Full Text

Keymessage

While other benign neoplastic lesions in cavernous sinus exhibit non-specific appearances, hemangiomas demonstrate consistent features on imaging. MRI has a valuable role in the follow-up of granulomatous and inflammatory lesions.

Introduction

The cavernous sinus (CS) is an intracranial dural venous sinus extending from the orbital apex and superior orbital fissure anteriorly, to the Meckel’s cave posteriorly. It is composed of a network of small venous channels that may arbitrarily be divided into different compartments. The main venous influx into the CS is the superior and inferior ophthalmic veins, pterygoid plexus, and sylvian vein and posteriorly into the superior and inferior petrosal sinuses. The internal carotid artery (ICA) is the most medial structure within the CS located in the carotid trigone. Cranial nerves III and IV and the first and second divisions of the cranial nerve V (from superior to inferior) are situated in the lateral dural wall of the CS (Figures 1 & 2). Sixth cranial nerve courses in the central part of the CS inferolateral to the ICA. These structures are involved by neoplastic, vascular, infective and infiltrative lesions. Neurogenic tumors, meningiomas, epidermoids, dermoids and cavernous hemangiomas commonly involve the cavernous sinus [1]. Patients present with oculomotor paresis, diplopia, sensory motor disturbances of facial muscles and rarely with symptoms and signs of cavernous sinus thrombosis by spread through the skull base and foramina (Figure 3). The two classifications based on anatomical location by Ishikawa and Jefferson demonstrated inflammatory lesions more in the anterior type and tumors were occupying mainly the posterior and the whole types. T2 hypointense signal pattern is noted in chronic inflammatory lesions, granulomas and lymphomas while cavernomas and hemangiomas demonstrate hyperintensisty on T2 sequence. T1 hyperintensity is typically observed in dermoids, adenomas, melanomas and hemorrhagic lesions (Figure 4). Extradural transcavernous or intradural surgical approaches are decided depending on the plane of the lesion and its extent of location. Though primary radiation is offered as an option, many lesions require biopsy for specific treatment. We retrospectively reviewed the medical records for clinical, imaging and histological diagnostic findings of lesions involving CS. The tumor site, size, morphology or signal characteristics and post intravenous (I.V) gadodiamide enhancement pattern were assessed. The present case series of twenty seven patients describes the clinical presentation and MRI features correlating with histopathology that helped selection of treatment modality.

Figure 1: Diagram of a coronal section through the cavernous sinuses and the sella turcica. Note that the medial wall of the cavernous sinus (CS) is constituted by only a single layer of the dura mater. ACA = anterior cerebral artery, ICA = cavernous segment of the ICA, III = oculomotor nerve, IV = trochlear nerve, MCA = middle cerebral artery, OC = optic chiasm, PIT = pituitary gland, SS = sphenoid sinus, VI = abducens nerve, V1 = ophthalmic nerve, V2 = maxillary nerve, black arrowhead = inner periosteal layer of the dura, white arrowhead = outer meningeal layer of the dura, * = subarachnoid.

Figure 2: Diagram of the segments and major branches of the cavernous segment of the ICA. Light blue structure represents the cavernous sinus (CS). The segments of the cavernous carotid artery from proximal to distal are the posterior vertical (PV), the posterior genu (PG), the horizontal (HORZ), the anterior genu (AG), and the anterior vertical (AV). ACP = anterior clinoid process, DDR = distal dural ring, OA = ophthalmic artery, PDR = proximal dural ring, PLL = petrolingual ligament.

Figure 3: Anatomic pathways related to the cavernous sinuses and the regional bone anatomy. Blue areas represent the position of the cavernous sinuses in the floor of the middle cranial fossa. Yellow lines outline the anatomic region set in boldface in each label. ACP = anterior clinoid process, OC = optic canal, PCP = posterior clinoid process, PPF = pterygopalatine fossa (outlined in orange on image at far left).

