Unmasking benign intradural-extramedullary spinal tumors: a systematic review on treatment outcomes and radiological features

Abstract

Introduction:

Benign intradural extramedullary (IDEM) spinal tumors constitute about 65% of primary spinal neoplasms. Certain primary malignant or metastatic intradural lesions, although infrequent, may resemble their primary counterparts. Moreover, despite significant advances in the operative techniques, morbidity following treatment of IDEM tumors remains a considerable problem. Therefore, the current review aimed to summarize the most important data about radiological features of IDEM tumors on MRI and discuss optimal treatment of these lesions.

Materials and methods:

A systematic search was conducted based on MEDLINE, Scopus, Web of Science, Clinical Trial.gov, CENTRAL, EBSCO, and Embase according to the PRISMA statement on 8 July 2024. Extracted data were synthesized only qualitatively since the obtained data were not suitable for conducting the meta-analysis.

Results:

A total of 1003 records were identified in the database search. After considering eligibility criteria, 33 articles were further analyzed, of which 5 discussed diagnostic aspects and 28 evaluated treatment outcomes.

Conclusion:

The multidisciplinary approach, combining advanced diagnostic imaging, surgical innovation, radiation therapy, and emerging technologies, provides a comprehensive framework for the effective management of intradural extramedullary tumors. Continued research and clinical trials are essential to refine these strategies, enhance diagnostic precision, and optimize treatment outcomes for patients with spinal tumors.

Introduction

Spinal tumors constitute around 15% of central nervous system (CNS) tumors^[1]. These tumors can be categorized according to malignancy (benign or malignant), origin (primary or metastatic), anatomical localization with regard to the dura mater (intradural or extradural), and spinal cord (intramedullary or extramedullary). Primary spinal neoplasms exhibit a relatively low incidence rate, accounting for approximately 2–4% (CNS), with an estimated 65% manifesting as intradural-extramedullary (IDEM) lesions^[2]. Most frequent IDEM tumors include meningiomas, schwannomas, and neurofibromas^[3]. Among the less commonly encountered nonmalignant IDEM tumors are ependymomas, hemangiomas, lipomas, paragangliomas, vascular neoplasms, myxomas of nerve sheaths, and other exceptionally rare entities, including the seldom-described glomus tumors of the spinal canal^[4].

HIGHLIGHTS

• Despite significant advances in the operative techniques, morbidity following treatment of IDEM tumors remains a considerable problem. Therefore, the current review aimed to summarize the most important data about radiological features of IDEM tumors on MRI and discuss optimal treatment of these lesions.

• The multidisciplinary approach, combining advanced diagnostic imaging, surgical innovation, radiation therapy, and emerging technologies, provides a comprehensive framework for the effective management of intradural extramedullary tumors.

• Continued studies are essential to refine these strategies, enhance diagnostic precision, and optimize treatment outcomes for patients.

IDEM tumors may cause significant spinal cord compression, leading to pain, loss of bladder control, and neurological deficits in sensory and motor functions^[5]. Although the clinical image of IDEM tumors may not invariably demonstrate pathognomonic traits, and their presentation most commonly is related to the affected spinal cord level, the utilization of magnetic resonance imaging (MRI) frequently facilitates the recognition of essential characteristic features. Given the anatomical proximity to spinal nerves or the dura mater, schwannomas and meningiomas emerge as the predominant primary tumors of the spinal canal^[6]. Certain malignant primary^[7] or^[8] intradural lesions, although infrequent, may resemble their primary counterparts. However, their initial management differs as it typically requires adjunct oncological treatment. Therefore, enhancing the diagnostic efficacy of MRI and improving knowledge about radiological features suggesting malignant lesions at the preoperative stage would be valuable for avoiding diagnostic mistakes, which aggravate treatment outcomes in consequence.

At present, total microsurgical resection of the tumor and decompression of the spinal canal constitute the gold standard for the management of IDEM tumors^[9]. However, despite remarkable improvements in the operative techniques and neuroimaging of IDEM tumors observed in recent decades, morbidity following treatment remains a considerable problem^[10]. As recent studies found, the use of intraoperative neuromonitoring, minimally invasive surgical techniques, and stereotactic radiosurgery (SRS) as an additional component of treatment may decrease postoperative neurological deficits and improve treatment outcomes^[11].

The current systematic review aimed to summarize the most important data about radiological features on MRI and the utility of MRI in the diagnosis of IDEM tumors. A critical aspect of this review is the potential for malignant spinal tumors to radiologically mimic benign ones, complicating diagnosis and treatment strategies. Furthermore, optimal treatment of these lesions was discussed. By synthesizing recent advancements and clinical trial data, we aim to provide a thorough understanding of the therapeutic landscape, emphasizing best practices and future directions in managing benign spinal tumors.

Materials and methods

Study design

The current systematic review complies with the standards of the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) Statement (Figure 1) and AMSTAR (Assessing the Methodological Quality of Systematic Reviews) Guidelines (supplemental Digital Content Files 1, available at: http://links.lww.com/MS9/A967 and supplemental Digital Content 2, available at: http://links.lww.com/MS9/A968, respectively)

Figure 1.

A flowchart demonstrating the process of studies selection according to the PRISMA statement.

Eligibility criteria

Inclusion criteria for the systematic review were as follows: (1) studies recruiting patients with benign, intradural-extramedullary spinal tumors; (2) studies focusing on radiological features or treatment outcomes of spinal tumors; (3) study design as the following: randomized controlled trial, single-arm study, case series, observational study; (4) article type as the following: conference proceeding and clinical protocol. Studies were excluded from the systematic review when fulfilled at least one of the following exclusion criteria: (1) sample size of <10 patients (2) articles written in a language other than English; (3) studies on pediatric populations; (4) studies on animal models; (5) preclinical studies; (6) article type as the following: review, systematic review, meta-analysis, case report, editorial, erratum, corrigendum.

Search strategy and article selection process

The study search was conducted on 8 July 2024 using MEDLINE, Scopus, Web of Science, Clinical Trial.gov, Cochrane Central Register of Controlled Trials (CENTRAL), EBSCO (CINAHL Ultimate), and Embase databases. During the search, the following terms were used in combination with Boolean operators (AND, OR): “benign,” “intradural,” “extramedullary,” “spinal tumors,” “intradural extramedullary,” “spinal cord neoplasm,” “spinal cord tumor.” The search strategies were modified to suit each specific database. Obtained records were included or excluded according to the eligibility criteria by one of the authors (M.S.). Any uncertainties regarding article inclusion were consulted with other authors. No automation tools were used during the article selection process. Searches were supplemented by checking reference lists.

