Medialthalamotomies were introduced in the late 1940s by various functional neurosurgical groups for the treatment of neuropathic (i.e., neurogenic) pain.1–7They were soon recognized as procedures with low complication rates and without risk for the development of iatrogenic pain manifestations or somatosensory deficits. This is in contrast to lesions located more laterally in the ventroposterior nucleus.8Pain relief was shown to be achieved for all body locations. With some exceptions, these early relatively small series showed frequent, more or less complete recurrence of the original pain. In 1911, Head and Holmes published their hypothesis of an essential medial thalamic center, localized medial to a pain-generating lesion in the posterolateral thalamic region.9Later, Sano1早期的证据关于荷兰的一代ormal impulses in the ventroposterior nucleus and their amplification in a reverberating circuit between lateral and medial thalamic nuclei. In an effort starting in the early 1990s, Jeanmonod and collaborators defined the posterior part of the central lateral nucleus (CLp) in the medial thalamus using multiarchitectonic studies and intraoperative single-cell recordings and showed that it is an effective surgical target.10–15The central lateral nucleus (CL) has been considered as part of the anterior group of the intralaminar nuclei, but in nonhuman primates and humans, there is a posterior extension of the CL between the posterior pole of the mediodorsal nucleus and the medial pulvinar nucleus (PuM). The CL has important cortical projections that extend to large cortical domains. As such, it is in position to transfer nociceptive information, conveyed through the spinothalamic and spino-reticulo-thalamic tracts, to large cortical domains, including areas involved in nociception.15The CL has known afferents from the spinothalamic tract.16It is distant from primary somatosensory nuclei and has a combination of "diffuse" (layer I) and "nondiffuse" (layers III and IV) cortical projections including areas mediating discriminative (primary somatosensory cortex, secondary somatosensory cortex, posterior insula, and posterior parietal), affective-motivational (anterior cingulate and anterior insula), cognitive (prefrontal), and motor (premotor cortex) aspects of pain. Advantages of the CLp over other surgical targets in the medial thalamus (center median/parafascicular [CM/Pf] complex, or PuM) were highlighted previously.15Intraoperative microelectrode recordings in the CLp were shown to display low threshold calcium spike bursts (average interburst frequency of 4 Hz) in as much as half of the recorded neurons, and < 1% of them responded to sensory or motor stimulations.12The rationale of performing a CLp thalamotomy (CLT) is based on the fact that this area has lost its normal function over time (< 1% cells with receptive fields and absence of deficits after CLT) and sustains/amplifies a deleterious low-frequency overproduction (approximately 4 Hz), which is the source of a thalamocortical dysfunctional mechanism, that is, thalamocortical dysrhythmia. Thalamocortical dysrhythmia is characterized by the coupled increase of low- and then high-frequency activities in cortical areas of the pain matrix, which were demonstrated by quantitative electroencephalography (EEG) recordings.17
Patient series of CLT in chronic and therapy-resistant neuropathic pain were performed first with radiofrequency lesioning11,13and later with the new technology of the transcranial MR-guided focused ultrasound (MRgFUS) surgery.18–20Gamma Knife technology was and still is used to perform medial thalamotomies in neuropathic pain21–25(CM/Pf complex or CLp more recently). Deep brain stimulation (DBS) of the CL was first tried very recently in a pilot study.26
MRgFUS has been shown to be a safe and accurate lesioning technique27,28removing the risk of infection, strongly reducing the risk of bleeding, and providing a targeting accuracy within 1 mm. This is a case series of patients with chronic and therapy-resistant neuropathic pain syndromes with long-term follow-ups after MRgFUS CLT performed in a single center over the past 11 years.
开云体育世界杯赔率
Study Context
This is a single-center, nonrandomized cross-sectional retrospective analysis of consecutive patients with chronic and therapy-resistant neuropathic pain followed after MRgFUS CLT between 2011 and 2022. The first 11 treatments were part of a study approved by the Ethics Committee of Aargau/Solothurn and Swissmedic and were sponsored by Insightec Ltd. All patients treated in this study signed an informed consent form after having been fully informed about the treatment, its results, and its risks.
