23-year review of spheno-orbital meningioma: clinical, radiological, and pathological insights from 100 cases

Abstract

Background

Spheno-orbital meningioma (SOM) represents a unique variant of sphenoid wing meningiomas, distinguished by its propensity for bone infiltration and cranio-orbital involvement. SOM exhibits a considerable incidence of misdiagnosis and recurrence.

Purposes

To elucidate the clinical, radiological, and pathological characteristics of SOM.

Methods

Review of electronic medical records, histopathology, radiological images and follow-up information of 100 SOM patients.

Results

Of the 100 patients (28 males, 72 females) with SOM, mean age was 46.8 ± 12.6 years and prevalent symptoms were proptosis (99%). All the CT scans showed hyperostosis with 89.3% of the hyperostosis having an irregular edge. In MRI scans, dural tail sign was observed across all patients and the cranio-orbital tumors often penetrated temporal muscle (74.1%), extraocular muscle (74.1%) and lacrimal gland (63%). All the 100 patients underwent surgical intervention, and among them, 62 individuals received postoperative radiotherapy. Grade I resections had a lower recurrence rate(16.7%), which further decreased with the addition of radiotherapy(13.9%). In contrast, all patients with grade II or higher grade resections without radiotherapy experienced recurrence, indicating a higher risk associated with less complete tumor removal. The pathological examination revealed that intraorbital sections exhibited comparable tumor density to intraorbital SOM tumors, along with increased fibrous density but decreased vascular distribution.

Conclusions

Radiological characteristics of SOM included cranio-orbital tumors, hyperostosis of the sphenoid wing with an irregular edge, and dural tail sign. Combination of gross total resection and adjuvant radiotherapy was recommended to minimize recurrence rate. Intracranial SOM tumors tended to be softer and more bleed-prone than intraorbital sections, necessitating surgical precision.

Highlights

Imaging shows sphenoid wing hyperostosis and the dural tail sign.

Complete tumor removal, followed by radiotherapy, reduces recurrence risk.

The intracranial part exhibits a lower fibrous and higher bleeding intention.

Introduction

Spheno-orbital meningioma (SOM) is a distinct subtype of sphenoid wing meningioma. SOM is distinguished by cranio-orbital involvement; however, research indicates that ocular symptoms frequently manifest as the initial signs of SOM [1,2,3,4]. Consequently, patients with SOM commonly initiate their medical consultation with ophthalmologists, emphasizing the necessity for ophthalmologists to possess a thorough comprehension of the disease’s clinical characteristics in order to avoid misdiagnosis. Nevertheless, the clinical and radiological manifestations of SOM are easily confused with other diseases, resulting in a significant rate of misdiagnosis during the initial presentation [1].

The primary treatment for SOM is surgical resection, which poses intricate challenges due to the tumor’s proximity to vital structures such as extraocular muscles, optic and oculomotor nerves, superior orbital fissures and cavernous sinuses. This infiltration presents notable obstacles to achieving a thorough resection, resulting in a higher recurrence rate compared to other subtypes of meningioma [3, 5]. Due to their specialized knowledge of the orbit’s anatomical structures, ophthalmologists, in collaboration with neurosurgeons who possess expertise in the intracranial region, are better equipped to perform joint surgeries that facilitate the complete removal of lesions and minimize complications. Consequently, ophthalmologists play a crucial role in promptly diagnosing and appropriately treating SOM.

In our study, we undertook an exhaustive review of medical records from 100 primary onset cases of SOM treated at our institution over the past 23 years. Our objective was to shed light on the clinical manifestations, imaging features, treatment modalities, recurrence patterns, and histopathological findings related to SOM. To our understanding, this retrospective analysis is one of the most extensive single-center clinical series studies on SOM to date.

Materials and methods

This study was a retrospective observational case series. This study was performed with informed consent and retrospectively approved by the ethics committee of the 3rd Medical Center of Chinese PLA General Hospital (KS2023-014).

