Efficacy of bevacizumab for posttraumatic choroidal neovascularization.

Abstract

Choroidal neovascularization (CNV) can occur in 15–30% of patients with choroidal rupture, with consequent haemorrhage or serous macular detachment with concurrent central vision loss (Chanana et al. 2009). Currently, anti-vascular endothelial growth factor (anti-VEGF) therapy has been extensively used in various diseases with CNV, including age-related macular degeneration (AMD), myopia, angioid streaks, posterior uveitis and trauma (Chang et al. 2008). However, because of the rarity of posttraumatic CNV, only case reports documenting the use of anti-VEGF therapy for this condition are available (Artunay et al. 2009; Chanana et al. 2009; Liang et al. 2009; Prasad et al. 2009). We performed a multicentre study including 10 eyes of 10 patients who received intravitreal bevacizumab injections for CNV secondary to traumatic choroidal rupture. Changes in visual acuity, morphological changes, the number of injections required for CNV regression and recurrence rate were analysed. The mean age of patients was 41.2 ± 17.0 years, and the male to female ratio was 8:2 (Table 1). The time interval between trauma and active CNV development ranged from 7 weeks to 9 years; nine (90%) of the 10 patients developed CNV within 12 months after choroidal rupture (median: 5.5 months), while the remaining one (patient 9) developed CNV 9 years after the trauma. Choroidal neovascularization (CNV) involved the fovea in eight eyes (80%). The mean follow-up period after intravitreal bevacizumab injection was 14.45 months. Subjective vision, best-corrected visual acuity (BCVA) and CNV-associated retinal oedema exhibited an improvement after bevacizumab injection, with no adverse events. A mean improvement of 0.28 ±0.22 logMAR (14 letters) in BCVA and a reduction of 70.2 ± 42.7 μm in retinal thickness were observed at 1 month after the first injection (p = 0.0089 and 0.0018, respectively). At the final visit, the mean total improvement of 0.45 ± 0.30 logMAR (22.5 letters) was achieved (p = 0.0035) with partial restoration of the photoreceptor layer. In five eyes (50%), a single injection of bevacizumab effectively inactivated CNV, with the disappearance of leakage on fluorescein angiography and no fluid collection on optical coherence tomography. The remaining five patients (50%) received multiple injections. Patients 7 and 9 received repeated bevacizumab injections according to the protocol of the clinic, although intraretinal and/or subretinal fluid had already disappeared after the first injection. In Patients 6, 8 and 10, intraretinal fluid remained after the first injection, and complete absence of fluid was noted after 1–2 additional injections. The actual number of bevacizumab injections for an intraretinal and/or a subretinal fluid-free state was less than three in all patients, with an average of 1.4 per patient. In seven patients (70%), a fluid-free state was achieved by only one injection. Once CNV was inactivated by a single or multiple injections, no recurrence was observed in any patient during the observation period. In this study, one or a few bevacizumab injections provide favourable visual outcomes without recurrence in active posttraumatic CNV, similar to the 10 available previous case reports. Therefore, a single intravitreal injection of bevacizumab, followed by close monitoring of CNV activity, might be adequate for the initial treatment of active posttraumatic CNV. Interestingly, posttraumatic CNV did not recur after the initial anti-VEGF therapy in our case series and previous reports. However, frequent recurrence is known to occur in other types of CNV, including neovascular AMD, myopic CNV and idiopathic CNV. These clinical features suggest that posttraumatic CNV may have the different nature from other types of CNV.