Etiology, Treatment Patterns, and Outcomes for Choroidal Neovascularization in the Pediatric Population: An Intelligent Research in Sight (IRIS®) Registry Study.

OBJECTIVE Choroidal neovascularization (CNV) is a rare, but devastating, cause of vision loss in children, with most current publications limited to small case series. Using a large clinical registry allowed us to understand the most common causes of this disease and the visual outcomes. DESIGN Retrospective analysis. PARTICIPANTS Patients younger than 18 years in the Intelligent Research in Sight Registry diagnosed with CNV between 2013 and 2019. METHODS Cases were identified based on International Classification of Diseases, Ninth and Tenth Revisions, diagnosis codes for CNV or CNV-related etiology and Current Procedural Terminology treatment codes. MAIN OUTCOME MEASURES Etiology of CNV, treatment patterns, and visual outcomes. RESULTS Two thousand three hundred fifty-three eyes with pediatric CNV were identified. The most common identifiable causes of pediatric CNV were posterior uveitis or inflammatory chorioretinal disease (19.4%), myopia (18.4%), hereditary dystrophy (5.4%), chorioretinal scar (4.2%), choroidal rupture (3.5%), optic nerve drusen (3.2%), osteoma (1.9%), and solar retinopathy (0.2%). In 38.2% of eyes, CNV was idiopathic, and in 5.7% of eyes, multiple causes were coded. One thousand forty-one eyes (44.4%) underwent treatment. The mean age of mean age of patients whose eyes received treatment 13.6 ± 3.5 years compared with 12.4 ± 4.1 years for the untreated group (P < 0.001). In 88.9% of eyes, anti-vascular endothelial growth factor (VEGF) injections were administered, 7.9% of eyes received laser therapy, 0.3% of eyes received photodynamic therapy, and 2.9% of eyes received combination therapy. In the eyes receiving anti-VEGF agents, 68.4% required 3 injections or fewer (P < 0.0001). Eyes undergoing treatment exhibited worse baseline visual acuity (VA) than eyes that did not undergo treatment (0.62 ± 0.50 logarithm of the minimum angle of resolution [logMAR] vs. 0.44 ± 0.50 logMAR; P < 0.0001). Visual acuity in the treatment group improved significantly from 0.62 ± 0.50 logMAR at baseline to 0.39 ± 0.43 logMAR at year 1 (P < 0.0001). Visual acuity in the untreated group improved significantly from 0.44 ± 0.50 logMAR at baseline to 0.34 ± 0.44 logMAR at year 1 (P < 0.001). Treated eyes showed a statistically significant higher odds of exhibiting a 2-line vision improvement or better compared with the untreated group at 12 months regardless of treatment type and after controlling for baseline VA (odds ratio, 2.4; P < 0.0001). CONCLUSIONS CNV is a rare, sight-threatening condition in children, with the most common causes being idiopathic, inflammatory chorioretinal disease, and myopia. Eyes undergoing treatment tended to be in older patients and showed worse baseline VA compared with eyes that did not undergo treatment. Those that were treated experienced significant improvement in vision that was maintained in the long term.