Challenging the osseous component of sphenoorbital meningiomas

Abstract

Background: Intraosseous growth is a unique feature of sphenoorbital meningiomas (SOM). Its close relation to neurovascular structures limits complete surgical resection and possibly contributes to the high recurrence rate.

Objective: To evaluate the growth behavior of intraosseous remnants and develop a protocol for precise intraoperative visualization of intraosseous SOM.

Methods: We included 31 patients operated for SOM from 2004 to 2017. The growth velocity of the intraosseous tumor component was volumetrically calculated in 20 cases. To improve accuracy of image guidance, we implemented a specialized bone surface-based registration algorithm. For intraoperative bone visualization, we included CT in multimodality continuous image guidance in 23 patients. The extent of resection (EOR) was compared with a standard MR-only navigation group (n = 8).

Results: In 11/20 cases (55%), a progressive regrowth of the intraosseous SOM remnant was noted during a mean follow-up of 52 months (range 20-132 months). We observed a mean increase of 6.2 cm³ (range 0.2-23.7 cm³) per patient and side during the follow-up period. Bone surface-based registration was significantly more accurate than skin surface-based registration (mean 0.7 ± 0.4 mm and 1.9 ± 0.7 mm, p < 0.001). The EOR of the intraosseous component was significantly higher using CT + MRI navigation compared with controls (96% vs. 81%, p = 0.044).

Conclusion: Quantitative assessment of the biological behavior of intraosseous remnants revealed a continuous slow growth rate independent of the soft tumor component of more than half of SOM. According to our data, application of a multimodal image guidance provided high accuracy and significantly increased the resection rate of the intraosseous component of SOM.

Keywords: Bone infiltration; Image guidance; Meningioma; Skull base; Sphenoorbital meningioma.

Fig. 1

Neurovascular limit for safe resection of the intraosseous component of SOM. To safely accomplish removal of the largest part of the intraosseous tumor component, we propose an anatomical limit based on critical neurovascular structures. From a standard pterional approach, this limit was defined by the frontal, ethmoid, and sphenoid sinuses (FrontS, EthmS, SphenS), the medial wall of the optic canal (OC), superior orbital fissure (SOF), foramen rotundum (FR), foramen ovale (FO), foramen lacerum (FL), and temporomandibular joint (TMJ) pan. FS = foramen spinosum

Fig. 2

Bone surface-based registration: 3D-model used for registration (a). Concurrent intraoperative view of the exposed bone (b) with attached bone-attached patient reference tracker and landmark check of the coronal suture after bone surface-based registration (c)

Fig. 3

Case 49f, left-sided SOM: intraoperative multimodality image guidance concurrently displaying (1) osseous tumor extension (CT-based segmentation, yellow) and (2) soft tumor extension (MR T1 CE-based, blue). The background image is a MR T1 CE fat suppression MRI. a Soft tumor extension into the orbit via the SOF with surrounding osseous SOM involvement of the anterior clinoid process, the middle fossa floor extending up to the lateral wall of the sphenoid sinus, and the temporal bone. b Soft tumor extensions at the temporopolar dura, the SOF, the lateral orbit, and the temporal muscle. Osseous tumor extension in the greater wing of the sphenoid bone

Fig. 4

Operative result of case 54f by merging preoperative (yellow) and postoperative (orange) CT scans. a The osseous tumor remnants (orange) are mainly confined medial to the proposed line. b Removal of the lateral parts of the greater sphenoid wing. Resection was limited by involvement of the foramen rotundum and lateral sphenoid sinus wall. c Removal of the middle cranial fossa infiltrate lateral to the oval foramen. d Osseous remnant covering the TMJ pan medially and anteriorly

Fig. 5

Growth rate of the intraosseous tumor component of SOM: in 11/20 cases (55%) with postoperative remnants and in all contralateral sides of the 3 bilateral tumors, a progressive regrowth was noted. Follow-up mean 52 months (range 20–132 months). Growth velocity mean 1.48 cm³ tumor increase per year (range 0.1–6.4 cm³/year), indicated by the interrupted line

Fig. 6

Comparison of EOR between CT + MRI and the MRI-only group. EOR was significantly higher for the osseous tumor component (p = 0.044; r = 0.36), but not for the soft tumor component (p = 0.286; r = 0.25) after application of the multimodality navigation protocol.

Fig. 7

Example for regrowth of osseous SOM remnant. Case 47f, meningothelial SOM (WHO I): right-sided SOM presenting with exophthalmus and visual disturbance. At initial presentation, the intraosseous meningioma component was confined to lateral of the proposed resection line. A complete resection of the soft component and a subtotal resection of the osseous component with restitution of exophthalmus and vision was achieved (yellow = remaining osseous tumor, 1.1 ccm).