In our research, we conducted a retrospective review of a database encompassing 206 patients diagnosed with cranio-orbital meningioma. These patients underwent craniotomies at the 3rd Medical Center of the Chinese PLA General Hospital from 2000 to 2022. From this analysis, we derived a refined set of screening criteria for patient inclusion. All patients exhibited symptomatic mass effect and/or progressive visual neurological deficits, leading to significant functional impairments such as vision loss, proptosis causing disfigurement, and other cranial neuropathies that impacted their quality of life. These conditions were in line with the indications for surgical resection. Criteria encompassed primary onset, histopathological confirmation of diagnosis, and meningioma originating from the sphenoid wing and involved the orbit. The tumor growth location was ascertained by integrating patients’ imaging data with lesion location descriptions from operative notes. From the initial cohort, only 100 patients satisfied all criteria. This study thus comprised these 100 patients, each of whom underwent preoperative ophthalmological and neurological assessments. We collated clinical data for each patient, which included demographics (age, gender), preoperative visual acuity, clinical signs and symptoms, misdiagnosis records, radiological imaging, surgical records, and pathological findings. It’s noteworthy that all 100 patients had CT scans, and 97 underwent contrast-enhanced MRI, as verified by their medical documentation. Experienced neuro-oncologists meticulously reviewed the CT scans of 28 patients and MRI scans of 27 patients.

All patients underwent a standardized surgical procedure. This procedure involves an initial skin incision starting from the upper edge of the zygomatic arch, extending upwards to the temporal line, and then curving forward to intersect both the midline and the hairline. The flap is reflected anteriorly. The temporalis fascia and periosteum are consistently incised in the anterior quarter of the temporalis muscle, separated up to the superior and lateral orbital rim. Concurrently, the deep temporalis fascia and periosteum are incised. The temporalis muscle is then dissected to the superior edge of the zygomatic arch and secured. A frontotemporal bone flap is subsequently created, and a precision grinding drill is employed to excise the sphenoid ridge and any abnormal skull hyperplasia. This reveals the superior orbital fissure, lateral orbital wall, anterior clinoid process, and greater wing of the sphenoid, thereby exposing the skull base and temporal floor. The electric oscillating saw or wire saw is utilized to separate the lateral edge of the orbit, facilitating the exposure of the orbital contents from the lateral side of the supraorbital foramen to the medial side of the frontozygomatic suture. The intraorbital meningioma segment is excised, followed by microscopic removal of the intracranial meningioma portion. During the procedure, it’s vital to avoid undue tension on the extraocular muscles and to dissect them along the muscle fibers to prevent hemorrhage or transection, which might result in ocular motility disturbances. It’s imperative to delineate and manage the tumor, typically adhered to the orbital periosteum layer, which has a clear demarcation from the surrounding eye muscle. Adhering to these surgical principles can mitigate complications and enhance postoperative results. Depending on the tumor’s extent, the lateral wall of the cavernous sinus may be incised. If required, based on the tumor’s overall spread, additional steps like anterior clinoidectomy, optic canal unroofing, and foramen ovale opening might be incorporated. Conclusively, a silicone drain is positioned to facilitate fluid drainage.

According to the operative note and postoperative imaging, we evaluated and classified the SOM surgeries through Simpson grading system [6]. Simpson Grade I necessitates the complete removal of the tumor, including the underlying bone and associated dura, as observed macroscopically. Grade II entails the complete removal of the tumor with coagulation of the affected dura only. Grade III is characterized by the complete removal of the tumor without resecting the dura or underlying bone. Grade IV is defined as subtotal tumor resection. Grade V indicates simple decompression, with or without biopsy.

The pathological analysis relied on the professional judgment of two experienced pathologists according to the WHO classification and specific anatomical location. All identifying patient information was removed, and each patient was assigned a unique number to maintain anonymity. Six hematoxylin and eosin (HE)-stained slides were selected for each patient, comprising three intracranial tissue slides and three intraorbital tissue slides. The two pathologists independently reviewed the intracranial and intraorbital slides for each patient. After evaluating the slides for a given patient, each pathologist immediately documented their observations. This evaluation process was repeated for all patients. Upon completion of the independent evaluations, the recorded findings from both pathologists were compared. When discrepancies were identified between the two sets of results, a third pathologist was consulted to provide an additional review and make the final determination. This systematic approach ensured a thorough and unbiased assessment of the pathological differences between intracranial and intraorbital tissues.

