Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Jun 1.
Published in final edited form as: J Psychiatr Res. 2020 Mar 21;125:113–120. doi: 10.1016/j.jpsychires.2020.03.011

Connectivity-Based Identification of a Potential Neurosurgical Target for Mood Disorders.

Jennifer A Sweet 1,2, Suraj Thyagaraj 2, Zhengyi Chen 2, Curtis Tatsuoka 2, Michael D Staudt 3, Joseph R Calabrese 2,4, Jonathan P Miller 1,2, Keming Gao 2,4, Cameron C McIntyre 2
PMCID: PMC7183327  NIHMSID: NIHMS1580742  PMID: 32272241

Abstract

Objective:

Stereotactic ablation (cingulotomy) and subcallosal cingulate deep brain stimulation (SCC DBS) of different regions of the cingulum bundle (CB) have been successfully used to treat psychiatric disorders, such as depression and bipolar disorder. They are hypothesized to work by disrupting white matter pathways involved in the clinical manifestation of these disorders. This study aims to compare the connectivity of different CB subregions using tractography to evaluate stereotactic targets for the treatment of mood disorders.

Methods:

Fourteen healthy volunteers underwent 3T-MR imaging followed by connectivity analysis using probabilistic tractography. Twenty-one anatomic regions of interest were defined for each subject: 10 CB subregions (including the classical cingulotomy and SCC DBS targets) and 11 cortical/subcortical structures implicated in mood disorders. Connectivity results were compared using Friedman and Bonferroni-corrected post-hoc Wilcoxon tests.

Results:

CB connectivity showed a high degree of regional specificity. Both of the traditional stereotactic targets had widespread connectivity with discrete topology. The cingulotomy target connected primarily to the dorsomedial frontal, dorsal anterior cingulate, and posterior cingulate cortices, whereas the SCC DBS target connected mostly to the subgenual anterior cingulate and medial/central orbitofrontal cortices. However, a region of the rostral dorsal CB, lying between these surgical targets, encompassed statistically equivalent connections to all five cortical regions.

Conclusions:

The CB is associated with brain structures involved in affective disorders, and the rostral dorsal CB demonstrates connectivity that is comparable to the combined connectivity of cingulotomy and SCC DBS neurosurgical interventions. The rostral dorsal CB represents a surgical target worthy of clinical exploration for mood disorders.

Keywords: Cingulotomy, Deep Brain Stimulation, Cingulum Bundle, Psychosurgery, Tractography, Neurosurgical Targeting, Mood Disorders

Introduction

Current theories of affective processing support the existence of two parallel networks that underlie mood and thought: a dorsal component associated with cognitive and attentional activities, and a ventral compartment responsible for emotional processing.[1–3] Data suggest that concurrent changes in both regions of the brain may occur in mood disorders such as depression, with hypoactivity of the dorsal region and hyperactivity of ventral structures.[1, 4] The anterior cingulate cortex (ACC) lies between these two circuits and has extensive connections to each,[3] largely via white matter (WM) fibers traveling within the cingulum bundle (CB).[5–8] As such, the ACC and the CB have been implicated in the pathophysiology of various psychiatric disorders and investigated as surgical targets for treatment.[4, 6, 9–24]

A cingulotomy is a destructive lesion traditionally made within a portion of the dorsal ACC (dACC) and the underlying dorsal CB.[25] While the exact neural pathways responsible for the therapeutic effects of cingulotomies remain incompletely understood, anatomical studies suggest that these lesions may strongly influence and/or involve key dorsal networks of the brain, described above.[20, 24, 25] Historically, cingulotomies have been used to treat a variety of psychiatric disorders, including treatment-refractory depression (TRD), obsessive-compulsive disorder (OCD), and chronic pain.[13–24] However, studies evaluating the clinical effects of these lesions, as well as the effects of acute stimulation of the dACC and dorsal CB, produce mixed results.[15, 18–24] Consequently, in an age when deep brain stimulation (DBS) has largely supplanted lesioning techniques,[9, 10] cingulotomies are one of the few lesioning targets that have not yet successfully found its DBS correlate, though investigations continue.[14]

Interestingly, more progress has been made examining DBS of a different region of the CB and ACC for TRD, the subcallosal cingulate (SCC),[9–12] which has also been described as a subgenual cingulotomy target for symptoms of OCD.[26, 27] Analyses of the neural structures affected by SCC DBS largely demonstrate involvement of critical ventral compartments of the brain, alluded to above.[9–12] In addition, the most recent modeling results of SCC DBS suggest that direct modulation of the CB is the primary determinant of therapetuic benefit.[28] Yet, while the clinical findings from such investigations are promising, results from randomized clinical trials fall short.[9, 11, 29, 30] Therefore, similar to the mixed results seen with cingulotomies, surgically targeting the subgenual CB may also yield variable outcomes.[9, 11, 20–24, 29]

Altogether, these findings suggest that, in isolation, neither the dorsal CB, with its dorsal cortical connections, nor the subgenual CB, with its largely ventral cortical associations, are ideal targets for the treatment of mood disorders. However, if a single surgical target possessed substantial WM connections to both the dorsal and ventral brain cortices implicated in affective processing, this may represent an alternative target with complementary therapeutic benefits. Such a discovery can only occur with a clearer understanding of the specific neural pathways responsible for the effects of cingulotomies and SCC DBS. Therefore, this study uses two types of probalistic tractography techniques in healthy, human subjects to assess and compare the WM tract connections involved in both cingulotomy lesions and SCC DBS. It then explores the connectivity of the CB subregions lying between these two areas to determine if a more encompassing surgical target, one that possesses substantial WM connections to both dorsal and ventral brain networks, exists for the treatment of psychiatric diseases, such as mood disorders.

