Article Text

Download PDFPDF
Original research
Interstitial laser anterior capsulotomy for obsessive–compulsive disorder: lesion size and tractography correlate with outcome
  1. David Satzer1,
  2. Anil Mahavadi2,
  3. Maureen Lacy3,
  4. Jon E Grant4,
  5. Peter Warnke1
  1. 1 Neurosurgery, University of Chicago Biological Sciences Division, Chicago, Illinois, USA
  2. 2 Neurosurgery, The University of Alabama at Birmingham, Birmingham, Alabama, USA
  3. 3 Psychaitry and Behavioral Neuroscience, University of Chicago Biological Sciences Division, Chicago, Illinois, USA
  4. 4 Psychiatry, University of Chicago Biological Sciences Division, Chicago, Illinois, USA
  1. Correspondence to Professor Peter Warnke, Neurosurgery, University of Chicago Biological Sciences Division, Chicago, IL 60637, USA; pwarnke{at}


Background Anterior capsulotomy is a well-established treatment for refractory obsessive–compulsive disorder (OCD). MRI-guided laser interstitial thermal therapy (LITT) allows creation of large, sharply demarcated lesions with the safeguard of real-time imaging.

Objective To characterise the outcomes of laser anterior capsulotomy, including radiographical predictors of improvement.

Methods Patients with severe OCD refractory to pharmacotherapy and cognitive–behavioural therapy underwent bilateral anterior capsulotomy via LITT. The primary outcome was per cent reduction in Yale-Brown Obsessive–Compulsive Scale (Y-BOCS) score over time. Lesion size was measured on postablation MRI. Disconnection of the anterior limb of the internal capsule (ALIC) was assessed via individual and normative tractography.

Results Eighteen patients underwent laser anterior capsulotomy. Median follow-up was 6 months (range 3–51 months). Time occupied by obsessions improved immediately (median Y-BOCS item 1 score 4–1, p=0.002). Mean (±SD) decrease in Y-BOCS score at last follow-up was 46%±32% (16±11 points, p<0.0001). Sixty-one per cent of patients were responders. Seven patients (39%) exhibited transient postoperative apathy. One patient had an asymptomatic intracerebral haemorrhage. Reduction in Y-BOCS score was positively associated with ablation volume (p=0.006). Individual tractography demonstrated durable ALIC disconnection. Normative tractography revealed a dorsal–ventral gradient, with disconnection of orbitofrontal streamlines most strongly associated with a positive response (p<0.0001).

Conclusions Laser anterior capsulotomy resulted in immediate, marked improvement in OCD symptom severity. Larger lesions permit greater disconnection of prefrontal–subcortical pathways involved in OCD. The importance of greater disconnection is presumably related to variation in ALIC structure and the complex role of the PFC in OCD.

  • disconnection
  • image analysis
  • neurosurgery
  • psychiatry

Data availability statement

Data are available upon reasonable request.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Obsessive–compulsive disorder (OCD) has a lifetime prevalence of 1%–3% and carries substantial morbidity, with high rates of anxiety, depression, unemployment and social isolation.1 First-line treatment for OCD includes serotonin reuptake inhibitors and cognitive–behavioural therapy.1 However, a significant subset of patients with OCD are refractory to medical and behavioural therapies.2

The symptoms of OCD are thought to arise from pathological activity in the circuits connecting the prefrontal cortex (PFC), striatum and thalamus.3 Multiple surgical procedures targeting these circuits have been designed for refractory OCD. Deep brain stimulation (DBS) has gained acclaim as a titratable therapy, but maximal response is not seen until 3–6 months after surgery.4 The appeal of ablative surgery lies in the potential for immediate symptom relief5 6 after a single treatment, without need for implanted hardware and the associated risk of device-related complications such as infection. Among ablative procedures, anterior capsulotomy leads to greatest relief from OCD symptoms.7

