Detecting Retinal Holes on Optos Ultra-Widefield Images

Optos ultra-widefield (UWF) imaging can capture retinal holes in a single, rapid photograph that covers far more of the retina than a standard fundus camera. In studies comparing Optos images to dilated examination by a retinal specialist, detection rates for peripheral retinal holes and tears range from roughly 57 percent to 78 percent, depending on the patient population and where the hole sits. That means Optos is a useful screening and documentation tool, but it does not replace a thorough dilated exam when the clinical suspicion for a retinal break is high.

What the Optos Actually Sees

A traditional fundus camera covers about 30 to 45 degrees of the retina in a single shot. The Optos scanning laser ophthalmoscope captures up to 200 degrees in one image, which includes much of the peripheral retina where holes and tears tend to form. The device uses two laser wavelengths, red and green, to produce a composite image. The green laser highlights inner retinal layers and blood vessels, while the red laser penetrates deeper to reveal choroidal features and pigment changes. This dual-wavelength approach lets clinicians see pigmented lattice degeneration, atrophic round holes, and horseshoe tears as dark or bright spots against the surrounding retina, often without dilating the pupil.

Retinal holes generally appear as small, round, well-defined defects, sometimes surrounded by a cuff of lighter tissue if there is localized fluid. On Optos images, they can look deceptively similar to other peripheral findings like cobblestone degeneration or vitreoretinal tufts, which is one reason the technology works best as a screening step rather than a standalone diagnostic tool.

Sensitivity Depends on Where the Hole Is

The location of a retinal hole on the retina matters enormously for whether Optos will catch it. A study published in Retina found that for lesions sitting behind the equator of the eye (the midpoint between the front and back), overall sensitivity was about 74 percent, and for lesions that would need treatment it was 76 percent. But for lesions in front of the equator, closer to the ora serrata, sensitivity dropped to 45 percent overall and just 36 percent for treatable lesions. Specificity, the ability to correctly identify a normal area as normal, was 85 percent.¹

A separate study of patients being screened before refractive surgery reported a detection rate of about 78 percent compared with dilated fundus exam, with specificity around 60 percent. That lower specificity means Optos flagged findings as abnormal more liberally, sometimes overcalling normal variants as possible lesions.²

The pattern is consistent across the literature: Optos performs reasonably well for lesions in the mid-periphery and posterior retina, but its reach falls off toward the far periphery. A retinal hole sitting near the very edge of the retina, where the retina meets the ciliary body, can easily hide behind the eyelid shadow or fall outside the scan’s effective range. If your doctor suspects a peripheral tear based on symptoms like new floaters or flashing lights, they will almost certainly still dilate your eyes and use indirect ophthalmoscopy or scleral depression to look at the far edges directly.

How High Myopia Changes the Picture

People with high myopia (severe nearsightedness) are at elevated risk for retinal holes and tears because their elongated eyeball stretches and thins the peripheral retina. In these patients, Optos screening before procedures like implantable collamer lens surgery is common. A study of 1,725 highly myopic eyes found that about 4 percent had retinal holes or tears and about 18 percent had peripheral retinal degenerations. Optos sensitivity for detecting holes and tears in this population was roughly 57 percent, with very high specificity near 100 percent.³

That 57 percent sensitivity is lower than the figures seen in non-myopic populations, and the reason is partly geometric. Highly myopic eyes are longer from front to back, which pushes the peripheral retina even further toward the edge of the imaging field. Myopic eyes also tend to have what’s called a “tessellated” fundus, where the stretched retinal pigment epithelium reveals the choroidal vessels underneath, creating a busy background that makes small holes harder to spot. Ultra-widefield imaging still improves the ability to detect and document peripheral pathology that is common in about half of all highly myopic eyes, but the miss rate for small individual holes is real.⁴

Artifacts That Fool the Image

Even under good conditions, Optos images are vulnerable to artifacts that can obscure a retinal hole or mimic one. A study examining how artifacts affect the detection of peripheral retinal tears found that eyelid and eyelash shadows, light reflections, and the face guard of the device all significantly reduced the chance of spotting a tear. Face guard artifacts dropped detection probability by about two-thirds, reflections by half, and lid or lash artifacts by roughly a third.⁵

In clinical practice, technicians try to minimize these problems by asking you to open your eyes wide, blink just before the capture, and position your face carefully in the chin rest. Some protocols use “steering,” where the device is aimed in different directions (up, down, left, right) to capture montage images of the far periphery, reducing the dead zones caused by the eyelid and nose bridge. The same study found that montaging multiple directed images improved detection compared to a single central capture, though artifacts still cropped up. If your Optos image comes back with a suspicious dark spot in an area partially obscured by an eyelash shadow, expect your doctor to say, “I need to look at that directly.”

Do Most Retinal Holes Need Treatment?