The cretiria of application of radiotherapy postoperative was recommended if postoperative pathology indicated a tumor proliferation index (Ki-67) greater than 5%. In cases of complete tumor resection but with intraoperative evidence of tumor invasion into the arachnoid or abnormal bone structure, radiotherapy was still recommended even if the postoperative Ki-67 was less than 5%. Additionally, radiotherapy was advised for incomplete tumor resection.

Long-term clinical and radiological follow-up data were available for 61 patients. This information was gathered through telephone conversations, outpatient consultations, or medical records. Patient characteristics were presented as means ± SD (standard deviation) or in percentages, as deemed suitable.

Results

Epidemiological and clinical characteristics

The study involved 100 patients diagnosed with SOM, consisting of 28 males and 72 females. The mean age at admission was 46.8 ± 12.6 years (range, 10–71 years). The predominant age group for onset was between 40 and 49 years, representing 38% of the cohort. Notably, 3 patients were below 15 years of age.

The most prevalent symptom was proptosis, observed in 99% of the patients. This was followed by eyelid edema in 63 patients (63%) and eye movement disorders in 21 patients (21%). A decline in visual acuity was reported by 43 patients (43%). In contrast, only 20 patients (20%) mentioned experiencing headaches and dizziness. Upon their first visit, only 48 patients (48%) received an accurate diagnosis. The most frequent misdiagnosis was fibrous dysplasia, accounting for 13.5% of the 52 misdiagnosed cases (Table 1).

Table 1 Epidemiological and clinical characteristics

Radiological findings

After excluding cases with incomplete or absent imaging data, CT scans from 28 patients and MRI scans from 27 patients diagnosed with SOM were analyzed. All 28 CT scans exhibited hyperostosis, with 96.4% displaying greater sphenoid wings hyperostosis (Fig. 1A), 39.3% showing temporal bone hyperostosis, 39.3% indicating lesser sphenoid wings hyperostosis, and 14.3% presenting anterior clinoid process hyperostosis. Remarkably, in 25 patients (89.3%), the edges of the hyperostosis were irregular. A unique instance of an osteolytic lesion on the greater sphenoid wing was identified (Fig. 1B). Additionally, a subdural ossification layer was present in one case (Fig. 1C). On the MRI T1 enhanced fat-suppressed sequence, 18 cases (66.7%) showed bone intensification of the sphenoid bones (Fig. 1D). It’s worth noting that none of the patients had dilated superior orbital fissures; however, 6 patients (21.4%) exhibited constricted fissures.

Fig. 1

Radiological Characteristics of SOM Patients: (A) illustrates hyperostosis of the sphenoid wing with irregular boundaries. (B) depicts osteolytic bone lesions. (C) highlights a subdural ossification layer. (D) displays the enhanced region of the sphenoid bone on T1-weighted MRI with fat suppression sequences using gadolinium enhancement, accompanied by tumor infiltration into the lacrimal gland and extraocular muscles

On CT scans, the intracranial segments of the tumors exhibited no discernible differences from the surrounding normal brain tissue. Conversely, MRI imaging demonstrated the tumors to be iso- to hypointense on T1-weighted sequences, hyper- to isointense on T2-weighted sequences, and exhibiting marked enhancement on the T1-weighted enhanced sequence, which allowed for the precise determination of the extent of tumor invasion. Among the 27 patients, 20 (74.1%) demonstrated penetration into the extraocular muscle and temporal muscle (Fig. 1D), while 17 (63%) exhibited involvement of the lacrimal gland. Additionally, 11 (40.7%) infiltrated into the cavernous sinus, 4 (14.8%) extended to the free edge of the cerebellar tentorium, and 2 (7.4%) involved the pituitary fossa. The presence of the dural tail sign was consistently observed in all patients (Fig. 1D). It is noteworthy that none of the SOM tumors exhibited necrosis or cystic alterations. Brain tissue edema was only detected in 2 patients (7.4%).