Data extraction

For qualified studies focusing on diagnostic aspects of IDEM tumors, extracted data included: (1) study and year of publication; (2) country of origin; (3) study design; (4) sample size; (5) main findings. In the case of eligible articles assessing treatment modalities and their outcomes, data extracted from each paper included: (1) study and year of publication; (2) country of origin; (3) study design; (4) sample size; (5) median length of follow-up; (6) intervention; (7) primary outcome; (8) results. No automation tools were used during the data extraction process.

Risk of bias assessment

The Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) checklist [https://www.acpjournals.org/doi/10.7326/0003-4819-155-8-201110180-00009] was used to assess the risk of bias of diagnostic studies. According to QUADAS-2, studies were classified as having a high, low, or unclear risk of bias across four key domains: patient selection, index test, reference standard, and flow and timing. A study was rated as low risk if all signaling questions were answered with “yes”; otherwise, it indicated potential bias. If there was not enough information to make a judgment, the risk was labeled as unclear.

For interventional studies, risk of bias was evaluated using the Methodological Index for Nonrandomized Studies (MINORS) developed by Slim et al^[12] [https://pubmed.ncbi.nlm.nih.gov/12956787/]. This assessment tool consists of 12 items for comparative studies, with the first 8 items also applicable to noncomparative studies. Each item is rated on a scale of 0 (not reported), 1 (reported but inadequate), or 2 (reported and adequate), leading to a maximum possible score of 16 for noncomparative studies and 24 for comparative studies.

Data synthesis

The data extracted from each eligible study were qualitatively synthesized in tables using Microsoft Excel software and the main text of the article. Due to high variability between included studies in terms of study design, objectives, and reported outcomes, a meta-analysis was not conducted.

Funding of included studies

Information regarding funding sources for the studies included in this review was extracted whenever available. However, the majority of the analyzed articles did not clearly report their sources of financial support or potential sponsorship. As such, the influence of funding on study outcomes could not be comprehensively assessed, representing a limitation of the current review.

Results

Study selection

A total of 1003 records were identified in the database search. Following the removal of duplicates, titles, and abstracts of 407 records were screened, from which 72 were searched for full-text obtainment. A total of 66 full-text articles were retrieved. After the full-text screening, 29 records were qualified in the current systematic review and 37 records were excluded due to the following reasons: necessary data impossible to obtain from the article (11), ineligible aim of the study (8), ineligible article type (7), language other than English (5), too small sample size (5). The screening of reference lists of included articles revealed an additional four eligible records. Finally, 33 articles were included in the current systematic review, of which 5 discussed diagnostic aspects and 28 evaluated treatment outcomes.

Radiological diagnosis of benign IDEM tumors

The main characteristics of the included studies focusing on aspects of radiological diagnosis of benign IDEM tumors are presented in Table 1. A cumulative sample size reached 831 patients. Three of the included studies were conducted retrospectively. The design of one study was cross-sectional, and another study included a prospective. Two of the studies analyzed radiological features of IDEM tumors. Two studies evaluated the diagnostic accuracy of MRI, and one remaining study assessed a predictive tool of outcomes following IDEM tumor resection. Table 2 demonstrates typical MRI features of benign IDEM tumors according to their histopathological diagnosis.

Table 1.

Characteristics of included studies focusing on aspects of radiological diagnosis of benign intradural-extramedullary tumors

No.	Author	Year	Country	Study design	Sample size	Aim of study	Main findings	References
1	Liu	2022	China	Retrospective	585	To evaluate multimodel weighted fusion framework based on MRI sequences and the patient’s age for the diagnosis of benign and malignant spinal tumors	Accuracy of model in differentiating benign and malignant tumors was higher than the accuracy of three clinicians	10.1186/S13244-022-01227-2/FIGURES/7
2	Saeed	2022	Pakistan	Cross-sectional study	140	To assess the diagnostic accuracy of MRI in intradural spinal tumors	Comparision of the results of MRI with histopathology demonstrated PPV = 99%, NPV = 95.5% and diagnostic accuracy = 97.7%	https://jcimcr.org/articles/JCIMCR-v3-2088.html
4	Singh	2021	India	Prospective	44	To analyze the effect of tumor occupancy ratio on the surgical outcomes of patients with IDEM tumors	surgical outcome is better in patients with tumors with a lower occupancy ratio in the MRI (P < 0.001)	10.14444/8077
5	De Verdelhan	2005	France	Retrospective	52	To compare MRI features of spinal schwannomas and meningiomas	On T2W-MRI, signal intensity was higher and more heterogeneous for schwannomas; diagnosis of schwannoma should be made when IDEM tumor demonstrates hyperintensity on T2W images without the dural tail sign	10.1016/S0150-9861(05)83 021-4
6	Lee	2009	South Korea	Retrospective	10	To present characteristic CT and MRI findings of calcified spinal meningiomas	The signal intensity of the calcified tumor varied on all imaging sequences	10.1007/s00586-015-4075-y

MRI – magnetic resonance imaging, CT – computed tomography, PPV – positive predictive value, NPV – negative predictive value

Table 2.

Magnetic Resonance Imaging features of most common intradural-extramedullary tumours

Tumor histology	T1WI features	T2WI features	Special features	Malignant lesions with similar appearance	References
Schwannoma	75% isointense, 25% hypointense; with contrast: virtually all enhanced	>95% hyperintense,	Vascular changes, hemorrhage, fatty degeneration, and adjacent bone remodeling	MPNST	63, 64
Meningioma	Isointense to slightly hypointense; with contrast: moderate homogeneous enhancement	Isointense to slightly hyperintense	Dural tail sign	Primary spinal lymphomas, hemangiopericytomas	59, 60
Neurofibroma	Isointense	Hyperintense, typically homogenous gadolinium enhancement	Fusiform or spherical morphology	MPNST	65, 66

T1WI – T1 weighted image, T2WI – T2 weighted image, MPNST – malignant peripheral nerve sheath tumours

Treatment of benign IDEM tumors

The characteristics of studies reporting on treatment outcomes of benign IDEM tumors qualified in the current systematic review are shown in Table 3. The total number of patients analyzed by eligible studies was 3444, including 2810 subjects with IDEM tumors. Five studies assessed the treatment effectiveness of SRS with or without surgical resection. In the remaining 23 studies, the authors treated the patients by surgery only, and six of these studies used additional intraoperative neuromonitoring. Mean follow-up in qualified studies ranged between 0.08 and 7.7 years. Twenty-three studies were conducted retrospectively, whereas five studies were prospective. Most of the included studies were performed in the United States (6), followed by India (5), South Korea (3), and Italy (3).

