Abstract
Purpose :
European studies of anti-VEGF therapy for neovascular AMD (NAMD) have shown that visual outcomes in the "real world" do not achieve those shown in randomized clinical trials (RCT). While the cause of this discrepancy is unknown, patient characteristics and under treatment associated with discontinuous treatment regimens are implicated. We explored this issue by analyzing a large database composed of aggregated, longitudinal electronic medical records (EMR) from a demographically diverse sample of US retina specialists.
Methods :
The HIPAA-compliant Vestrum Health Retina Database was utilized. Inclusion required at least 3 monthly anti-VEGF injections for naive NAMD patients in the first 4 months between January 2011 and July 2013 with availability of 6 to 24 months of follow-up data, ending July 2015 (n=2,213). The eyes were divided into 3 cohorts: those with records including visual acuity (VA) measurements up to and including 6-months but not beyond (“6-month cohort”, n=97), up to 12-months but not beyond ("12-month cohort”, n=195), and those up to and including 24 months (“24-month cohort”, n=1,921), with each cohort being mutually exclusive of one another. VA outcomes were assessed on each cohort as a whole, and also stratified by baseline VA.
Results :
Best VA outcomes were evident in the 24-month cohort, followed by the 12-month and finally the 6-month cohorts. When stratified by baseline VA of 20/200 or worse, 20/70-20/200, 20/40-20/70, and 20/40 or better, the final mean change in number of letters gained or lost in the 24-month cohort was +19.9, +2.6, -1.2, and -5.2 letters respectively, in the 12-month cohort +8.9, -1.3, -5.6, and -4.5 letters respectively, and in the 6-month cohort +8.8, -4.6, -1.9, and -5.2 letters respectively.
Conclusions :
This analysis corroborates prior observations that visual outcomes following anti-VEGF therapy for NAMD in the “real world” do not achieve those seen in RCT. Eyes with better baseline VA are disproportionately affected, with loss of vision observed in eyes with baseline VA of 20/70 or better. Duration of treatment and/or follow-up appear to correlate with best visual outcome. This observation highlights the need for more intensive therapy and treatment compliance in NAMD, especially those patients with better baseline VA.
This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.