Purchase this article with an account.
David Rivera, Aida Jimenez-Corona, Sergio Hernandez-Jimenez, Paloma Almeda-Valdes, Liliana Perez-Peralta, Veronica Yazmin Zurita-Cortes, Sandra Evelin Cardenas-Gonzalez, Enrique O Graue-Hernandez; Non-mydriatic camera vs OCT for the detection of diabetic macular edema. Invest. Ophthalmol. Vis. Sci. 2018;59(9):6155.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
OCT has now become a more useful tool in assessing diabetic macular edema (DME) since it can measure and map the whole macula quantitatively. It has been described that OCT can detect DME better than clinical fundus examination by an ophthalmologist. Majority of programs use fundus photography for screening, most of them do not use OCT as part of the screening process.The objective of this study was to compare the diagnostic efficacy of fundus photography compared with OCT the reference study for the diagnosis of diabetic macular edema.
We included patients with diabetes, non-mydriatic camera (NMC) and OCT were acquired and interpreted by certified readers according to the Scottish diabetic retinopathy grading scale. We excluded patients with non-readable images and those with other pathology that could mislead the diagnosis.For each NMC vs OCT comparison, the parameters studied were the following: Sensitivity, Specificity, positive predictive value (PPV), negative predictive value (NPV) and area under the receiver operating characteristic curve (AUROC). The diagnostic performance of NMC was then stratified by diabetes duration, HbA1c and glomerular filtration rate (GFR).
Three hundred and fifteen eyes were included, mean age was 55.1 years, 56.8% were female, mean diabetes duration was 13.9 years and mean HbA1c was 8.5%. Systemic hypertension was present in 56%.The results of Sensitivity, specificity, PPV, NPV and AUC-ROC for NMC vs. OCT for any DME were 87.6%, 81.9%, 49.1%, 97.1%, 0.85.Results were stratified by diabetes duration, HbA1c and glomerular filtration rate (Table 1).For any DME, value of AUC-ROC for the group of <10 years of DM duration was 0.93 while for >15 years it was 0.81. For any DME, value of AUC-ROC for the group of <7%HbA1c was 0.94 while for ≥9% HbA1c it was 0.78. For any DME, value of AUC-ROC for the group of ≥60 GFR was 0.83 while for <60 GFR it was 0.90
The non-mydriatic camera is sufficiently sensitive and specific for detecting DME, but 13% of patients with DME will be missed when using NMC, however, in the stratified analysis, the diagnostic performance of NMC decreases for patients with a larger time of diagnosis, higher HbA1c levels, and higher GFR. In patients with these characteristics, a high suspicion of DME should be maintained since sensitivity drop to 81%.
This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.
This PDF is available to Subscribers Only