Compassionate use of dexamethasone implant for the treatment of macular edema secondary to central retinal vein occlusion in a clinical setting

Abstract

Editor, The GENEVA trial has demonstrated the beneficial effects of a sustained-release, dexamethasone delivery system for macular oedema (ME) secondary to retinal vein occlusion (Haller et al. 2010). We report the results in patients with ME related to central retinal vein occlusion (CRVO) treated with dexamethasone implant in a clinical setting. An open-label, compassionate use program case series was approved by the local IRB. Written informed consent was obtained from all patients. Patients affected by CRVO-related ME were prospectively enrolled from January 2008. Inclusion criteria were as follows: nonischemic CRVO, ME with central foveal thickness (CFT) ≥250 μm on optical coherence tomography (OCT), best-corrected visual acuity (BCVA) between 1.30 and 0.30 logMAR (approximately 20/400–20/40 Snellen equivalent). Exclusion criteria included any other ocular disorder, ocular hypertension or glaucoma, media opacities, and signs of ocular infection. A 700 μg dose of dexamethasone was implanted in the operating theater. The protocol included a monthly complete ocular examination over a 12-month follow-up, with retreatments carried out on a pro-re-nata (PRN) basis starting from the fourth month. Retreatment criteria: documented positive response after the first treatment (BCVA improvement of at least five letters, CFT reduction by at least 20%), at least five letter-BCVA deterioration, associated with CFT >250 μm. Best-corrected visual acuity measurement and OCT were performed by an ophthalmologist masked to the patient’s condition. Primary outcome was the BCVA change at the end of the follow-up. Overall, 10 patients (six men) were enrolled, (mean age of 61.8 ± 6.2, mean CRVO duration of 29.6 ± 11 months and mean ME duration of 26.4 ± 9.3 months). Mean BCVA was 1.10 ± 0.26 logMAR at baseline, changing to 0.79 ± 0.30 logMAR at the end of the follow-up, with a statistically significant difference with respect to the baseline value from the third month on (Fig. 1). The mean number of Early Treatment Diabetic Retinopathy Study (ETDRS) lines gained was 3.10 ± 1.45 (9 patients gaining at least three lines at the end of the follow-up). Mean CFT passed from 620 to 273 μm at the 12-month examination, with a statistically significant difference. The mean number of treatments was 1.7 ± 0.4 (range: 1–2). Two patients developed a nuclear cataract, while one patient experienced increased intraocular pressure, which was effectively controlled with topical antiglaucoma therapy. Change in the mean best-corrected visual acuity (BCVA) (LogMAR) and in central foveal thickness (CFT, μm) during the 12 months of follow-up. With regard to the BCVA, a statistically significant difference in comparison with the baseline value was registered from the 3-month examination up to the end of the scheduled follow-up. Considering the CFT, a statistically significant improvement in the mean CFT in comparison with the baseline value was registered at 1-month examination and following in each time-point until the end of the scheduled follow-up. The most appropriate management of ME related to CRVO is a matter of debate. Intravitreal anti-VEGF drugs have yielded promising results, as has intravitreal injection of triamcinolone acetonide (Ferrara et al. 2007; Ip et al. 2009, Brown et al. 2010). The GENEVA trial has demonstrated that dexamethasone implant can both reduce the risk of vision loss and improve the incidence of visual improvement (Haller et al. 2010). Nevertheless, the effect in the CRVO subgroup turned out to be partially unsatisfactory. Indeed, BCVA improvement was statistically significant for both 350 and 700 μg subgroups at the second month and for the 350 μg subgroup at the third month too, losing the statistically significant difference afterwards. A possible explanation may emerge from the study design, which provides for retreatment only after the sixth month. The design of this compassionate program study provided for retreatment from the fourth month on a PRN basis. Our results show that a sustained effect can be obtained for both functional and anatomical aspects. It is worth remarking that our patients displayed older CRVO forms, with longer ME duration in comparison with the GENEVA trial (13 versus 5 months, respectively), lower mean BCVA at baseline (20/320, versus 20/80, respectively) and higher CFT (620 versus 550 μm, respectively) (Haller et al. 2010). Although obvious limitations include small number of patients and absence of a control group, this study illustrates, with all due caution, the positive effect of dexamethasone implant for ME secondary to CRVO in a clinical setting, demonstrating a sustained effect if a PRN basis is followed. Further studies are needed to ascertain the best therapeutic approach and the most appropriate timing of the treatment.