Exploring the promise of retinal prostheses

As many as 25 million people around the world suffer from retinal degenerative diseases such as macular degeneration and retinitis pigmentosa. These conditions destroy the retina's light-detecting photoreceptors, causing significant vision impairment.

Although photoreceptors are lost, retinal output cells are mostly preserved—a fact that retinal prosthetic devices are designed to capitalize on. This technology uses implanted electrodes to inject a charge into tissue to activate the retina's surviving neurons, such as retinal ganglion cells.

Researchers have conducted 10 clinical trials of various types of retinal implants. The trials differ in factors such as the number of patients, the number of electrodes and location of the implant. In a review of this research published in September 2015 in Clinical and Experimental Optometry (from Optometry Australia), the authors found that "prosthetic vision has progressed well beyond" its early, rudimentary stages of the 1960s. "Clinical evaluation in blind human subjects has demonstrated improvements in spatial localisation, object or character recognition, and motion detection and mobility." Still, the authors concluded, "there is much that needs to be done."

In fact, current devices provide only limited vision restoration, concurs David Lewerenz, O.D., associate professor and chief of Low Vision Services at Northeastern State University Oklahoma College of Optometry, and AOA Vision Rehabilitation Section chair. Patients perceive spots of light and high-contrast edges, but not more detailed images such as faces and landscapes.

"One of the challenges is that these devices work on the principle of the digital camera—a pixel-by-pixel representation of the world—whereas the retina works on a very different principle that's far more complex," Dr. Lewerenz says. "Retinal prostheses stimulate the retina in a way that doesn't really work very well with that system."

One device, the Argus II, has received Food and Drug Administration approval in the U.S. and is eligible for Medicare coverage. It's known as an epiretinal prosthesis. It uses electrodes placed between the retina and the vitreous cavity and targets surviving retinal ganglion cells. A video camera mounted on eyeglasses converts an image of a visual scene into instructions that are sent wirelessly to the prosthetic device. In one study, Argus II patients showed marked improvement in locating a door-like object or tracking a white line in a high-contrast setting.

Another type of device, called a subretinal prosthesis, uses light-detecting photodiodes placed between the outer retina and the retinal pigment epithelium. It stimulates the bipolar cells, rather than the ganglion cells. In one study, patients using a subretinal prosthesis could delineate white dining table objects against a black background.

The research review reveals that no device has been shown to be clearly superior. Each has advantages and disadvantages, says Bhavani Iyer, O.D., director of the Center for Visual Rehabilitation at the University of Texas Medical School in Houston, and AOA Vision Rehabilitation Section vice chair-elect.

"The jury is still out," she says.

Epiretinal implants, for example, can be adjusted by calibrating the external components rather than by further surgery. But if patients want to see something to their right, they have to turn their head to the right to take the image in. By comparison, "[Patients] wearing a subretinal could move their eye rather than their head," Dr. Lewerenz says, "which is a vision that is a lot more natural." Patients with subretinal implants, on the other hand, can risk retinal detachment and photoreceptor loss.

Goals for the future

Dr. Iyer hopes prosthetic developers and researchers will focus on improving the spatial resolution of the images that the patient receives, as well as ease of implantation.

"Detail vision is really what would make a huge difference in these patients," she adds. "The ultimate goal would be to give patients vision that is pretty close to what they would have if they didn't have the disease. To simulate that, we have quite some way to go."