Therapy resistance and thus the indication for MRgFUS CLT was based on the lack of efficacy and/or side effects of antiepileptic and antidepressant drugs during at least 1 year. Pain taxonomy followed the updated system of the International Association for the Study of Pain for neuropathic pain.29Diagnostic criteria for classical and idiopathic trigeminal neuralgia were as previously published.20The case report of a patient with cluster headache included in this analysis was recently published.30
Surgical Procedure and Target Determination
All treatments were performed using a 3T MRI system (GE, Discovery 750, GE Healthcare) using the Exablate Neuro (Insightec). CLT was planned on maps from theAtlas of the Human Thalamus and Basal Ganglia.14Its 3D coordinates in relation to the anterior commissure–posterior commissure line were transferred onto operative MR images and were modified according to individual anatomy as seen on the preoperative high-resolution MR images cut in the stereotactic planes. Accuracy determination and target reconstruction were performed according to the method of Moser et al.28based on findings on MRI performed 2 days after MRgFUS CLT. Determination and coverage of the CLT target evolved over years of experience with the MRgFUS system. At first, only one sonication spot was placed 6 mm dorsal to the intercommissural plane and 8 mm from the medial thalamic border where CLp output fibers converge.18The present targeting strategy has as a goal of including the largest part of the CLp and consists of a set of four target subunits placed at 6 mm (2 subunits) and 8–9 mm (2 subunits) dorsal to the intercommissural plane. The anteroposterior position of the target subunits is determined based on visualization of the junction between the mediodorsal and PuM nuclei on preoperative axial T2-weighted images. This corresponds to the anteroposterior position of the CLp, centered in our experience between 3 mm anterior and 1 mm posterior to the posterior commissure. In the mediolateral dimension, two subunits are placed to cover the mediolateral extent of the CLp, from the medial thalamic border to 10 mm laterally (e.g., 5 and 8 mm laterally to the thalamic border for mediolateral positioning of the target subunit centers) (Fig. 1). The operative setup and strategy were described previously.18Target coverage was deemed sufficient based on the achieved thermal dose volume coverage at 240 cumulative equivalent minutes at 43°C, for all target subunits, as described for other targets.31,32This usually corresponds to temperatures at a focal point between 55°C and 58°C and sonication durations between 8 and 31 seconds.
Axial T1-weighted MR images obtained 2 days after the intervention, showing the four MRgFUS CLT target subunits, as described in the开云体育世界杯赔率, with superimposed modified atlas maps of Morel’sStereotactic Atlas of the Human Thalamus and Basal Ganglia146.3 (A) and 8.1 (B) mm dorsal to the intercommissural plane. ac = anterior commissure; D = dorsal; ic = internal capsule; LP = lateral posterior nucleus; MD = mediodorsal nucleus; mtt = mammillothalamic tract; pc = posterior commissure; VA = ventral anterior nucleus; VLa = ventral lateral anterior nucleus; VLp = ventral lateral posterior nucleus.
In most unilateral neuropathic pain syndromes, we perform CLT bilaterally on the basis of the presence of bilateral spinothalamic projections, the recording of bilateral pathological thalamic bursting activity, and the increase of pain relief brought by the addition of an ipsilateral CLT.33
Normal coagulation and blood pressure were checked for all patients prior to surgery. A dose of 10 mg domperidone (Motilium lingual) was given before starting the sonications. Since 2017, all patients received 20 mg of intravenous methylprednisolone in the hour after the end of the operation, 20 mg after 12 hours, and 2 mg of dexamethasone three times daily for 3–4 days to control/limit the perifocal edema of the lesion. Anesthesiological support was never needed.
Baseline Assessment and Follow-Up
疼痛评估与一个完整的神经examination including assessment of esthesia and algesia and evaluation for any side effects were performed after 2 days, 3 months, and 1 year. Later follow-up assessments were mostly performed through email and telephone conversations. Baseline and follow-up assessments included the items of the McGill Pain Questionnaire. Only 6 different pain qualities were systematically recorded (electricity, stinging, burning, shooting, tearing, and compression). Pain intensity was noted on a visual analog scale (VAS) for the least, the worst, and the mean pain intensities on a scale between 0 and 100 for both continuous pain and pain attacks. If not provided by the patient, the arithmetical mean pain intensity was calculated. A pain attack was defined as a sharp and short-lived (seconds to a few minutes) increase in pain.
患者要求percentage value of postoperative pain relief compared with their preoperative state (baseline). Good outcome was defined as pain relief ≥ 50%. A recurrence was defined as an initial good outcome (pain relief ≥ 50%) and later a decrease of pain relief to < 50% and/or recurrence of pain attacks.
The Mini–Mental State Examination and later the Montreal Cognitive Assessment (MoCA) were performed preoperatively and after 2 days and 1 year follow-up. All patients were asked to complete the Hospital Anxiety and Depression Scale (HADS)34at baseline and each follow-up examination. Three questions of World Health Organization Quality of Life Questionnaire Brief Version (WHOQOL-BREF)35were recorded at baseline and at the follow-ups.
Repeat MRgFUS CLT was offered when patients met the criteria of symptom recurrence (defined above) and evidence (radiology and intraoperative thermal doses) for insufficient target coverage.
Statistical Analysis
Statistical analysis was performed using Analyse-it for Microsoft Excel version 6.15 (Analyse-it Software Ltd.), as well as with SigmaPlot version 15.0 (Systat Software Inc.). Statistical comparisons of means were obtained by ANOVA (Kruskal-Wallis ANOVA on ranks when normal distribution and equal variance failed) and Dunn’s method for post hoc pairwise multiple comparison. If not specified otherwise, values given in the text are means ± standard deviation; p ≤ 0.05 indicates statistical significance.
