Posterolateral Sulcus Approach for Intramedullary Hemorrhage Associated With Cavernous Malformation of High Cervical Spine: Operative Technique and Outcomes
Article information
Abstract
Objective
Intramedullary hemorrhage (IH) associated with cavernous malformation (CM) of the high cervical spine remains a significant challenge for neurosurgeons. This study aimed to evaluate the efficacy and safety of the posterolateral sulcus (PLS) approach in managing these complex cases.
Methods
This single-center retrospective study included 58 cases of spinal intramedullary tumors treated surgically over the past 4 years. The PLS approach on the side of the IH was applied for the removal of CM. Neurological function was assessed using the modified McCormick functional scale (MMCS) before surgery, one week after surgery, and at the most recent follow-up.
Results
Six patients with IH associated with CM above the C3 level were identified from the database. The mean age was 31.2 years, and 4 of the 6 patients were female. Symptom duration prior to surgery ranged from 0 to 48 months. Total removal of the CM was achieved in all 6 cases without any serious adverse events including respiratory complications. The average follow-up duration was 21.7 months. The mean MMCS score was 3.0 before surgery, maintained at 2.5 in the early postoperative period, and improved further to 2.2 at the most recent follow-up. One patient of ventral-type CM experienced recurrent hemorrhage at the same level 30 months after the initial surgery. This patient subsequently underwent a second surgery using the anterolateral sulcus approach, which was well tolerated.
Conclusion
The PLS approach enables safe removal of CM even in the high cervical spine. However, ventral-type CMs remain a major surgical concern.
INTRODUCTION
Intramedullary spinal cord tumors (IMSCTs) or vascular lesions, such as ependymomas or cavernous malformations (CMs), are often at risk of causing intramedullary hemorrhage (IH) [1-3]. When IH occurs in the high cervical spine, it poses a significant risk of severe neurological deterioration. The spinal cord in this region blends imperceptibly into the medulla oblongata of the brainstem, a structure critical for maintaining vital life functions [4,5]. Even minimal surgical manipulation of intramedullary lesions may result in devastating neurological deficits [6]. Therefore, safe and highly precise surgical techniques are essential to minimize postoperative complications.
The posterolateral sulcus (PLS) approach, also known as dorsal root entry zone (DREZ) myelotomy, is considered one of the safer entry zones for IMSCTs or intramedullary spinal CMs when the lesion is laterally localized within the spinal cord [7-11]. One advantage of the PLS approach is a reduced risk of posterior column injury compared to the posterior median sulcus (PMS) approach. However, a known disadvantage is the increased risk of motor dysfunction on the side of the myelotomy, due to the proximity of surgical manipulation to descending fiber tracts in the lateral column of the spinal cord. To date, there is limited surgical evidence supporting the use of the PLS approach for intramedullary lesions of the high cervical spine. This study aimed to evaluate the efficacy and safety of the PLS approach for treating intramedullary CMs with hemorrhage in this challenging anatomical region.
MATERIALS AND METHODS
1. Study Design
This study was conducted as part of a retrospective, multicenter observational study approved by the institutional review board of Osaka Medical and Pharmaceutical University (approval number: 2020-219). Patients with IMSCTs who underwent surgical removal at our institution between January 2021 and February 2025 were included. A total of 58 surgical cases of IMSCTs, including CMs, were treated during this period. Among them, patients with IMSCTs located in the high cervical spine (above the C3 level) were selected and retrospectively reviewed. This study specifically focused on surgical cases of IH associated with CM above the C3 level.
2. Data Collection
Patient demographic information—including age, sex, spinal level of the CM, surgical data, and neurological status before and after surgery—was obtained from medical records. Surgical details, including the surgical approach, extent of resection, operative time, and estimated blood loss, were collected from operative records. Neurological symptoms were categorized using the Ogilvy classification12: (1) Discretely occurring episodes of neurological decline with varying degrees of recovery; (2) Acute onset with rapid neurological deterioration; (3) Acute onset of mild symptoms followed by progressive worsening over weeks to months; (4) Acute onset of mild symptoms followed by a stable, asymptomatic, transient, or minimally symptomatic state lasting months; (5) Slowly progressive neurological decline.