Contemporary techniques for capsulotomy include radiofrequency (RF) ablation, stereotactic radiosurgery (SRS) and focused ultrasound (FUS).8 RF yields immediate and sustained OCD relief,5 6 but lesions are variable due to unpredictable thermal spread.9 SRS results in gradual improvement in OCD symptoms over 1–3 years due to the slow development of radionecrosis.10–12 FUS incorporates real-time MRI guidance but is substantially limited by energy transfer across the skull and the small focus of the ultrasound beam.13

Laser interstitial thermal therapy (LITT) is a percutaneous ablative method like RF but permits the creation of larger, more demarcated14 lesions with the guidance of real-time thermal MRI.9 Thermal imaging allows the surgeon to continuously monitor which tissues are subjected to irreversible thermal injury, allowing optimal target ablation while avoiding off-target damage. Capsulotomy is an ideal application for LITT due to the need for robust prefrontal disconnection while preserving neighbouring structures, such as the corticobulbar tract in the genu of the internal capsule. The present study details the outcomes of laser anterior capsulotomy for OCD at the University of Chicago, where the first case reported in the literature was performed in 2015.15


Study protocol

This retrospective case series included patients who underwent laser anterior capsulotomy for OCD from 2015 to 2020. Patients with less than 3 months of follow-up were excluded. Consent for chart review was obtained from recently treated patients and waived for patients no longer receiving care at the University of Chicago.

Patient selection

To be considered for capsulotomy, patients were required to have severe OCD (Yale-Brown Obsessive–Compulsive Scale (Y-BOCS) score ≥30), refractory to pharmacotherapy (minimal response to a 12-week trial of clomipramine and three selective serotonin reuptake inhibitors at maximum doses) and cognitive–behavioral therapy (minimal response using weekly exposure response prevention for at least 16 weeks). Surgical candidacy was evaluated by a panel with representation from neurosurgery, psychiatry and medical ethics examining all available medical, psychiatric and neuropsychological records.

Surgical procedure

Under general anaesthesia, the head was immobilised in a stereotactic frame (CRW; Integra, Plainsboro, New Jersey, USA). A 1 mm intraoperative CT (BodyTom; Neurologica, Danvers, Massachusetts, USA) was obtained and merged with T2 and contrast-enhanced T1 MRI sequences (StealthStation; Medtronic, Minnesota, USA). Bilateral trajectories through the anterior limb of the internal capsule (ALIC) were planned based on direct visualisation of ALIC on T1 and T2 sequences (figure 1A). For each trajectory, a 1.6 mm diameter polycarbonate cooling catheter, containing a 0.4 mm diameter optical fibre with a 10 mm aperture tip (Visualase, Medtronic) was implanted through a 3.2 mm drill hole. Placement accuracy was confirmed with a second intraoperative CT. The patient was brought to a 3 Tesla MRI scanner (Philips, Amsterdam, Netherlands). After a 3–5 W test, ablation was performed at 7.5–9.3 W. The laser fibre was pulled back within the catheter as needed to allow additional ablation. Ablation was guided by magnetic resonance (MR) thermography with an irreversible damage threshold of 70°C superimposed on a T1 or T2 background (figure 1B,C). Immediate postablation contrast-enhanced T1 images were acquired (figure 1D). The catheter and laser fibre were removed, and the incision was closed with a single suture. Patients received dexamethasone intraoperatively and for 3 days after surgery to minimise ablation-related oedema.

Figure 1

Laser anterior capsulotomy technique. (A) Intraoperative, preablation oblique coronal T1-weighted MRI showing laser catheters in the anterior limb of the internal capsule bilaterally. (B) Real-time magnetic resonance thermography illustrating temperature map during ablation. (C) Damage map of tissue subject to temperatures above irreversible damage threshold of 70°C. (D) Immediate postablation MRI showing peripherally enhancing ablation volume.

Psychiatric outcomes and general functional status

Patients were categorised by OCD subtype according to a five-subtype system.16 Y-BOCS was administered at outpatient visits. Additionally, a modified Y-BOCS item 1 (time occupied by obsessive thoughts in the past day rather than week) was assessed on postoperative day 1. Patients with ≥35% decrease in Y-BOCS score at last follow-up were classified as responders.8 Employment status and ability to leave the home were determined before surgery and at last follow-up.