Methods

Study Population

After obtaining approval by the Institutional Review Board, healthy volunteers were recruited at a single institution from 2015–2017 for participation in this cross-sectional study, to obtain MR imaging with diffusion-weighted imaging (DWI) sequences (clinicaltrials.gov, NCT02655978). Following informed consent, subjects were screened for eligibility into the study. Eligibility included age ≥18 years, without any current and/or lifetime psychiatric disorders as assessed by the Mini International Neuropsychiatric Interview for DSM-5.[31] Subjects were excluded if they used illicit drugs, steroids, stimulants, opioids, or nicotine. Urine toxicology screening was used to confirm the absence substance use. The presence of structural brain lesions, progressive neurological disease, preexisting implanted electrical device, or contraindications to MRI were also exclusion criteria. MRIs were acquired within one week of evaluation.

Imaging Acquisition & DWI Processing

Subjects underwent 3-T MRIs (Siemens) with 3D T1-weighted, non-contrasted sequences as reference data (number of slices 160, flip angle 9°, slice thickness 1mm, pixel size 0.67×0.67mm, TR 1600msec, TE 3.05msec) and 3D T2-weighted FLAIR images (number of slices 160, flip angle 120°, slice thickness 1mm, pixel size 0.5×0.5mm, TR 6000msec, TE 356msec). DWI sequences were obtained with 8 receiver channels and sensitivity encoding echo planar imaging with pulse sequences (number of slices 72, flip angle 90°, voxel resolution 1.8mm isometric with pixel size 1.8×1.8mm, slice thickness 1.8mm, matrix 128×128, gradient directions 30, b-values 0 and 1000sec/mm2, TR 11,600msec, TE 99msec). Two DWI sets were acquired with opposing phase-encoding directions to correct susceptibility-induced distortions. Image processing of DWI data for tractography was performed using MRTrix (http://www.mrtrix.org, version 3.0). This included corrections for motion and eddy-current induced distortions,[32, 33] denoise and unringing,[34–36] EPI-distortion,[37] and bias field.[38] Anatomically constrained tractography (ACT) was prepared[39] to create 10 million streamlines, which were then filtered to 1 million streamlines to reduce long-track bias using spherical-deconvolution informed filtering of tracks (SIFT).[40]

Regions of Interest Creation

To assess WM connections potentially involved in both cingulotomies and SCC DBS, and to evaluate the connections of the CB segments situated between these two surgical targets, the anterior CB was divided into 10 equally sized 3mm-radius spherical subregion Regions of Interest (ROIs; Figure1A) bilaterally. Efforts were made to include only the WM fibers of the CB, rather than the overlying gray matter of the cingulate cortex. The center of each subregion ROI sphere was manually created in the Montreal Neurological Institute (MNI) brain template space and co-registered to the subject-specific DWI space. This was then transformed onto the subject-specific T1 datasets, from which coordinates for each subregion center were obtained. A 3mm-radius sphere was then created around the center coordinates for each subregion ROI. CB1–3 (Figure 1B), represented the CB subregions of three standard cingulotomy lesion locations, as described by Yang, et al.[24] Although standard cingulotomies include both CB WM and cingulate cortex gray matter, only the WM component of these lesions was evaluated for the purposes of this study. CB9–10 (Figure 1C) denoted the CB subregions most closely involved with SCC DBS, as described by Mayberg, et al.[10] CB4–8 (Figure 1D) characterized the area of the CB lying between the two surgical targets.

Figure 1.

Figure 1.

Open in a new tab

A) Regions of interest (ROIs) of ten anterior cingulum bundle (CB) subregions. B) CB1–3 ROIs represent the CB subregions commonly involved in standard cingulotomy lesions. C) CB9–10 ROIs represent the CB subregions most closely associated with subcallosal cingulate (SCC) deep brain stimulation (DBS). D) CB4–8 ROIs represent the CB subregions lying between the cingulotomy and SCC DBS surgical targets.

Eleven cortical and subcortical brain areas commonly implicated in the pathophysiology of mood disorders[6] were created bilaterally into target ROIs using FreeSurfer[41] (Figure 2). These cortical and subcortical target ROIs included the amygdala, thalamus, dorsolateral frontal cortex (dlFC), dorsomedial frontal cortex (dmFC), dorsal ACC (dACC), subgenual ACC (sACC), posterior cingulate cortex (PCC), frontal pole (FP), medial/central orbitofrontal cortex (m/cOFC), lateral orbitofrontal cortex (lOFC), and ventrolateral prefrontal cortex (vlPFC). The thalamus was parcellated whole.

Figure 2.

Figure 2.

Open in a new tab

Cortical/subcortical target regions of interest (ROIs), represented in coronal (top row), sagittal (middle row), and axial (bottom row) views. A) Dorsomedial Frontal Cortex (dmFC); B) Dorsal Anterior Cingulate Cortex (dACC); C) Posterior Cingulate Cortex (PCC); D) Amygdala; E) Thalamus; F) Dorsolateral Frontal Cortex (dlFC); G) Lateral Orbitofrontal Cortex (lOFC); H) Ventrolateral Prefrontal Cortex (vlPFC); I) Subgenual Anterior Cingulate Cortex (sACC); J) Frontal Pole (FP); K) Medial/Central Orbitofrontal Cortex (m/cOFC).

