Ocular Ischemic Syndrome Secondary to Inflammatory Pseudotumor of the Skull Base Masquerading as Giant Cell Arteritis

A 77-year-old man was referred to neuro-ophthalmology because of 2 months of intermittent visual disturbances, consisting of flashes of light affecting the left eye more than the right, lasting 1 minute at a time, and recurring every 2 days. More recently, he had also developed 1-hour episodes of a kaleidoscopic “plaid” pattern of colors obscuring central vision in his right eye, usually occurring when he went outside on a bright day, and resolving on returning indoors. Over the previous month, he had also noticed new left-sided headache, scalp tenderness, left-sided jaw claudication with prolonged chewing, and shoulder and neck myalgias. He did not have any fevers, chills, sweats, or weight loss. His medical history included Type 2 diabetes with associated retinopathy requiring laser panretinal photocoagulation (PRP), hypertension, dyslipidemia, coronary artery disease, and atrial fibrillation. He had previously undergone left carotid endarterectomy. On examination, visual acuity was 20/25 in each eye. Color vision, pupils, motility, and alignment examinations were normal, and there was no relative afferent pupillary defect. Slit-lamp examination was notable for prominent conjunctival vessels in both eyes. Intraocular pressures were low, at 5 mm Hg in each eye. Funduscopy showed normal optic discs, maculae, and vessels; peripheral retinas were notable for proliferative diabetic retinopathy and past PRP. Humphrey perimetry was unremarkable bilaterally. Palpation of his temporal arteries revealed weak pulses bilaterally and tenderness on the left. Bloodwork revealed an ESR of 38 mm/h, and CRP was 4.9 mg/L; however, despite these unremarkable inflammatory markers, there was a heightened suspicion for giant cell arteritis (GCA), given his new history of jaw claudication, headaches, proximal myalgias, and transient visual disturbances. He was started on a 60 mg daily dose of oral prednisone and he was scheduled for a temporal artery biopsy (TAB) in 7 days' time. In the interim, fluorescein angiography revealed proliferative diabetic retinopathy and PRP in both eyes. Retinal transit time was prolonged, at approximately 40s, prompting suspicion for ocular ischemic syndrome (OIS). In this context, the hypotony was felt most likely to be due to ciliary body hypoperfusion. Doppler ultrasonography was undertaken to investigate the OIS and showed a chronic occlusion of the right internal carotid artery (ICA) and a calcified plaque in left ICA, which was not causing a hemodynamically significant stenosis. Because bilateral carotid occlusive disease was not detected, it was felt that any underlying OIS was likely artertic in origin, and therefore the TAB was allowed to proceed. The TAB, undertaken 7 days after starting prednisone, was negative for GCA. By 2 weeks, he had noticed significant improvement in symptoms, with no further visual disturbances, myalgias, or jaw claudication, and an “80%” improvement in his headaches. Prednisone was discontinued, given the normal erythrocyte sedimentation rate (ESR)/C-reactive protein (CRP), negative TAB, and risk of exacerbating his diabetes; however, his symptoms recurred shortly after its cessation. MRI head and neck (Fig. 1A) showed an extensive infiltrative mass lesion involving the left nasopharynx, left parapharyngeal space, and central skull base. The muscles of the left medial pterygoid and soft palate were infiltrated by the lesion. MR angiography (Fig. 1B) revealed 2 tandem stenoses of the left ICA caused by the mass, the tightest of which was at the junction between the carotid canal and the cavernous sinus (with threatened near occlusion) and the second of which was at the skull base.FIG. 1.: MRI. A. A coronal T1 postgadolinium MRI demonstrates the ill-defined mass involving the left nasopharynx, left parapharyngeal space, and central skull base (arrow). B. A contrast-enhanced MRA of the upper neck and circle of Willis demonstrates the 2 focal stenoses (arrows) of the left internal carotid artery (ICA). Of note, the right ICA is not visible due to a known long-standing occlusion.These imaging findings were suspicious for nasopharyngeal carcinoma; however, despite 3 operative biopsies, histopathology was notable only for fibroinflammatory tissue containing a mixed/polymorphous population of well-differentiated lymphocytes (B and T cells), plasma cells, histiocytes, and neutrophils (Fig. 2). Granulomas and perivascular inflammation were absent, and there was no necrosis. Lymphoid follicles with reactive germinal centers were not seen. No neoplastic or dysplastic cells were seen. Immunohistochemical staining was negative for IgG4, and there was no storiform fibrosis or obliterative thrombophlebitis to suggest IgG4-related disease. There were no fungal organisms or mycobacteria. Infectious and inflammatory bloodwork returned negative.FIG. 2.: Hematoxylin and eosin (H&E)-stained biopsy specimens. A. ×10 magnification, depicting loose fibrovascular connective tissue with associated prominent mixed inflammation. There is no granulomatous inflammation, and necrosis is absent. Storiform fibrosis and occlusive thrombophlebitis are not seen. B. ×40 magnification, better depicting the inflammatory cells, which include numerous neutrophils, plasma cells, and well-differentiated lymphocytes (confirmed on subsequent CD3, CD5, and CD20 stains to be a mixed population of B and T cells). No neoplastic or dysplastic cells are evident. Although reactive changes are present in the blood vessel