Structural imaging

Catheter tip location was determined by MRI artefact. MRI sequences were coregistered in ITK-SNAP V.3.8.0 ( Ablation volume was segmented on immediate postablation T1 sequences as the contrast-enhancing volume including the hypointense centre. Fractions of ALIC ventral and dorsal to the ablation were measured along a dorsal–ventral line bisecting ALIC in the plane of the laser catheters.

Preablation T1 sequences were automatically segmented in BrainSuite V.19b ( The dorsomedial prefrontal cortex (DM-PFC, anterior superior frontal gyrus), dorsolateral prefrontal cortex (DL-PFC, anterior middle frontal gyrus), ventrolateral prefrontal cortex (VL-PFC, pars orbitalis and triangularis), orbitofrontal cortex (OFC; orbitofrontal gyri, gyrus rectus and transverse frontal gyrus) and anterior cingulate cortex (ACC, anterior and subgenual cingulate gyrus) were defined based on USCBrain Atlas regions.18 For each PFC region, cortical volume was normalised to total brain volume.

Individual tractography

Diffusion tensor imaging (DTI) was obtained on postoperative day 1 with a Philips scanner using 32 gradients, b value 1000 s/mm3, echo time of 90–104 ms, repetition time of 4.6–6.5 s, slice thickness of 2 mm and voxel size of 2.0 mm3. DTI acquired on other scanners was excluded from quantitative analysis. Preoperative and delayed postoperative DTI sequences were obtained in some patients using 32 gradients, b value 1000 s/mm3, echo time of 78–85 ms, repetition time of 8.0–8.3 s, slice thickness of 4 mm and voxel size of 1.9–2.0 mm3.

Deterministic tractography was performed in BrainSuite with one seed per voxel, step size of 0.25 mm, maximum of 500 steps, maximum angle of 30°, minimum streamline length of 30 mm and minimum fractional anisotropy of 0.20. The number of streamlines between each PFC region and the ipsilateral thalamus was determined.

Normative tractography

Ablation volumes were normalised to the Montreal Neurologic Institute-152 atlas in SPM12 ( Ablation volumes were averaged to create heat maps. A difference map based on responder status was calculated by subtracting the non-responder heat map from the responder heat map. The 10% of voxels most associated with responder status (‘hottest voxels’) and the 10% of voxels most associated with non-responder status (‘coldest voxels’) were identified using SPM and ImageJ V.1.52q (National Institutes of Health).19

Normative tractography was performed using Lead-DBS,20 and a group connectome was derived from 19 healthy subjects.21 PFC streamlines within ALIC were identified based on the 10% of voxels with the greatest number of streamlines per voxel for each PFC region. Discriminant streamlines were identified from the hottest and coldest voxel maps.

Statistical analysis

Preoperative and postoperative Y-BOCS scores were compared with the Wilcoxon sign-rank test. Due to the repeated Y-BOCS observations with variable number and timing, a generalised linear mixed model was used to assess potential predictors of outcome. Per cent improvement in Y-BOCS score was the dependent variable; one independent variable at a time was analysed as a fixed effect; and a random effect was introduced to account for variation between patients. A χ2 test was used to compare discriminant streamline counts across PFC regions. An alpha level of 0.05 was used for all significance testing. Statistical analysis was performed with SAS OnDemand for Academics (SAS Institute Inc., Cary, NC, USA, V. 9.4.).



Eighteen patients were included. Seven patients (39%) were female. Mean (±SD) age was 35±16 years (range 19–61 years). The primary OCD subtype was doubt/checking in nine patients, contamination/cleaning in five patients and taboo thoughts in four patients. Three patients had a history of depression; no patients met the criteria for major depressive disorder at the time of surgery.

Fourteen patients (78%) had undergone at least two times per week of exposure response prevention and/or an intensive OCD-focused intensive outpatient programme. Eleven patients (61%) had done residential OCD programmes. One patient had undergone DBS of ALIC without benefit and had his stimulator removed immediately prior to capsulotomy.

Median follow-up was 6 months (range 3–51 months).

