MRI of Diabetic Retinopathy

Diabetic retinopathy (DR) is the leading cause of blindness in working-age adults and occurs in about 45% of diabetic patients. DR is often diagnosed by late manifestations of the disease, such as blurred vision and abnormal appearance of the fundus. By the time a clinical diagnosis is made, vision loss or blindness is often the inevitable outcome. Results from clinical trials have documented that strict glycemic control in the early stages of diabetes is effective in reducing the risk of DR. Thus, early identification of diabetic individuals who are at risk of DR will enable timely intervention to minimize the risk of vision loss. The retina consists of three major cell layers (photoreceptor, bipolar, and ganglion) and two vascular (retinal and choroidal) layers bounding the retina, each of which is affected differently by DR. Our laboratory has pioneered innovative layer-specific blood-flow and functional MRI techniques based on echo-planar imaging at ~80x80x1000 µm. Echo-planar imaging of the eye suffers from susceptibility artifacts and image blurring, which precludes the earliest and most reliable DR detection. Moreover, although fluorescein angiography is widely used for visualization of vessels and diagnosis of DR, it does not have layer resolution and fluorescein (an optical contrast agent) can cause severe negative reactions, especially in diabetics who often have compromised renal function. Alternative approaches that can provide vascular layer resolution without needing a contrast agent will also have important clinical significance. We have recently explored non-echo-planar imaging approaches to study the retina. Our preliminary data show that balanced steady-state free precession (bSSFP) MRI provides markedly improved image quality and spatial resolution for blood-flow and functional MRI, and that MR angiography (MRA) provides layer-specific retinal and choroid vessels in the rat retina without using a contrast agent. In this proposal, we aim: i) to further develop bSSFP and MRA methods to study the rat retina with laminar specificity without using a contrast agent, ii) to evaluate these approaches to detect the earliest changes in retina of streptozotocin-induced diabetes, and iii) to test the hypothesis that strict glycemic control reverses these very early pre-vision loss manifestations of retinal changes in diabetes. Our central hypothesis is that abnormal retinal and choroidal blood flow and neurovascular coupling in the early pre-clinical stage of retinal changes in diabetes are detectable by these novel MRI methods prior to anatomic changes, and are reversed by strict glycemic control.