10.3389/fninf.2011.00023 [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.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

RESOURCES