Symptom duration was defined as the period from symptom onset to the time of surgery. Neurological status before and after surgery was assessed using the modified McCormick functional scale (MMCS) [13] and manual muscle testing (MMT), particularly of the upper limbs. The MMCS is defined as follows: (1) neurologically normal; (2) presence of sensory deficits with or without mild gait difficulty; (3) presence of sensory deficits with or without moderate gait difficulty; (4) more severe neurological deficits requiring a cane/brace for ambulation or involving bilateral upper limb impairment, with partial or full independence; (5) requiring a wheelchair or cane/brace with bilateral upper limb impairment, usually not independent [9]. The location of the CM was determined using axial magnetic resonance imaging (MRI) and classified into 2 types: if the CM and associated hemorrhage extended to the ventral surface of the spinal cord, it was classified as ventral type; otherwise, it was considered dorsal type.
3. Surgical Indication and Technique
Surgery was indicated for symptomatic cases with progressive or persistent neurological deficits, or radiological evidence of recurrent hemorrhage or lesion enlargement.
The PLS approach on the side of the IH was applied for the removal of CM. The patient was placed in the prone position under general anesthesia. The thorax was elevated by 15°, and the head was maintained in a neutral flexed position without rotation. Just before the start of surgery, 500 mg of methylprednisolone was routinely administered intravenously. After the opening of wound, laminectomy was performed using the standard en bloc technique. The laminectomy was extended sufficiently to expose the entire lesion and widened to the medial pedicular surfaces. The dura mater was opened while preserving the arachnoid membrane. The arachnoid membrane was carefully incised to avoid injury at sites of arachnoid adhesion or vascular connection. The dorsal surface of the spinal cord was initially inspected with the aid of indocyanine green videoangiography (1) to visualize the dorsal surface vasculature and exclude abnormal flow patterns, (2) to confirm patency of the posterior spinal artery, and finally (3) to define the safest entry zone for myelotomy [14-17]. A linear incision was made along the PLS, just medial to the dorsal nerve roots, to initiate the myelotomy. The shallow vertical groove of the PLS naturally continues into the posterolateral tract of Lissauer. Any crossing vessels over the PLS were gently coagulated at low power under continuous saline irrigation. The myelotomy was extended rostrally and caudally by splaying the spinal tissue using a microdissector. Careful dissection of the posterolateral tract of Lissauer revealed the cleavage plane between the dorsal and lateral columns, which was further opened to expose the dorsal surface of the CM. Upon encountering the CM, the gliotic layer—covered with hemosiderin from prior hemorrhage—was gently dissected. This step was critical to preserving spinal cord function. In ventral-type lesions, dissection was continued cautiously until the ventral pia mater was identified. The entire procedure was performed with minimal bleeding. After complete removal of the CM, the spinal cord shape was restored by suturing the pial edges. The arachnoid membrane was then closed together with the dura mater to reduce the risk of postoperative arachnoid adhesions. The resected cervical laminae were reconstructed using a basket titanium plate and screws.
4. Intraoperative Neurophysiological Monitoring
Somatosensory evoked potentials, transcranial motor evoked potentials (MEPs) from extremity muscles, or D-waves were routinely used for intraoperative neurophysiological monitoring (IONM), except in emergency cases (Neuromaster MEE-2032GI; Nihon Kohden, Japan). The level of neuromuscular blockade was monitored by recording a train-of-four (TOF) response from the abductor hallucis longus muscles following tibial nerve stimulation. MEP recording was initiated after recovery of a TOF ratio of 1.0, prior to dural incision. Suprathreshold stimulation was performed using an intensity exceeding 100 mA to establish a baseline. Stimulation was delivered as a train of 5 to 7 pulses at intervals of 0.5 msec. IONM was conducted under total intravenous anesthesia, and no muscle relaxants were administered after the induction of general anesthesia. Neurophysiological data were displayed and stored throughout the procedure. In the case of muscle MEPs, a reduction in amplitude by more than 50% or complete disappearance was considered a positive alarm indicating electrophysiological abnormalities [18,19]. IONM was supervised by a dedicated neurophysiological technician.