Symptom reduction

Median Y-BOCS item 1 score decreased from 4 (range 3–4) before surgery to 1 (range 0–3) on postoperative day 1 (p=0.002, figure 2A).

Figure 2

Postoperative improvement in Y-BOCS. (A) Y-BOCS item 1 (time occupied by obsessive thoughts) scores before surgery (preop) and on POD#1. *Note that when assessed on POD#1, Y-BOCS item 1 refers to obsessive thoughts in the last 24 hours rather than the last week. (B) Per cent decrease in full Y-BOCS score over time after surgery for each patient. Responders (≥35% improvement in Y-BOCS at the last follow-up) are represented in red, while non-responders are represented in blue. The responder threshold is illustrated with a dotted line. for the patient who underwent repeat laser anterior capsulotomy, scores after the second surgery are not shown. POD#1, postoperative day 1; Y-BOCS, Yale-Brown Obsessive Compulsive Scale.

Mean Y-BOCS score decreased from 34±2 before surgery to 18±11 at last follow-up (p<0.0001, figure 2B). Mean decrease in Y-BOCS score from baseline to last follow-up was 46%±32% or 16±11 points (figure 2B). Eleven patients (61%) were responders. There was a marginally significant decrease in Y-BOCS improvement over time (−1.3% per month, p=0.05). The patient with prior ineffective DBS had 77% improvement in Y-BOCS at last follow-up.

One patient did not improve (Y-BOCS score increased from 36 to 40) and underwent repeat capsulotomy 1 year later. Three years after reoperation, this patient had a Y-BOCS score of 31. Data from this reoperation were excluded from analysis.

General functional status

Data regarding general functional status were available for 17 patients. Four patients (24%) were employed before surgery and at last follow-up. An additional four patients (24%) became employed after surgery. One patient (6%) was a college student on leave and returned to school after surgery. Eight patients were homebound before surgery, of whom three (18%) became able to leave the home regularly.

Adverse events

One patient had an asymptomatic tract haemorrhage. No focal neurological deficit was observed. Seven patients (39%) exhibited transient postoperative apathy, which resolved after 3–4 months. No patients developed major depressive disorder after surgery.

Preoperative variables and clinical outcome

There was no relationship between Y-BOCS change and preoperative Y-BOCS score (p=0.71) or age at surgery (p=0.69). Improvement in Y-BOCS score did not vary with OCD subtype (contamination/cleaning vs doubt/checking, p=0.57; contamination/cleaning vs taboo thoughts, p=0.19; doubt/checking vs taboo thoughts, p=0.34). There was no relationship between Y-BOCS improvement and cortical volume for DM-PFC (p=0.26), DL-PFC (p=0.76), VL-PFC (p=0.16), OFC (p=0.06) or ACC (p=0.46).

Lesion geometry and energy delivery

Mean laser catheter tip coordinates were 13.9±1.9 mm lateral, 6.9±3.9 mm anterior and 3.4±2.0 superior to the anterior commissure. One pull-back of the laser fibre was necessary in 26 hemispheres (72%), whereas no pull-back was needed in 10 hemispheres (28%). Mean ablation power was 8.0±0.5 W; ablation time was 39±13 s/volume; and total energy delivery was 1.41±0.31 kJ.

Mean total ablation volume (left plus right) was 2.4±0.6 cc. In the plane through the laser catheters, 83%±18% of the dorsal-ventral ALIC length was included in the ablation; 16%±17% of ALIC length was ventral to the ablation; and 1%±3% of ALIC length was dorsal to the ablation.

Improvement in Y-BOCS was positively related to ablation volume (33%/cc, p=0.006) but not energy delivered (p=0.15) or length of ALIC dorsal (p=0.34) or ventral (p=0.18) to the ablation. Volumes of caudate (p=0.64), putamen (p=0.17), globus pallidus (p=0.09) and nucleus accumbens (p=0.16) ablated were not related to Y-BOCS improvement.

Individual tractography

Among 15 patients who underwent standardised postoperative DTI, Y-BOCS improvement was not related to number of streamlines between the thalamus and the DM-PFC (p=0.17), DL-PFC (p=0.46), VL-PFC (p=0.40), OFC (p=0.35), ACC (p=0.20) or overall PFC (0.90).