Treatment and prognosis

In this study, ophthalmologists and neurosurgeons collaboratively performed the surgeries for the 100 SOM patients. The resection outcomes were categorized according to the Simpson grading system, with 88 patients achieving a grade I resection, 9 achieving a grade II resection, 1 achieving a grade III resection, and 2 achieving a grade IV resection. Among the patients, 62 received postoperative radiotherapy, with 58 of them having undergone a grade I resection, 3 having undergone a grade II resection, and 1 having undergone a grade IV resection (Fig. 2).

Fig. 2

Treatment and follow-up of SOM

We conducted a follow-up on various treatments to determine the recurrence rate among patients receiving different interventions. During a mean follow-up of 78 ± 51.7 months, 61 patients were successfully contacted and a total of 13 patients (21.3%) experienced recurrence (Table 2). Among the 18 patients who underwent Simpson grade I resection, 3 patients (16.7%) experienced recurrence in 125.5 ± 33.4 months. Conversely, among the 36 patients who underwent Simpson grade I resection and received postoperative radiotherapy, only 5 patients (13.9%) experienced recurrence in 72 ± 33.5 months. In the case of patients undergoing Simpson grade II resection, 100% experienced recurrence, while 33.3% (1 out of 3) of patients who underwent Simpson grade II resection and received postoperative radiotherapy experienced recurrence. It is worth noting that all patients who underwent Simpson grade III resection (n = 1) and Simpson grade IV resection, regardless of postoperative radiotherapy, experienced recurrence (Fig. 2). The average recurrence time for patients who relapsed was 67.2 ± 48.8 months. For patients who underwent surgical resection alone, the average recurrence time was 79.1 ± 58.8 months. In contrast, for those who received radiotherapy following surgical resection, the average recurrence time was 57 ± 40.2 months. However, the P-value between the two groups is greater than 0.05, indicating that the difference is not statistically significant. The shortest recurrence time after surgery was 11.2 months, which was observed in a patient who underwent Simpson Grade II resection with postoperative radiotherapy (Table 2). Worthy of note is that the meningioma in this patient is pathologically graded as WHO Grade II. Conversely, the longest duration to recurrence, amounting to 171.1 months, was observed in a patient who underwent Simpson I resection (Table 2). Notably, the patient who underwent Simpson IV resection with postoperative radiotherapy encountered 5 subsequent recurrences.

Table 2 Recurrence characteristics

Furthermore, we conducted a study on 96 SOM patients to assess the alterations in clinical symptoms following surgery. Our findings indicated a significant improvement in headache and dizziness symptoms in 20 patients preoperatively. Additionally, postoperative vision either improved or remained stable in 79.1% of patients (n = 76). Nevertheless, it is important to acknowledge that not all outcomes were favorable. Adverse outcomes were also identified, including visual acuity decrease in 20.8% (n = 20) of patients, ptosis in 17.7% (n = 17), and eye movement disorders in 19.8% (n = 19). Only one patient developed brain herniation after surgery. Fortunately, no fatalities were reported in relation to the surgical interventions performed for SOM.

Pathology

Among the 100 cases of meningioma that were pathologically confirmed, 95 cases were found to have distinct subtypes according to the WHO classification. Of these, 92 cases were classified as WHO grade I, with the majority being of the meningothelial subtype (71 out of 92, accounting for 77.2% of the cases), followed by the transitional subtype (7 cases, 7.6%). Two cases were classified as WHO grade II (atypical meningioma), while a single case was classified as WHO grade III (anaplastic meningioma). It is worth noting that 8 tumors (8.4% of the cases) exhibited mixed WHO subtypes, as indicated in Table 1. Upon examination of 47 hyperostotic bones, tumor cells were detected in 43 samples, accounting for 91.5% of the cases.