5. Outcome Measures
Neurological status was evaluated using the MMCS prior to surgery, one week postoperatively, and at the most recent followup. Additionally, each patient completed a questionnaire regarding their return to social activities after surgery. In addition, neurological symptoms after surgery such as posterior column dysfunction or neuropathic pain were assessed at outpatient clinic.
6. Ethical Statement
This study was approved by the local ethics board of Osaka Medical and Pharmaceutical University (approval No. 2020-219). All procedures were conducted in accordance with the ethical standards of the Institutional Review Board and the 1964 Declaration of Helsinki. Written informed consent was not required due to the retrospective study design and the use of anonymized clinical data, as all patients had previously provided written consent for their treatment. However, an opt-out approach was adopted to ensure consent for participation in the study.
RESULTS
1. Demographic and Surgical Data of Patients
A total of 15 patients underwent removal of IMSCTs in the high cervical spine (Table 1). CM was the most common pathology, observed in 6 of the 15 patients (40%), followed by ependymal tumors. Both patients with astrocytic tumors had highly malignant histology. Table 2 summarizes the data for all 6 CM cases. The mean age was 31.2±11.5 years, and 4 of the 6 patients were female. Three patients experienced acute onset of mild symptoms followed by progressive worsening over weeks to months. The duration of symptoms prior to surgery ranged from 0 to 48 months. Patient 1 presented with acute onset and rapidly declining neurological status and underwent surgery on the same day as their emergency room visit (published data) [5]. All 6 patients demonstrated IH on MRI. Five CMs were located on the dorsal surface of the spinal cord, while the remaining lesion (patient No. 2) was ventrally located. All patients underwent surgical removal of the CM via the PLS approach on the side of the IH. Total removal of the CM was achieved in all 6 cases without any serious adverse events. None of the patients had any postoperative respiratory complications. The average followup duration after surgery was 21.7 months. The mean MMCS score was 3.0 before surgery, maintained at 2.5 in the early postoperative period, and improved further to 2.2 at the most recent follow-up. During the follow-up period, patient No. 2, who had a ventral-type CM, experienced recurrent hemorrhage at the same level 30 months after the initial surgery and subsequently underwent a second surgery. No other patients experienced recurrent hemorrhage. No patients developed a new deep sensation deficit related to posterior column dysfunction. Four patients experienced neuropathic numbness or slight pain in the ipsilateral upper limb or neck region after surgery, with management of conservative medication whereas the remaining 2 patients did not require medication. A questionnaire survey conducted at the outpatient clinic indicated that all patients returned to social activities.
Demographic data of the patients with spinal intramedullary tumors of high cervical spine (n=15)
Summary of the patients with spinal intramedullary cavernous malformation of high cervical spine
2. Intraoperative MEP Data
Two of the 6 patients experienced worsening MMT scores in the upper limb on the side of the myelotomy 1 week after surgery, while the remaining patients either maintained or improved their scores. IONM was performed in 5 of the 6 cases; Case 1 did not undergo IONM due to the emergency nature of the surgery. Following myelotomy via the PLS approach, a reduction in muscle MEP amplitude of more than 50% was observed in 4 cases, and complete disappearance of the MEP waveform was noted in 1 case. The MEP waveform gradually recovered in 3 cases toward the end of the surgeries, while 2 cases were not assessed. Of the 4 cases with reduced amplitude, 3 were ultimately determined to be false positives, while 1 was a true positive. The case with complete disappearance of the MEP waveform was also classified as a true positive. However, final assessments at the most recent follow-up indicated that all patients either maintained or improved their MMT scores. The relationship between intraoperative muscle MEP waveforms and postoperative outcomes is summarized in Table 3.
Relationship between intraoperative muscle MEP waveforms and postoperative outcomes
3. Illustrative Cases
1) Dorsal type: acute onset and rapidly declining symptoms and neurological status
Case 5: A 20-year-old female was admitted to our institution with acute onset and progressively worsening motor weakness of the right limbs (MMT grade 2) lasting one month. The MMCS just before surgery was assessed as grade 4. Preoperative MRI revealed IH located from the central to dorsal aspect at the C1 level (Fig. 1A–D). The lesion was classified as a dorsal-type CM with hemorrhage at the C1 level. The PLS approach was successfully applied on the left side to achieve complete removal of the CM (Fig. 1E–H). One week postoperatively, MMT on the side of the myelotomy remained unchanged. The MMCS slightly improved to grade 3 early after surgery and further improved to grade 2 at the most recent follow-up. No residual lesion was confirmed (Fig. 1I and J).