Results
Patient characteristics are summarized inTable 1. Sixty-three consecutive MRgFUS CLT interventions were performed in 55 patients. The ages of the patients ranged between 38 and 80 years (mean 62 ± 11 years). The mean duration of the pain syndrome between onset and MRgFUS CLT was 14 ± 11 years (range 3–47 years). Nine patients were included in earlier publications.20,30
Patient characteristics
| Value | |
|---|---|
| No. of interventions | 63 |
| No. of pts | 55 |
| Mean age at treatment, yrs (range) | 62 ± 11 (38–80) |
| Mean pain duration, yrs (range) | 14 ± 11 (3–47) |
| Female sex, % | 32% |
| No. of CLT targets | 112 |
| Lt | 53 |
| Rt | 59 |
| No. of pts w/ bilat CLT | 48 |
| No. of pts w/ contralat CLT only | 7 |
| Side effects, % of targets | 0.9 |
| No. of repeat MRgFUS CLTs | 8 |
| No. of pts w/ previous pain ops (%) | 26 (47) |
| No. of pain ops after MRgFUS CLT | 1 |
| No. of deaths | 5 |
| Follow-up, mos | |
| Mean ± SD (range) | 55 ± 40 (3–132) |
| Median | 46 |
| No. of interventions in each neuropathic pain subgroup | |
| Spinal cord injury | 14 |
| Secondary trigeminal neuralgia | 11 |
| Classical & idiopathic trigeminal neuralgia | 8 |
| Postdiscectomy radiculopathy | 6 |
| Thalamic infarction | 5 |
| Neuropathy | 5 |
| Cortical & basal ganglia infarctions | 3 |
| Brainstem injury | 3 |
| Plexus avulsion | 3 |
| Amputation | 2 |
| Postherpetic neuralgia | 2 |
| Cluster headache | 1 |
Pt = patient.
Prior to MRgFUS CLT, 26 patients (47%) had a functional neurosurgical intervention for their neuropathic pain syndrome (discectomies, spine decompression, and fusion were not counted). After MRgFUS CLT, 1 patient underwent pain surgery (1 glycerol rhizotomy). Overall, 112 CLT targets were performed (53 left and 59 right CLTs). One international patient was lost to follow-up (i.e., the patient was examined 2 days after MRgFUS but did not keep contact with our clinic for later assessments, either in person or remotely).
Overall, 30% of patients came from outside of Switzerland. Five patients had died of unrelated causes at the time of this report. The follow-up duration ranged between 3 and 132 months (mean 55 ± 40 months, median 46 months). The Kaplan-Meier estimator of follow-up durations is depicted inFig. 2.
Kaplan-Meier estimator of the follow-up durations in this cross-sectional retrospective study.
Surgery
The vast majority of patients received bilateral MRgFUS CLT (48 patients) in one session. In 7 patients, MRgFUS CLT was performed on the side contralateral to their pain syndrome only. Repeat MRgFUS interventions were performed in 8 patients (5 bilaterally and 3 on one side only).
The mean skull density ratio of the patients was 0.56 ± 0.1 (range 0.37–0.77, median 0.56). The mean operation duration from the start of stereotactic frame fixation to its removal was 5.3 ± 1.9 hours (range 3.2–10 hours). For the final 20 treatments, the duration was 4.1 ± 0.6 hours (range 3.2–5.3 hours). The mean lesion volume measured on axial T2-weighted images 2 days after treatment was 160 ± 96 mm3(range 13–452 mm3). The sonication duration lasted between 8 and 31 seconds. The average power of the final sonications was 1036 ± 190 W. Accuracy of targeting was determined on axial T2-weighted images 2 days after CLT and showed similar values as published previously, with a mean targeting accuracy for each of the three dimensions below 0.5 mm.27,28All patients were discharged after a 1-night hospital stay.
Morbidity
There was no bleeding or infection. A serious neurological adverse event was recorded in 1 patient (1.6% of interventions or 0.9% of the MRgFUS CLT targets performed). This happened during a repeat MRgFUS CLT in a patient previously treated with radiofrequency ablation. The patient reported numbness on his upper lip laterally, and we found a hypoalgesia of moderate intensity in a 1- to 2-cm2area. The numbness was regredient but did not disappear at follow-up 2 years after MRgFUS CLT. Sonications were painful for a few seconds in 35 cases (56%), but a change in the procedure or sedation was never required.
Pain Relief
The mean pain relief rated by patients was 42% ± 32% at 3 months (n = 61), 43% ± 36% at 1 year (n = 57), and 42% ± 37% at the last follow-up (mean 55 months, n = 63). All patients who could not be reached 1 year after MRgFUS CLT were assumed to have 0% pain relief.Figure 3illustrates the pain relief distribution across the patient series. Pain relief was ≥ 30% in 65% of cases at 3 months (n = 61), in 63% of cases at 1 year (n = 57), and in 61% of cases at the last follow-up (mean 55 months) (n = 63). Good outcomes (≥ 50% pain relief) were found in 54% of cases at 3 months (n = 61), in 49% of cases at 1 year (n = 57), and in 51% of cases at the last follow-up (n = 63). An absence of pain relief was reported in 28% of cases at 3 months (n = 61), in 28% of cases at 1 year (n = 57), and in 35% of cases at the last follow-up (n = 63).
Quantile boxes. The median is plotted as avertical thick line, 1st and 3rd quartiles as abox, mean pain relief as alarger blue square, and all individual results as small squares (light gray). The confidence interval of the box plots is 99%. Figure is available in color online only.