Tractography

Probabilistic tractography was performed using two distinct and validated techniques: FSL (http://www.fmrib.ox.ac.uk/fsl, version 5.0.10)[42] and MRTrix (http://www.mrtrix.org, version 3.0).[39, 40] Both techniques utilized diffusion tensor-fitted DWI data that was preprocessed with MRTrix, as described above. Tractography was run bilaterally from each CB subregion ROI (CB1–10; Figure 1A) to each ipsilateral cortical and subcortical ROI (Figure 2). The number of streamlines (tractography-derived fiber tracts representing WM axons) projecting from each seed ROI to each target ROI was calculated and assessed; values were averaged across right and left hemispheres for statistical analysis. The WM connections most significantly associated with the cingulotomy lesion ROIs (CB1–3) were determined, as were the WM fibers most highly connected to the SCC DBS ROIs (CB9–10). The connectivity of each CB subregion ROI lying between the two surgical targets (CB4–8) to cortical and subcortical target ROIs was then analyzed and compared to that of the cingulotomy and SCC DBS ROIs.

Data Analysis

A Friedman’s rank sum test was used to analyze the streamline connections between each CB subregion ROI and each cortical and subcortical target ROI. Type I error level for overall connectivity tests for each of the 11 cortical and subcortical target ROIs that connected to all of the 10 CB ROIs was 0.0045 (0.05/11=0.0045, there were a total of 11 overall tests). Type I error level for overall connectivity tests for each of the 10 CB ROIs that connected to all of the 11 cortical and subcortical target ROIs was 0.005 (0.05/10=0.005, there were a total 10 overall tests). Post-hoc pairwise comparisons were then done using paired Wilcoxon signed-rank tests to assess and characterize these differences statistically. The Type I error level was adjusted to be 0.0011 for multiple testing using Holm-Bonferroni correction for post-hoc analysis for each of the 11 cortical and subcortical target ROIs that connected to all of the 10 CB ROIs (0.05/45=0.0011, there were 45 post hoc comparisons). The Type I error level was adjusted to be 0.00091 for multiple testing using Holm-Bonferroni correction for post-hoc analysis for each of the 10 CB ROIs that connected to all of the 11 cortical and subcortical target ROIs (0.05/55=0.00091, there were 55 post hoc comparisons). All statistical analyses were performed using R (R Foundation for Statistical Computing, Vienna, Austria, version 3.4.4).

Results

Overview

Four healthy male and ten healthy female subjects, with a mean age of 34.8 +/− 11.7 years, were included in the study. Statistically significant differences (p<0.0045) in overall connectivity were found between the 10 CB subregion ROIs (CB1–10) and the 11 cortical/subcortical target ROIs using Friedman’s tests. A post-hoc analysis with Holm-Bonferroni correction demonstrated statistically significant differences in (p<0.0011) connectivity between the cingulotomy and SCC DBS targets. In addition, CB 5 was found to have connectivity findings that did not differ significantly from either surgical target, therefore encompassing the connectivity of both targets more than any other CB subregion evaluated (Figure 3).

Figure 3.

Figure 3.

Open in a new tab

Box plots of the number of streamlines resulting from FSL tractography of various cingulum bundle (CB) subregion Regions of Interest (ROIs) to the 11 cortical and subcortical ROIs. A) Box plots of CB1–3 (cingulotomy target) streamlines, demonstrating that these subregions had the highest connectivity to the dorsomedial prefrontal cortex (dmFC), the dorsal anterior cingulate cortex (dACC), and the posterior cingulate cortex (PCC). B) Box plots of CB9–10 (subcallosal cingulate deep brain stimulation target; SCC DBS) streamlines, illustrating that these subregions had the highest connectivity to the subgenual anterior cingulate cortex (sACC) and the medial/central orbitofrontal cortex (m/cOFC). C) Box plots of CB4–6 streamlines, showing that these subregions had the highest connectivity to all five dorsal (dmFC, dACC, PCC) and ventral (sACC, m/cOFC) cortices connected to by the cingulotomy and SCC DBS targets.

Cingulotomy Connectivity

In assessing the streamline connectivity of the cingulotomy ROIs (CB1–3) to cortical and subcortical target ROIs, all cingulotomy ROIs were found to have the highest connectivity to the same three dorsal cortical areas, the dmFC, dACC, and PCC (Figure 3A), with both the FSL and MRTrix tractography techniques. This difference in connectivity between CB1–3 to the dmFC, dACC, and PCC was statistically significant when compared to that of the CB1–3 connectivity to the amygdala, dlFC, lOFC, vlPFC, and FP using FSL (p<0.000911). Findings were comparable using the MRTrix tractography method. The exact p-values for each tractography technique can be found in the Supplemental Materials.

SCC DBS Connectivity

In assessing the streamline connectivity of the SCC DBS ROIs (CB9–10) to the cortical and subcortical target ROIs, the SCC DBS ROIs were found to have the highest connectivity to the same two ventral cortical areas, the sACC and the m/cOFC (Figure 3B), using both FSL and MRTrix. With FSL, this difference in CB9–10 connectivity to the sACC and m/cOFC was statistically significant when compared to CB9–10 connectivity to the amygdala, thalamus, dlFC, and vlPFC, as well as to the lOFC for CB9 (p<0.00091). The MRTrix technique also showed analogous results (see Supplemental Materials).

