Lecture: Retinopathy of Prematurity in 2026 – Evaluation, Management and Disease Burden

In this webinar, Dr. Gise will discuss the current strategies for the evaluation of patients with retinopathy of prematurity (ROP) and their management. We will touch upon the screening criteria and strategies, the use of tele-screening, a review of current treatments such as intravitreal injection and laser photocoagulation, and the visual outcomes of nano and micro premature cohorts. The webinar will feature a case-based approach to these topics with a focus on the strategies used at Boston Children’s Hospital for evaluation and management of these infants, specifically in-person rounding and tele-screening. (Level: Intermediate and Advanced)

Lecturer: Dr. Ryan Gise, Ophthalmologist, Boston Children’s Hospital/Harvard Medical School, USA

Transcript

DR. GISE: Hi, everyone. Thank you so much for coming. My name is Dr. Gise, I’m one of the pediatric ophthalmologists and neuro-ophthalmologists at Boston Children’s hospital and talking today about retinopathy of prematurity. It’s near and dear to my heart and one of my favorite things that I do. I hope I can share that with you guys today. I have no financial disclosures. My goals for the talk are to review the current strategies for evaluation of patients with retinopathy and prematurity and their management. To review the updated classification system for international classification of retinopathy of prematurity, the third edition, just a few wears ago. It’s important to start with what ROP is. The pathophysiological underpinnings dictate how we screen and treat these patients. It’s the arrest of normal retinal vascularization in the preterm infant with compensatory mechanisms that lead to blindness. The epidemiology, it’s the leading cause of preventable childhood blindness worldwide. In the United States, 80,000 infants are at risk of ROP each year and approximately 1500 have disease. And word wide, it’s estimated at 32,000 patients a year. The agenda, talk about the history, pathophysiology, classification, screening and finally treatment with updates for 2026. In terms of history, in the 1940s was the first epidemic of retinopathy of prematurity, and it was called retrolental fibroplasia. It was a retinal detachment. Obviously at this point there was no recovering of vision and they really didn’t have treatment options. And this is what you might see at the bedside back then. So, that was in the 1940s. In the 1950s, the role of oxygen was elucidated in the development of retinopathy of prematurity. They were looking at various different causes, fluctuations in temperature, ambient life, and oxygen. Finally it was discovered that high oxygen levels were causing blindness, and unfortunately it was also noted that high oxygen levels were important for the survival of these children. That was in the 1960s, while they could cut down on the oxygen, they had worse adverse systemic events. That’s more important, brain development and other organs. And so, it was really where a balance needed to be struck and it was also very hard because there were really no treatment options at that time. In 1984, the first ICROP classification was released. And then in 1986, the CRYO-ROP study demonstrated that we did have treatment methods and if eyes reached a certain level, which at that time was called threshold, where you had five contiguous clock hours of stage 3 or eight total clock hours of stage 3 plus disease, then treatment with cryotherapy would help improve visual outcomes. Although visual outcome were still not what we want them to be today. With cryo, it’s the same thing you’re doing with laser. You’re basically destroying the avascular retina or the cells that are causing the neovascularization to grow. It’s obviously very destructive. And then in 2000, the early treatment of ROP study came into existence and revolutionized the way we care for ROP and still the underpinnings of the way that we care and treat ROP today. The idea of type 1 ROP was introduced where we have a level of ROP where treatment was definitively better than observation. And type 2 ROP were treatment with laser versus the level of ROP that kids have. And we’ll get into what that is in a few minutes was basically equivocal. And below that, the risk of treatment was actually greater than the risk of observation. And then in 2008, the BEATROP study was published, that looked at bevacizumab on a wide scale for retinopathy prematurity. Babies in Zone 1 and posterior Zone 2, but mostly Zone 1, were found to have better outcomes with treatment and obviously there were some issues with this study, as with every study, but it was really the first one that brought it into the main stream. And then in the 2010s, we started to have other studies looking at the treatment of ROP with Anti-VEGF agents. The ROP1, looking at the VEGF levels systemically and the ability to treat retinopathy prematurity. And the RAINBOW trial, and finally FIREFLEYE, BUTTERFLEYE, looking at aflibercept into the treatment of retinopathy prematurity, and led toot FDA approval, our Federal drug Administration, governmental approval for the use for the retinopathy of prematurity. The first drug deemed so. In terms of pathophysiology, normal retina development. The choroidal circulation vascularizes around 6 weeks of gestational age. Retina becomes increasingly metabolic, leading to hypoxia. And stimulates VEGF around 13 weeks of gestational age. And here’s where it’s important to the clinical practice, nasal retinal vascularization is complete around 36 weeks of gestation. Temporal retinal vascularization usually completes around 40 weeks. This is kind of our benchmarks for when we’re looking at premature infants what to expect and when to expect. Every now and