The number of pain attacks decreased from a mean of 113 ± 231 (n = 39) per day at baseline to 11 ± 29 (n = 35, p < 0.001, 90% reduction) at the 3-month follow-up and 9 ± 30 (n = 30, p < 0.001, 92% reduction) 1 year after MRgFUS CLT. The mean percentage of pain attack reduction was 66% at 3 months and 76% at the 1-year follow-up. At the 1-year follow-up, pain attacks were suppressed in 33% (12/36).
In addition to the analysis of the whole patient group, 4 subgroups with more than 5 interventions were analyzed separately. Subgroup analysis of interventions against classical and idiopathic trigeminal neuralgia (n = 8) revealed a mean pain relief of 76% ± 19% at a mean last follow-up of 80 ± 31 months, and all 8 patients had ≥ 50% pain relief. For secondary trigeminal neuralgia (n = 11), the mean pain relief was 31% ± 42% at the last follow-up (59 ± 35 months), and 4 patients had ≥ 50% pain relief. Interventions for spinal cord injury (n = 14) revealed a mean pain relief of 35% ± 24% at the last follow-up (36 ± 32 months), and 5 patients had ≥ 50% pain relief. For postdiscectomy radiculopathy (n = 6) the mean pain relief was 68% ± 37% at the last follow-up (70 ± 49 months), and 5 patients had ≥ 50% pain relief.
VAS Results
At baseline, a continuous pain component was reported in 90% of cases and pain attacks in 71%. The evolution of mean VAS scores from baseline to 3 months and 1 year after CLT is shown inTable 2. At the 1-year follow-up, the mean maximum VAS score for continuous pain was reduced by 34%, and its mean VAS was reduced by 41%. The mean maximum VAS score for pain attacks at 1 year was reduced by 46% and its mean VAS by 49%. Drug intake (opiates, antiepileptics, and antidepressants) at baseline and at follow-up examinations are detailed inTable 2.
Pain intensity according to VAS scores and drug intake at baseline and follow-up
| Baseline | 3-mo FU | p Value | 1-yr FU | p Value | |
|---|---|---|---|---|---|
| Continuous pain | |||||
| Average of mean VAS score | 54 ± 16 (n = 57) | 43 ± 23 (n = 49) | 0.017* | 32 ± 25 (n = 42) | <0.001*** |
| Mean minimum VAS score | 30 ± 22 | 22 ± 21 | 0.15 | 16 ± 20 | 0.002** |
| Mean maximum VAS score | 77 ± 20 | 62 ± 28 | 0.009** | 51 ± 37 | <0.001*** |
| Suppressed continuous pain | 10% (5/49) | 29% (12/42) | |||
| Pain attacks | |||||
| Average of mean VAS score | 76 ± 16 (n = 45) | 47 ± 35 (n = 40) | <0.001*** | 39 ± 35 (n = 36) | <0.001*** |
| Mean minimum VAS score | 62 ± 25 | 41 ± 35 | 0.014* | 31 ± 30 | <0.001*** |
| Mean maximum VAS score | 88 ± 14 | 52 ± 37 | <0.001*** | 48±40 | <0.001*** |
| Suppressed pain attacks | 25% (10/40) | 33% (12/36) | |||
| Mean no. of pain attacks per day | 113 ± 231 (n = 39) | 11.4 ± 29 (n = 35) | <0.001*** | 9.2 ± 30 (n = 30) | <0.001*** |
| Drug intake | |||||
| Opiates | 32% (20/63) | 13% (8/60) | 16% (9/56) | ||
| Antiepileptics | 56% (35/63) | 33% (20/60) | 46% (26/56) | ||
| Antidepressants | 49% (31/63) | 45% (27/60) | 36% (20/56) |
FU = follow-up.
*p ≤ 0.05;
**p ≤ 0.01; and
***p ≤ 0.001.
Numbers in parentheses indicate the number of interventions.
Pain Characteristics
The pain was localized in 30% of cases on the left, in 37% on the right, and in 33% on both sides of the body. A total of 243 pain qualities were recorded at baseline (mean 3.9 ± 1.3, median 4 per patient). There were 129 pain qualities (mean 2.3 ± 1.3, median 2 per patient) at the 3-month follow-up and 104 at the 1-year follow-up (mean 2.1 ± 1.2, median 2 per patient).Table 3shows the 6 recorded pain qualities at baseline and their evolution at 3 months and 1 year. At 3 months, 2 patients reported a new burning pain quality and 1 patient a new tearing quality. The 2 new burning qualities were not reported again at 1 year. At 1 year, 4 patients reported a newly developed pain quality (electricity [n = 1], stinging [n = 1], and tearing [n = 2]).
Presence of pain qualities at baseline and percentage of patients for whom they were suppressed at follow-up
| Pain Quality | Baseline | 3-mo FU (n = 57) | 1-yr FU (n = 49) |
|---|---|---|---|
| Electricity | 62% | −49% | −59% |
| Stinging | 87% | −51% | −47% |
| Burning | 81% | −29% | −45% |
| Shooting | 38% | −75% | −79% |
| Tearing | 57% | −39% | −61% |
| Compressive | 60% | −53% | −68% |
Allodynia was found in 32 of the 63 cases at baseline. At 3 months (n = 61), it was suppressed (16%) or strongly reduced (34%) in 50% of cases. At 1 year (n = 54), allodynia was suppressed in 40% (10/25) or reduced in 28% (7/25) of cases. No patient developed new allodynia after MRgFUS CLT.