Finding a retinal hole on an Optos image does not automatically mean you need laser or surgery. Most asymptomatic round holes in otherwise healthy eyes are low-risk. A long-term study of untreated asymptomatic retinal breaks in natural (non-operated) eyes found that they rarely progress to clinical retinal detachment, with only about 1 to 2 percent of cases developing enough change to potentially justify treatment.⁶ A Cochrane systematic review likewise concluded that retinal tears and holes not associated with acute symptoms are significantly less likely to be the sites of breaks responsible for later retinal detachment.⁷

The risk calculation shifts when a hole is symptomatic, meaning it showed up alongside new floaters, flashes of light, or a curtain-like shadow in your vision. It also shifts with the type of break: fresh horseshoe tears carry a higher risk of progression than round atrophic holes. A classic study on the natural history of retinal breaks found that progression to detachment was unusual for round holes or operculated holes (where a small plug of retina has separated), occurring in roughly 5 to 10 percent of cases, compared to a much higher rate for fresh symptomatic horseshoe tears.⁸

For asymptomatic holes found incidentally on Optos screening, especially small round atrophic holes without surrounding fluid, the standard approach is usually observation with periodic re-imaging. Laser retinopexy, which creates a ring of adhesion around the hole to wall it off, is reserved for higher-risk situations. Recent evidence suggests that prophylactic laser may deserve more consideration for asymptomatic fellow eyes with lattice degeneration, particularly when the other eye has already had a detachment, though the supporting data remain below the level of a randomized trial.⁹

Adding OCT to the Peripheral View

Standard Optos images are two-dimensional, which means they show you the surface of the retina but not what is happening in its layers. A round dark spot could be an atrophic hole, a patch of thinning, or just pigment variation. This is where newer swept-source optical coherence tomography (SS-OCT) systems capable of reaching the peripheral retina add a layer of information that Optos alone cannot provide. Wide-field SS-OCT can produce cross-sectional slices through a peripheral retinal hole, revealing whether there is vitreous traction pulling on the edges, whether the edges are everted (curled back), and whether subretinal fluid has begun to collect underneath. Those details directly influence whether a hole is classified as atrophic and stable versus tractional and potentially progressive.¹⁰

A study evaluating navigated wide-field SS-OCT found that it significantly improved diagnostic capability by confirming full-thickness holes, identifying vitreoretinal attachment, and detecting subretinal fluid that was not apparent on the color image alone.¹¹ More recent work with navigated peripheral SS-OCT has demonstrated clear visualization of retinal holes and tears alongside the vitreous and choroid, providing high-quality cross-sectional images of structures that were previously only assessable by direct clinical exam.¹²

In practical terms, if your Optos image shows a questionable peripheral finding, your doctor might follow up with peripheral OCT to determine whether the finding is a true full-thickness hole, whether there is traction that could widen it, and whether fluid has started to seep underneath. That three-dimensional detail can be the difference between “let’s watch this” and “let’s treat this today.”

Using Optos Images to Plan Surgery

When a retinal hole or tear does cause a detachment and surgery is needed, Optos images have a practical role in preoperative planning. Scleral buckle surgery, one of the traditional approaches to repairing a retinal detachment, requires the surgeon to locate the retinal break from the outside of the eye and place a silicone band over it. Historically, surgeons relied on indirect ophthalmoscopy drawings and mental geometry to translate what they saw inside the eye to a position on the outside. A study investigating whether ultra-widefield fundus photography could help estimate retinal hole location before scleral buckle surgery found no statistically significant difference between the hole-to-limbus distance estimated from preoperative Optos images and the measurement taken with calipers during the actual procedure. The researchers concluded that preoperative UWF imaging is a feasible way to localize retinal holes before surgery and may help reduce the learning curve for less experienced surgeons.¹³

This is especially useful in training settings, where a resident performing a scleral buckle can study the Optos image beforehand to understand the clock-hour position and approximate distance from the limbus of each break. It does not replace intraoperative visualization, but it shortens the search time once the eye is open and the surgeon is working under a time constraint.

AI-Assisted Detection of Retinal Holes on Widefield Images

Researchers have been training deep-learning algorithms to automatically flag retinal breaks and detachments on ultra-widefield images, and the early results are promising. One deep-learning classification system tested on fundus photographs achieved very high accuracy for detecting lesions overall and for distinguishing retinal detachment from retinal breaks specifically. For retinal breaks, the system reached an area-under-the-curve (a measure of how well it discriminates positive from negative cases, where 1.0 is perfect) of 0.913.¹⁴

A pilot study specifically targeting tessellated (highly myopic) eyes, where the busy background makes human interpretation harder, developed a combined deep-learning system that achieved an area-under-the-curve of 0.888 for lattice degeneration and 0.953 for retinal breaks.¹⁵ These numbers suggest that AI could eventually serve as a reliable second reader, particularly in high-volume screening settings where thousands of images pass through a clinic and not every one gets reviewed by a retinal specialist. The technology is not yet in routine clinical use for this purpose, but it is moving in that direction.