Table 3.

Characteristics of included studies assessing treatment outcomes of benign intradural-extramedullary tumors

No.	Author	Year	Country	Study design	Sample size (IDEM tumors)	Follow-up, years	Intervention	Primary outcome	Histopathological types of IDEM (No of cases)	Main results	Reference
1	Randhawa	2024	India	Retrospective	130 (130)	0.5	Surgery + IONM	Improvement in functional status measured by Frankel scale	Schwannoma (56), meningioma (37), neurofibroma (16), epidermoid cyst (6), ependymoma (5), dermoid cyst (5), arachnoid cysts (3), metastasis (2), paraganglioma (1)	49.3% of patients improved of ≥2 points in Frankel score; significant decrease in DPS postoperatively; 7 cases of recurrence	10.25259/SNI_689_2023
2	Singh (1)	2024	Sweden	Retrospective	180 (146)	4.4	surgery	neurological improvement in mMC	schwannoma	significant improvement in mMC scores, pain, gait, motor, sensory, and bladder functions; 12 cases of recurrence (9 following STR)	10.3390/cancers16030519
3	Singh (2)	2024	Sweden	Retrospective cohort	94 (n/a)	7.3	Surgery	HRQoL	Schwannoma	HRQoL equal between patients with spinal schwannoma and the general population; 94% of patients returned to work	10.3390/cancers16101882
4	Prabhuraj	2023	India	retrospective	457 (320)	1	surgery ± IONM	recurrence rate and predicting factors	schwannoma (305), neurofibroma (15)	12 cases of recurrence; cervicothoracic location was a significant risk factor for recurrence	10.1016/j.jocn.2023.06.019
5	Yu	2023	China	Retrospective	87 (87)	5.2	Surgery (minimally invasive keyhole approach) + IONM	Neurological improvement in mMC	Schwannoma (49), meningioma (38)	Stable or improved mMC in all patients; 93% of patients reported excellent or good outcome according to Odom’s criteria; no cases of recurrence	10.1186/s13018-023-04074-z
6	Harsoori	2022	India	Prospective	44 (44)	N/R	Surgery	Functional improvement in Frankel scale and mMC	Schwannoma, meningioma, ependymoma, neurofibroma, lipoma, angiolipoma, ganglioma, astrocytoma, extradural metastasis, hematopoietic tumors	59% of patients demonstrated partial recovery, 25% of patients showed significant recovery, and in the remaining patients, functional status did not improve	https://www.worldwidejournals.com/international-journal-of-scientific-research-(IJSR)/fileview/clinical-evaluation-of-intradural-extramedullary-spinal-tumors-in-a-tertiary-care-hospital-india_January_2023_1791417162_5620691.pdf
7	Arab	2022	Egypt	Retrospective	23 (23)	1.9	Surgery ± IONM	Functional improvement in Frankel scale; improvement of pain in VAS	Meningioma (8), schwannoma (7), neurofibroma (5), ependymoma (2), arachnoid cyst (1)	95.7% of patients improved in the Frankel scale; pain decreased in 91.3% of patients	10.21608/ejhm.2022.262323
8	Wang	2022	China	Retrospective	40 (40)	8	Surgery (modified dura preservation technique)	Neurological improvment in mMC	Meningioma	87.5% of symptomatic cases improved in mMC; one case of recurrence	10.1016/j.wneu.2021.12.118
9	Ampie	2021	United States	Retrospective	46 (45)	4.4	Surgery + IONM ± adjuvant radiotherapy	Neurological improvement, length of hospital stay	Meningioma	Symptoms resolved or improved in 30 patients; one case of recurrence	10.3171/2021.2.FOCUS20977
10	Dobran	2021	Italy	Retrospective	40 (40)	0.08	Surgery (laminectomy versus hemilaminectomy)	Improvement of pain in VAS	Schwannoma	Significantly lower VAS in hemilaminectomy than laminectomy patients at last follow-up; significantly faster operative time, shorter postoperative bed stay, and shorter hospitalization in hemilaminectomy group	10.1055/s-0041-1 722 968
11	Naranyan	2021	India	Prospective	35 (35)	1.33	Surgery	Functional improvement measured by Nurick grade, Frankel scale, and DPS	Schwannoma (12), meningioma (9), neurofibroma (6), ependymoma (3), epidermoid cyst (2), neuroenteric cyst (1), paraganglioma (1), dermoid cyst (1)	77% of patients improved in the Frankel scale, mean DPS, and mean Nurick score demonstrated a significant decrease	10.7759/cureus.15733
12	Newman	2021	United States	Prospective cohort	57 (57)	1.28	Surgery	Quality of life improvement in BPI and MDASI	Schwannoma (35), meningioma (18), neurofibroma (2), paraganglioma (1), mixed (1)	Statistically significant improvement in BPI and MDASI scores	10.1093/neuros/nyaa561
13	Olguner	2021	Turkey	Retrospective	49 (34)	N/R	Surgery + sodium fluorescein	Neurological improvement in mMC	Schwannoma (19), meningioma (12), ependymoma (1), hemangiopericytoma (1), ganglioneuroma (1)	No singnificant difference in mMC between the preoperative and postoperative period	10.3389/fonc.2020.618579
14	Patel	2021	India	Retrospective	31 (31)	1	Surgery	Improvement in VAS, Frankel scale, and KPS	Schwannoma (16), meningioma (11), lipoma (2), hemangioma (1), ependymoma (1)	Statistically significant improvement of KPS, VAS, and Frankel scale; one case of recurrence	10.25259/SNI_839_2020
