Purpose
To evaluate the first implementation of UWFI within the IHS DR ocular telehealth program in terms of DR identification and ungradable rates compared to previously used nonmydriatic multi-field fundus photography (NMFP).
Methods
Review of the IHS-Joslin Vision Network (JVN) program deployed in 72 sites across the 26 states serving American Indian and Alaska Native communities from May 27 to Nov. 25, 2014. Patients underwent JVN imaging either with NMFP (NWS6S, Topcon; stereoscopic pairs of 3-45 degree, 2-30 degree retinal fields, and 1 external image) or UWFI (Daytona, Optos, plc; 200 degree stereoscopic pair). Images were evaluated by JVN validated protocol on identical color calibrated LCD high resolution monitors by certified graders.
Results
A total of 7,460 subjects were imaged by NMFP and 456 were imaged by UWFI. Small statistically significant differences existed between groups in age (mean NMFP 54 vs 51, p<0.001) and gender (%female NMFP 57% vs 64%, p=0.004). No difference existed in baseline diabetes duration. The eye ungradable rate for DR and DME was lower with UWFI compared to NMFP (DR: 5% vs 31%, p<0.001; DME: 6% vs 30%, p<0.001). DR identification and referable DR (>moderate nonproliferative DR or DME) was increased with UWFI from 12% to 23% (p<0.001) and from 6% to 14% (p<0.001), respectively. In eyes with DR imaged with UWFI (N=206 eyes of 120 subjects), the distribution of peripheral retinal lesions outside ETDRS fields was evaluated. H/Ma, venous beading, IRMA and NVE were present in the periphery in 67%, 6%, 1% and 1%, respectively. Peripheral lesions suggested a more severe DR diagnosis in 20% of eyes.
Conclusions
In a standardized DR ocular telehealth program serving Native Americans across 26 states, nonmydriatic UWFI reduced ungradable image rate by 85% to less than 5%. In addition, there was a nearly 2-fold increase in DR identification and identification of peripheral lesions suggested a worse DR severity in 20%.These data suggest that UWFI may significantly improve DR identification and acquisition of gradable images in a geographically diverse population-based teleophthalmology DR programs.