Remote Screening and Teleophthalmology

One of the most practical applications of Optos imaging for retinal holes is in teleophthalmology, where images are captured at a primary care or optometry office and transmitted to a remote retinal specialist for interpretation. A study comparing ultra-widefield imaging to in-person ophthalmoscopy for screening found that imaging actually detected more abnormal findings overall: at least one abnormal finding was present on ophthalmoscopy in about 11 percent of eyes, compared with 27 percent on adjudicated image interpretation. A total of 187 findings were picked up by imaging but missed on examination, versus 42 detected on examination but missed by imaging.¹⁶

Those numbers deserve some context. The in-person exam in that study was an undilated ophthalmoscopy, not a full dilated retinal examination with scleral depression, so the comparison is not between Optos and the gold standard. But the point for teleophthalmology is that Optos captures a permanent, transmittable record that multiple specialists can review, whereas a brief office exam relies entirely on the skill and attention of the examiner present. For retinal holes specifically, this means a patient in a rural clinic without a retinal specialist can have an Optos image sent to an expert who looks for lattice degeneration, atrophic holes, and other peripheral findings and decides whether a referral is warranted.

Screening in Premature Infants and Children

Retinopathy of prematurity (ROP) screening traditionally involves a bedside exam with an indirect ophthalmoscope, which can be stressful for fragile premature infants. Optos imaging has been explored as a way to capture the peripheral retina in older children with a history of prematurity, offering a faster, more comfortable alternative that does not require anesthesia. Research has suggested that wide-angle fundus imaging with an Optos device can facilitate screening and follow-up for persistent avascular retina in these patients and is compatible with telemedicine, allowing remote expert review.¹⁷

This application is still developing. In very young infants, positioning in the Optos device is impractical, and dedicated neonatal cameras remain the standard for acute ROP screening. But for follow-up visits in toddlers and older children who need peripheral retinal monitoring, Optos offers a quick capture that avoids the discomfort of a speculum and scleral depression. Parents who have been through the traditional ROP screening process tend to welcome it.

When Observation Includes Repeat Imaging

If an asymptomatic retinal hole is found on Optos and the decision is to observe rather than treat, the imaging itself becomes part of the follow-up plan. The advantage of having a baseline Optos image is that subsequent visits can produce comparison images under nearly identical conditions. A hole that has grown, developed surrounding fluid, or acquired new pigmentation is easier to detect on serial images than on serial clinical exams, where the examiner’s subjective impression introduces variability. The refractive surgery screening study noted that accuracy between different readers of the same Optos images ranged from 72 to 87 percent, which is imperfect but allows standardized documentation over time.¹⁸

For patients, this means the Optos image from your initial visit becomes a reference point. If you return in six months or a year with new symptoms, the doctor can pull up the old image side by side with the new one and see whether anything has changed. That kind of objective comparison is harder to achieve with examination notes alone, where descriptions like “small round hole at 2 o’clock, posterior to equator” leave room for ambiguity.

Laser treatment, when it is eventually needed, can also be guided by the Optos image. Demarcation laser photocoagulation, which places burns in a ring around the break to create a scar that walls off fluid, benefits from knowing the exact size and position of the hole before the patient is under the laser. For shallow retinal detachments associated with peripheral holes, laser barricade has been described as a reasonable alternative to surgical repair in selected cases, provided the patient understands that the adhesion takes up to two weeks to form and further intervention might be needed if fluid continues to spread.¹⁹

What an Optos Image Cannot Tell You on Its Own

For all its utility, the Optos image of a retinal hole is a surface photograph. It cannot tell you the depth of the hole, whether vitreous gel is tugging on its edges, or whether a thin layer of fluid has begun to separate the retina underneath. These are precisely the details that determine risk, and they require either a careful clinical exam with scleral depression or cross-sectional imaging with peripheral OCT. A review of peripheral retinal degenerations and anomalies using ocular imaging technologies emphasized that a range of conditions, including retinal holes, vitreoretinal tufts, lattice degeneration, and snail-track degeneration, all benefit from multimodal imaging rather than reliance on a single modality.²⁰

The bottom-line reality is that Optos ultra-widefield imaging has become a standard first step in evaluating retinal holes because it is fast, comfortable, and captures an enormous field of view in a permanent digital record. It catches the majority of holes in the mid-periphery and flags many that would otherwise go undocumented. But its sensitivity drops for far-peripheral lesions, artifacts can obscure critical areas, and it cannot assess the three-dimensional features that drive treatment decisions. The technology works best as one tool in a layered diagnostic approach: Optos for screening and documentation, dilated exam for clinical confirmation, and peripheral OCT when the details of a specific lesion need to be worked out.