Management and Outcomes for Neovascular Age-Related Macular Degeneration: Analysis of United States Electronic Health Records.

PURPOSE

To assess anti-vascular endothelial growth factor (VEGF) management patterns and anatomic and visual acuity (VA) outcomes among patients with neovascular age-related macular degeneration (nAMD) in United States clinical practice.

DESIGN

Retrospective observational cohort study.

PARTICIPANTS

Patients (N = 30 106) initiating intravitreal anti-VEGF treatment for nAMD between October 2009 and November 2016.

METHODS

Analysis of longitudinal electronic health records from USRetina.

MAIN OUTCOME MEASURES

Number of intravitreal injections, OCT examinations, and fluorescein angiography (FA) examinations per study eye during the first 12 months; corrected VA and central retinal thickness (CRT) at 12 months; and number of ophthalmologist visits, stratified by index anti-VEGF agent.

RESULTS

Over the first 12 months, patients made a mean of 8.1 (range, 1-39) ophthalmologist visits, received a mean of 6.0 (range, 1-27) anti-VEGF injections, and underwent 7.2 OCT and 5.3 FA examinations per study eye. For eyes with paired baseline and 12-month readings, mean CRT declined from 320 to 271 μm (mean change, -48 μm), and mean VA increased from 60.3 to 61.0 approximate Early Treatment Diabetic Retinopathy Study (ETDRS) letters (mean change, +0.6 letters). Twelve months after initiating index treatment with bevacizumab, ranibizumab, and aflibercept, 19.3%, 15.8%, and 15.5% of eyes, respectively, showed greater than 10-letter gain, whereas 13.2%, 14.7%, and 14.4% of eyes, respectively, showed greater than 10-letter loss. Mean change from baseline VA at 12 months increased linearly with cumulative anti-VEGF injection count: +1.79 versus -0.95 approximate ETDRS letters for eyes receiving 7 or more injections versus fewer than 7 injections. Similarly, the magnitude of the reduction from baseline CRT at 12 months tended to increase linearly with increasing number of anti-VEGF injections. Multivariate linear regression analysis, adjusted for covariates, indicated a significant association between cumulative number of anti-VEGF injections and change from baseline in VA at 12 months, with each unit increase producing an estimated gain of 0.37 approximate ETDRS letters.

CONCLUSIONS

This analysis of combined morphologic and functional outcomes of anti-VEGF therapy, the largest conducted to date in nAMD, identified relatively low anti-VEGF injection frequencies, coupled with moderate anatomic and limited VA improvements, in United States clinical practice.