Purchase this article with an account.
Sandra Halim, Amy A Kirkwood, Tunde Peto, Usha Chakravarthy, Jayashree Ramu, Philip G Hykin, Sobha Sivaprasad; Treatment response based on topography of retinal neovascularization in proliferative diabetic retinopathy. Invest. Ophthalmol. Vis. Sci. 2019;60(9):6555. doi: https://doi.org/.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To evaluate the spatial distribution of foci of neovascularization of the optic disc (NVD) and elsewhere in the retina (NVE) in the CLARITY study which compared aflibercept to panretinal photocoagulation (PRP) in proliferative diabetic retinopathy patients.
Color images of the fundus were graded for NVD and NVE at baseline, 12 and 52 weeks and classified by location into superior, inferior, temporal and nasal with reference to the fovea at the NetwORC of UK reading centers by trained graders. Statistical analyses were performed using multi-level logistic regression to assess the relationship between neovascularization location, treatment arm, and time and their interaction. Patient and location were included as levels in the model.
A total of 232 eyes were analyzed (116 per group). There were 88 NVD lesions and 355 NVE lesions, the majority being in the nasal quadrant. Baseline lesion numbers were similar in the aflibercept and PRP arms (Table 1).The total number of NVD and NVE lesions reduced in the aflibercept arm compared to PRP both at 12 weeks (OR 0.26, p<0.001) and 52 weeks (OR 0.27, p<0.001).On examining change in lesion frequency over time, NVD foci decreased significantly at 12 and 52 weeks with both treatments. There was a significant improvement from baseline in all NVE locations at 12 weeks for both treatments with maximum effect in the nasal and superior quadrants (Table 2). The reduction was persistent, though smaller, at 52 weeks in the aflibercept arm, but disappeared by week 52 in the PRP arm. However, there was no significant interaction between treatment and location (p=0.52), therefore the treatment effect (afliberept versus PRP) was similar at all locations.
Aflibercept appears to be a better treatment than PRP at both time points and in all disease locations. Both treatments had shorter-term positive effects in NVE locations with a reduced effect by week 52 for aflibercept and no improvement from baseline at 52 weeks in the PRP arm. This treatment effect for both arms was similar at all locations.
This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.
Table 1. Distribution of new vessel fronds at baseline by location in study eyes of patients enrolled in the CLARITY study.
Table 2. Odds ratios for change in NVD and NVE lesions from baseline to 12 and 52 weeks for each treatment arm.
This PDF is available to Subscribers Only