Abstract
Purpose:
This paper reports a case of an older white woman presenting with recalcitrant bilateral vitreoretinal inflammation that was ultimately proven to be primary vitreoretinal lymphoma by vitreous biopsy who subsequently developed a branch retinal artery occlusion (BRAO) following an intravitreal injection of methotrexate.
Methods:
Case summary.
Results:
The patient was treated with serial intravitreal methotrexate injections and subsequently developed a BRAO immediately following her seventh injection.
Conclusions:
A full systemic evaluation to rule out other causes of the BRAO was negative and given the timing of her symptoms after the injection it was determined that the BRAO was most likely realted to the intravitreal methotrexate injection.
Introduction
Intravitreal methotrexate (MTX) is effective in the treatment of primary vitreoretinal lymphoma. 1 Ophthalmic vaso-occlusive events have not been reported as a known complication of intravitreal MTX injection. We report a case of branch retinal artery occlusion (BRAO) following intravitreal MTX injection for the treatment of primary vitreoretinal lymphoma.
Case
A white woman presented with a 3-month history of recalcitrant panuveitis. She had been treated with topical steroid eye drops, oral prednisone, and periocular steroid injections, all of which had marginal or no effect. She presented for a second opinion.
On presentation, her vision was 20/30 in the right eye and 20/25 in the left eye. She had 3+ cell and 2+ haze without snowballs or snowbanks in the right eye, and rare vitreous cells in the left eye (Figure 1A). Her widefield fluorescein angiography demonstrated bilateral retinal vascular leakage (Figure 1B). B-scan ultrasonography did not demonstrate extraocular masses. Laboratory evaluation for sarcoidosis, syphilis, and tuberculosis, and other inflammatory conditions was negative. CT of the chest was normal.

Initial (A) ultra-widefield fundus photos and (B) fluorescein angiography demonstrating bilateral asymmetric inflammation.
Given the patient’s history of recalcitrant inflammation, there was concern for a malignant process. MRI of the brain and orbits was negative. Vitreous biopsy was performed; flow cytometry of the sample demonstrated monotypic B-cells expressing CD19, CD20, and monotypic lambda surface light-chain immunoglobulin, consistent with diffuse large B-cell lymphoma.
Given her isolated ophthalmic disease, she opted for intravitreal injections of MTX 0.1 mL into the right eye and observation of the minimally involved left eye. The patient’s inflammation improved after 7 injections of intravitreal MTX. Immediately following her seventh injection, she complained of blurry vision that progressed into a scotoma the following day. No anterior-chamber paracentesis was performed immediately following the injection; acute intraocular pressure spike was not suspected as the patient complained only of blurred vision rather than complete vision loss.
She presented 3 days later with an area of retinal whitening in the macula, and her fluorescein angiography showed a cilioretinal artery occlusion (Figure 2A); spectral-domain optical coherence tomography demonstrated inner retinal hyperreflectivity corresponding to the distribution of the occlusion, and swept-source optical coherence tomography angiography demonstrated flow abnormality of the cilioretinal artery (Figure 2B).

(A) Fluorescein angiography 3 days following intravitreal methotrexate administration demonstrating delayed perfusion of the cilioretinal artery; (B) optical coherence tomography demonstrating inner retinal thickening and increased reflectivity in the distribution of the cilioretinal artery; optical coherence tomography angiography demonstrating flow abnormality of the cilioretinal artery. Arrows designate the location of the branch retinal artery occlusion.
The patient was admitted for workup to rule out thrombotic or embolic phenomena; echocardiogram and carotid ultrasound were normal, and bloodwork revealed no evidence of coagulopathy.
Following evaluation and discussion with the patient’s oncologist, it was determined that her BRAO was unlikely to be related to her lymphoma, and was more likely linked to the intravitreal MTX.
The decision was made to stop intravitreal MTX and switch to orbital radiation. Radiation therapy of 3600 cGy in 20 fractions was administered over 4 weeks, which the patient tolerated well. Six months later, the patient’s right eye demonstrates atrophic changes related to her BRAO (Figure 2B), with final visual acuity of 20/40.
Discussion
Although other intravitreal agents, such as antivascular endothelial growth factor (anti-VEGF) medications, have been linked to retinal artery occlusion following injection, 2,3 arterial occlusive events have not been reported in association with intravitreal MTX. Retinal artery occlusion has additionally been reported in the setting of primary central nervous system lymphoma, 4 but in our patient the event seems less likely associated with her underlying disease given the timing of onset. Systemic MTX has also not been linked to retinal arterial occlusive events.
The cause of our patient’s BRAO is unclear. Retinal artery occlusion following anti-VEGF injection is theorized to be related to disruption of the protective effect of VEGF on the endothelium, 2 a theory that is not applicable in the setting of MTX injection. Given the almost immediate onset of symptoms following intravitreal injection in this patient, and the negative systemic workup for hypercoagulability, it is more feasible that the arterial occlusion was caused by direct compression as a result of sudden elevation of intraocular pressure, leading to reduced ocular perfusion pressure. Although this mechanism more commonly results in retinal vein occlusion, 5 it may be a feasible mechanism of artery occlusion in the setting of higher-volume intravitreal injections.
Footnotes
Ethical Approval
Cleveland Clinic Institutional Review Board approval was obtained under IRB #13-788: Medical Records Imaging in Uveitis Patients.
Statement of Informed Consent
Informed consent was not obtained as no identifiable patient information was used in the writing of the manuscript.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: An unrestricted grant award from Research to Prevent Blindness to the Department of Ophthalmology, Cole Eye Institute (RPB1508DM).
