Abstract
Purpose :
To estimate the frequencies of, and risk factors for, intra-lesional (ILMA) and extra-lesional macular atrophy (ELMA; synonymous with geographic atrophy) at the extended IVAN follow up visit.
Methods :
Following ethics approval, data from 528 of the 532 IVAN participants eligible for inclusion in the extended follow up were acquired either at a study visit or passively collected from routine follow up visits. Visual acuity (VA), clinical visits and injections given to both study and fellow eyes were obtained from the clinical records and available images from the most recent visit were submitted to the reading center. Images were assessed by trained graders who identified ILMA and ELMA and outlined the area occupied by atrophy. Trial allocations to drug and treatment frequency, injection rate during extended follow up, and otherILMA/ELMA risk factors that had been identified at IVAN exit (CNV type at IVAN entry, subretinal fluid and pigment epithelial detachment (PED) at IVAN exit) were included in the models. The primary analysis compared eyes which developed ILMA during extended follow up with those that did not. A secondary analysis concentrated on eyes in which ILMA developed or worsened during follow-up and on eyes without ILMA or with ILMA that had not worsened. Similar approaches were used to construct the models for ELMA.
Results :
The median length of follow up (FU) was 5.3 (inter-quartile range 4.8-5.8) years. ILMA developed in 197 of 319 eyes without ILMA at IVAN exit. In the primary analyses, older age (per decade, OR=1.57, 95% CI 1.08-2.28) and the absence of PED at the fovea (OR=2.45, 1.17-5.12) increased the risk of ILMA; injection rate reduced the risk of ILMA (OR=0.87, 0.78-0.98). The presence of ELMA did not significantly increase the risk of ILMA but the effect size was large (OR 3.46). Risk factors for ELMA were age, absence of PED at the fovea and ELMA present at IVAN exit; as with ILMA, injection rate tended to reduce the risk of ELMA but was not significant (OR=0.90, 0.81-1.01). The secondary analyses had very similar findings.
Conclusions :
We found no evidence to suggest that higher injections rates caused more ILMA or GA in study eye over the extended follow up. PED presence at the fovea was protective. At IVAN exit SRF and classic CNV were associated with a reduction in the risk ILMA; these associations were not significant at extended follow-up but remained in the same direction.
This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.