Rostral Dorsal CB Subregion Connectivity

The streamline connectivity of CB 4–8 subregion ROIs was then evaluated and compared to that of the cingulotomy target and the SCC DBS target (Figure 3C). Given that CB1–3 showed the highest connectivity to dorsal cortices belonging to the dmFC, dACC, and PCC, and CB9–10 showed highest connectivity to the ventral cortices of the sACC and m/cOFC, the connectivity of all CB subregions to these five dorsal and ventral cortical targets was also analyzed more in depth (Figure 4). The findings were again similar using both tractography techniques (as shown in the Supplemental Materials).

Figure 4.

Figure 4.

Open in a new tab

Box plots of the number of streamlines resulting from FSL tractography of the 10 cingulum bundle (CB) subregion Regions of Interest (ROIs) to the five dorsal and ventral cortices connected to by the cingulotomy and subcallosal cingulate deep brain stimulation (SCC DBS) targets. A) Box plots of CB1–10 streamlines to the key dorsal cortical targets, the dorsomedial prefrontal cortex (dmFC), the dorsal anterior cingulate cortex (dACC), and the posterior cingulate cortex (PCC), demonstrating that CB1–3 (cingulotomy target) connectivity is not significantly different from that of CB4–6 connectivity to these regions, except for the connectivity of CB1 versus CB6 to the PCC. B) Box plots of C1–10 streamlines to the key ventral cortical targets, the subgenual anterior cingulate cortex (sACC) and the medial/central orbitofrontal cortex (m/cOFC), showing that CB9–10 (SCC DBS target) connectivity is not significantly different from that of CB4–8 connectivity to these regions, except for the connectivity of CB9 and CB10 versus CB4 to the m/cOFC. (See Table 1 for exact p-values.)

With respect to the dorsal cortical connections using the FSL technique, CB4–8 were not significantly different in connectivity to the dmFC compared to CB1–3 connectivity to the dmFC (p>0.0011; Table 1A, left). Likewise, CB4–7 all had connectivity that was not significantly different from the dACC compared to CB1–3 (p>0.0011; Table 1A, middle). CB4–5 had connectivity to the PCC that was not significantly different compared to CB1–3, while CB6 had connectivity to the PCC that was not significantly different compared to CB2–3 (p>0.0011; Table 1A, right). These findings are also shown graphically in Figure 4A. In assessing the ventral connections using the FSL technique, the connectivity of CB4–8 to the sACC was not found to be statistically different from that of CB9–10 to the sACC, (p>0.0011; Table 1B, left). Additionally, CB5–8 connectivity to the m/cOFC was not statistically different to that of CB9–10 compared to the m/cOFC (p>0.0011; Table 1B, right). These findings are illustrated in Figure 4B. Similar findings were found using the MRTrix tractography technique (see Supplemental Materials).

Table 1.

P-values from post-hoc comparisons using paired Wilcoxon signed-rank tests of FSL streamline connectivity between cingulum bundle (CB) pairs to specified cortical areas that were found to be significant on Friedman’s rank sum tests with p<0.0045; Type I error level for Wilcoxon signed-rank tests was adjusted to 0.0011 for multiple testing using Holm-Bonferroni correction. A) Pairwise post-hoc analysis comparing the connectivity of CB1–3 (cingulotomy target) versus CB4–8 to the key dorsal cortical regions of the dorsomedial prefrontal cortex (dmFC), dorsal anterior cingulate cortex (dACC), and posterior cingulate cortex (PCC), found to be not significant (black; p>0.0011) or significant (red; p<0.0011) on the paired Wilcoxon signed-rank tests. B) Pairwise post-hoc analysis comparing the connectivity of CB9–10 (subgenual cingulate deep brain stimulation, SCC DBS, target) versus CB4–8 to the key ventral cortical regions of the subgenual anterior cingulate cortex (sACC) and the medial/central orbitofrontal cortex (m/cOFC), found to be not significant (black; p>0.0011) or significant (red; p≤0.0011) on the paired Wilcoxon signed-rank tests.

A
dmFC dACC PCC
CB1 CB2 CB3 CB1 CB2 CB3 CB1 CB2 CB3
CB4 1.0000 1.0000 1.0000 0.5562 0.9999 1.0000 0.0813 0.6009 0.9953
CB5 0.9999 0.9998 1.0000 0.4242 0.9992 0.9998 0.0319 0.3826 0.9644
CB6 0.8934 0.8671 0.9355 1.0000 0.8041 0.7290 0.0002 0.0110 0.2373
CB7 0.9984 0.9972 0.9996 0.7679 0.0389 0.0260 <0.0001 0.0005 0.0319
CB8 0.0964 0.0813 0.1333 0.0813 0.0003 0.0002 <0.0001 <0.0001 0.0009
B
sACC m/cOFC
CB9 CB10 CB9 CB10
CB4 0.6009 0.9984 0.0001 <0.0001
CB5 1.0000 0.9953 0.0211 0.0043
CB6 0.8671 0.1553 0.0137 0.0026
CB7 0.3826 0.0171 0.4242 0.1800
CB8 0.9162 0.2073 0.9882 0.8934
Open in a new tab

Taken together, when including both FSL and MRTrix tractography techniques, CB5 consistently demonstrated comparable connectivity to the three key dorsal cortical targets (dmFC, dACC, PCC) as CB1–3 and both ventral cortical targets (sACC, m/cOFC) as CB9–10. Based on the MNI152 1mm x, y, z coordinate space, the coordinates for the left CB5 subregion center are −8, 29, 18 in mm coordinates or 98, 155, 90 in voxel coordinates. On the right hemisphere, the CB5 subregion center is at 8, 29, 18 in mm coordinates and 82, 155, 90 in voxel coordinates. CB4 and CB6 subregions showed the next most consistent connectivity to all five dorsal and ventral cortical targets.