Bodily pain distribution as drawn by the patients was compared with the baseline drawings. Its extension was reduced in 51% of cases at 3 months (6% with ≥ 50% pain reduction and 45% with < 50%) and in 57% at the 1-year follow-up (24% with ≥ 50% pain reduction and 33% with < 50%). Interestingly, among the 16 cases without any pain relief at 1 year, 6 (38%) drew a clearly smaller painful area. In 3 cases, the reduction of pain extension was > 50%. In 2 cases, the reduction in pain extension preceded a significant pain reduction (70% and 100% at the last follow-up).Figure 4exemplifies the variability between pain relief and extension of bodily pain distribution, both reported by the patient himself.
Bodily pain distribution as drawn by the patient at baseline (A, C, and E) and 1 year after bilateral MRgFUS CLT (B, D, and F).A and B:Postherpetic neuralgia (dermatomes T8 and T9 with extension to T10–T12 on the left) 6 years prior to MRgFUS CLT with 80% pain relief.C and D:Peripheral neuropathic pain after cervical hernia and microdiscectomy (C5–6 and C6–7) 6 years prior to MRgFUS with 70% pain relief.E and F:Neuropathic pain after spinal cord injury at the thoracic level T4 34 years prior to MRgFUS CLT with 0% pain relief. Figure is available in color online only.
Secondary Outcome Measures
Answers to questions 1, 2, and 17 of the WHOQOL-BREF showed statistically significant improvement of these 3 items of quality of life (Table 4). HADS scores were reduced (improvement) from baseline 15.0 ± 8.0 (over a maximum of 42) to 12.2 ± 8.0 (p = 0.05) at 3 months and 11.7 ± 8.0 (p = 0.02) at the 1-year follow-up. The anxiety component of the score was 7.5 ± 4.2 at baseline, 6.2 ± 4.5 (p = 0.09) at 3 months, and 6.0 ± 4.3 (p = 0.05) at 1 year. The depression score was 7.2 ± 4.1 at baseline, 6.0 ± 4.2 (p = 0.3) at 3 months, and 5.6 ± 5.2 (p = 0.03) at 1 year.
Quality of life at baseline and follow-up
| Baseline | 3-mo FU | p Value | 1-yr FU | p Value | |
|---|---|---|---|---|---|
| How would you rate your quality of life? (very poor = 1, very good = 5) | 2.7 ± 1.0 (n = 52) | 3.4 ± 0.9 (n = 43) | 0.009** | 3.4 ± 0.8 (n = 43) | 0.001** |
| How satisfied are you with your health? (very dissatisfied = 1, very satisfied = 5) | 2.2 ± 1.0 (n = 52) | 2.8 ± 1.1 (n = 44) | 0.007** | 3.1 ± 1.1 (n = 43) | <0.001*** |
| How satisfied are you with your ability to perform your daily living activities? (very dissatisfied = 1, very satisfied = 5) | 2.6 ± 1.2 (n = 51) | 3.3 ± 1.1 (n = 40) | 0.024* | 3.5 ± 1.1 (n = 40) | 0.003** |
*p ≤ 0.05;
**p ≤ 0.01; and
***p ≤ 0.001.
Questions 1, 2, and 17 of the WHOQOL-BREF.35Numbers in parentheses indicate the number of interventions.
The mean MoCA score was 27.9 ± 2.6 at baseline (n = 50), 28.9 ± 1.8 (p = 0.02) 2 days after MRgFUS CLT (n = 50), and 29.3 ± 1.1 (p = 0.01) at 1 year (n = 37). The MMST at baseline was 29.4 ± 0.8 (n = 13), 29.5 ± 0.5 (n = 13) at 3 months, and 29.8 ± 0.4 at 1 year (n = 13).
Discussion
This case series of 63 consecutive MRgFUS CLT interventions, performed in a single center and followed for a mean of 55 months, confirms the safety profile of the intervention18,20,27and shows sustained pain relief over time.13,20A notion present in the neurosurgical pain literature is that medial thalamotomies are safe but their efficiency does not hold over time.8Mean pain relief as reported by the patients was shown to remain remarkably stable (42% at 3 months, 43% at 1 year, and 42% at the last follow-up). The percentage of patients with "good" pain outcomes (≥ 50% pain relief) was 54% at 3 months, 49% at 1 year, and 51% at the last follow-up. Only 1 patient was lost to follow-up. Thus, contrary to reports in the literature, pain relief did not decrease over time.