Case 5: A 20-year-old female presented with acute onset and progressively worsening upper limb weakness. Preoperative magnetic resonance imaging (MRI) revealed intramedullary hemorrhage located from the central to dorsal aspect at the C1 level (A: T1-weighted image; B and D: T2-weighted images; C: T2*-weighted image). In intraoperative findings, (E) indocyanine green videoangiography confirmed the absence of abnormal vascular structures. (F) Myelotomy was initiated by dissecting the posterolateral sulcus. (G) Careful dissection was performed between the gliotic tissue covered by the hemosiderin-deposited layer (white arrow) and the cavernous malformation (white asterisk). (H) Complete removal of the lesion was achieved. Postoperative MRI obtained immediately after surgery (I) and at the final follow-up (J) confirmed complete removal of the lesion.
2) Dorsal type: acute onset of mild symptoms with subsequent stable, asymptomatic, transient, or minimally symptomatic status lasting months
Case 6: A 47-year-old male presented with a history of at least 2 episodes of sudden-onset sensory disturbance in the left upper limb over the past 2 years. The most recent symptomatic episode occurred approximately 6 months prior to surgery. The MMCS score before surgery was grade 2. Preoperative MRI revealed IH located from the central to dorsal aspect at the C3 level (Fig. 2A). The lesion was classified as a dorsal-type CM with associated hemorrhage. The PLS approach was successfully applied on the left side, and complete removal of the CM was achieved (Fig. 2B–E). Although transient disappearance of the MEP waveform occurred on the side of the myelotomy, the surgery was not interrupted or discontinued with gradual recovery as the same level to the control (Fig. 2F). One week postoperatively, MMT on the side of the myelotomy showed slight deterioration compared to the preoperative score. The MMCS remained at grade 2 in the early postoperative period and was unchanged at grade 2 at the most recent follow-up. No residual lesion was confirmed (Fig. 2G).
Case 6: A 47-year-old male had a history of at least 2 episodes of intramedullary hemorrhage over the previous 2 years. (A) Preoperative T2-weighted magnetic resonance imaging (MRI) revealed intramedullary hemorrhage located from the central to dorsal aspect at the C3 level. In intraoperative findings, (B) indocyanine green videoangiography showed no abnormal vascular structures. (C) Myelotomy was initiated by dissecting the posterolateral sulcus. (D) Gentle dissection was performed to remove the cavernous malformation (white asterisk). (E) The lesion was completely resected. (F) Complete disappearance of the motor evoked potential (MEP) waveform during the posterolateral sulcus approach. At the time of myelotomy (red arrow), intraoperative MEP waveforms on the myelotomy side disappeared, with gradual recovery observed during surgery. This patient experienced slight upper limb weakness on the side of the myelotomy early after surgery, which fully resolved within 3 months. (G) Postoperative T2-weighted MRI obtained early after surgery confirmed complete removal of the lesion.
3) Ventral type: recurrence 30 months after the initial surgery
Case 2: A 43-year-old female presented with acute and progressively worsening motor weakness in the left upper extremity (MMT grade 4), accompanied by neck pain and sensory disturbance on the same side, lasting for 3 months. The MMCS score before surgery was assessed as grade 3. Preoperative MRI revealed IH located on the ventral surface of the spinal cord at the C2–3 level (Fig. 3A and B). The lesion was classified as a ventral-type CM with hemorrhage. The PLS approach was successfully applied on the left side, and the CM was completely removed until the ventral pia mater was exposed (Fig. 3C–F). One week postoperatively, the MMT score on the side of the myelotomy remained unchanged. The MMCS improved to grade 2 in the early postoperative period (Fig. 3G). However, the patient developed recurrent symptoms, and IH was detected again at the same level 30 months after the initial surgery (Fig. 4A and B). The MMCS prior to the second surgery had worsened to grade 4. A C3 corpectomy was performed, and the dura mater was opened, revealing a recurrent CM on the ventral surface of the spinal cord. Myelotomy via the anterolateral sulcus (ALS) was performed to completely remove the recurrent lesion (Fig. 4C–F). The patient initially experienced worsening motor weakness in the left upper extremity immediately after the second surgery. However, her neurological status gradually improved. At the most recent follow-up, her MMCS score was assessed as grade 3. No residual lesion was confirmed (Fig. 4G).