Variation in prefrontal–thalamic streamlines was observed among three patients with preoperative tractography (figure 3A–C). While some persistent OFC-thalamic streamlines were seen in all three patients, unilateral preservation of OFC-thalamic streamlines was seen in the patient with the least Y-BOCS improvement (figure 3A).

Figure 3

Serial prefrontal-thalamic tractography. Streamlines between the thalamus and prefrontal regions (see inset), reconstructed from DTI acquired at different time points, are shown for the eight patients (A–H) who underwent multiple DTI scans. Preoperative tractography is superimposed on intraoperative, preablation coronal T1-weighted MRI. Tractography from POD#1 is overlaid on the immediate postablation contrast-enhanced T1 image. Delayed (3–10 months) tractography is displayed on T2-weighted MRI. The Y-BOCS item 1* score corresponding to each timepoint is shown in the lower right of each image in the preop and POD#1 columns. Per cent decrease in overall Y-BOCS score is shown in the bottom right of each image in the columns of 3 and >3 months. *Note that when assessed on POD#1, Y-BOCS item 1 refers to obsessive thoughts in the last 24 hours rather than the last week. ACC, anterior cingulate cortex; DL-PFC, dorsolateral prefrontal cortex; DM-PFC, dorsomedial prefrontal cortex; DTI, diffusion tensor imaging; NR, Non-Responder; OFC, orbitofrontal cortex; POD#1, postoperative day 1; VL-PFC, ventrolateral prefrontal cortex; Y-BOCS, Yale-Brown Obsessive Compulsive Scale.

Five patients underwent DTI 3–10 months after surgery (figure 3D–H). Comparison with DTI from postoperative day 1 shows stable appearance of PFC-thalamic streamlines.

Normative tractography

Group connectome analysis demonstrated expected22 23 ALIC organisation, with OFC streamlines most ventral and DM-PFC streamlines most dorsal (figure 4B). Substantial overlap was observed between regional streamline groups. ACC streamlines were not identified in ALIC.

Figure 4

Normative volumetry and tractography. (A) Axial T1-weighted MRI section of the MNI-152 atlas showing coronal planes through the ALIC (B–E). (B) Structural organisation of ALIC shown by normative streamlines terminating in the PFC regions. Streamlines were calculated from a group connectome derived from 19 healthy subjects.21 (C) Heat map of ablation volumes normalised to MNI space. (D) Difference map based on responder status (≥35% improvement in Y-BOCS at the last follow-up), obtained by subtracting the non-responder heat MAP from the responder heat map. (E) Normative streamlines passing through the 10% of voxels most associated with responder status (red) or the 10% of voxels most associated with non-responder status (blue). Additional (F) sagittal and (G) coronal images illustrate gradient of ventral/responder streamlines and dorsal/non-responder streamlines. (H) PFC surface rendering colour coded by ratio of responder to non-responder streamlines for each PFC region. ACC, anterior cingulate cortex; ALIC, anterior limb of the internal capsule; DL-PFC, dorsolateral prefrontal cortex; DM-PFC, dorsomedial prefrontal cortex; MNI, Montreal Neurologic Institute; OFC, orbitofrontal cortex; PFC, prefrontal cortex; VL-PFC, ventrolateral prefrontal cortex.

A heat map of all ablations is shown in figure 4C. Y-BOCS improvement was not related to number of streamlines between the ablation volume and the DM-PFC (p=0.29), DL-PFC (p=0.14), VL-PFC (p=0.10), OFC (p=0.35), ACC (p=0.56) or overall PFC (p=0.09).

Voxels associated with responder status were predominantly located in ventral ALIC (figure 4D). ‘Hottest’ (responder) streamlines were located more ventrally within ALIC and the frontal lobe, compared with ‘coldest’ (non-responder) streamlines (figure 4E–G). The ratio of hottest to coldest streamlines decreased from ventral to dorsal PFC (p<0.0001, figure 4H).