then, vascularize before 36 weeks, sometimes the gestational age is off, they’re estimations. And sometimes around 40 weeks. Sometimes take longer if they have active retinopathy of prematurity and other comorbidities. It’s important to know so you can anticipate what’s going to happen in the retina. And then gestational age, postnatal age, and menstrual age. Gestational age, measured from the first day of mother’s last menstrual period to birth. Some are born at 23 weeks of gestational age. And postnatal age, how old they are chronologically. James is 11 weeks postnatal, and post menstrual age, James was 23 weeks gestation and now is a 34 week infant. And so this is kind of language that we use to talk. So, now that we know the normal retinal vascular development, it’s really important to look at what happens in retinopathy of prematurity. In utero, it’s a hypoxic environment, you have normal vessel growth and normal levels of IGF-1 and VEGF. When you have premature birth, the hypoxic that the child was in becomes hyperoxic. And once it becomes hyperoxic, then the signaling of vessel — to promote vessel growth, the VEGF and the IGF-1 levels decrease and vessel growth arrests. Then as the retina starts to mature and become more metabolically active, it starts to send out increased signaling to help the blood vessels grow. Here is how you get the spike in VEGF. And unfortunately you get that spike, not only do the normal blood vessels grow, but you also get retinal neovascularization, this is what leads to retinopathy of prematurity. And either the retinopathy of prematurity will resolve on its own in the majority of cases, or end up becoming a proliferative retinopathy and lead to retinal detachment if we don’t treat it. So, here is a diagram from the BEATROP study published in the New England Journal of medicine. It can take use along the way. You can see around 24 weeks and greater is when we start — is when we have these NICU baby who is we’re dealing with. And they don’t usually get examined until 31 weeks of life in the United States and at that point, some of these very, very young babies very retinopathy of prematurity. Some won’t. And so, we’re following them to see when they get to that level of type-1 ROP where treatment becomes warranted. And then obviously there are multiple different ways that we can treat now. Cryotherapy is still in existence, although not widely used in the United States. There’s also laser therapy and bevacizumab therapy. Now it’s anti-VEGF therapy because we have three agents that can be used are used depending on your center. So, classification. So, the way that I think about ROP is Zone, Stage, and Plus disease. The where, the what and how bad is it? Or is it bad? So in the zone where this is the diagram we typically use. This is taken from the third ICROP study. So, in the third ICROP study what was introduced was this concept of posterior Zone II, just about in Zone I, a little bit further out, but highly concerning.S way to look at it clinically, zone I or zone I, if you can see the nerve in the 28 diopter lens doing indirect ophthalmoscopy and still see the vessels, you’re in Zone I. It’s device the difference from the ocular nerve to the macula, but that can be hard to see and kind of map out when you’re just looking in and the baby is obviously moving. RetCam photos can be really helpful for figuring this out when you get a chance to. And you look at the clock hours and Zone I circumstance basically finished at nasal vascularization. When I’m looking for Zone II versus Zone III, the first place is nasally, it’s aural nasally, it’s usually Zone III. They don’t have to vascularize at the same pace. Usually they’re pretty equal and tell the story together. So, then we have the stage. And one of our fellows taught me this one this is DR FIT as a way to remember what the different classification is. So, the demarcation line is stage I. So, you tart to see that there is a physical difference. It can be hard to tell stage zero or immature retina from stage 1. Stage 2, you feel a raised ring. Running your thumb over it, you feel a bump. Stage 3, you get fibrovascular proliferation into the vitreous. Stage 4 is incomplete or partial retinal detachment, stage 4a and 4 and stage 5, total retinal detachment. Our goal is to prevent stage 4 and stage 5. Pictures from the ICROP diagram and the most recent to elucidate what immature retina looks like. You can follow the blood vessels out and you can see at edge of the blood vessels, there’s nothing there. The blood vessels just stop. And looking at this, you can see the distance from the nerve to the macula and twice that. So, this is into Zone II. But really the blood vessels just stop so there’s nothing that we need to worry about. Once we start to see stage 1, this is more the ROP is present. It’s still very mild. But you can see where there’s a clear demarcation between vascular and avascular retina. And that’s denoted by the white arrows. One of the things that ICROP 3 also introduced is this concept of the notch. So, a notch is temporal. It is often right in line with the macula. And when we’re classifying ROP zone, we will classify it by the most posterior zone that it is. So, remember everything is backwards. So, anterior is far out in the peripheral retina. Posterior is more towards the back and the optic nerve. And what will happen with the notch is the notch can bring it back a zone. So, this would be stage 1, Zone 1 by notch, even though most of the vascularized retina appears to be in Zone 2 or posterior Zone 1. The black arrow heads is in Zone 1, the macula. It is the most posterior or Zone that we label it as. And then stage 2 is a raised ridge. So, you can, again, you see it. It’s a little bit thicker. It’s hard to tell in a picture, obviously, without