Abstract
Purpose :
Severity of DR, measured by the ETDRS DR severity scale (DRSS), is a validated predictor of risk of vision-threatening complications (e.g., Davis et al, 1998). Retinal function (RF), as measured by the ERG, has also been shown to decrease with DR severity (e.g., Bresnick et al, 1984). In this observational trial, we evaluated the impact of adding baseline (BL) measures of retinal function to DRSS on the prediction of surgical or medical intervention.
Methods :
468 diabetic patients were enrolled between 2013 and 2014 at 2 centers. Patients previously treated for macular edema or proliferative DR and those with other retinopathy were excluded. At BL, RETeval flicker electroretinograms (ERG) and pupillary light reflex data were obtained and combined into a DR score of RF validated to detect vision threatening DR (VTDR DRSS ≥ Level 53 or clinically significant macular edema) (Maa et al, 2016). 7-field color fundus photographs were obtained at BL and graded using DRSS. Charts from 268/279 patients at one site were reviewed for visual interventions related to VTDR (vitrectomy, anti-VEGF injection, pan-retinal photocoagulation, focal laser). Incidence of events within 1, 2, and 3 years of BL were summarized using Kaplan-Meier analysis.
Results :
Patients had a 6%, 10%, 16% probability of having an ocular intervention in the subsequent 1, 2, or 3 years. For patients with VTDR, the probability of an intervention was 11%, 19%, and 30%, whereas for patients without VTDR, the probability was 3%, 5%, and 6% (p<0.001). If RF was abnormal in patients with VTDR there was a 24%, 44%, 53% probability of an intervention in the subsequent 1, 2, or 3 years, whereas normal RF reduced the risk to 2%, 2%, and 14% (p < 0.001). For patients with no, mild or moderate DR the risk of intervention with normal RF was 3%, 4% and 6% at 1, 2 and 3 years but was increased to 12%, 20% and 20% in patients with an abnormal RF (p = 0.01).
Conclusions :
Results suggest that retinal function adds prognostic value to structural information obtained from fundus photography. The ERG procedure does not require dark adaptation and is completed in <5 minutes. Adding this measure to funduscopic appearance can help in the management of patients with diabetic retinopathy.
This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.