Diabetic macular edema (DME) is the most common cause of vision loss in diabetic patients. While anti-VEGF treatments and to a lesser extent corticosteroid and macular photocoagulation have improved outcomes in patients with DME, no single therapy is universally effective and currently there is no a priori means of determine which patients will respond best to any given therapy. The purpose of this study is to determine whether specific parameters of ocular imaging studies including optical coherence tomography and fluorescein angiography can predict response to treatment in patients with DME. This is a prospective observational cohort study that will collect clinical data and imaging studies obtained as standard of care. Up to 150 subjects with clinically significant DME will be enrolled at Duke Eye Center or its satellite offices. These imaging studies will be analyzed to determine whether specific parameters are associated with poor or favorable response to specific treatments. There will be no intervention as part of this observational trial, thus the primary risk to subjects is loss of confidentiality, which will be minimized by the study team.

The investigators hypothesize that DME is a common endpoint caused by differing pathobiology which varies between patients. Clinically fluorescein angiography (FA) and optical coherence tomography (OCT) are used to diagnose and assess diabetic retinopathy including DME. These two imaging modalities provide different biological information with FA giving functional data regarding the perfusion and integrity of the retinal water homeostasis system where OCT gives structural information including the quantity and location of fluid in patients with DME. The investigators believe that careful analysis of both the leakage pattern on FA and the characteristics of intraretinal fluid on OCT have potential to provide insight into the predominant mechanism of DME in an individual patient. Specifically, it has been proposed that at least two forms of DME exist, focal and diffuse. In focal DME, leakage seen on FA originates predominantly from leaking microaneurysms present in the retinal microvasculature, which are markers for endothelial dysfunction and focal breakdown of the blood retinal barrier. Similarly, it has been hypothesized that focal leakage imaged by OCT will result in accumulation of noncystic fluid in the extracellular space. In contrast, diffuse pattern leakage has no discernable source on FA and it is believed that this pattern represents failure of the Müller and RPE cell pump function resulting in accumulation of intracellular fluid in the retina. Diffuse leakage imaged with OCT may appear as cystic fluid accumulation which represents swollen Müller or other retinal cells.

In clinical practice, most patients with DME have a mixture of focal and diffuse leakage with one type being predominant. The investigators hypothesize that, because they are driven by disparate pathobiology, different DME subtypes will respond differently to treatment. Thus, it may be possible to use fluorescein angiography and/or optical coherence tomography to predict the optimal treatment for an individual patient, thereby improving patient outcomes and possibly reducing treatment burden. To date, there are no prospective studies correlating FA and OCT imaging parameters with response to specific therapies, nor is there prospective data on using imaging parameters to guide choice of treatment modality in subjects with DME. As a first step toward determining whether imaging parameters predict treatment response, the investigators will prospectively collect imaging, treatment and outcome data in patients with diabetic macular edema treated at Duke Eye Center.

Inclusion Criteria:

• Able to provide written informed consent
• Diagnosis of DME in one or both eyes which is visually significant in the opinion of the clinician.

Exclusion Criteria:

• Macular edema secondary to causes other than diabetes
• Known or suspected sensitivity or allergy to fluorescein dye
• Significant media opacity (e.g. cataract or vitreous hemorrhage)
• Prior history of vetrectomy surgery

Patients with a diagnosis of DME in one or both eyes which is visually significant in the opinion of the clinician/investigator.