Two-Year Risk of Exudation in Eyes with Non-Exudative, Neovascular AMD Detected with Swept Source OCTA

Written by: Brian K. Do, MD

August 2019

Yang J, Zhang Q, Motulsky EH, Thulliez M, Shi Y, Lyu C, de Sisternes L, Durbin MK, Feuer W, Wang RK, Gregori G, Rosenfeld PJ. Two-Year Risk of Exudation in Eyes with Non-Exudative AMD and Subclinical Neovascularization Detected with Swept Source OCT Angiography. American Journal of Ophthalmology (2019). Doi:

While ICG has historically been demonstrated to be able to identify macular neovascularization (MNV) in eyes with AMD in the absence of exudation, we know now that OCT angiography (OCTA) is as good, if not better than traditional dye-based angiography at detecting these neovascular membranes. Especially in cases of AMD with significant sub-retinal hyperreflective material (SHRM), OCTA is valuable in determining whether or not neovascular disease has developed. However, until recently, the clinical implications of being able to reveal MNV in non-exudative AMD have remained somewhat unclear.

In this particular prospective cohort study, Yang et al reported on the prevalence and incidence of MNV and exudation in eyes with initially non-exudative AMD, in individuals in whom the fellow eye had been diagnosed with exudative AMD. Eyes with non-exudative AMD were classified as either intermediate AMD (iAMD), or late AMD (geographic atrophy). A total of 227 patients who met the criteria were enrolled over the course of a nearly four-year period: 154 eyes with iAMD (61% women), 73 eyes with late AMD (60.3% women).

Subclinical neovascularization was seen in 30 of 227 eyes with non-exudative AMD at the time of initial examination (13.2% of eyes): 19 eyes with iAMD (12.3%) and 11 eyes with late AMD (15.1%). All of these lesions were classified as type 1 MNV. Of these 227 eyes, 191 had follow-up visits. (127 iAMD, 64 late AMD) Follow-up ranged from 1 to 47 months; 16 of the 30 eyes diagnosed with MNV at baseline did not develop exudation during the study (although, as previously noted, follow-up was highly variable).

19 of 191 eyes with any type of follow-up ultimately developed exudation – 11.4% – at two years. 13 of these 19 eyes had subclinical type 1 MNV before exudation; 1 had subclinical type 3 MNV; 5 had no prior diagnoses of MNV before exudation. The relative risk of exudation after detection of subclinical MNV at two years of follow-up was 13.6 times (95% CI, 4.9 – 37.7) greater than in the absence of subclinical MNV. Additionally, it was determined via Kaplan-Meier analyses amongst the patients with multiple visits that the incidence of exudation (not surprisingly) was higher in eyes with subclinical MNV at the initial visit (34.5%) than in those without (6.3%).

While the prevalence rate of subclinical MNV reported in this study (13.2%) was fairly high, the overall likelihood of exudation in eyes with non-exudative AMD included was quite low, and there appeared not to be a difference between eyes with iAMD and late AMD. (the study may not have been adequately powered to detect such a difference) The authors did not recommend treatment of these subclinical lesions, largely because of the small percentage of cases in which exudation occurs. In this series, it appeared as though visual acuity was preserved as long as the affected maculae remained fluid free; further they theorize that elimination of the MNV by anti-VEGF therapy might accelerate macular atrophy (since these non-exudative lesions MAY represent the body’s attempt at recapitulation of the choriocapillaris closer to the RPE). Close follow-up is recommended in these patients, and treatment warranted once exudation is detected.