being able to see three dimensions. But you can tell it’s a little bit thicker. You can also start to see some little white, pink circles that are in front of the ridge. And so, you can see them right around there right in front of the arrow. And that is some neovascularization that is posterior to the ridge. It’s not touching the ridge. We call those popcorn if they’re circles and hot dogs if they’re cylinders. That’s a sign of disease activity, but not included in our treatment criteria or anything else. It’s more just something that we need to know about because it means that the stage 2 is more active than if it wasn’t there. And then finally, we get to stage 3 where treatment becomes warranted which we’ll talk about in a little bit. And so here at the white arrows you can see that even though it’s hard to tell in three dimensions, it looks like a wall of pinkish tissue and then the blood vessels appear to be much more dilated and tortuous, this is type 3 and requires treatment. Given the location, in mid- to posterior Zone II based on the distance. Here is the nerve and the fovea is right about here. And then it’s probably about here would be posterior Zone II-ish. And it’s obviously not an exact science. And so, depending on the age and other comorbidities of this child, you would have a lot of treatment options at our institution. Laser might be considered with the child was older or injection if the child was more sick currently. It all depends on the situation that the child is in and we’ll talk about that a little bit more later. And then unfortunately when ROP is not treated adequately or quickly enough, it can result in things like this where this is a 4A detachment. You can see the few bro vascular ring, it almost looks like a scleral buckle inside the eye and it’s raised. Fortunately the macula is not involved with this child so the vision would be hopefully somewhat preserved. And you can get a 4B, you have a partial retinal detachment and the macula is involved in vision would be greatly effected. And here is another incidence of a tractional retinal detachment. And if it continues to progress, it can lead to total retinal detachment and you can see it just like in the picture of the retrolental fibroplasia and that’s on the stand. The plus disease is the last piece of the ROP classification of is it bad? And Plus disease is part of the art as much as of the science of retinopathy prematurity. There can be a lot of disagreement between investigators. What ICROP 3 has done in order to help us better describe Plus disease is introduced the importance of the P score which I’ll talk about in a second. Which is essentially a grading scheme for Plus disease. And I was fortunate enough to see Michael Chang from the National Eye Institute to talk about this research, what they found is with Plus disease, when even world experts will disagree on whether something is or isn’t Plus disease. Especially if it’s kind of borderline. So, if you give any picture of some of these borderline Plus disease patients to ten experts, what they found was 6 out of 10 would agree that it either is or isn’t Plus disease, but no more than that. It’s about 60%. That number goes way up when you start giving it a score. And you say is it better or worse than this image or that image? And so, whether there’s a comparison, the ability of the experts to reach a consensus went way up, which is why the P score was introduced. And so, here you have some pictures of the P score. And they did a really nice job of just highlighting the blood vessels and also putting a color to it. Green obviously meaning normal. Green to yellow. Yellow, so those top three. The top three pictures are P scores 1 to 3. The middle three are 4 to 6, and the bottom three are 7 to 9. And so, 7 to 9 generally is indicative of Plus disease. 1 to 3 is generally not. 4 is generally not. 5 and 6 is where you start to kind of wonder, is it progressing? Is it active disease? It there a retinopathy of prematurity going along with this? So, this idea that we have some comparative images has been instrumental in helping us better describe what we’re seeing in addition to RetCam photographer. And — or fundus photos, I should say. Because they’re different cameras. And getting agreement among experts. The other thing that ICROP 3 has introduced is it changed the classification of what was known as aggressive posterior retinopathy of prematurity, A-ROP, or Rush disease for those of us who a little bit older. So, the change that was made was to call it aggressive retinopathy of prematurity. They noted that depending on where you are in the world and what your NICU care is like, you may have aggressive retinopathy of prematurity outside of zone I, which is the AP-ROP definition was in Zone I. And it’s important to know it’s different wherever you are in the world. And so, the nice thing about retinopathy of prematurity, it tends to follow a cadence. I know that children in the United States who are born between 22 and 26 weeks are at risk of developing this aggressive ROP. It usually is developed around 33 to 35 weeks. So, I know in that time frame when I’m screening to be really attentive to this. And what you’re looking for is Plus disease that’s out of proportion to the retinopathy that you see. You see tortuated vessels like in this picture, but no neovascularization, it’s inside the retina itself. It’s important to recognize this because it comes on very quickly and can lead to retinal detachment very quickly. Here with the white arrows you can see that kind of junction between the vascular and avascular retina. And one of the hallmarks of this aggressive posterior — in our — I say it’s posterior — but aggressive retinopathy of prematurity, it tends to skip stages. Meaning that we don’t see the