Outcomes in Neovascular Age-Related Macular Degeneration when Neovascular Lesion Activity Is Uncertain: Observational Study

Purpose: To determine the characteristics and outcomes of eyes with neovascular age-related macular degeneration (nAMD) in which choroidal neovascular (CNV) lesion activity was graded as uncertain.

Design: Prospective, database, observational study.

Participants: Treatment-naïve patients with nAMD tracked by the Fight Retinal Blindness! (FRB!) registry initiating treatment between January 1, 2008, and January 1, 2014, with 2 years of follow up.

Methods: Lesion activity was determined by the clinical judgment of the providing physician on the basis of clinical examination with ancillary testing. Eye-based analysis investigated clinical characteristics and outcomes by proportion of visits with uncertain CNV lesion activity. Encounter-based analysis investigated CNV lesion grading over time with corresponding treatment and follow up.

Main outcome measures: Change in mean visual acuity (number of letters read on a logarithm of the minimum angle of resolution chart) at 2 years. Secondary outcomes included the number of visits and injections.

Results: We identified 1631 eyes of 1419 patients with 27 974 visits to 46 retinal specialists. The CNV lesion activity was uncertain at 4601 encounters (16.4%) and experienced by the majority of eyes (52%) and providers (72%). Uncertainty of CNV lesion activity did not significantly decline with increasing number of visits in individual patients (P = 0.97) but did decline from 32% in 2010 to 4% in 2015 (P < 0.001). Eyes having no visits with uncertain CNV lesion grading gained more letters (mean, 7.0) than eyes with moderate or high levels of uncertainty (mean, 3.98 and 3.37; P = 0.03 and 0.02, respectively). The mean probability of receiving an injection was higher in visits with active (0.89) compared with inactive (0.68, P < 0.001) and uncertain (0.69, P < 0.001) CNV lesion activity. Subsequent follow-up length was also decreased in visits with active (mean ± standard deviation [SD], -0.9±39 days) compared with inactive and uncertain CNV lesion activity (mean ± SD, 1.9±39 days, P < 0.001, and 2.4±39 days, P < 0.001, respectively). Conclusions: Rates of uncertainty regarding CNV lesion activity were high, and eyes with higher rates of uncertain CNV activity had worse outcomes. Further studies are warranted to establish whether eyes should be treated more aggressively when the activity status of the CNV lesion is uncertain.