This series of 18 patients expands on early case reports15 24 and constitutes the largest series to date of laser anterior capsulotomy for OCD. Time occupied by obsessions improved immediately. Mean improvement in Y-BOCS score was 46% (16 points) at last follow-up, and 61% of patients were responders. Greater improvement was seen with larger lesions. A gradient was seen between ventral PFC streamlines associated with clinical response when ablated and dorsal PFC streamlines associated with non-response.

Outcomes of existing surgical treatments for OCD

Anterior capsulotomy has been performed since the 1940s25 using a variety of techniques. Mechanical leucotomy has a low response rate and relatively high risk of intracerebral haemorrhage.8 RF yields immediate results with long-term stability.6 26 Gradual improvements in Y-BOCS scores have been observed after SRS and FUS due to the inherent limitations of energy delivery with both modalities.10–12 A recent meta-analysis found a 46% Y-BOCS reduction with a 59% response rate across capsulotomy techniques,7 comparable to 46% and 61%, respectively, in the present study. Like RF and unlike SRS and FUS, immediate marked improvement was seen after LITT. While assessment of immediate treatment response may be confounded by perioperative factors such as recent anaesthesia, this finding is corroborated by dramatic improvement in Y-BOCS scores observed in the first 1–2 months after surgery.

Other surgical treatments for OCD include anterior cingulotomy and DBS. Cingulotomy is associated with a 32% Y-BOCS reduction and 36% response rate, inferior to capsulotomy.7 DBS for OCD most commonly targets the striatal region, including ALIC.27 The maximal effects of DBS are not seen for at least 3–6 months after surgery.4 Large series report 40%–42% mean reduction in Y-BOCS score with a 52%–60% responder rate at 1 year after DBS of ALIC.4 28 One patient in the present case series had previously undergone DBS of ALIC without any clear benefit and had a robust (77% reduction in Y-BOCS score) response to laser anterior capsulotomy.

Previous experience with laser anterior capsulotomy

Prior analysis of the first four cases in this series revealed strong relationships between ablation volume, energy deposition and Y-BOCS change.24 This finding led to the use of large ablation volumes in subsequent cases. While the correlation between ablation volume and clinical improvement persisted, no relationship between energy deposition and outcome was observed. The latter finding likely reflects the linear relationship between time and energy deposition but not volume of tissue reaching 70°C.29

The Brown University group recently published a series of nine patients treated with laser anterior capsulotomy targeting the ventral third of ALIC.30 Mean decrease in Y-BOCS score was 37%, and 78% of patients were classified as responders. A 3 mm aperture laser was used. Mean total ablation volume was 1.0±0.3 cc and did not correlate with outcome. There was a trend towards greater reduction in Y-BOCS scores with more ventral ALIC coverage. While the clinical outcomes of the Brown study are comparable to those of the present study, smaller ablation volumes may have obscured the significance of large lesions encompassing more dorsal ALIC regions that were linked to clinical improvement in the present study. The composition of the Brown and University of Chicago cohorts also differed; a third of patients in the Brown study previously underwent SRS, whereas 1 of 18 patients in the present study previously underwent DBS and none underwent SRS.

Adverse effects

The safety of anterior capsulotomy is well documented. Aggregate published rates of haemorrhage (1.8%) and hemiparesis (0.8%) are very low. Weight gain has been reported in 13% of patients. Reoperation due to small lesion size occurred in 9% of patients.8 In the present case series, one patient had an asymptomatic intracerebral haemorrhage; no focal neurological deficits were observed; and one patient required reoperation.

Apathy and disinhibition have been reported in 22%–40% of patients undergoing anterior capsulotomy, are often temporary, and have been associated with multiple procedures and high radiation doses but not lesion volume per se.5 8 30 31 In the present cohort, seven patients (39%) demonstrated transient apathy which resolved within 3–4 months of surgery despite the relatively large lesion volumes. Transient apathy might relate to temporary changes in local blood–brain barrier permeability.32

Neuropsychological testing after capsulotomy has shown improved overall psychosocial function, cognition, executive function and memory.5 33 In this study, four previously unemployed patients became employed; one college student returned to school; three previously homebound patients were able to leave home regularly; and no patients had a decline in general functional status.

