Long-segment intraspinal schwannomas resection: What is the minimum number of laminectomy levels required? Authors: Zhiyu Xi, Li Jia, Yingfeng Wang, Qiyu Jia, Chengyu Xia, Jiang Liu Published: 2025 ABSTRACT AbstractBackground and objectivesIn canonical knowledge, long-segment intraspinal schwannomas (SCHs) require laminectomies at all involved levels for complete exposure. To evaluate a minimally invasive, limited-exposure strategy for gross total resection (GTR) of multi-level intraspinal SCHs.MethodsA retrospective analysis was conducted on 11 patients with intraspinal SCHs involving ≥ 3 vertebral segments who underwent laminectomy confined to critical vertebrae, achieving gross total resection without full exposure of all involved levels. The surgical approach involved focused laminectomy targeting critical vertebrae, intracapsular decompression, and meticulous microsurgical dissection. Follow-up included clinical assessments and MRI evaluations at 3 months and 1-year post-surgery.ResultsThe cohort comprised 4 females and 7 males, with a median age of 50 years. Lesions were predominantly located in the cervical spine. Ten cases exhibited cystic lesions, while one presented a fully solid tumor. The average surgical time was 119 ± 15.6 minutes. Laminectomy was performed within 2 levels (6 cases) or 1 level (5 cases). All patients achieved complete symptom resolution at discharge, with no perioperative complications. During a mean follow-up of 23.5 ± 13.0 months, no cases of postoperative spinal instability or tumor recurrence were identified.ConclusionFor long-segment intraspinal SCHs, particularly cystic lesions, minimizing the number of laminectomy levels to achieve GTR is technically feasible and clinically safe.Supplementary InformationThe online version contains supplementary material available at 10.1007/s00423-025-03932-0.Keywords: Schwannomas, Long-segment, Intraspinal, Minimally invasive, Laminectomy, Gross total resection Introduction Spinal schwannomas (SCHs), originating from Schwann cells of the myelinated nerve sheath, are among the most common intradural extramedullary spinal tumors [1]. These slow-growing benign tumors have an annual incidence of 0.3 to 0.5 per 100,000 and typically appear during the fourth and fifth decades of life [2]. SCHs occur in the sheath around the nerve root and may involve any vertebral segment. Tumors can extend to the paravertebral space through the bone foramen, forming a dumbbell shape. Due to nerve root or spinal cord compression, sensory or motor dysfunction is the most common symptom in these patients. The optimal treatment is gross total resection (GTR) of the tumor with preservation of the nerve. A minimally invasive resection is pursued to minimize bony defects, reduce tissue trauma, and decrease the incidence of spinal instability. Most spinal SCHs grow with limited volume in the spinal canal, likely due to early symptom presentation. However, lesions involving more than one vertebral segment are not uncommon. Alvarez-Crespo DJ et al. [3] reviewed 2,542 cases in the literature, finding that multiple-segment involvement accounted for 13.8%. In general, long-segment intraspinal SCHs require laminectomies at all involved levels for complete exposure. However, we often find that tumors can be easily removed after dissection with the nerve root due to their soft texture and clear boundaries without adhesions. This frequently leads us to reconsider whether full exposure of all involved levels is necessary. Even when hemilaminectomy or other methods are performed to decrease damage, multi-segmental fenestration significantly increases the risk of postoperative complications and spinal instability, particularly when involving more than three levels—a scenario that does not align with the concept of minimally invasive surgery. Accordingly, we implemented a limited-exposure strategy for gross total resection (GTR) in multi-level spinal SCHs, presenting technical insights from 11 cases. Patients and methods Patients This retrospective analysis was approved by the ethics committee of the First Affiliated Hospital of the University of Science and Technology of China. Eleven cases of spinal SCHs involving multiple segments were enrolled from February 2021 to July 2024. Inclusion criteria were: (1) intraspinal lesions spanning ≥ 3 vertebral segments; (2) laminectomy confined to critical vertebrae, attaining gross total resection without full exposure of all involved levels; and (3) histopathologically confirmed schwannoma diagnosis. Data were retrospectively extracted from electronic medical records. The institutional review board approved this study under a waiver of informed consent due to its observational design involving anonymized data. Surgical procedure The tumor-bearing nerve root is localized through multimodal integration of neurological symptoms and preoperative MRI findings. Combined somatosensory evoked potentials (SSEP) and motor evoked potentials (MEP) monitoring is routinely employed to verify the functional integrity of the tumor-bearing root and differentiate the tumor-nerve interface using triggered electromyography. A focused laminectomy plan is formulated to target critical laminae demonstrating tumor origin while preserving adjacent anatomical levels. For paramedian intraspinal lesions, a limited-width