Discussion

This study has three important findings. First, it confirms that currently used cingulotomy targets predominantly affect dorsal cortical pathways connecting to the dmFC, dACC, and PCC. Second, it demonstrates that SCC DBS largely involves ventral cortical networks, including WM tracts traveling to the sACC and m/cOFC. Third, this study shows that certain regions of the rostral dorsal CB, most notably CB5, which lies in between standard cingulotomy and SCC DBS targets, have comparable connections to all five key dorsal and ventral cortices. Therefore, this region of the rostral dorsal CB may represent a surgical target with more comprehensive connectivity to critical areas implicated in mood disorders than either of the currently used surgical targets alone.

It has been postulated that depressive states are characterized by three concurrent neural processes: 1) underactivity of dorsal associative structures, including the dACC, PCC, and premotor cortices; 2) hyperactivity of ventral paralimbic regions, such as the sACC and basal frontal areas; and 3) preservation of the integrity of connections from the rostral ACC to both dorsal and ventral cortices.[1, 4, 10] Interestingly, destructive cingulotomies seem to only address the first of these three components,[21, 24] likely due to releasing the inhibitory effects of the CB on such dorsal structures.[1] Similarly, SCC DBS may mostly address the second and third components via modulation of such WM tracts, with lesser effects on the first component.[10] Yet, DBS of the rostral dorsal CB, specifically CB5, may effectively respect all three aspects of neural processing. Such findings corroborate anatomic and radiographic connectivity analyses of the CB, which show that the CB is heterogeneous in its composition[5–8] with the region of the rostral dorsal CB having greater connectivity than other CB segments to notable cortical and subcortical structures.[5–8] Therefore, these findings imply that surgically targeting this region of the rostral dorsal CB, may better encompass key dorsal and ventral networks than either the cingulotomy or SCC DBS target in isolation. Moreover, modulation of this target via DBS, as opposed to lesioning, would maintain the integrity of these critical networks.

Several other studies have attempted to perform cingulotomy lesion analyses. These studies aimed to describe the clinical benefits seen in patients who were treated with cingulotomies, and help determine which anatomical structures may be responsible for such effects.[15, 18–24] For OCD, the therapeutic effects of the cingulotomy have been largely attributed to involvement of the cognitive and emotional regions of the ACC, as well as to the WM tracts connecting to prefrontal, orbitofrontal, anterior cingulate, and striatal subcortical regions.[18, 24] For chronic pain and depression, the benefits have been postulated to be from either direct ablation of the ACC and/or from ablation of the CB WM connections to the ACC.[14, 21, 23, 43] Yet, many studies investigating the effects of cingulotomies for these various diseases are disappointing. With respect to the treatment of medically refractory OCD, cingulotomies were reported to be successful in only 30–53% of patients.[22, 24, 44] Likewise, the reported benefit in patients with chronic pain and depression were also mixed.[20–23]

However, there is also extensive variability in the surgical techniques of these investigations, including lesion size, location, and the number of lesions performed.[19, 21, 23, 24] In fact, patients who did not fare well after their initial cingulotomy procedures frequently underwent subsequent surgeries to extend the size of the cingulotomy and/or to create additional lesions.[19, 21, 24, 44] Notably, patients with treatment-refractory OCD and depression who failed to respond to a single bilateral cingulotomy, showed clinical improvement with additional cingulotomies placed more anteriorly.[15, 21, 24, 44] Similarly, Steele et al. assessed treatment-refractory depression patients treated with a single bilateral cingulotomy, and showed that lesions that were placed more anteriorly in position produced improved therapeutic responses more than posterior lesions.[23] These findings suggest that more anteriorly-located lesions, which approach the location of CB5, may have increased therapeutic efficacy.

With respect to SCC DBS, which is used as an investigational target for TRD, local and remote network effects of stimulation have been found.[10, 11] In a study by Mayberg et al.[10] utilizing PET imaging preoperatively and postoperatively, the preoperative hyperactive sACC region, was found to have reduced activity at three and six months after DBS stimulation. Likewise, the previously underactive dorsal cortical activity of the prefrontal cortex and dACC was found, to some degree, to have increased metabolic activity following SCC DBS.[10] Yet, despite such impressive findings of network activation, the outcomes of clinical trials have been less noteworthy.[9, 11, 29] As a result, more recent SCC DBS investigations have incorporated tractography to aid in surgical targeting, in an effort to improve clinical outcomes by ensuring the inclusion of key WM pathways in the field of stimulation.[11] These studies suggest that incorporating WM pathways that connect to dorsal brain structures, as well as to ventral structures, may yield better clinical results.[11] This lends further support to the notion that selection of a surgical target for mood disorders should include targets with WM connections to both dorsal and ventral brain regions, such as CB5.