In a subsequent comparative analysis of 27 SOM samples, intriguing observations were made regarding the intracranial and intraorbital sections. The tumor density was found to be consistent between the intraorbital and intracranial segments in 21 cases, as depicted in Fig. 3A and B. Furthermore, the fibrous density was higher in the intraorbital segment for 18 cases, representing 66.7% of the sample, in contrast to the intracranial section, as shown in Fig. 3C and D. Additionally, 9 cases exhibited equivalent fibrous densities in both sections. In 63% of the cases, the vascular distribution in the intraorbital segment was found to be less dense compared to the intracranial segment, as illustrated in Fig. 3C and D. Additionally, 24.9% of the cases exhibited similar vascularity in both sections. Notably, a discrepancy in tumor classification between the intraorbital and intracranial segments was observed in 5 cases (18.5%).

Fig. 3

Photomicrographs highlight the distinctions between the intraorbital and intracranial sections of a single tumor. (A) and (B) depict similar tumor cell densities in both the intraorbital and intracranial regions. Conversely, (C) and (D) reveal a heightened fibrous density in the intraorbital tumor compared to its intracranial counterpart, with the former also exhibiting a less dense vascular distribution

Discussion

Understanding the clinical characteristics of SOM is of utmost importance in facilitating timely diagnosis and intervention. Our data revealed that SOM primarily impacted middle-aged women, with a majority of them experiencing ocular symptoms. Specifically, proptosis was observed in 99% of patients, eyelid edema in 63%, and decreased visual acuity in 43%. In contrast, only 20% of patients reported cranial symptoms like dizziness and headaches. These findings are consistent with previous research [2,3,4,5, 7].

Radiological imaging, particularly CT and enhanced MRI, plays a pivotal role in diagnosing and treating SOM [1]. CT scans primarily identify bone alterations, whereas MRI scans determine the tumor’s extent. Our study revealed that all SOM patients (100%) exhibited hyperostosis on CT, with 88.9% displaying irregular hyperostosis edges. This CT imaging features may distinguish SOM from fibrous dysplasia, which is characterized by hyperostosis with smooth edges [8]. Prior research findings propose that the hyperostosis of SOM arises due to tumor cell infiltration into the bone interior [3, 9]. Consistently, our histopathology findings indicate that 91.4% of the analyzed bones exhibited the presence of meningioma cells, thereby emphasizing the urgent necessity for surgical excision of hypertrophic bone. Additionally, we have made a novel discovery that malignant SOM can occasionally manifest as osteolytic lesions spanning the entire layer, albeit infrequently (1 out of 27 cases). Our observation of the dural tail sign in all SOM patients is consistent with previous research [7]. In Song-tao Qi’s study on convexity meningiomas, it was suggested that tumor invasion could be associated with specific types of dural tail signs [10]. Consequently, a comprehensive removal of the dura may help minimize the risk of tumor recurrence. Although the superior orbital fissure serves as the primary communication channel between the two SOM parts, none of our patients exhibited its widening. Instead, some (15.4%) showed fissure narrowing, possibly due to the proliferation of enclosing bony structures, such as the sphenoid bone’s greater and lesser wings and body. However, a limitation of this study is that complete imaging records were only available for 28 patients due to data retention challenges over the extended timeframe. These challenges arose from multiple upgrades to the hospital’s imaging system and patients’ loss of personal copies. More robust data management and retention strategies will be implemented to ensure comprehensive data retention for future research.SOM is prone to recurrence, and the recurrence rate varies depending on the treatment method. The reported recurrence rate ranges from 21 to 50% [11,12,13]. In our study, the recurrence rate for patients who underwent Simpson grade I surgery in conjunction with postoperative radiotherapy was determined to be 13.9%, representing the most favorable outcome among all treatment modalities. Conversely, among patients who underwent incomplete resection (Simpson grade II and higher), a substantial majority of 100% (3 out of 3 individuals) experienced recurrence. Furthermore, the recurrence rate for patients who underwent incomplete resection combined with postoperative radiotherapy was equally significant, reaching 50% (2 out of 4 cases). This finding suggested that a comprehensive resection of SOM, along with postoperative radiotherapy, could effectively reduce the recurrence rates. Notably, SOM recurrence tends to occur slowly [14, 15]. In the present study, it was observed that the minimum duration for recurrence following the surgical procedure was 11.2 months, while the maximum duration recorded was 171.1 months.