减少意味着脉管分数显示出类似的体育rcentage reductions as the pain relief percentage (−41% for continuous pain and −49% for pain attacks) at the 1-year follow-up. Although the mean intensity of the pain attacks as rated by the VAS was only reduced by 49%, the mean frequency of pain attacks was reduced by 92%. Allodynia was reduced or suppressed in 68% of cases and never appeared de novo after MRgFUS CLT. More than half of the patients taking opiates at baseline had stopped their intake at the 1-year follow-up. Bodily distribution of pain as drawn by the patients globally correlated with the pain relief reported, although some exceptions were observed. InFig. 4Eand F, the patient presented with a substantial reduction of his pain area along with 0% pain relief. The pain relief percentages given by patients provide a global assessment of the obtained relief, which comprises the perception of pain and its psycho-emotional (paralimbic/multimodal) modulation. The relevance of this modulation factor is well recognized and has been addressed, among other studies, in the study by Michels et al.36In contrast to their patient subgroup with ≥ 50% pain relief, EEG overactivity was maintained 1 year after radiofrequency ablation in the subgroup with insufficient pain relief (< 50%). This happened in parallel with a positive correlation between a frustration scale and an increase in EEG cortical current source density. The latter was located in the bilateral frontopolar, prefrontal interhemispheric and orbitofrontal paralimbic/multimodal cortical areas. The reduction or suppression of a dysrhythmic cortical overactivity at the source of neuropathic pain can thus be hindered by emotions, and, in that case, frustration.
The reduction of pain qualities was higher in general at the 1-year than at the 3-month follow-up assessment. Among the 6 recorded pain qualities mostly associated with neuropathic pain in our experience, shooting, electricity, tearing, and compression were reduced in more than 50%. Stinging (−47%) and burning (−45%) pain qualities were more resistant to MRgFUS CLT.
The MoCA showed a slight but nevertheless statistically significant improvement in the mean score at the 1-year follow-up compared with baseline. Some learning effect of recurrent testing cannot be ruled out. The mean HADS score was significantly improved at the 1-year follow-up. Selected questions of the WHOQOL-BREF (seeTable 4) revealed statistically significant improvements.
四个亚组分析显示相似pain relief for secondary trigeminal neuralgia and spinal cord injury, with long-term mean pain relief of 31% and 35%, respectively. Better results were obtained for the subgroups of postdiscectomy radiculopathy and classical and idiopathic trigeminal neuralgia, where long-term pain relief was 68% and 76%, respectively. Particularly good results of CLT for the treatment of classical and idiopathic trigeminal neuralgia were published in 2020.20This observation was also shared by Franzini et al.24using the Gamma Knife to perform CLT for trigeminal pain syndromes.
DBS was first explored for the treatment of pain before being dominantly applied against Parkinson’s disease. However, the US FDA withdrew its approval for the treatment of pain in 1989. Two industry-sponsored trials (in 1993 and 1997, reported in 200137) showed only an approximately 20% response rate, which led the manufacturer to withdraw its FDA application.38In 2006, Hamani et al.39reported that 24% of patients maintained long-term benefit of their DBS implantation against pain. At the 1-year follow-up, 76% of patients received no definitive implant or their device was switched off. Later studies showed some benefits, even in the long term but without overcoming the threshold of ≥ 50% pain relief in more than 50% of patients.40–43Targets used in DBS were the ventral posterior medial nucleus (VPM) and ventral posterior lateral nucleus (VPL), the periaqueductal gray matter, the CM/Pf complex, and only very recently one trial of 6 patients with the combined targets of CL and VPM and VPL.26One major difficulty when analyzing results in the context of medical implants is the number of patients who undergo trials but do not definitively undergo implantation. Rasche et al.,42for example, applied very strict criteria for successful electrode test trials. Thus, only 57% of patients qualified for definitive internalization of their DBS system.
Results of radiosurgical medial thalamotomies were recently reviewed by Franzini et al.44池6研究(共有118名患者的结果) reported a "meaningful" pain reduction in 38% of patients at the last follow-up. However, the presence of 51% of cancer pain patients in these series brings uncertainty as to the obtained pain relief for neuropathic pain cases.
Side Effects
As expected from an incisionless technique, there was no bleeding and no infection. One patient developed a somatosensory deficit in the trigeminal area most likely due to encroachment of the therapeutic lesion on the VPM nucleus during MRgFUS CLT. This happened in the context of previous radiofrequency lesioning in his medial thalamus, with probable gliotic tissue displacement. The patient had not fully recovered from his deficit at the last follow-up. This remains the only significant side effect recorded in 112 lesions of the CLT in this series. This compares favorably with the side effects of any implanted devices.45Pooled Gamma Knife studies on medial thalamotomy reported an adverse event rate of 5% (4% related to radiation).44
Study Limitations
This consecutive surgical case series has the limitation of its retrospective nature and not being randomized and sham controlled. In addition, various neuropathic pain conditions were analyzed as a single group. A larger study would allow more subgroup analyses as presented in a recent contribution.20
Conclusions
The overall side-effect profile of MRgFUS CLT, retrospectively studied in 63 interventions, speaks for a safe neurosurgical approach. The efficacy in relieving pain in chronic and therapy-resistant neuropathic pain syndromes, although modest as shown by its average pain relief (42%), proved to be remarkably stable over time. At a mean follow-up duration of 55 months, more than 50% of patients reported ≥ 50% pain relief. Patients with classical and idiopathic trigeminal neuralgia reported a higher average pain relief (76%) compared with the whole patient group.
