Intravitreal Triamcinolone for Radiation-Induced Macular Edema

Despite a completely different origin, radiation-induced retinopathy exhibits features similar to diabetic retinopathy. Microvascular damage leads to retinal nonperfusion, neovascularization, and macular edema. While neovascular glaucoma usually can be prevented by panretinal laser coagulation, focal or grid laser to reduce macular edema and improve vision has limited success.¹

A 64-year-old man was seen in the medical retina clinic of the Sydney Eye Hospital (Sydney, Australia) for evaluation and treatment of radiation-induced retinopathy in his better eye (left eye). The vision in his right eye was impaired owing to traumatic optic neuropathy. A left parotid gland carcinoma had been treated with total parotidectomy and external beam irradiation (5400 rad [54 Gy] in 30 fractions [5 fractions per week]) 6 years previously, resulting in a left facial nerve palsy with lagophthalmos, cataract, and radiation-induced retinopathy.

In July 2001, the visual acuity was 20/70 OD and 20/50 OS. There was a left facial nerve palsy with lagophthalmos, punctuate keratopathy, and pseudophakia. Funduscopy showed moderate disc pallor, abnormal shunt vessels on the temporal aspect of the disc, and diffuse, cystoid macular edema. The patient was treated with macular grid laser (argon green, 21 burns, 0.1 seconds, 110 m W, 100 µm). Seven months later, the best-corrected visual acuity was 20/50 OS (Early Treatment Diabetic Retinopathy Study [ETDRS] vision score, 68; ETDRS Log MAR chart at 4 m, 38 letters). Funduscopy showed some microaneurysms, persistent retinal thickening, and exudates in the temporal macula. Fundus fluorescein angiography revealed areas of capillary nonperfusion in the lower and temporal midperiphery, without signs of neovascularization, as well as prominent leakage from the microaneurysms and dilated capillaries temporal to the fovea. Late-phase studies revealed significant edema of the central fovea (Figure 1A and B).

In view of encouraging, albeit anecdotal, reports of its efficacy for chronic diabetic cystoid macular edema, treatment with intravitreal triamcinolone acetonide was considered. After extensive discussion with the patient and informed consent was obtained, and 0.1 mL (4 mg) of triamcinolone acetate was injected intravitreally through the pars plana using a 27-gauge needle. The injection was performed in a minor-procedure room with topical and subconjunctival anesthesia after the eye had been prepared with half-strength betadine, one drop of 0.5% timolol, and digital massage.

Three months after the injection, the best-corrected visual acuity had improved from 20/50 to 20/25 OS (ETDRS vision score, 68 to 79; ETDRS Log MARchart at 4 m, 38 to 49 letters). Funduscopy showed some residual retinal thickening in the temporal macula as well as a deranged capillary bed in this area. Fundus fluorescein angiography demonstrated the same areas of capillary nonperfusion and altered capillaries in the temporal macula. However, there was significantly less leakage from these diseased vessels and only minimal edema temporal to the fovea (Figure 1C and D).

Nine months after the intravitreal injection of triamcinolone, there was rapid decline of visual acuity in the left eye. The best-corrected visual acuity was 20/80 (ETDRS vision score, 55; ETDRS Log MAR chart at 4 m, 25 letters). Contact lens examination showed a relapse of the edema in the temporal macula, extending into the central fovea. Optical coherence tomography, which was not available at the time of the first treatment, showed a circumscribed area of retinal edema in the temporal macula involving the central fovea (Figure 2A), corresponding to the area of leakage demonstrated on fluorescein angiography at baseline (Figure 1B).

Treatment with intravitreal triamcinolone acetonide (0.1 mL/4 mg) was repeated. One month later, best-corrected visual acuity had improved again to 20/40 (ETDRS vision score, 70; ETDRS Log MAR chart at 4 m, 40 letters). Optical coherence tomography showed a marked regression of the retinal edema(Figure 2B). Three months after retreatment, the visual acuity was 20/25 (ETDRS vision score, 78; ETDRS Log MARchart at 4 m, 48 letters), and there was complete resolution of the macular edema on optical coherence tomography (Figure 2C). The intraocular pressure remained within normal limits during follow-up.

Intravitreal triamcinolone has been shown to be effective for ocular neovascularization in animal models and for uveitic macular edema in humans.² There is some evidence that this treatment might be effective for macular edema in diabetic retinopathy³ and central retinal vein occlusion⁴ as well. Our patient with radiation-induced retinopathy showed significant improvement in visual acuity and reduction of vascular leakage after focal laser had failed to improve the situation.

The mechanism of this effect of intraocular steroid treatment is poorly understood, but it appears to restore a compromised inner blood-retina barrier.⁵ Our case illustrates that the beneficial effect of intravitreal triamcinolone on vasogenic macular edema may be independent of the initial insult. As this is only a case report with limited follow-up, more research is needed to establish if there is a beneficial long-term effect and if there are any unique safety concerns in cases of radiation-induced retinopathy, such as an increased susceptibility to the development or reactivation of intraocular malignancies. Repeated injections seem to be needed to maintain the initial effect. Furthermore, it remains unclear whether this treatment will alter the course of the ischemic and neovascular processes associated with radiation-induced retinopathy.

The authors have no relevant financial interest in this article.