Case 2 (first surgery): A 43-year-old female presented with acute and progressively worsening symptoms. Preoperative magnetic resonance imaging (MRI) revealed intramedullary hemorrhage on the ventral surface of the spinal cord at the C2–3 level. (A and B) T1- and T2-weighted images (T2WIs), respectively. (C) In intraoperative findings, no visible abnormalities were observed on the spinal cord surface. (D) Myelotomy was initiated by dissecting the posterolateral sulcus. (E) The intramedullary cavernous malformation (CM) located on the ventral side (white asterisk) was gently dissected from the spinal cord. (F) The CM was removed, revealing the ventral pia mater (white arrow). (G) T2WI after the first surgery.
Case 2 (second surgery): (A and B) T1- and T2-weighted images (T2WIs) at the time of recurrence. (C) In intraoperative findings, after corpectomy and dural incision, recurrent cavernous malformation (white asterisks) was identified. Indocyanine green videoangiography (ICG-VA) confirmed no direct arterial supply from the anterior spinal artery (white arrow) to the lesion. (D) Myelotomy was initiated by dissecting the anterolateral sulcus. (E) No residual lesion remained following gross total resection. (F) Pial suture was performed; subsequent ICG-VA demonstrated good circulation of anterior spinal arteries (white arrow). (G) T2WI after the second surgery. No residual lesion was observed.
DISCUSSION
1. Summary of the Study
This study focused specifically on surgical cases of CMs in the high cervical spine—a region that continues to present significant challenges for neurosurgeons in achieving successful surgical outcomes. In this case series, the PLS approach was applied in all cases. A known complication associated with the PLS approach is the risk of motor dysfunction on the side of the myelotomy during the early postoperative period. In the present study, 2 of the 6 patients experienced worsening MMT scores in the upper limb on the side of the myelotomy 1 week after surgery. However, all patients eventually maintained or improved their motor function without any serious adverse events. The average score on the MMCS was 3.0 before surgery, maintained at 2.5 in the early postoperative period, and further improved to 2.2 at the most recent follow-up. One patient (patient No. 2), who had a ventral-type CM, experienced recurrent hemorrhage at the same level 30 months after the initial surgery. She subse-quently tolerated a second surgery well, which was completed using the ALS approach. No other patients experienced recurrent hemorrhage. A questionnaire administered during outpatient follow-up indicated that all patients returned to social activities. The IONM results were characteristic of the PLS approach. Amplitude changes observed during surgery were directly related to manipulation of motor pathways—such as the corticospinal tract—on the same side as the myelotomy. Nevertheless, no long-term neurological complications were observed.
2. Surgical Anatomy of PLS Approach and Anatomical Considerations of the High Cervical Spine
The PLS approach, also known as DREZ myelotomy, was originally developed as a surgical technique for selective destruction of the posterolateral aspect of the spinal cord—an area corresponding to the entry point of the dorsal nerve roots [20,21]. Moreover, the PLS has been recognized as one of the safe entry zones for IMSCTs, particularly those located laterally within the spinal cord, as previously reported [5,8,9]. One of the primary advantages of the PLS approach is a lower risk of injury to the posterior columns compared to the PMS approach [8,9,11]. Indeed, in this study, there were no patients who developed symptoms related to posterior columns. In contrast, a notable disadvantage is the increased risk of motor dysfunction on the side of the myelotomy. This risk is primarily attributable to anatomical characteristics of the spinal cord. The PLS defines the plane between the dorsal and lateral columns of the spinal cord and is composed of pia mater. This plane extends into the dorsolateral fasciculus (Lissauer tract) and the Rexed laminae I to V of the dorsal horn of the gray matter [22]. Anatomical studies of the PLS have revealed several structural variations, including clear, short, absent, or irregular types [23]. The average width and height of the dorsal horn—including Lissauer’s tract and laminae I to V—were reported as 0.46 mm and 3.0 mm, respectively, at the C1–4 spinal levels. Each lateral half of the spinal cord contains 3 white matter columns—dorsal, lateral, and ventral—surrounding the central gray matter. The descending fiber systems of the lateral corticospinal and rubrospinal tracts are located within the lateral column. This fiber system in the lateral column is not clearly delimited and may overlap within cross-sectional areas, whereas the ascending fiber system in the dorsal column is sharply defined by glial septa [24]. Due to this anatomical organization of the white matter, the descending fiber systems—particularly the lateral corticospinal and rubrospinal tracts—may be more susceptible to surgical stress during the PLS approach compared to the PMS approach, which accesses the spinal cord via the dorsal midline.
