Comparisons of handheld retinal imaging devices with spectral domain optical coherence tomography (SDOCT) in the identification of macular pathology

Purpose : To compare monoscopic macula centered images taken by mydriatic handheld retinal imaging with SDOCT for detection of macular pathology in diabetic patients.
Methods : Mydriatic macular images of 177 eyes of 92 diabetic patients were taken with 3 handheld retinal imaging devices [Aurora (AU), Smartscope (SS), RV700 (RV)] and compared with the Cirrus 6000 SDOCT taken during the same visit. Images were evaluated for the presence of diabetic macular edema (DME) on monoscopic fundus photographs adapted from Early Treatment Diabetic Retinopathy Study (ETDRS) definitions [no DME, noncenter-involved DME (non-ciDME) and center-involved DME (ciDME)]. Presence of DME on SDOCT used DRCR Retina Network Cirrus gender-based thresholds of central subfield thickness. Sensitivity and specificity were calculated for each device with the SDOCT as gold standard.
Results : Mean age was 56.6±10.8 years and 38% were male. Severity by ETDRS grading: No DR 40.1%, mild NPDR 19.2%, moderate 14.7%, severe 10.2%, proliferative 15.8%, ungradable for DR 0%; no DME 72.9%, non-ciDME 6.8%, ciDME 17.0%, ungradable for DME 3.4%. Gradable images by SDOCT (N=174, 98.3%) showed no DME in 84.7%, non-ciDME in 4.9%, and ciDME in 10.4%. Epiretinal membranes (ERM) were the second most common pathology, present in 11.9% of eyes. Ungradable rate for images (poor visualization in >50% of the macula), was AU:12.3%, SS:16.0%, RV:8.6%. Summary results are shown in table 1. For non-ciDME, sensitivity and specificity were similar across devices (0.71 – 0.78, 0.93 – 0.96) but sensitivity for ciDME was highest with AU(0.76). For nondiabetic macular pathology across all devices, sensitivity was highly variable (0.13 – 0.67) but highly specific (0.99 – 1.00). Sensitivity for ERM was lowest across all devices (0.13 – 0.46).
Conclusions : Compared to SDOCT, monoscopic handheld macular imaging attains high specificity but low sensitivity in identifying macular pathology. Without stereopsis, 22-29% of eyes without DME on monoscopic photos have DME on SDOCT, and 28-37% of eyes with DME on monoscopic handheld imaging will have no DME on SDOCT. Additionally 54-87% of eyes with macular ERM are missed without SDOCT imaging. This suggests the importance of SDOCT integration to improve detection of macular pathology, leading to appropriate referrals in large-scale DR screening programs.