Finally, it must be stated that in cases of failed cingulotomy and other difficult to treat patients, limbic leucotomies have also been described.[45, 46] A limbic leucotomy includes two lesions, a dorsal cingulotomy lesion and a ventral subcaudate tractotomy lesion. Effectively, a combination of these two lesions incorporates dorsal pathways primarily involved in cingulotomies and more ventral circuitry associated with caudate tractotomies,[46] which also overlaps with the networks associated with SCC DBS.[11] Thus, once again, the proposed target of CB5 may be more successful at encompassing the desired dorsal and ventral networks with a single target than other targets alone or in combination.

Ultimately, these data suggest that utilizing a surgical target that incorporates key dorsal and ventral brain networks, such as CB5, may represent an effective target for the treatment of affective mood disorders. Moreover, modulation of this target, rather than lesioning, would preserve the integrity of these critical pathways, and the use of tractography could further enhance outcomes by enabling a patient-specific targeting methodology.

Limitations

There are several limitations to the study. First, this study includes a small number of subjects. In addition, the use of healthy human subjects makes the applicability of the findings to patients with psychiatric disorders unclear. Errors of MR data acquisition are also worth noting, particularly the limited number of directions obtained for the DWI sequences, which may limit the accuracy of tractography. ROI masks can also affect tractography results, which is why the CB subregion masks were created in a standard MNI space, the center of which were transferred to the patient-specific space, around which new spheres were created. Additionally, though efforts were made to restrict the CB ROIs to the WM fibers, the possibility of incorporating a small portion of the overlying cingulate cortex cannot be entirely excluded. Similarly, the FreeSurfer cortical and subcortical target ROI masks have limitations in that the results are contingent on the structures defined by the FreeSurfer software, which may or may not accurately represent the corresponding true anatomical structures. The use of two distinct tractography techniques were intended to reduce technical limitations and variability inherent in tractography, however, these limitations still exist. Finally, it is difficult to determine from this study the role that the non-WM portion of cingulotomy and SCC DBS interventions play in the efficacy of the procedures.

Conclusion

In healthy, human subjects, tractography analysis of the CB and the surgical targets used for cingulotomy and SCC DBS demonstrated significant differences in connectivity to key dorsal and ventral cortices associated with affective processing. While cingulotomies were found to significantly connect to three dorsal brain regions, SCC DBS significantly connected to two ventral brain areas. However, a region of the CB lying between these surgical targets, CB5, significantly connected to all five critical dorsal and ventral structures, potentially representing a more effective future surgical target for the treatment of mood disorders.

Supplementary Material

1

Acknowledgments

This work was supported by the National Institutes of Health Grant 2KL2TR000440. The authors thank Nicole Woods, BA, Brittany Brownrigg, BS, Carla Conroy, MPH, and the Mood Disorders Program at University Hospitals Cleveland Medical Center and Case Western Reserve University, as well as the generous contributions of John C. Morley, David and Karen Crane, and the Marguerite Wilson Foundation.

This publication was made possible by the Clinical and Translational Science Collaborative of Cleveland, KL2TR002547 from the National Center for Advancing Translational Sciences (NCATS) component of the National Institutes of Health and NIH roadmap for Medical Research. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the NIH.

Footnotes

Conflict of Interest:

Dr. Jennifer Sweet is currently on the scientific advisory board of Koh Young Inc. and was previously on the scientific advisory board and received stock options from Helius Medical Technologies. She also received grant support from the National Institutes of Health for the work in this study (2KL2TR000440; clinicaltrials.gov, NCT02655978).

Dr. Keming Gao was on a speaker’s bureau of AstraZeneca, Pfizer and Sunovion, and an advisory board of Sunovion and Otsuka, and received grant supports from AstraZeneca, Brain and Behavior Research Foundation, and Cleveland Foundation.

Dr. Joseph R. Calabrese has received federal funding from the Department of Defense, Health Resources Services Administration and National Institute of Mental Health as well as grant support from: Abbott Laboratories; AstraZeneca; Bristol-Myers Squibb Company; Cephalon, Inc. (now Teva Pharmaceutical Industries Ltd.); Dainippon Sumitomo Pharma Co., Ltd.; GlaxoSmithKline; Janssen Pharmaceuticals, Inc.; Eli Lilly and Company; Intra-Cellular Therapies, Inc.; Pfizer, Inc; H. Lundbeck A/S; Sunovion Pharmaceuticals Inc.; Takeda Pharmaceutical Company Limited. Dr. Calabrese has served as a consultant/advisory board member/speaker for: Abbott Laboratories; Allergan; AstraZeneca; Bristol-Myers Squibb Company; Cephalon, Inc. (now Teva Pharmaceutical Industries Ltd.); Dainippon Sumitomo Pharma Co., Ltd.; GlaxoSmithKline; Janssen Pharmaceuticals, Inc.; H. Lundbeck A/S; Merck & Co., Inc.; Otsuka Pharmaceutical Co., Ltd.; Pfizer, Inc; Repligen Corporation; Servier; Sunovion Pharmaceuticals Inc.; Solvay Pharmaceuticals, Inc.; Takeda Pharmaceutical Company Limited.

Dr. Cameron McIntyre receives grant support from the National Institutes of Health (NIH P50 NS098573 “Circuit-Based Deep Brain Stimulation for Parkinson’s Disease” or NIH R01 NS105690 “Augmented Reality Platform for Deep Brain Stimulation”) and is a paid consultant for Boston Scientific Neuromodulation and Kernel, as well as a shareholder in the following companies: Surgical Information Sciences, Autonomic Technologies, Cardionomic, Enspire DBS, Neuros Medical.