15	Nemeiko	2019	Denmark	Retrospective cohort	279 (100)	N/R	Surgery	Improvement of clinical symptoms	Schwannoma (49), meningioma (38), neurofibroma (6), paraganglioma (2), lipoma (2), epidermoid cyst (2), teratoma (1)	Motor deficit improved in 16/100 patients, sensory deficit in 12/100 patients, neurogenic bladder/bowel dysfuntion in 11/100 patients, ataxia in 12/100 patients and pain in 43/100 patients	https://ugeskriftet.dk/dmj/characteristics-and-outcomes-patients-primary-intraspinal-tumours
16	Bhimani	2018	United States	Retrospective	991 (991)	0.08	Surgery	Surgical complications	N/R	6.7% of patients required readmission, 4.5% returend to the operating room; 1.2% of patients died	10.1016/j.wneu.2018.06.014
17	Gilard	2018	France	Retrospective	87/87	7.7	Surgery	Neurological improvement in mMC and identification of predictive factors	Meningioma	17 patients deteriorated postoperatively – 75% of them partially recovered; 6 cases of recurrence	10.1007/s11060-018-2929-y
18	Harel	2017	Israel	Retrospective cohort	111 (111)	N/R	Surgery + IONM	Neurological improvement in mMC	Meningioma (76), nerve sheath tumor (35)	Statistically significant improvement in mMC; no statistical difference between IONM and control group regarding new neurological deficit; IONM predicts postoperative neurological deficits	10.1007/s10143-017-0815-2
19	Afathi	2015	France	Retrospective	18 (18)	2	Surgery (minimally invasive transmuscular approach)	Length of hospital stay, complication rate, tumor recurrence	Meningioma (11), neurinoma (6), myxopapillary ependymoma (1)	No complication in any patient; no tumor recurrence; mean hospital stay: 6 days	10.1016/j.neuchi.2015.05.001
20	Shin	2015	South Korea	Retrospective	58 (58)	0.91	SRS ± surgery	Local tumor control rate	Schwannoma (50), neurofibroma (4), MPNST (4)	Local tumor control rate of beign tumours was 95.4%, signficant improvement of pain symptoms	10.3171/2015.1.SPINE14910
21	Marchetti	2013	Italy	Retrospective	18 (18)	3.58	SRS ± surgery	Local tumor control rate, neurological improvement, pain relief in VAS	21 lesions: meningioma (11), schwannoma (9), neurofibroma (1)	None of the lesions showed progression; 29% of symptomatic patients demonstrated neurological improvement; 64% of patients have significant pain relief	10.1007/s00701-013-1756-3
22	Iacoangeli	2012	Italy	Retrospective cohort	65 (65)	2	Surgery – minimally invasive (hemilaminectomy and preservation of the outer dural layer) versus laminectomy	Surgical complications, mean hospital stay, long-term symptoms improvement	Meningioma	No significant diferences in extent of tumor removal, hospital stay and surgical complications; better postoperative course im minimally invasive group	10.2147/CIA.S38923
23	Sachdev	2011	United States	Retrospective	87 (87)	2.75	SRS ± surgery	Local tumor control, complications, and improvement of clinical symptoms	103 lesions: schwannoma (47), meningioma (32), neurofibroma (24)	Overall control rate of 95; 91% of meningiomas, 67% of neurofibromas and 86% of schwannomas were clinically stable; one case of myelopathy	10.1227/NEU.0b013e318218db23
24	Yeo	2011	South Korea	Retrospective	25 (20)	1.79	Surgery (hemilaminectomy)	Neurological improvement	Schwannoma (12), ependymoma (5), meningioma (3)	Neurological improvement in all cases except one patient with stable symptoms, no cases of recurrence	10.3340/jkns.2011.50.3.195
25	Song	2009	South Korea	Retrospective	12 (12)	2	Surgery	Functional improvement in Frankel scale, improvement of neurological deficit	Meningioma (4), schwannoma (4), epidermoid cyst (2), ependymoma (1)	8 patients improved in Frankel scale, neurological deficit imrpoved in all cases; 2 cases of CSF leak and 2 cases of paresthesia	10.4055/cios.2009.1.2.74
26	Gerszten	2008	United States	Prospective	73 (73)	3.08	SRS ± surgery	Local tumor control, pain relief in VAS	Schwannoma (35), neurofibroma (25), meningioma (13)	Long-term pain relief in 73% of cases; long term tumor control in all cases; 3 cases of radiation-induced myelopathy	10.1227/01.neu.0000318174.28461.fc
27	Dodd	2006	United States	Prospective	51 (51)	3	SRS ± surgery	Local tumor control, pain relief	55 lesions: schwannoma (30), meningioma (16), neurofibroma (9)	All lesions were stable or decreased size postradiosurgery; 25-50% of patients reported pain reduction; one case of radiation-induced myelopathy	10.1227/01.NEU.0000204128.84742.8 F
28	Chiou	1989	Germany	Retrospective	256 (87)	N/R	Surgery (laminectomy versus hemilaminectomy)	Improvement of neurological deficits, urinary incontinence, subjective complaints	Meningioma (41), neurinoma (27), others (19)	Subjective complaints improved in 59/87 patients, neurological deficit improved in 51/87 patients, urinary incontinience improved in 32/87 patients; hemilaminectomy is more advatageous for extramedullary tumours	10.1007/BF01403599