The surgical removal of CM in the high cervical spine may carry the anatomical and surgical risks specific to the spine level, such as: (1) proximity to the medulla oblongata, and (2) risk to the respiratory pathway to the spinal cord. The spinal cord blends indistinguishably into the medulla oblongata of brainstem. In the brainstem, there are cranial nerve nuclei, major descending fiber bundles, such as the corticospinal tract or corticonuclear tract, as well as major ascending fiber bundles, such as the medial lemniscus or spinothalamic tract. Some of the major tract fibers cross at the level of the medulla oblongata [25]. Therefore, the rostral extension of myelotomy at this high cervical spine is anatomically limited, unlike at the middle or lower cervical spine [5]. Another critical consideration of neuroanatomy is the respiratory center in the brainstem. The respiratory center is located in the medulla oblongata and pons. The respiratory center is divided into 3 major groups, 2 in the medulla and 1 in the pons. The 2 groups in the medulla are the dorsal respiratory group (DRG) and the ventral respiratory group (VRG), respectively [26]. The VRG maintains a constant breathing rhythm by stimulating the diaphragm and external intercostal muscles to contract, resulting in inspiration. The DRG has the most fundamental role in the control of respiration, initiating inspiration (inhalation). Axons from those respiratory groups can decussate in the medulla to bilaterally innervate the phrenic nuclei that are located bilaterally in the caudal C3 to rostral C6 spinal cord. It should be noted that the surgical removal of intramedullary CM in the high cervical spine may carry a higher risk of not only motor or sensory neurological dysfunction but also respiratory disorders compared to the middle or lower cervical spine.
3. Literature Review of Surgical Access to Intramedullary CM of High Cervical Spine
Few clinical studies have specifically focus on intramedullary CMs of high cervical spine; however, existing data suggested that surgical outcomes at this level may be poorer compared to other spinal regions. Liu et al. [27] conducted a multicenter prospective clinical study evaluating surgical outcomes in symptomatic spinal CM cases. The study included a total of 268 patients with symptomatic spinal CMs, of which 28 cases were located in the high cervical spine (C1–4). Among all spinal levels, high cervical lesions demonstrated the lowest rate of favorable outcomes—75% at the high cervical level, compared to 86% at the lower cervical, 79% at the thoracic, and 100% at the lumbar level. Although these differences were not statistically significant, the trend suggests that high cervical lesions may present greater surgical challenges. Cai et al. [28] reported outcomes from 29 consecutive cases of symptomatic spinal CMs, including lesions at the craniomedullary junction and cervical spine. In their cohort, 19 patients (65.5%) improved, 4 patients (13.8%) remained stable, and 6 patients (20.7%) experienced worsening neurological function. Asimakidou et al. [29] performed a systematic review of surgical outcomes in intramedullary spinal cord CMs. Their analysis included 21 studies involving 1,091 patients, of whom 1,005 (92.1%) underwent surgical resection and 86 (7.9%) were managed conservatively. The authors found no significant difference in long-term recovery between thoracic and cervical lesions; however, outcomes for the high cervical spine specifically were not analyzed separately. Ren et al. [10] conducted a single-center retrospective study involving 219 surgically treated patients with spinal intramedullary CMs. They concluded that DREZ myelotomy—equivalent to the PLS approach—is optimal for dorsolateral deep lesions, although it may be associated with moderate to severe local pain. While DREZ myelotomy involves dissection of the substantia gelatinosa and the posterolateral tract of Lissauer, it often leads to postoperative worsening of superficial sensation in the dermatomes innervated by the nerve root emerging from the entry zone. Wang et al. conducted the literature review with focusing on the surgical outcome of intramedullary ependymal cysts that were treated via either PMS or DREZ (PLS) fenestration [11]. While not focused on CM, their analysis provided valuable insight into the anatomical and functional implications of surgical selection for intramedullary lesions. DREZ approach, equivalent to PLS approach, may reduce posterior column injury and transient postoperative deficits compared to PMS myelotomy. PLS approach presented here, although not novel, remains a valuable corridor for dorsolateral intramedullary lesions of the spinal cord. We emphasize that no single approach is universally superior, and that ventral or central lesions may require a PMS or ALS approach depending on their orientation.