laminectomy spanning the midline is performed. Spinous process resection is routinely implemented to optimize tumor delivery. Following staged intracapsular decompression, the tumor-bearing nerve root and adhesive regions are meticulously identified. Microscopic dissection is conducted with controlled manipulation. Adequate exposure of adhesive areas is maintained despite the small laminotomy, and additional levels of laminotomy are performed if visualization is insufficient. A representative surgical video illustrating the procedure is included in the Supplementary Materials. Follow-up All patients were received the follow-up routine, consisted of a clinical assessment 3–6 months after surgery at the outpatient department, as well as MRI 3 months and 1 year after surgery. Results Clinical baseline data The baseline characteristics of the 11 patients are listed in Table 1 and summarized in Table 2. The cohort comprised 4 females and 7 males, with a median age of 50 years (range: 20–72 years). Lesions were predominantly located in the cervical spine (5 cases), followed by the thoracic spine (2 cases) and lumbar spine (4 cases). The most common clinical symptom was pain, which was present in 10 cases, accompanied by sensory disturbances at corresponding innervated levels in 7 cases. Table 1Clinical data details of 11 casesNo.AgeGenderSymptoms and durationLesion locationCystic (Y/N)Quantities of opened levelsOperation time (min)Follow-up time (month)151MNeck pain and right arm numbness for 3 yearsC2-C7Y212512245FWaist and left leg pain for 2 monthsT12-L2Y110936333MNeck pain and left arm numbness for 1 yearC3-C7Y21386452FBack pain for 2 yearsT7-L1Y211048520MWaist pain and hip numbness for 6 monthsT12-L3Y111812671MWaist pain and hip numbness for 10 monthsT11-L2Y110824750FTwo legs numbness with pain for 1 yearL1-L3N214930827MNeck pain and left arm numbness for 8 monthsC4-T1Y211918956MWaist and left leg pain for 1 yearL1-L3Y197121072FNeck and right arm numbness for 6 monthsC3-C7Y2128241139MNeck and left arm pain for 2 yearsC4-C7Y110336 Clinical data details of 11 cases Table 2Demographic data and surgical outcomes for 11 patientsItemValueAge (year)Median 50 (range 20–72)Sex Female4 (36.4%) Male7 (63.6%)Preoperative symptoms N (%) Spinal pain10 (90.9%) Sensory disturbance of the lower extremities5 (45.5%) Pain of the lower extremities3 (27.3%)Duration of symptoms (month)Median 12 (range 2–36)Tumor location N (%) Cervical5 (45.5%) Thoracic2 (18.2%) Lumbar4 (36.4%)Cystic lesion N (%) Y10 (90.9%) N1 (9.1%)Tumor involved vertebrae number N (%)Median 4 (range 3–7) 33 (27.3%) 44 (36.4%) 52 (18.2%) 61 (9.1%) 71 (9.1%)Opened laminae number N (%) 15 (45.9%) 26 (54.5%) Operative time (min)average 119 ± 15.6Tumor resection Gross total resection11(100%) Subtotal resection0Postoperative sensory deficit0Postoperative motor deficit0Following up (month)average 23.5 ± 13.0Tumor recurrence0 Demographic data and surgical outcomes for 11 patients Among the 11 cases, 10 exhibited cystic lesions, which demonstrated mixed signals on T2-weighted imaging (T2WI), while only 1 case presented a fully solid tumor texture. The longest tumor spanned 7 vertebral segments. The average surgical time was 119 ± 15.6 min. Laminectomy was performed, within 2 levels (6 cases) or 1 level (5 cases). Notably, in lesions involving ≤ 4 vertebral segments, a single level lamina was typically opened, whereas lesions involving > 4 segments required the opening of 2 levels (with one exception, details mentioned below). The ratio of opened levels to involved levels ranged from 1/4 (smallest) to 2/5 (largest), excluding the exceptional case with a ratio of 2/3. All patients achieved complete symptom resolution at discharge, with no perioperative complications observed. During the mean follow-up period of 23.5 ± 13.0 months (range: 6–36 months), no cases of postoperative spinal instability or tumor recurrence were identified on serial MRI examinations. Case examples Case 1 A 51-year-old male presented with a 3-year history of cervicalgia and persistent right upper extremity paresthesia for over 2 years. Neurological examination revealed hypoesthesia to light touch in the right C5-C7 dermatomes. Preoperative MRI demonstrated an intradural extramedullary mass extending from C2 to C7, causing significant spinal cord compression (Fig. 1, A, B and C). The lesion exhibited hypointensity on T1-weighted imaging (T1WI) and mainly hyperintensity on T2WI, with a net-like septum and heterogeneous contrast enhancement. Axial MRI sequences revealed anatomical continuity between the tumor and the C5 nerve root (Fig. 1 and D). Surgical intervention was initiated with a C5 laminectomy, confirming the tumor origin at the C5 nerve root (Fig. 1 and E). Following intracapsular decompression, the rostral tumor component was readily delivered, while the upper portion demonstrated dense adhesions. A subsequent C4 laminectomy enabled complete en bloc resection (Fig. 1 and F). Case 1 images. (A-C) Preoperative sagittal MRI demonstrating an intradural extramedullary mass spanning C2-C7 with spinal cord compression (D) Axial contrast-enhanced T1-weighted MRI at the C4 level, with a white arrow denoting tumor encapsulation of the C5 nerve root (E) Intraoperative photograph showing the tumor-bearing C5 nerve root (black arrow) prior to