Abstract
Purpose :
Optical coherence tomography offers insights in the retinal structure and allows for noninvasive monitoring of retinal changes. Unfortunately many ophthalmologists do not have access to OCT technology and therefore depend on their clinical examination skills. We performed a prospective observational clinical study to investigate how clinical decision-making of specialists and residents is impacted by the use of OCT imaging when treating neovascular AMD.
Methods :
44 eyes from 38 patients were assesed for this study. All patients presented for re-evaluation 4-6 weeks after the loading dose of VEGF-Inhibitors (3 intravitreal injections every 4 weeks after diagnosis). The patients were examined twice: by a resident (with less than 5 years clinical experience in ophthalmology) and then by a specialist (with more than 5 years clinical experience in ophthalmology). The first evaluation was purely clinical, the second comprised OCT imaging. After each evaluation, each examiner independently filled out a form stating wheather further anti-VEGF treatment was recommended or not. The "true outcome" was defined as the decision of a specialist using clinical evaluation AND OCT. We calculated the true positive, true negative, false positive and false negative rate as deviation from this "true outcome". We used Fisher's exact test to evaluate differences between the different rates for different groups.
Results :
Specialists and residents without or with the aid of OCT did not differ significantly in their accuracy to decide on the correct treatment option (P = 0.705 and P = 1 respectively). However, both groups benefited from using the OCT to support their recommendations (P = 0.001237 and P = 0.000173 for specialists and residents respectively). The gain of accuracy was similar for both groups (P = 1 for difference).
Conclusions :
Both specialist and residents benefit from using OCT imaging to substantiate their recommendation on continuing or pausing treatment for neovascular AMD after the loading dose.
This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.