Abstract
Purpose :
To test the feasibility of remote ophthalmic imaging technology for early identification of referable retinal pathology in the setting with relatively low disease prevalence.
Methods :
Prospective, non-randomized study on 635 patients (1270 eyes) conducted on diabetic patients in Duke primary care clinic. Adult patients with type 1 and 2 diabetes underwent retinal screening using color fundus (CFP) and optical coherence tomography (OCT) camera (MaestroCare, Topcon) on un-dilated pupils. Obtained images were graded by masked readers for interpretability and the presence of predetermined retinal pathology, with each eye graded independently. Retinal pathology was defined as referable to a retina specialist, requiring further intervention or follow-up, and incidental findings referable to the retina or comprehensive ophthalmology.
Results :
The mean age of screened patients was 66.1 (SD±13.5), the mean A1c 7.6 (SD±1.7), with an average disease duration of 5.9 years (demographics presented in Table 1). Remote image interpretability was significantly better by the OCT relative to CFP (97.9% vs. 83.5%, p<0.0001). We have identified 58 patients with different stages of Diabetic Retinopathy (Table 1 and Table 2 A), 114 patients with other retinal pathologies that required further retina evaluation, and 50 patients with incidental retinal findings that needed comprehensive ophthalmology follow-up (Table1 and the most common incidental findings are presented in Table 2 B). Considerable agreement in the final diagnosis was found between the gold standard (dilated exam by the Duke ophthalmology) and the screening outcome for a limited number of patients assessed at Duke. The screening significantly improved providers' HealthCare Effectiveness Data and Information Set (HEDIS, from 4 to 5 star; 74% to 84%).
Conclusions :
Ophthalmic imaging technologies at the point of service may harbor sufficient diagnostic information to enable accurate referral and timely treatment of retinal pathology. In return, this might lead to improved clinical and cost-effectiveness. Additionally, this approach might be attractive to primary care and endocrinology clinics, as it might significantly improve their quality performance (HEDIS) measures and bonus insurance reimbursements.
This is a 2021 ARVO Annual Meeting abstract.