Acknowledgments
We thank Dr. Payam Pourtehrani and colleagues at Rodiag Diagnostic Centers Solothurn for MRI, Dr. Alexander Arnold for internal medicine support, Drs. Maja Strasser and Robert Bühler for neurological examinations, Tanja Thalmann and Samuel Ryser for nursing care, and Franziska Rossi for administrative support.
Disclosures
Dr. Jeanmonod reported grants from Insightec Ltd. (the first 11 treatments were sponsored by Insightec Ltd.) during the conduct of the study.
Author Contributions
概念和设计:年度最佳。采集的数据: Gallay, Magara, Moser, Kowalski, Jeanmonod. Analysis and interpretation of data: Gallay, Magara, Kaeser, Jeanmonod. Drafting the article: Gallay, Jeanmonod. Critically revising the article: all authors. Reviewed submitted version of manuscript: all authors. Approved the final version of the manuscript on behalf of all authors: Gallay. Statistical analysis: Kaeser. Administrative/technical/material support: Moser. Study supervision: Gallay, Jeanmonod.
References
-
1 ↑
SanoK.Intralaminar thalamotomy (thalamolaminotomy) and postero-medial hypothalamotomy in the treatment of intractable pain. In:KrayenbühlH,MaspesPE,SweetWH, eds.Pain—Its Neurosurgical Management.Vol 8.Karger Publishers;1977:50–103.
-
2
HécaenH,TalairachJ,DavidM,DellMB.Coagulations limitées du thalamus dans les algies du syndrome thalamique. Résultats thérapeutiques et physiologiques.Rev Neurol (Paris).1949;81(11):917–931.
-
3
FairmanD.Unilateral thalamic tractotomy for the relief of bilateral pain in malignant tumors.Confin Neurol.1967;29(2):146–152.
-
4
RichardsonDE.Thalamotomy for intractable pain.Confin Neurol.1967;29(2):139–145.
-
5
LaitinenLV.Mesencephalotomy和thalamotomy for chronic pain. In:LunsfordLD, ed.Modern Stereotactic Neurosurgery.Springer;1988:269–277.
-
6
GybelsJM,SweetWH.Neurosurgical Treatment of Persistent Pain.Physiological and Pathological Mechanisms of Human Pain.Vol 11.Karger;1989.
-
7 ↑
HitchcockER,TeixeiraMJ.A comparison of results from center-median and basal thalamotomies for pain.Surg Neurol.1981;15(5):341–351.
-
8 ↑
MarkVH,ErvinFR,HackettTP.Clinical aspects of stereotactic thalamotomy in the human. Part I. The treatment of chronic severe pain.Arch Neurol.1960;3(4):351–367.
-
10 ↑
JeanmonodD,SarntheinJ,MagninM,et al.Chronic neurogenic pain: thalamocortical dysrhythmic mechanisms and their surgical treatment.Thalamus Relat Syst.2005;3(1):63–70.
-
11 ↑
JeanmonodD,MagninM,MorelA.Thalamus and neurogenic pain: physiological, anatomical and clinical data.Neuroreport.1993;4(5):475–478.
-
12 ↑
JeanmonodD,MagninM,MorelA.Low-threshold calcium spike bursts in the human thalamus. Common physiopathology for sensory, motor and limbic positive symptoms.Brain.1996;119(Pt 2):363–375.
-
13 ↑
JeanmonodD,MagninM,MorelA,SiegemundM.Surgical control of the human thalamocortical dysrhythmia: I. Central lateral thalamotomy in neurogenic pain.Thalamus Relat Syst.2001;1(1):71–79.
-
15 ↑
JeanmonodD,MorelA.The central lateral thalamotomy for neuropathic pain. In:LozanoAM,GildenbergPL,TaskerRR, eds.Textbook of Stereotactic and Functional Neurosurgery.Springer;2009:2081–2096.
-
17 ↑
SarntheinJ,SternJ,AufenbergC,RoussonV,JeanmonodD.Increased EEG power and slowed dominant frequency in patients with neurogenic pain.Brain.2006;129(Pt 1):55–64.
-
18 ↑
JeanmonodD,WernerB,MorelA,et al.Transcranial magnetic resonance imaging-guided focused ultrasound: noninvasive central lateral thalamotomy for chronic neuropathic pain.Neurosurg Focus.2012;32(1):E1.
-
19
MartinE,JeanmonodD,MorelA,ZadicarioE,WernerB.High-intensity focused ultrasound for noninvasive functional neurosurgery.Ann Neurol.2009;66(6):858–861.
-
20 ↑
GallayMN,MoserD,JeanmonodD.MR-Guided Focused Ultrasound Central Lateral Thalamotomy for Trigeminal Neuralgia.Front Neurol.2020;11:271.
-
21 ↑
YoungRF,JacquesDS,RandRW,CopcuttBR.Medial thalamotomy with the Leksell Gamma Knife for treatment of chronic pain.Acta Neurochir Suppl.1994;62(105):110.
-
22
UrgosikD,LiscakR.Medial Gamma Knife thalamotomy for intractable pain.J Neurosurg.2018;129(suppl 1):72–76.