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

References

  • 1.Mayberg HS, Limbic-cortical dysregulation: a proposed model of depression. J Neuropsychiatry Clin Neurosci, 1997. 9(3): p. 471–81. [DOI] [PubMed] [Google Scholar]
  • 2.Phillips ML, et al. , Identifying predictors, moderators, and mediators of antidepressant response in major depressive disorder: neuroimaging approaches. Am J Psychiatry, 2015. 172(2): p. 124–38. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Sakas DE and Panourias IG, Rostral cingulate gyrus: A putative target for deep brain stimulation in treatment-refractory depression. Med Hypotheses, 2006. 66(3): p. 491–4. [DOI] [PubMed] [Google Scholar]
  • 4.Mayberg HS, et al. , Reciprocal limbic-cortical function and negative mood: converging PET findings in depression and normal sadness. Am J Psychiatry, 1999. 156(5): p. 675–82. [DOI] [PubMed] [Google Scholar]
  • 5.Sweet JA, et al. , Clinical Evaluation of Cingulum Bundle Connectivity for Neurosurgical Hypothesis Development. Neurosurgery, accepted for publication, 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Heilbronner SR and Haber SN, Frontal cortical and subcortical projections provide a basis for segmenting the cingulum bundle: implications for neuroimaging and psychiatric disorders. J Neurosci, 2014. 34(30): p. 10041–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Jones DK, et al. , Distinct subdivisions of the cingulum bundle revealed by diffusion MRI fibre tracking: implications for neuropsychological investigations. Neuropsychologia, 2013. 51(1): p. 67–78. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Wu Y, et al. , Segmentation of the Cingulum Bundle in the Human Brain: A New Perspective Based on DSI Tractography and Fiber Dissection Study. Front Neuroanat, 2016. 10: p. 84. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Lozano AM, et al. , Subcallosal cingulate gyrus deep brain stimulation for treatment-resistant depression. Biol Psychiatry, 2008. 64(6): p. 461–7. [DOI] [PubMed] [Google Scholar]
  • 10.Mayberg HS, et al. , Deep brain stimulation for treatment-resistant depression. Neuron, 2005. 45(5): p. 651–60. [DOI] [PubMed] [Google Scholar]
  • 11.Riva-Posse P, et al. , A connectomic approach for subcallosal cingulate deep brain stimulation surgery: prospective targeting in treatment-resistant depression. Mol Psychiatry, 2018. 23(4): p. 843–849. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Riva-Posse P, et al. , Defining critical white matter pathways mediating successful subcallosal cingulate deep brain stimulation for treatment-resistant depression. Biol Psychiatry, 2014. 76(12): p. 963–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Ballantine HT Jr., et al. , Treatment of psychiatric illness by stereotactic cingulotomy. Biol Psychiatry, 1987. 22(7): p. 807–19. [DOI] [PubMed] [Google Scholar]
  • 14.Boccard SG, et al. , Targeting the affective component of chronic pain: a case series of deep brain stimulation of the anterior cingulate cortex. Neurosurgery, 2014. 74(6): p. 628–35; discussion 635–7. [DOI] [PubMed] [Google Scholar]
  • 15.Bourne SK, et al. , Beneficial effect of subsequent lesion procedures after nonresponse to initial cingulotomy for severe, treatment-refractory obsessive-compulsive disorder. Neurosurgery, 2013. 72(2): p. 196–202; discussion 202. [DOI] [PubMed] [Google Scholar]
  • 16.Dougherty DD, et al. , Prospective long-term follow-up of 44 patients who received cingulotomy for treatment-refractory obsessive-compulsive disorder. Am J Psychiatry, 2002. 159(2): p. 269–75. [DOI] [PubMed] [Google Scholar]
  • 17.Leiphart JW and Valone FH 3rd, Stereotactic lesions for the treatment of psychiatric disorders. J Neurosurg, 2010. 113(6): p. 1204–11. [DOI] [PubMed] [Google Scholar]
  • 18.Rauch SL, et al. , Volume reduction in the caudate nucleus following stereotactic placement of lesions in the anterior cingulate cortex in humans: a morphometric magnetic resonance imaging study. J Neurosurg, 2000. 93(6): p. 1019–25. [DOI] [PubMed] [Google Scholar]
  • 19.Richter EO, et al. , Cingulotomy for psychiatric disease: microelectrode guidance, a callosal reference system for documenting lesion location, and clinical results. Neurosurgery, 2004. 54(3): p. 622–28; discussion 628–30. [DOI] [PubMed] [Google Scholar]
  • 20.Schoene-Bake JC, et al. , Tractographic analysis of historical lesion surgery for depression. Neuropsychopharmacology, 2010. 35(13): p. 2553–63. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Shields DC, et al. , Prospective assessment of stereotactic ablative surgery for intractable major depression. Biol Psychiatry, 2008. 64(6): p. 449–54. [DOI] [PubMed] [Google Scholar]
  • 22.Spangler WJ, et al. , Magnetic resonance image-guided stereotactic cingulotomy for intractable psychiatric disease. Neurosurgery, 1996. 38(6): p. 1071–6; discussion 1076–8. [PubMed] [Google Scholar]