4. The Significance and Limitation of IONM
In this study, except for 1 case of emergency surgery without intraoperative monitoring, positive alarm of MEP following the PLS approach was noted in the remaining 5 cases. Only 2 of 5 patients exhibited motor weakness on the side of myelotomy after surgery and were regarded as true positive. A previous study showed that MEP had higher sensitivity and specificity for detecting neurological injury during IMSCT resection than SEP (90% vs. 85% and 82% vs. 72%, respectively) [30]. Another study reported that MEP had the highest sensitivity and specificity in various neuromonitoring modalities [31]. Although 50% amplitude decrease of MEP is widely accepted as positive alarm, it may not always reflect permanent injury—particularly in the lateral columns accessed via the PLS approach. This is because the corticospinal tract in the lateral columns is located just adjacent to the myelotomy by PLS approach. The reduction of MEP amplitude may reflect that the myelotomy by PLS approach was correctly made within the lateral column. Furthermore, MEP responses do not necessarily accurately reflect motor dysfunction after surgery. Because motor function of the spinal cord is regulated by not only corticospinal tract but also other descending fiber systems such as rubrospinal tract or reticulospinal tract. In principle, MEP changes should not be ignored, but surgical procedures should not be interrupted or discontinued. We believe that the surgeon should carefully understand both the results of IONM and the findings of microscopic observation of the spinal cord parenchyma. It looks like of great importance for surgeons to understand the significance and limitation of IONM.
5. Study Limitations
This study has several limitations. First, the total number of patients with CMs was small. However, to our knowledge, no previous studies have specifically focused on the use of the PLS approach for CMs located in the high cervical spine. Second, the generalizability of the findings is limited. The surgical outcomes described may be difficult to replicate in other settings, as technical factors vary significantly. Third, all patients in this study underwent surgery using the PLS approach. As such, it remains unclear how outcomes may have differed with alternative approaches. Although the optimal approach for resecting symptomatic spinal CMs should be selected on a case-by-case basis, prospective comparative studies evaluating different myelotomy approaches are warranted.
CONCLUSION
Surgical removal of intramedullary CMs of the high cervical spine was successfully performed using the PLS approach. With the exception of one ventrally located lesion, no cases of recurrence were observed, and neurological function improved postoperatively in all patients. Although the criteria for IONM remains difficult to standardize, surgeons should be aware that transient motor deficits are a potential risk associated with the PLS approach. Nonetheless, this technique offers a viable surgical route for resecting intramedullary CMs in the high cervical spine. However, ventral-type lesions continue to pose significant surgical challenges. Surgeons should select an optimal surgical approach based on the location of CMs.
Notes
Conflict of Interest
The authors have nothing to disclose.
Funding/Support
This study received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Acknowledgments
The authors would like to thank Aishi Okazaki for his technical support of intraoperative monitoring. The authors would also like to thank Editage for English language editing.
Author Contribution
Conceptualization: TT; Data curation: YF; Formal analysis: YF; Methodology: YF, HK, MF, RY, RH, MK, NN, MF, SK, TT, MW; Project administration: TT, MW; Visualization: YF, TT; Writing – original draft: YF; Writing – review & editing: TT.