Lesioning and white matter disconnection

OCD is thought to arise from pathological activity in pathways that pass through ALIC and connect the PFC, striatum and thalamus.3 Increased activity and decreased volume of OFC have been consistently reported in OCD and normalise with treatments including anterior capsulotomy.34–36 ACC and DL-PFC participate in the cognitive and affective features of OCD.3 35 Preoperative DL-PFC-thalamic and ACC-thalamic streamline counts have been found to predict response to capsulotomy,37 and DBS of streamlines connecting ACC, VL-PFC, thalamus and the subthalamic nucleus has been associated with clinical improvement.38 39

ALIC topography varies by individual, but there is a consistent organisation in which medial OFC projections are most ventral, followed by lateral OFC, ACC and VL-PFC, followed most dorsally by DL-PFC and DM-PFC.22 23 Individual surgeons have elected to target ventral ALIC only, ventral-ALIC and mid-ALIC, or the entire ALIC. A meta-analysis found no effect of dorsal lesion extent on OCD symptom reduction or adverse effects.7 However, this analysis included few ventral lesions and did not account for precise lesion size location, size or continuity.

Few authors have analysed lesion size with respect to outcome. One of the few studies to do so found no effect of lesion volume of improvement in Y-BOCS score. However, this study highlights the small size of RF and SRS lesions, with a mean total lesion volume of 1.6±0.8 cc despite including the dorsal ALIC.31 Mean ablation volume after LITT was 50% larger (2.4±0.6 cc) in the present study.

Given the larger ablation volumes in this study and the positive relationship between ablation volume and outcome, one might expect a higher responder rate and greater reduction in Y-BOCS scores compared with other studies. Sparing of OFC projections could have negatively affected outcomes, since many described targets extend inferiorly to or below the level of the anterior commissure,5 6 23 and sparing of the ventral-most ALIC was common in this cohort (mean 16% of ALIC length ventral to ablation). However, ventral ALIC coverage did not correlate with improvement in Y-BOCS score.

The association of lesion volume with clinical improvement is presumably related to individual variation in ALIC structure and the roles of different PFC regions in OCD. Preoperative tractography obtained from a subset of patients in the present study illustrated how different ablation volumes would be necessary to completely disconnect OFC projections in different patients. Moreover, disconnection of OFC alone may be insufficient. Creation of larger lesions may be beneficial to ensure coverage of the various and variable PFC projections, particularly in the absence of evidence to suggest any adverse neurocognitive effects of dorsal lesion extent.7 LITT permits creation of large, sharply demarcated lesions with the safeguard of real-time MRI.


Long-term outcome data are not yet available for most patients. Likewise, most patients have not yet completed formal neuropsychiatric testing after surgery. Objective measures of depressive symptoms are not reported, though no patients developed major depressive disorder after surgery. A follow-up study is currently planned to assess for change in cognition, mood and anxiety after laser capsulotomy. Quantitative comparison of individual tractography is challenging, and streamline counts are known to be imprecise measure of connectivity.40 Normative tractography overlooks individual variation in connectivity but nonetheless has been successfully employed to predict outcomes of DBS for OCD.38


Laser anterior capsulotomy is an effective surgical therapy for severe refractory OCD. Immediate benefit in OCD symptoms has only been reported for LITT and RF. Y-BOCS improvement and response rates are comparable between LITT and other techniques. LITT permits the creation of large lesions under MRI guidance, facilitating disconnection of the cortical–striatal–thalamic pathways implicated in OCD. Furthermore, LITT yields sharply demarcated lesions with real-time MRI guidance, protecting critical neighbouring tracts.

Data availability statement

Data are available upon reasonable request.

Ethics statements

Patient consent for publication

Ethics approval

This study was approved by the University of Chicago institutional review board (protocol IRB20-1245).



  • Contributors Conception or design of the work, critical revision of the article and final approval of the version to be published: all authors. Data collection: DS, AM and JEG. Data analysis and interpretation: DS. Drafting the article: DS.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Linked Articles