-
23
Pérez de la TorreRA,Rodríguez HernándezJJ,Al-RamadanA,GharaibehA.Management of phantom limb pain through thalamotomy of the centro-median nucleus.Neurol Int.2021;13(4):587–593.
-
24 ↑
FranziniA,AttuatiL,ZaedI,et al.Gamma Knife central lateral thalamotomy for the treatment of neuropathic pain.J Neurosurg.2021;135(1):228–236.
-
25 ↑
LovoEE,TorresB,CamposF,et al.Stereotactic gamma ray radiosurgery to the centromedian and parafascicular complex of the thalamus for trigeminal neuralgia and other complex pain syndromes.Cureus.2019;11(12):e6421.
-
26 ↑
NowackiA,ZhangD,BarlateyS,et al.Deep brain stimulation of the central lateral and ventral posterior thalamus for central poststroke pain syndrome: preliminary experience.Neuromodulation.10月17日在线发表2022. doi:10.1016/j.neurom.2022.09.005
-
27 ↑
GallayMN,MoserD,JeanmonodD.Safety and accuracy of incisionless transcranial MR-guided focused ultrasound functional neurosurgery: single-center experience with 253 targets in 180 treatments.J Neurosurg.2019;130(4):1234–1243.
-
28 ↑
MoserD,ZadicarioE,SchiffG,JeanmonodD.MR-guided focused ultrasound technique in functional neurosurgery: targeting accuracy.J Ther Ultrasound.2013;1:3.
-
29 ↑
FinnerupNB,HaroutounianS,KamermanP,et al.Neuropathic pain: an updated grading system for research and clinical practice.Pain.2016;157(8):1599–1606.
-
30 ↑
MagaraAE,GallayMN,MoserD,JeanmonodD.Complete resolution of chronic cluster headache following central lateral thalamotomy using incisionless MRI-guided focused ultrasound with 6 years of follow-up: illustrative case.J Neurosurg Case Lessons.2022;4(22):CASE22259.
-
31 ↑
GallayMN,MoserD,FederauC,JeanmonodD.Radiological and thermal dose correlations in pallidothalamic tractotomy with MRgFUS.Front Surg.2019;6:28.
-
32 ↑
GallayMN,MoserD,JeanmonodD.MR-guided focused ultrasound cerebellothalamic tractotomy for chronic therapy-resistant essential tremor: anatomical target reappraisal and clinical results.J Neurosurg.2021;134(2):376–385.
-
33 ↑
JeanmonodD,MagninM,MorelA.A thalamic concept of neurogenic pain. In:Proceedings of the 7th World Congress on Pain. Progress in Pain Research and Management.2卷.IASP Press Seattle;1994:767–787.
-
34 ↑
ZigmondAS,SnaithRP.The hospital anxiety and depression scale.Acta Psychiatr Scand.1983;67(6):361–370.
-
35 ↑
The WHOQOL Group.Development of the World Health Organization WHOQOL-BREF quality of life assessment.Psychol Med.1998;28(3):551–558.
-
36 ↑
MichelsL,Moazami-GoudarziM,JeanmonodD.Correlations between EEG and clinical outcome in chronic neuropathic pain: surgical effects and treatment resistance.Brain Imaging Behav.2011;5(4):329–348.
-
37 ↑
CoffeyRJ.Deep brain stimulation for chronic pain: results of two multicenter trials and a structured review.Pain Med.2001;2(3):183–192.
-
38 ↑
SchwalbJM,HamaniC.The history and future of deep brain stimulation.Neurotherapeutics.2008;5(1):3–13.
-
39 ↑
HamaniC,SchwalbJM,RezaiAR,DostrovskyJO,DavisKD,LozanoAM.Deep brain stimulation for chronic neuropathic pain: long-term outcome and the incidence of insertional effect.Pain.2006;125(1-2):188–196.
-
40 ↑
BoccardSGJ,PrangnellSJ,PycroftL,et al.Long-term results of deep brain stimulation of the anterior cingulate cortex for neuropathic pain.世界Neurosurg.2017;106:625–637.
-
41
BoccardSG,PereiraEA,MoirL,AzizTZ,GreenAL.Long-term outcomes of deep brain stimulation for neuropathic pain.开云体育app官方网站下载入口.2013;72(2):221–231.
-
42 ↑
RascheD,RinaldiPC,YoungRF,TronnierVM.Deep brain stimulation for the treatment of various chronic pain syndromes.Neurosurg Focus.2006;21(6):E8.
-
43 ↑
AbdallatM,SaryyevaA,BlahakC,et al.Centromedian-parafascicular and somatosensory thalamic deep brain stimulation for treatment of chronic neuropathic pain: a contemporary series of 40 patients.Biomedicines.2021;9(7):731.
-
44 ↑
FranziniA,RossiniZ,MoosaS,et al.Medial thalamotomy using stereotactic radiosurgery for intractable pain: a systematic review.Neurosurg Rev.2022;45(1):71–80.
-
45 ↑
OhMY,AboschA,KimSH,LangAE,LozanoAM.Long-term hardware-related complications of deep brain stimulation.开云体育app官方网站下载入口.2002;50(6):1268–1276.