IDEM – intradural extramedullary, IONM – intraoperative neuromonitoring, DPS—Denis pain scale, mMC—modified McCormick scale, STR—subtotal resection, HRQoL—health-related quality of life, AIS—American Spinal Injury Association impairment scale, VAS—visual analog scale, MEP—motor evoked potentials, BPI—Brief Pain Inventory, MDASI—MD Anderson Symptom Inventory, KPS—Karnofsky Performance Score, SRS—stereotactic radiosurgery, MPNST—malignant peripheral nerve sheath tumors, n/a—not available/not reported; “ +” indicates additional procedure was performed in all IDEM patients, “ ±” in part of the study group

Risk of bias assessment

Table 4 provides a summary of the risk of bias assessment of diagnostic studies according to the QUADAS-2 tool. The majority of studies demonstrated a low risk of bias in the patient selection domain. All included studies showed a low risk of bias in the index test domain. However, one study raised concerns in the reference standard domain. In terms of flow and timing, two studies demonstrated low risk of bias.

Table 4.

Results of risk of bias assessment of included studies according to QUADAS-2 Tool

Study	Year	Patient selection	Index test	Reference standard	Flow and timing
Liu	2022	Low	Low	Unclear	Unclear
Saeed	2022	Unclear	Low	Low	Low
Singh	2021	Low	Low	High	Unclear
Lee	2009	Low	Low	Low	Low
De Verdelhan	2005	Low	Low	Unclear	Unclear

The results of risk of bias assessment of interventional studies according to the MINORS tool are presented in Table 5. Noncomparative studies received scores ranging from 2 to 11 out of a possible 16, while comparative studies scored between 10 and 12 out of a maximum of 24.

Table 5.

Results of risk of bias assessment of interventional studies according to the MINORS Tool. Noncomparative studies were presented in lines 1–24, comparative studies in lines 25–27

No.	Study	Year	Clearly stated aim	Inclusion of Consecutive Patients	Prospective Data Collection	Endpoints Appropriate to Study Aim	Unbiased Assessment of Study Endpoint	Follow-Up Period Appropriate to Study Aim	Less than 5% Lost to Follow-Up	Prospective Calculation of Study Size	Adequate control group	Contemporary group	Baseline equivalence of groups	Adequate statistical analyses	Total Score
1	Randhawa	2024	2	2	0	1	0	1	2	0	N/A	N/A	N/A	N/A	8
2	Singh (1)	2024	2	2	0	2	0	2	2	0	N/A	N/A	N/A	N/A	10
3	Singh (2)	2024	2	2	0	2	0	2	2	0	N/A	N/A	N/A	N/A	10
4	Prabhuraj	2023	1	1	0	1	0	1	0	0	N/A	N/A	N/A	N/A	4
5	Yu	2023	2	2	0	2	0	2	2	0	N/A	N/A	N/A	N/A	10
6	Harsoori	2022	1	1	1	2	0	0	0	0	N/A	N/A	N/A	N/A	5
7	Arab	2022	1	0	0	1	0	2	2	0	N/A	N/A	N/A	N/A	6
8	Wang	2022	2	2	0	2	0	2	2	0	N/A	N/A	N/A	N/A	10
9	Ampie	2021	1	2	0	1	0	2	2	0	N/A	N/A	N/A	N/A	8
10	Naranyan	2021	2	2	1	1	0	1	2	0	N/A	N/A	N/A	N/A	9
11	Newman	2021	2	2	1	2	1	1	1	0	N/A	N/A	N/A	N/A	10
12	Olguner	2021	1	2	1	2	0	2	2	0	N/A	N/A	N/A	N/A	10
13	Patel	2021	1	0	0	1	0	1	0	0	N/A	N/A	N/A	N/A	3
14	Nemeiko	2019	1	2	0	0	0	1	1	0	N/A	N/A	N/A	N/A	5
15	Bhimani	2018	2	2	1	2	0	0	0	0	N/A	N/A	N/A	N/A	7
16	Gilard	2018	2	1	0	2	0	1	0	0	N/A	N/A	N/A	N/A	6
17	Afathi	2015	1	0	0	0	0	1	0	0	N/A	N/A	N/A	N/A	2
18	Shin	2015	2	0	0	1	0	0	2	0	N/A	N/A	N/A	N/A	5
19	Marchetti	2013	1	1	0	1	1	1	2	0	N/A	N/A	N/A	N/A	7
20	Sachdev	2011	1	0	0	1	0	1	1	0	N/A	N/A	N/A	N/A	4
21	Yeo	2011	2	1	0	1	0	1	2	0	N/A	N/A	N/A	N/A	7
22	Song	2009	1	1	0	1	0	2	2	0	N/A	N/A	N/A	N/A	7
23	Gerszten	2008	2	2	1	1	1	2	2	0	N/A	N/A	N/A	N/A	11
24	Dodd	2006	2	2	2	1	0	2	2	0	N/A	N/A	N/A	N/A	11
25	Chiou	1989	1	1	0	0	0	0	2	0	N/A	N/A	N/A	N/A	4
26	Dobran	2021	1	0	0	1	1	0	2	0	2	2	1	2	12
27	Harel	2017	2	1	0	2	0	1	1	0	2	0	1	1	11
28	Iacoangeli	2012	1	1	0	1	0	2	2	0	2	0	1	0	10

Discussion

Radiological diagnosis of IDEM tumors

Radiological examination serves as the primary diagnostic approach for all intradural extramedullary tumors, with gadolinium-enhanced MRI being the imaging modality of choice. It also serves as a highly precise, nonintrusive, secure, and convenient imaging method as it proves invaluable for timely detection, strategizing treatment, and assisting in surgical biopsy procedures^[13].

Meningiomas mainly develop in the thoracic spine and are located lateral to the spinal cord, and 60–70% present the dural tail sign^[14]. Typically, present symptoms are associated with compression of the spinal cord or spinal nerve roots. MRI most commonly reveals T1: Typically isointense to slightly hypointense, potentially heterogeneous, T2: Generally isointense to slightly hyperintense, T1 with contrast: shows moderate homogeneous enhancement^[15]. Calcified meningiomas may appear hypointense on T1 and T2, with minimal contrast enhancement^[16].

Neurofibromas and schwannomas, though similar, have distinct features. Schwannomas frequently cause radicular pain and sensory issues and are often associated with adjacent bony remodeling^[17]. They often exhibit vascular changes, hemorrhage, and fatty degeneration, which are rare in neurofibromas^[18]. Signal characteristics include: T1: 75% isointense, 25% hypointense, T2: >95% hyperintense, often mixed, T1 C +: Virtually all enhance^[19]. Neurofibromas typically manifest as compact tumors, often displaying a fusiform or spherical morphology and typically enveloping rather than displacing nerve roots^[20]. On MRI, they appear isointense on T1-weighted images and markedly hyperintense on T2-weighted images, exhibiting pronounced gadolinium enhancement, typically homogeneously, although sometimes peripherally^[21]. Plexiform neurofibromas typically present as substantial masses with a curved structure, resulting in a distinctive “bag of worms” appearance^[13]. Melanotic schwannomas diverge and exhibit, in 64.7 and 70.6% of cases, respectively, hyperintensity on T1-weighted images and hypointensity on T2-weighted images^[22].

Primary melanomas of the spine exhibit T1 hyperintensity on MRI due to paramagnetic free radicals^[23]. Malignant peripheral nerve sheath tumors (MPNST) mimic schwannoma imaging but demonstrate a more aggressive demeanor. They exhibit characteristics such as size, peripheral enhancement, surrounding edema-like area, and internal cystic lesions. In cases linked to neurofibromatosis 1, heterogeneity on T1-weighted images is vital for distinguishing between neurofibroma and MPNST^[24].