The collaboration of ophthalmologists and neurosurgeons is imperative in order to attain maximal tumor resection and mitigate surgical complications. In our study, wherein both specialists jointly performed the surgeries, we observed an impressive Simpson grade I resection rate of 88% and a relatively low postoperative complication rate of 27.5%. We recommend combined pterygopoint and orbital wall dissection approach, supplemented by a frontotemporal method. This integrated methodology provides enhanced tumor visibility, allowing for multi-angle observation, broader resection, and improved decompression of orbital content. Moreover, it significantly mitigates stress on the temporal lobe. By resecting the anterior bed process and opening the optic canal, the length of the optic nerve within the internal carotid triangle is effectively doubled, expanding its area of exposure. This facilitates decompression of the optic canal, which has been suggested by previous studies to potentially improve postoperative patient outcomes [12, 16]. When the tumors are fully exposed, it is advised that the ophthalmologist initially excise the intraorbital tumor, followed by the sequential removal of epidural and intradural tumors by the neurosurgeon. This procedural approach not only enhances the anterior surgical field, ensuring an unobstructed visual field, but also enables a thorough examination of the tumor from various perspectives and facilitates appropriate directional separation. To achieve complete eradication of SOM lesions and minimize the likelihood of recurrence, it is imperative to excise not only the primary tumor, but also the degenerated bone and affected meninges.

Postoperative pathological results showed that the predominant histopathological subtype of SOM is endothelial, corroborating earlier studies [17, 18]. Uniquely, we conducted a comparative analysis between the intracranial and intraorbital segments of the SOM tumors, revealing that the intracranial component exhibited a lower fibrous composition and a higher degree of vascularization in contrast to its orbital counterpart. This observation suggested that the intracranial tumor segment possessed a softer consistency, heightened vulnerability to compression, and an increased propensity for bleeding when compared to the orbital section. As a result, surgeons are advised to exercise caution during the management of intracranial tumors in SOM cases, emphasizing delicate manipulation techniques and minimizing direct clamping of tumor tissue. The Ki-67 value, the degree of tumor invasion into surrounding tissues, and the extent of surgical resection all play crucial roles in determining the necessity of postoperative adjuvant therapy.

This study is subject to inherent limitations characteristic of retrospective research, such as potential selection bias and reliance on the accuracy and completeness of medical records. It is worth noting that only 61 patients in this group had access to comprehensive long-term medical records, which is a common occurrence in retrospective reviews and highlights the possibility of missing or incomplete data. Furthermore, it is important to acknowledge that the data collection was exclusively conducted at a single medical center, which might constrain the generalizability of our findings to broader populations or regions. Nevertheless, it is noteworthy that our patient cohort is diverse, encompassing individuals from various regions across China, partially mitigating this limitation. Additionally, this single-center approach ensures standardized surgical procedures, enhancing the consistency of our treatment data.

Taken together, SOM predominantly affects middle-aged women and the primary clinical presentation is proptosis. Despite this limitation of data retention challenges, the data from the 28 patients with complete imaging records provided valuable insights. On CT scans, SOM often presents as hyperostosis of the sphenoid wing, characterized by roughened surfaces. Enhanced MRI delineates the tumor’s extent, displaying iso- to hypointense signals on T1-weighted images and hyper- to isointense signals on T2-weighted images. Gross total resection combined with adjuvant radiotherapy could minimize recurrence rate. Meningothelial tumors are the most prevalent histological subtype. Notably, the intracranial component of SOM tumor, being less fibrous and more vascular, tends to be softer and more fragile, making it bleed-prone.