Retinal implants: can electronics give back some sight?

A retinal prosthesis, often called a bionic eye, is an electronic implant that stimulates the retina, the light-sensitive layer at the back of the eye. It is designed for people who have lost the cells that detect light but still have some working cells behind them. In one design a camera image is turned into electrical signals and passed to the surviving cells, so the brain receives a rough picture. Losing these cells can make it hard to find objects or move around, and a device that gives a coarse picture has been tested for that purpose. The results so far are real but modest, and they come from very small studies.

This article is general information, not medical advice. If you have a sight condition, speak to an eye specialist or your GP about your options.

What has been shown in trials

Argus II. The Argus II, a device with 60 microelectrodes, was approved for use in Europe in 2011, and in 2013 it was granted US Food and Drug Administration approval after Humanitarian Device Exemption status (AAO EyeNet, December 2019). The FDA announcement of that approval is dated 19 February 2013 (FDA, reprinted by ScienceDaily). The FDA says it approved the system as a humanitarian use device, a pathway limited to devices that treat or diagnose fewer than 4,000 people in the United States each year. The intended users were adults aged 25 or older with severe to profound retinitis pigmentosa, an inherited condition, who have bare light perception or no light perception in both eyes, evidence of intact inner retina function and a previous history of being able to see forms. The FDA was clear about the limits: the system "will not restore vision to patients", though it may let them detect light and dark and so help them identify the location or movement of objects or people.

The interim international trial was a single-arm, prospective, multicentre study with no comparison group. It enrolled 30 people in the United States and Europe between 6 June 2007 and 11 August 2009, with follow-up of at least six months and up to 2.7 years. With the system on rather than off, subjects did statistically better at object localisation (96 per cent of subjects), motion discrimination (57 per cent) and discrimination of oriented gratings (23 per cent). The best visual acuity recorded was 20/1260. The most common serious adverse event was conjunctival erosion or dehiscence over the part of the implant outside the eye; it was treated successfully in all but one subject, whose device was removed (Humayun and colleagues, Ophthalmology, 2012).

Longer follow-up brought more problems. The AAO summary reports that two of the 30 Argus II implants failed about four years after implantation, when the telemetry link between the antennae broke down. In a separate study of 20 eyes, 50 per cent developed a fibrosis-like (scar-like) tissue between the electrode array and the retinal surface, and in most of these it advanced to retinoschisis, a split in the retina, although no change in visual function was noted. The same summary says more than 350 patients worldwide had been implanted with the Argus II by then (AAO EyeNet, December 2019).

A subretinal implant in inherited disease (2013). Stingl and colleagues reported a wirelessly powered chip placed under the retina of nine people who were blind from untreatable hereditary retinal degeneration. The chip carried 1,500 microphotodiode-amplifier-electrode elements on a 9 square millimetre chip. Light perception was restored in eight of the nine, light localisation in seven, motion detection in five, and grating acuity in six, up to 3.3 cycles per degree. Two reached a Landolt C acuity of up to Snellen 20/546. Three subjects read letters spontaneously. Identifying, locating and discriminating objects improved significantly in repeated tests over a nine-month period in the eight patients analysed (Stingl and colleagues, Proceedings of the Royal Society B, 2013; abstract on PMC).

PRIMA (2025). PRIMA is a subretinal photovoltaic implant designed for geographic atrophy, an advanced form of age-related macular degeneration in which patches of the light-sensing layer are lost. Users wear glasses that project near-infrared light onto the implant. The study was open-label, multicentre, prospective and single-group, with each participant compared against their own baseline and no separate comparison group. Participants had a visual acuity of at least 1.2 logMAR at the start. Vision was assessed with and without the glasses at six and 12 months. A total of 38 people received an implant, and 32 were assessed at 12 months. Of those 32, 26 (81 per cent, 95 per cent confidence interval 64 to 93) had a clinically meaningful improvement in visual acuity, which the authors defined in advance as at least 0.2 logMAR. Of the six not assessed, three had died, one had withdrawn and two were unavailable for testing. Using multiple imputation for the missing data, the authors estimated that 80 per cent (95 per cent confidence interval 66 to 94) of all 38 would have had such an improvement. The paper appeared in the New England Journal of Medicine, with a publication date of 20 October 2025 (Holz and colleagues, 2025; record at Erasmus University Rotterdam). The study was funded by Science Corporation and the Moorfields NIHR Biomedical Research Centre.

What is reported but not confirmed by peer review

• Argus II withdrawal. The AAO article reports that Second Sight announced in May 2019 that it planned to accelerate work on its Orion brain implant while temporarily suspending production of the Argus II. An IEEE Spectrum investigation published on 15 February 2022 reports that the company told Argus patients in July 2019 that it was phasing out the retinal implant technology, that a former engineer said no more were sold or made after that, and that in March 2020 the company laid off most remaining staff and announced an intention to wind down operations. It reports that more than 350 people worldwide have the implants and that the company proposed a merger with Nano Precision Medical in February 2022 (IEEE Spectrum, 2022). This is journalism, not a regulatory record. We have not found a source confirming that Argus II production or support has resumed, so the device should not be treated as currently available.
• PRIMA approval. Science Corporation announced that PRIMA has received a CE mark and is being launched commercially in Europe, covering 30 European countries, with the first commercial implant expected in Germany. The company describes the indication as vision restoration in dry age-related macular degeneration, and says it is still working with the FDA towards US marketing, with breakthrough device and humanitarian use device designations in hand. This is reported from the company's press release in trade coverage dated 24 July 2026 (Healio, 2026). We have not read the certificate or a regulator's own statement, and the 2025 trial paper does not mention approval. Approval and availability can change, so this status should be rechecked before relying on it.

What is speculative

Future implants with more electrodes might give finer images. Software that processes the camera image might help people recognise faces or read large print. The approach might also suit other inherited retinal diseases. None of these ideas has been tested in the studies reviewed here.

Limits and open problems

• Small numbers. The Argus II interim trial had 30 people, the 2013 subretinal study had nine, and PRIMA had 38. PRIMA had no separate comparison group, so its gains cannot be fully separated from other factors.
• Surgery. The devices described here all need an operation. PRIMA recorded 26 serious adverse events in 19 participants, and 21 of them occurred within two months of surgery, of which 20 resolved within two months of onset.
• Durability. Two Argus II implants failed after about four years. The PRIMA results cover 12 months, so longer-term durability is unknown.
• Resolution. An acuity of 20/1260 means a person must be 20 feet away to see what someone with normal vision sees at 1,260 feet. The best results in these trials are far from everyday sight.
• Cost and access. The AAO summary says Argus II pricing in countries where it was approved ranged from 115,000 to 150,000 US dollars, excluding medical and surgical care and training. We have found no published price for PRIMA in the sources read here. Argus II eligibility was limited to adults aged 25 or older with severe to profound retinitis pigmentosa and bare or no light perception in both eyes.
• Commercial support. The Argus II case shows that a device can lose its maker's support, and its supply of spare parts, even after regulatory approval.

What to watch next

• Longer follow-up for PRIMA, covering durability and serious adverse events beyond 12 months.
• A regulator's own published record for PRIMA, in Europe or the United States.
• Trials that measure reading, safe movement or quality of life, not only acuity charts.
• Larger studies with a comparison group.
• Whether any manufacturer offers long-term support and spare parts for implants already in use.