Spinal solitary fibrous tumors (SFTs) are rare, accounting for about 10% of CNS SFTs. They typically appear as oval lesions on MRI, with iso- to hypointensity on T1-weighted images and hyperintensity on T2-weighted images^[25]. Malignant variants may have less-defined margins and heterogeneous enhancement on gadolinium. Large feeding arteries observed on angiography or as vascular voids on MR imaging can help narrow the diagnosis^[26]. Within the cauda equina region, paraganglioma hemangioblastoma, spinal PNET, and myxopapillary ependymoma originate from the filum terminale, each displaying distinctive imaging characteristics^[27].

Spinal metastases, predominantly vertebral and epidural, occur in about 20% of cancer patients, with intradural and extramedullary involvement being rare (5%)^[8]. Lung (48%) and breast (16%) cancers are the most common sources. Differentiating primary spinal cord tumors from metastases via imaging alone is challenging; however, clinical factors like known primary cancer, rapid neurological decline, and concurrent systemic lesions can assist in diagnosis^[28].

In a conventional MRI evaluation of a glomus tumor of the spine, discernible characteristics are evident. Specifically, T1-weighted images depict a well-defined hypointense nodule, whereas T2-weighted images illustrate hyperintensity^[29]. Following intravenous administration of gadolinium contrast, pronounced enhancement is observed on T1-weighted images. This contrast examination additionally facilitates the identification of a sheath enveloping the majority of glomus tumors, delineated by a low signal intensity margin^[30]. Of note is the resemblance observed in MRI characteristics between glomangioma and schwannoma. Nevertheless, a distinctive attribute is discernible with the presence of a signal-void fistular lesion within the tumor, a feature documented in MRI assessments of several glomus tumor cases but notably absent in schwannomas^[31]. However, the limited number of documented patients precludes definitive categorization of this observation as a consistent hallmark.

Sathyanathan et al^[32] investigated intradural spinal cord lesions using magnetic resonance spectroscopy (MRS) and correlated the results with histo-pathological examination (HPE). While routine MRI alone detected 58% of cases, combining it with MRS increased detection to 84%. Statistical analysis revealed a significant difference (P = 0.007) between conventional MRI and MRS. However, the study specifically concentrated on intramedullary lesions. The authors concluded that MRS of the spinal cord enhances research and diagnosis by providing supplementary information to existing imaging techniques. Additionally, it offers valuable biomarker data for distinguishing between benign and malignant lesions, potentially reducing unnecessary procedures.

Data regarding the utility of MRI spectroscopy in the diagnostic process of glomangiomas are lacking in the literature. Diagnosing glomangiomas in deep anatomical locations may present challenges and can be readily conflated with neurilemmomas or paragangliomas based on imaging observations^[33]. In such cases, it is recommended to proceed with a biopsy for diagnostic purposes. Clinical presentation of spinal glomus tumors is indistinguishable from analogic intradural and extradural spinal tumors and involves neurological deficits secondary to compression of the spinal cord or nerve roots, prompting diagnostic imaging investigations. In certain instances, it may present as an incidental, sporadic discovery^[34].

In the recent study, Liu et al^[35] proposed a multimodel weighted fusion framework based on MRI and age data for enhancing the effectiveness of benign and malignant spinal tumors diagnosis. The accuracy of the model was higher than the accuracy of one radiologist and two spine surgeons, regardless of the consideration of age information. Diagnostic accuracy was improved when age information was included in the model.

Diagnostic challenges: malignant spinal tumors mimicking benign lesions

Accurate diagnosis of spinal tumors is complicated by the potential for malignant lesions to mimic benign ones on imaging studies. MPNSTs, for instance, often resemble schwannomas or neurofibromas but exhibit more aggressive features such as larger size, peripheral enhancement, surrounding edema-like areas, and internal cystic lesions^[36]. Similarly, primary spinal lymphomas can appear as homogeneously enhancing masses like meningiomas but may show more rapid growth and lack the characteristic dural tail sign^[37]. Chordomas, typically located in the sacrococcygeal region, can mimic giant cell tumors or osteoblastomas with their well-defined, enhancing masses, but tend to be more destructive^{[38], [39]}. There are also reports of chordomas of other regions mimicking schwannomas^[40]. Metastatic tumors often resemble vertebral hemangiomas on MRI with varied intensity and significant postcontrast enhancement, but are typically multifocal with associated soft tissue masses^[41–43]. Ependymomas, whether benign myxopapillary variants or malignant forms, can show enhancement and cystic components, though malignant types exhibit more irregular borders, hemorrhage, and necrosis^[44,45]. Hemangiopericytomas can resemble meningiomas with well-circumscribed enhancing masses but tend to be more aggressive with extensive vascular involvement^[46]. Chondrosarcomas, mimicking enchondromas, often exhibit cortical destruction and soft tissue extension, and Ewing sarcomas can present with mixed lytic and sclerotic lesions, periosteal reaction, and significant soft tissue masses, resembling an aggressive benign tumors^[47,48]. Recognizing these malignant tumors requires careful assessment of imaging features such as lesion size, border characteristics, presence of necrosis, hemorrhage, and associated soft tissue components, along with clinical correlation and, often, histopathological confirmation. This complexity underscores the importance of comprehensive imaging evaluation and multidisciplinary collaboration in diagnosing and treating spinal tumors.