dissection (F) En bloc tumor specimen measuring 8.5 cm in length (G) Postoperative CT reconstruction confirming C4-C5 laminectomy defects (white arrows) (H) Postoperative MRI demonstrating complete resection with preserved spinal alignment Postoperative MRI at 3 months confirmed gross total resection without residual tumor (Fig. 1 and H). The patient reported complete resolution of cervicalgia and paresthesia. One-year follow-up demonstrated maintained neurological improvement without evidence of recurrence or spinal instability. Case 2 A 27-year-old male presented with a 1-year history of intermittent neck pain and progressive left arm weakness (grade 4/5) evolving over 8 months. Preoperative contrast-enhanced cervical MRI revealed an intradural extramedullary heterogeneously enhancing mass spanning C4-T1 (Fig. 2, A and B). The initial surgical approach focused on C7 laminectomy; however, intraoperative exploration failed to localize the tumor-bearing nerve root. Subsequent C5 laminectomy revealed the lesion originating from the C6 nerve root, confirmed by triggered electromyography. Meticulous microsurgical dissection was performed along the tumor-spinal cord interface under intraoperative neuromonitoring, achieving en bloc resection through a two-level laminoplasty (C5/C7). Postoperatively, the patient regained full motor strength (grade 5/5) within one week. Surveillance MRI at 18 months demonstrated no evidence of recurrence while maintaining physiological cervical alignment (Fig. 2 and D). Case 2 images. (A-B) Preoperative MRI showing heterogeneous enhancement of a C4-T1 intradural mass. (C) Postoperative 3D-CT showing C5 and C7 laminectomy corridors (white arrows). (D) 18-month axial T1-weighted MRI confirming a tumor-free spinal canal Case 3 A 56-year-old male reported 1 year of progressive lumbar pain radiating to the left buttock and posterior thigh. Neurological examination revealed preserved lower extremity motor function (grade 5/5), intact sensation to light touch and pinprick, and normal deep tendon reflexes. Lumbosacral MRI demonstrated a “string-of-pearls” intradural lesion extending from L1-L3, characterized by T1-hyperintense, T2-hypointense nodules with homogeneous contrast enhancement (Fig. 3, A and C). The surgical approach included midline L1 and L3 osteoplastic laminotomy, identification of tumor nodes connected via nerve root fascicles, circumferential dissection preserving adjacent rootlets, and en bloc removal through a “skip” corridor (Fig. 3, D and E). The one-year follow-up demonstrated no evidence of recurrence or spinal instability. Case 3 images. (A-C) Preoperative lumbosacral MRI revealing a “string-of-pearls” intradural lesion (L1-L3) with inter-nodal connectivity (D) En bloc resected multinodular tumor complex (E) Postoperative CT showing L1 and L3 laminectomy defects (white arrows) with intact intermediate segment Discussion Spinal schwannomas (SCHs) arise from the nerve root sheath and exhibit diverse morphological configurations due to anatomical constraints imposed by the rigid vertebral column. Some authors [4, 5] have categorized spinal SCHs into five distinct types based on spatial distribution patterns: Type I, an intraspinal tumor less than two vertebral segments in length; Type II, an intraspinal tumor greater than two vertebral segments in length (giant tumor); Type III, an intraspinal tumor with extension into the nerve root foramen; Type IV, an intraspinal tumor with extraspinal extension (dumbbell tumors); and Type V, a tumor with erosion into the vertebral body (giant invasive tumor). In our study, all cases belonged to Type II. Theoretically, long-segment SCHs with extraspinal extension demonstrate a lower incidence, likely due to early clinical detection of extraspinal components through radicular symptom manifestation. Both our findings and existing literature suggest a predilection of Type II tumors for the cervical region [6, 7]. This anatomical preference may be attributed to the relatively larger cerebrospinal fluid (CSF) space and more elastic ligamentum flavum in the cervical canal, providing greater compensatory capacity for longitudinal tumor growth. For better visualization and easier manipulation, more levels of lamina often need to be exposed. However, the complications associated with multilevel exposure present a significant challenge for recovery. Surgical management of these extensive lesions typically requires multi-level laminectomies to achieve adequate visualization. However, such extensive exposures increase the risks of postoperative instability and delayed recovery. An interesting histological feature is that the tumor cells exhibit two distinct patterns, named Antoni A and Antoni B [8, 9]. The Antoni A pattern is typically associated with solid tumor portions and consists of denser cells, while the Antoni B pattern is related to cystic portions and contains looser cells. For cystic portions with an intact capsule and loose adhesions, we found that they could commonly be easily and smoothly removed from the canal. Extensive opening of the lamina to achieve full exposure may not always be necessary. We propose that it is practicable to achieve gross total resection safely without opening all levels. Only one previous study [10] reported a case of a giant cervical SCH resected with a single