  • 23.Steele JD, et al. , Anterior cingulotomy for major depression: clinical outcome and relationship to lesion characteristics. Biol Psychiatry, 2008. 63(7): p. 670–7. [DOI] [PubMed] [Google Scholar]
  • 24.Yang JC, et al. , Lesion analysis for cingulotomy and limbic leucotomy: comparison and correlation with clinical outcomes. J Neurosurg, 2014. 120(1): p. 152–163. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Cosgrove GR RS, Stereotactic cingulotomy. Neurosurgery Clinics of North America, 2003. 14(2): p. 225–35. [DOI] [PubMed] [Google Scholar]
  • 26.Huotarinen A, K. R, Hariz M, Laitinen’s Subgenual Cingulotomy: Anatomical Location and Case Report. Stereotact Funct Neurosurg, 2018. 96(5): p. 342–346. [DOI] [PubMed] [Google Scholar]
  • 27.Laitinen LV, V. J, Stereotaxic ventral anterior cingulotomy in some psychological disorders Psychosurgery, ed. Hitchcock ELL, Vaermet K. 1972, Springfiled: Charles C. Thomas. [Google Scholar]
  • 28.Howell B, et al. , Quantifying the axonal pathways directly stimulated in therapeutic subcallosal cingulate deep brain stimulation. Hum Brain Mapp, 2019. 40(3): p. 889–903. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Holtzheimer PE, et al. , Subcallosal cingulate deep brain stimulation for treatment-resistant unipolar and bipolar depression. Arch Gen Psychiatry, 2012. 69(2): p. 150–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Holtzheimer PE, et al. , Subcallosal cingulate deep brain stimulation for treatment-resistant depression: a multisite, randomised, sham-controlled trial. Lancet Psychiatry, 2017. 4(11): p. 839–849. [DOI] [PubMed] [Google Scholar]
  • 31.Sheehan DV, et al. , The Mini-International Neuropsychiatric Interview (M.I.N.I.): the development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. J Clin Psychiatry, 1998. 59 Suppl 20: p. 22–33;quiz 34–57. [PubMed] [Google Scholar]
  • 32.Andersson JL, Skare S, and Ashburner J, How to correct susceptibility distortions in spin-echo echo-planar images: application to diffusion tensor imaging. Neuroimage, 2003. 20(2): p. 870–88. [DOI] [PubMed] [Google Scholar]
  • 33.Andersson JLR and Sotiropoulos SN, An integrated approach to correction for off-resonance effects and subject movement in diffusion MR imaging. Neuroimage, 2016. 125: p. 1063–1078. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Kellner E, et al. , Gibbs-ringing artifact removal based on local subvoxel-shifts. Magn Reson Med, 2016. 76(5): p. 1574–1581. [DOI] [PubMed] [Google Scholar]
  • 35.Veraart J, Fieremans E, and Novikov DS, Diffusion MRI noise mapping using random matrix theory. Magn Reson Med, 2016. 76(5): p. 1582–1593. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Veraart J, et al. , Denoising of diffusion MRI using random matrix theory. Neuroimage, 2016. 142: p. 394–406. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Holland D, Kuperman JM, and Dale AM, Efficient correction of inhomogeneous static magnetic field-induced distortion in Echo Planar Imaging. Neuroimage, 2010. 50(1): p. 175–83. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Tustison NJ, et al. , N4ITK: improved N3 bias correction. IEEE Trans Med Imaging, 2010. 29(6): p. 1310–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Smith RE, et al. , Anatomically-constrained tractography: improved diffusion MRI streamlines tractography through effective use of anatomical information. Neuroimage, 2012. 62(3): p. 1924–38. [DOI] [PubMed] [Google Scholar]
  • 40.Smith RE, et al. , SIFT: Spherical-deconvolution informed filtering of tractograms. Neuroimage, 2013. 67: p. 298–312. [DOI] [PubMed] [Google Scholar]
  • 41.Fischl B, FreeSurfer. Neuroimage, 2012. 62(2): p. 774–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Jenkinson M, et al. , Fsl. Neuroimage, 2012. 62(2): p. 782–90. [DOI] [PubMed] [Google Scholar]
  • 43.Becerra L, et al. , Analogous responses in the nucleus accumbens and cingulate cortex to pain onset (aversion) and offset (relief) in rats and humans. J Neurophysiol, 2013. 110(5): p. 1221–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Bourne SK, S. S, Neal J, et al. , Beneficial effect of subsequent lesion procedures after nonresponse to initial cingulotomy for severe, treatment-refractory obsessive-compulsive disorder. Neurosurgery, 2013. 72(2): p. 196–202. [DOI] [PubMed] [Google Scholar]
  • 45.Shields DC, A. W, Eskandar EN, Jain FA, Cosgrove GR, Flaherty AW, Cassem EH, Price BH, Rauch SL, Dougherty DD., Prospective assessment of stereotactic ablative surgery for intractable major depression. Biol Psychiatry, 2008. 64(6): p. 449–54. [DOI] [PubMed] [Google Scholar]
  • 46.Yang JC, Ginat DT, Dougherty DD, Makris N, & Eskandar EN, Lesion analysis for cingulotomy and limbic leucotomy: comparison and correlation with clinical outcomes. J Neurosurg, 2014. 120(1): p. 152–163. [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

1
NIHMS1580742-supplement-1.docx (29.7KB, docx)

RESOURCES