Treatment of IDEM tumors

Standard surgical approaches

Generally, the cornerstone of treatment for all intradural, extramedullary lesions remains radical surgical excision, which typically results in favorable long-term outcomes with minimal risk of recurrence^[49]. However, approximately 10% of patients experience a recurrence after surgery due to incomplete removal^[50]. In 2001, Saito et al^[51] proposed a method for achieving complete resection of anterior-based lumbar spinal meningioma by selectively excising the involved inner layer of the dura while preserving the outer layer for closure. The primary objective of meningioma surgery is to attain a Simpson I total resection, thereby minimizing the risk of recurrence. However, complete resections, particularly for anteriorly situated spinal lesions, are frequently avoided due to the potential for postoperative cerebrospinal fluid leakage^[51,52]. Some authors recommend prospective use of autologous fat grafting during intradural spinal tumor surgery, claiming it effectively prevents cerebrospinal fluid-related complications, which are reported in up to 18% of patients^[32]. In instances where total resection is unattainable, observation may be considered for small, asymptomatic tumors. While the most commonly employed surgical technique involves traditional microscopic tumor excision, this classical method entails making an extended midline incision (typically two levels above and below the tumor), dissecting the paraspinal muscles subperiosteally, and performing multilevel laminectomies or facetectomies. This approach damages a significant portion of the structures within the posterior column, leading to decreased spinal stability and the potential for spinal deformity^[53]. Due to that, minimally invasive endoscopic approaches are increasingly utilized in contemporary practice. According to some authors, panoramic illumination and visualization provided by endoscopy offer clear benefits compared to microscopy in the dissection of tumors from nearby neural and vascular structures. This necessitates even smaller bony and dural exposures. Thus, in conjunction with microscopy, endoscopy utilizes a multiangled endoscope to directly observe and excise the lesion through a small bone window that may not be visible or accessible under the microscope^[54].

Minimally invasive and endoscopic techniques

Burtcher et al^[55] were the first to employ an endoscope-assisted microneurosurgical technique for the removal of multilevel IDEM tumors in the lumbar region. Following this, Barami et al^[56] documented the use of an endoscope-assisted posterior approach in two cases for resecting ventral IDEM tumors. Later, Caballero-García et al^[57], in a retrospective cohort study, evaluated the efficacy of a minimally invasive full endoscopic approach for extra- and intradural spinal tumor removal, reporting excellent clinical outcome^[57]. Similar observations have been noted by Şentürk et al^[58] who used a translaminar endoscopic approach.

A novel endoscopic technique known as unilateral biportal endoscopy (UBE) in minimally invasive spinal surgery (MISS) permits the precise manipulation of instruments with an enhanced, magnified perspective. It has found extensive use in various spinal degenerative conditions and is now being more commonly used in tumor treatment in recent years. Compared to traditional surgeries and uniaxial endoscopic approaches, UBE offers greater flexibility, convenience, minimal tissue trauma, reduced blood loss, and decreased postoperative pain^[59,60]. Peng et al^[61] reported the first case when UBE was successfully utilized for intradural extramedullary tumor excision.

Role of intraoperative monitoring

Although intraoperative monitoring is commonly utilized during spine tumor surgeries, and no prospective trials have confirmed its effectiveness, Kujur et al^[62] analyzed surgical outcomes for schwannoma and meningioma cases without employing intraoperative nerve monitoring (IONM). Despite this, their findings underscore the potential benefit of IONM in enhancing safety during surgery by providing real-time assessment of spinal cord function, especially in meticulous operations, thereby contributing to safer procedures and improved gross resection outcomes. Nonetheless, the absence of validated procedures to address intraoperative alerts underscores a crucial gap in knowledge. Future research efforts should concentrate on establishing treatment methodologies, validating procedural protocols, and consolidating clinical recommendations^[63].

Robotic and hybrid surgical techniques

At present, robotic surgery is primarily utilized when there is suspicion of spinal instability and a requirement for precise screw placement^[64]. Novel literature also provides instances of hybrid surgeries in cases involving extensive spinal tumors that extend into other body cavities, utilizing a combination of endoscopic and robotic surgery. Iga et al^[65] employed a blend of minimally invasive techniques for a dumbbell-shaped tumor, utilizing spinal endoscopy in conjunction with robot-assisted intrathoracic surgery with the da Vinci robot.

Emerging technologies (AR/VR)

Recently, there has been a growing scientific interest in the utilization of emerging technologies such as augmented reality (AR), virtual reality (VR), and mixed reality (MR)^[66]. Carl et al^[67] employed AR assistance in microscopic operations on ten patients diagnosed with intradural spinal tumors. They utilized preoperative imaging data to segment the objects visualized by AR, and nonlinear image registration was employed to accommodate spine flexibility. They found this method to be a reliable tool for intradural spinal tumor surgery. The automatic registration of AR ensures exceptional accuracy and provides an intuitive representation of both the tumor’s extent and the surrounding structures. Similar conclusions were reached by Sommer et al^[68].

Radiation therapy (SRS/SRT)

Radiation therapy is a viable option for managing spinal cord neoplasms, especially when facing unresectable lesions or residual disease postsurgery, or in cases where surgery is contraindicated due to medical comorbidities, performance status, lesion location, or rapidly recurring/progressive tumors^[69]. Conventional radiation therapy for spinal cord neoplasms often yields modest outcomes due to limitations such as multiple tumors or proximity to organs-at-risk^[70]. Hypofractionated approaches, with larger fraction sizes, offer potential benefits in tumor control and treatment duration reduction, particularly for metastatic disease in the bony spine^[71]. However, it may impact hormonal balance in women, potentially affecting fertility and adverse pregnancy outcomes, while abdominopelvic radiation is linked to uterine dysfunction^[72,73]. Conventional radiotherapy can induce neurocognitive impairment in cancer survivors by damaging neural progenitor populations responsible for adult hippocampal neurogenesis^[74,75]. However, the application of stereotactic radiation therapy (SRT) for spinal cord tumors is less standardized, with challenges such as unique radiobiological responses and close proximity to the spinal cord^[76]. While experiences from treating intracranial tumors guide some approaches, the efficacy and safety of SRT for spinal cord neoplasms require further exploration^[77,78].

Several studies have explored the use of SRT or SRS for the treatment of intradural extramedullary (IDEM) tumors, with generally positive outcomes and few transient complications reported. For instance, Selch et al^[79] treated 25 benign tumors, including intradural lesions, with SRS to 12 Gy in a single fraction and observed transient worsening of pretreatment neuropathies in two patients, which resolved with conservative treatment Similarly, Marchetti et al^[80], while treating meningiomas and schwannomas (11 lesions treated 10–13 Gy in 1 frac; 10 lesions 18.5–25 Gy in 4–6 frac) reported acute worsening of sciatica in one patient posttreatment but found no late toxicity among 18 patients with stable or decreased lesions on radiological follow-up.