vertebra opening, where the authors initially planned for a multilevel laminectomy but unexpectedly found that the tumor could be removed through the first level opened. Their case also demonstrated that it is not mandatory to perform removal of multiple levels of spinal lamina. Our study is the first to summarize this surgical approach applied in multiple cases. In our series, we performed a one-level laminectomy for lesions spanning four segments and a two-level laminectomy for lesions spanning five or more segments, successfully achieving gross total resection. The opening length generally occupied approximately 30% of the lesion’s length. The key to minimizing the number of levels opened is identifying the tumor-bearing nerve during the surgical planning stage. In the case mentioned above, intraoperative findings revealed the tumor arising from the C6 nerve root, which was initially misidentified as the C7 root through conventional imaging interpretation. This erroneous localization prompted a two-level laminectomy (C5 and C7), whereas an isolated C5 laminectomy might have sufficed. This case clearly underscores the critical importance of accurate preoperative identification of the tumor-bearing nerve root. In addition to identifying the specific intervertebral foramen related to the tumor on preoperative MRI, another important clue is the initial symptoms, which reflect the level of the lesion. If the exact tumor-bearing nerve is not definitively identified, attempting to open one spinal segment for verification is an option. After performing the laminectomy on one vertebra, the spinal canal may be filled with the tumor, and the nerve root might not be clearly visible. By carefully separating the tumor from surrounding structures and using intraoperative electrophysiological monitoring, the tumor-bearing nerve root can be accurately identified. Following intracapsular decompression, more space is created, facilitating the identification of surrounding anatomical structures. Typically, the remaining portion of the tumor tends to deliver out with the CSF pulsation after intracapsular decompression and adhesion dissection, making it easier to achieve total resection. If the tumor-bearing nerve is still not exposed or the manipulation window is insufficient after intracapsular decompression, an expanded opening will be necessary. If this segment exposure does not reveal the tumor-bearing nerve, it is commonly located at the upper level based on our experience. For instance, if a laminectomy on C5 does not reveal the nerve, opening C4 will likely expose it with high probability. Our cases predominantly involved cystic tumors, with one exception. The cystic components are relatively soft and less likely to cause fatal compression of the spinal cord prematurely, which may also explain why they can grow longer within the spinal canal. Cystic tumors also have looser adhesions to their surroundings, and MRI scans show that the upper and lower extremities of the lesions are more rounded and blunter. Therefore, cystic tumors are more suitable for our method. However, one exception in our cases was a completely solid lesion that presented as multiple small, spherical lesions connected to each other with a “string”. This is known as bead-like schwannomas, which have been reported in several literature sources [11–13]. We opened the first and third vertebral bones, performing a “skip” fenestration to expose the cranial and caudal boundaries, and separated the tumor from its surrounding adhesions to achieve complete resection. This resection concept is the same as for other cystic cases, so we included it in this study. Conclusions In summary, minimally invasive principles should be prioritized in the surgical management of spinal SCHs, given their benign biological nature. For long-segment intraspinal SCHs, particularly cystic lesions, minimizing the number of laminae opened to achieve gross total resection is technically achievable and clinically safe. Critical to this approach is the preoperative identification of the tumor-bearing nerve root, which can be optimized through multimodal neuroimaging correlation and physical signs. When combined with meticulous microsurgical techniques emphasizing gentle tissue manipulation, the feasibility of complete resection through limited exposures becomes demonstrably achievable, challenging traditional paradigms that favor multiple laminectomies. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Author contributions Zhiyu Xi: Conceptualization, Formal analysis, Data Curation, Writing - Original Draft; Li Jia: Investigation, Methodology, Data Curation; Yingfeng Wang: Data Curation; Qiyu Jia: Data Curation; Chengyu Xia: Conceptualization, Methodology; Jiang Liu: Conceptualization, Project administration, Supervision, Writing - Review & Editing. Funding None. Data availability No datasets were generated or analysed during the current study. Declarations Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Zhiyu Xi and Li Jia contributed equally to this work. Contributor Information Chengyu Xia, Email: xichyu@yahoo.com.cn. Jiang Liu, Email: jiangliuahmu@163.com. References Associated Data Supplementary Materials Supplementary Material 1 Data Availability Statement No datasets were generated or analysed during the current study.