Figure 4: Etiologic classification of cavernous sinus lesions. IgG4 = immunoglobulin G4.

Materials and methods

This retrospective study of data was approved by the hospital internal review board. Of the total 27 patients presenting with lesions in the cavernous sinus, 26 consecutive patients with CS lesions underwent MRI examination between 2008 and 2019 (age range, 13-63 years; mean age 41.12, SD ± 14.49; 16 males, 11 females with female to male ratio of 0.60). Both 1.5-T and 3-T systems were used for MR imaging. The imaging technique included conventional spin-echo, echo-planar diffusion-weighted imaging (DWI) and FLAIR sequence. DW images in 21 patients and GRE sequence in 24 were available. Axial T1- weighted (T1-W) contrast enhancement with Omniscan (Gadodiamide, GE Healthcare, Oslo, Norway) was supplemented with and without fat suppression pulse in addition to post contrast SPGR sequence.

Technical parameters

(a) Spin-echo T1- W imaging: repetition time msec/echo time msec, 2000/20; inversion time 400, flip angle 90°; and matrix 296× 205; (b) Fast spin echo T2-weighted imaging (T2-W): 3000/90; flip angle 90°; and matrix 492 × 479 for; and (c) Fluid-attenuated inversion recovery imaging: 10,000/125; 90°; and matrix, 300 × 205. Other parameters: Slice gap was 1.5-2 mm with a 5-mm section thickness, and FOV 230 × 250 mm. Echo-planar DW MR imaging was performed in the axial plane before contrast enhancement. T1-W 3 D spin-echo sequences without fat suppression and post contrast SPGR sequences were obtained using a matrix of 492 × 479; slice thickness 1mm without slice gap; flip angle, 9°; repetition time/echo time 5/10 msec and FOV 240 x 240. Intravenous dose of gadodiamide was 0.2 mL/kg (0.1 mmol/kg) body weight. T1-W, T2-FSE and FLAIR signal intensities were tabulated as hyper, hypo, iso intensities. Blooming and restriction of diffusion on the GRE and DWI sequences and intensity of contrast enhancement on T1- fat suppressed sequences were evaluated. Two patients underwent digital subtraction angiography. Histopathological diagnosis of lesion in the cavernous sinus was verified from the notes in the medical case files retrospectively.

Results

The study includes 27 patients and 59% were males. Visual disturbances were the predominant feature (62%) and about two thirds of the patients had no visual symptoms despite large size of the lesion. Cranial nerve involvement was observed in 55 % of the patients (Tables 1). Extra cavernous sinus involvement of Meckel’s cave, cerebral peduncles and basal cisterns were seen in all except one. Of the 26 patients who received surgery, complete excision in 9 patients was possible and subtotal excision or biopsy in the others. Trace images of DWI demonstrated facilitated diffusion in 11 and restricted (hyperintense in DWI and hypointense in ADC) in 10 patients. Mean ADC value in 17 patients was 0.09176, wherein lesions with restricted diffusion showed 0.7914 and those with facilitated demonstrated 0.776 × 10⁻³ mm²/s. Apart from meningioma and schwannoma, blooming on GRE sequence was observed in pituitary tumour and an epidermoid cyst. Varying patterns of enhancement of the lesions were noted following intravenous contrast enhancement (Table 2&3). Follow up MRI examination of cavernous hemangiomas after two years demonstrated total regression in one patient and significant reduction with thrombosis of the centre of the lesion in another. Homogeneous hyperintensity on T2-W sequences, lack of diffusion restriction and intense contrast enhancement were the hallmarks of cavernous hemangiomas (CSH) in contrast to the other lesions.

Table 1: Patient summary shows clinical spectrum in 27 patients.

 Note: III, IV, V, VI -- Cranial nerves; NA – Not available; ATT – Antituberculous treatment.

Table 2: MRI features of 27 lesions and mean ADC values are shown.