Shin et al^[81] reviewed 66 spinal SRS procedures for 110 tumors in 58 patients, including schwannomas, neurofibromas, and MPNSTs, with additional analysis on MPNST patients’ survival. The median prescribed dose for benign neurogenic tumors was 13 Gy in a single session, but large tumors received hypofractionated SRS (median 25 Gy in 5 fractions). Malignant tumors were treated with a higher dose of 40 Gy in 5 fractions. After treatment, most patients saw stability or improvement in symptoms, with no notable radiation-related complications post-SRS. Among those with moderate-to-severe pain before treatment, significant improvement in VAS scores was observed (mean score decreased from 6.0 to 1.0 over 8.1 months). Patients with less severe pain also experienced stability or improvement posttreatment. Neurologically compromised patients who underwent surgery followed by SRS showed improvement. No significant radiation-induced complications were observed during the follow-up period post-SRS^[81].

Gerszten et al^[82] studied 73 benign intradural extramedullary (IDEM) tumors treated with CyberKnife SRS, including meningiomas, neurofibromas, and schwannomas. The study involved single treatment sessions for most patients and a median follow-up of 37 months, and a maximum radiation dose of 15–25 Gy (mean dose: 21.64 Gy). They reported 100% long-term tumor control and pain relief in 73% of patients, with few complications, including three cases of radiation-induced myelitis leading to Brown–Séquard syndrome on MRI. Other studies reported different complications, such as radiation-induced myelopathy, posterior column dysfunction^[78].

Systemic therapies and chemotherapy

Adjuvant chemotherapy plays a very limited role in the treatment of benign spinal tumors treatment. Several studies evaluating various medical approaches aimed at inhibiting or stabilizing tumor growth were unable to show any clinical benefits for patients. Thus, adjuvant chemotherapy is not included in the standard therapy, regardless of the tumor’s location.

Chemotherapy drugs may be utilized in the management of advanced or unresectable tumors, specifically. In cases of recurrent SM, treatment with chemotherapy should be evaluated on a case-by-case basis in a multidisciplinary environment^[83]. Youssef et al^[84] utilized everolimus–octreotide combination for recurrent spinal meningiomas, demonstrated antitumor efficacy and was well-tolerated, with no significant long-term side effects observed. In a phase II clinical study, Kumthekar et al.^[85] investigated the efficacy of vascular endothelial growth factor (VEGF) monoclonal antibody – bevacizumab. In the studied patient cohort with various grades of malignancy (G1M, G2M, and G3M), bevacizumab demonstrated anti-tumor activity along with manageable adverse effects. Progression-free survival (PFS) at 6 months (PFS-6), median PFS, and median overall survival (mOS) reported as 87%, 22 months, and 35 months for Grade 1 meningiomas (G1M); 77%, 23 months, and 41 months for Grade 2 meningiomas (G2M); and 46%, 8 months, and 12 months for Grade 3 meningiomas (G3M). The use of bevacizumab in schwannomas is also well documented^[86]. A phase II clinical trial assessing Sunitinib for recurrent anaplastic meningiomas revealed a mean PFS of 5.2 months across all patients. However, the limited number of participants in (only six infratentorial and spinal meningiomas) restricts its applicability and precludes drawing definitive conclusions^[87]. Peptide radioreceptor therapy, known for its effectiveness in treating neuroendocrine tumors by targeting somatostatin receptors, has been considered for meningioma treatment and has shown promising activity, albeit solely within individual clinical trials^[88].

Current literature lacks evidence supporting the use of chemotherapy for benign glomus tumors, and no malignant transformation or metastases have been reported in spinal glomangiomas. However, in rare cases of glomus tumors with uncertain malignant potential, such as those described by Li et al^[91] and Nagata, adjuvant radiotherapy and reoperation were required due to tumor progression or recurrence, highlighting the complexity of managing these tumors^[89,90].

Limitations

The current systematic review is burdened with some limitations. Broad research objectives and heterogeneous outcomes measured by the authors of particular studies precluded conducting quantitative analyses. Furthermore, the articles written in a language other than English were excluded, limiting the analyzed data.

Conclusion

The diagnosis and treatment of intradural extramedullary tumors, particularly spinal glomus tumors, present unique challenges and require a multidisciplinary approach. Radiological examination, with gadolinium-enhanced MRI as the preferred modality, remains essential for precise diagnosis, treatment planning, and surgical guidance. Despite MRI’s efficacy in identifying tumors such as glomangiomas, similarities with other lesions like schwannomas necessitate additional diagnostic tools, such as MRS, to improve detection and differentiation rates. Surgical resection remains the cornerstone of treatment for these tumors, often resulting in favorable outcomes. However, the rise of minimally invasive techniques, including endoscopic and robotic-assisted surgeries, offers promising alternatives with reduced morbidity and improved recovery times. The use of intraoperative nerve monitoring and novel imaging technologies such as AR further enhances surgical precision and safety. Radiation therapy, including SRS, provides an effective adjunct or alternative treatment, particularly for unresectable tumors or residual disease. Emerging technologies and refined techniques in radiation therapy show potential for better tumor control and reduced side effects, although further research is needed to establish standardized protocols. Chemotherapy plays a limited role in the treatment of benign spinal tumors but may be considered for malignant or recurrent cases. Current literature indicates a lack of evidence for the efficacy of chemotherapy in benign glomus tumors, highlighting the need for individualized treatment plans based on tumor behavior and patient-specific factors. Advancements in deep learning and weighted fusion frameworks on MRI for the benign and malignant diagnosis of spinal tumors promise to revolutionize diagnostic accuracy and treatment planning. As research progresses, the integration of these advanced technologies and techniques will likely improve patient outcomes and reduce the need for invasive procedures.

In conclusion, the multidisciplinary approach, combining advanced diagnostic imaging, surgical innovation, radiation therapy, and emerging technologies, provides a comprehensive framework for the effective management of intradural extramedullary tumors. Continued research and clinical trials are essential to refine these strategies, enhance diagnostic precision, and optimize treatment outcomes for patients with spinal tumors.

Ethical approval

“Ethics approval was not required for this review.”

Consent

Informed consent was not required for this review article.

Sources of funding

This research received no external funding.

Author contributions

Writing the paper –W.C., M.S. Study concept or design, data collection, data analysis or interpretation – J.L., J.S., M.S.,B.C.,T.M., K.T., G.S.

Conflicts of interest disclosure

The authors declare that there is no conflict of interest regarding the publication of this manuscript. No funding was received by the authors for conducting this systematic review.

Guarantor

Bipin Chaurasia

Research registration unique identifying number (UIN)

Not applicable

Provenance and peer review

Ok.

Data availability statement

None.