endothelial cells, there is no evidence of thrombosis or vasculitis. Immunohistochemical staining for IgG4 was negative.Because there was sufficient evidence to rule out alternative diagnoses such as IgG4-related disease, lymphoma, granulomatosis with polyangiitis, fungal or mycobacterial infection, sarcoidosis, vasculitis, foreign body reaction, and dysplasia/neoplasia, the patient was diagnosed with idiopathic inflammatory pseudotumor of the skull base—a diagnosis of exclusion. Corticosteroids were resumed, with an IV pulse followed by daily oral prednisone 30 mg by mouth. Azathioprine 100 mg by mouth daily was also initiated. The vascular surgery service recommended medical management over a carotid revascularization procedure. Symptoms had improved at a 1-month follow-up visit, and MRI at 4 months revealed a decrease in the size of the mass and the degree of ICA compression, with a further reduction on MRI at 12 months. The patient has since continued to have no visual disturbances and normal visual fields. Two years later, he remains stable on azathioprine and oral prednisone 5 mg daily. DISCUSSION Although our patient presented with symptoms highly suggestive of GCA, his final diagnosis was nonarteritic OIS secondary to idiopathic inflammatory pseudotumor of the skull base. Three aspects of this case merit mention: 1) non-GCA entities may closely mimic jaw claudication and other classic GCA symptoms; 2) inflammatory pseudotumor of the skull base is a rare etiology for OIS; 3) knowledge of the limitations of Doppler ultrasonography can reduce false negatives in the detection of ICA stenoses. Jaw claudication is a highly specific, but not pathognomonic, symptom of GCA: it may also occur with non-GCA vasculitides, with nonvasculitic vascular conditions, and even with nonvascular etiologies (1). Our patient's jaw claudication was likely from a combination of nonarteritic near-occlusive compression of his only remaining ICA and direct inflammation and infiltration of the left medial pterygoid and palatal muscles. Although near-occlusive intraluminal ICA stenoses are the usual cause of OIS, our patient's OIS was caused by extrinsic compression of his only patent ICA by an unusual chronic inflammatory mass. Our review of the literature has not found any other reports of OIS secondary to occlusion by a mass. One report exists of a 39-year-old Chinese man with a history of nasopharyngeal carcinoma who developed OIS, but this developed as a consequence of delayed radiation-induced stenosis and not direct mass effect from tumor (2). Although most of our patient's episodes of visual disturbance were fleeting, he did have longer episodes, lasting up to 1 hour at a time. These episodes, although somewhat atypical in duration for OIS, nevertheless correlated closely with prolonged exposure to bright sunlight, and their near-immediate resolution with a return indoors was consistent with the retinal demand ischemia (“angina of the eye”) seen in OIS (3). The decision to perform a TAB was made after careful discussion of the risks and benefits with the patient. His age and his new symptoms, including headache, jaw claudication, proximal myalgias, and transient monocular visual disturbances, were highly concerning for GCA. For this reason, prednisone therapy had been initiated; however, his history of chronic diabetes with numerous existing diabetic complications meant he was at high risk for toxicity from long-term steroid use. Temporal artery biopsy was therefore felt to be an important investigation. Although he had OIS, his Doppler study indicated—falsely, in retrospect—that his carotid occlusive disease was unilateral (and not bilateral), and this was consistent also with his having previously undergone left carotid endarterectomy. As GCA is itself a known independent cause of OIS, we felt the TAB would not only be safe in this situation, but imperative (4). It is important to note that in situations of known bilateral carotid occlusive disease, the external carotid circulation should be carefully assessed before the biopsy is performed to rule out the situation in which the anterior cerebral circulation is supplied by the superficial temporal artery via retrograde flow through the ophthalmic artery or other collaterals. Although TAB is a mostly innocuous procedure, TAB in this rare situation can lead to devastating stroke or blindness. This highlights an important matter regarding the diagnostic accuracy of ultrasound. Carotid Doppler ultrasound has a sensitivity of 65% and a specificity of 95% in detecting surgically amenable lesions in the carotid arteries (5), whereas MRA offers a sensitivity and specificity of ≥82% and ≥95%, respectively (5). Although carotid Doppler ultrasound is a less invasive and expensive technique than MRA, it is limited by sonographer experience and patient factors (e.g., body habitus and anatomical configuration) (6). Furthermore, only the cervical component of the ICA can be assessed, as exemplified by our patient, whose very distal tandem occlusions were apparent only with MRA (6). STATEMENT OF AUTHORSHIP Category 1: a. Conception and design: J. A. Fraser and M. Gostimir; b. Acquisition of data: J. A. Fraser, B. Wehrli, and M. Gostimir; c. Analysis and interpretation of data: J. A. Fraser, B. Wehrli, and M. Gostimir. Category 2: a. Drafting the manuscript: J. A. Fraser and M. Gostimir; b. Revising it for intellectual content: J. A. Fraser, B. Wehrli, and M. Gostimir. Category 3: Final approval of the completed manuscript: J. A. Fraser, B. Wehrli, and M. Gostimir.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.