development of the Stage III. It go from one to two to three like older children do, it immediately goes to 3 and kind of explodes very quickly. And here’s another example of it. And so you can see this flat neovascularization that can be very tricky to deal with and is one of the places where Anti-VEGF agents, and we’ll talk about their benefits later in the talk, have really come into effect and been very helpful. And this zone I disease is what the BEATROP study, Anti-VEGF agents being especially help NFL treatment of. Screening. Who gets ROP? The two main factors for getting ROP that we’ve discover ready low birth weight and low gestational age. Look at the cryo ROP ski and the ETROP study, it was much highering the lower the birth weight and the gestational age. At the time of the ETROP study, less than 2 weeks was a fairly early child — 27. And now in the United States, we have 22 to 23, and I have one 21 treatment that needs. The lower the birth weight, the higher the risk of developing retinopathy can and prematurity. In terms of onset, CRYO-ROP and ETROP, around 34 weeks, Stage 1. And between — was around 36 to 37 weeks. That was the past. It’s important to know, so these studies were done in the United States. So, this is the retinopathy of prematurity that we see. It’s not necessarily what you see because it really does depend on the balance between NICU care, eye exams, et cetera. It may be slightly different. And it’s why it’s important to, especially with some of these screening algorithms that we have now like WINROP or GROP, they’re wonderful, but it’s very important to make sure that they’re validated in your patient population. Because if they’re not validated in your patient population, they may media allow in the wrong direction. So, in the itself, the birth weight and gestational age that we typically screen for is less than or equal to 1500 grams and less than or equal to 30 weeks. 30 weeks and 6 days is considered requiring treatment. Namesh Patel at Mass looks at what he calls the two ROP strategy. Mean for you don’t meet criteria for one of these criteria and you’re gonna get screened, but what is your risk of developing ROP? And it’s generally very, very low. At children’s, when we validated his idea, we actually only had a handful, like one or two children, that ended up requiring treatment who were only positive for one of these two screening criteria. And they had retinopathy of prematurity by 37 weeks so they obviously were caught. But it does beg the question of: Is there a way we can start screening less babies? Because there is some morbidity with the exams. They don’t like them. They can cause bradycardia and desaturations and things like that. In Canada, there’s a little more loose in their criteria. They actually screen 1250 grams or less instead of 1500 grams. And then the criteria worldwide varies and it varies depending on where you are and your NICUs. Because obviously kids may be older, but more at risk. And here’s that question of WINROP and G-ROP. What they incorporate is post-natal weight gain to help reduce the number of screenings. And it’s just very important, again, to make sure that the — they are validated in your patient population before you use them. One of the ways that we’re consider them, too, that there are some babies who are outside of our screening criteria who would have become positive by G-ROP, for example. And so, we may be missing babies that develop retinopathy of prematurity that needs treatment. A handful across the United States. That’s one of the questions that we’re hopefully going to try to tackle when we work with the American Academy of Pediatrics for our new screening guidelines. The more premature an in. Is, the longer the time to develop Severe ROP. Mean for they’re a 22 weeker, we still screen them around 31 weeks. What our studies have shown at Children’s, it’s in the they develop more severe ROP earlier than that. It can be hard to look in and see because the vitreous tends to be very hazy. But it’s rare that we’re seeing a baby for their first exam and treating them on that exam. And then one of the things I wanted to talk about was tele-screening, which was something that we’re doing a lot more at Boston Children’s right now and across the world. So, traditionally, obviously, screening has been in person. But there are some benefits to tele-screening. It really does improve access to care. So, there are some places in the United States and across the world where there aren’t ophthalmologists who are skilled and trained to screen for retinopathy of prematurity, or there aren’t enough of them. And so, by tele-screening, you are looking at patterns. You’re looking at the posterior pole, mostly. But you can get further out, depending on the skill of the photographer. And treatment-warranted ROP that would indicate the need for referral to a sister that will treat for retinopathy of prematurity. It’s been reliably indicating it, and it’s particularly good for finding Plus disease. There are some drawbacks for it, though. You can’t see the whole retina into zone III, babies cannot be graduated from screening or transitioned to the next area of care, follow-up to a pediatric ophthalmologist for strabismus and amblyopia. And it’s harder to convey the screening to patients and families because there’s no direct cant. When we’re rounding in person, we explain why we’re doing what we’re doing and how important it is and the risk of blindness. Sometimes given how many problems these babies have and how many specialists can be involved it can be hard for the NICU to convey every specialist problem. We have found that the importance of following up can be a little bit harder to convey that message. The equipment is expensive for the NICUs. I know there are some smartphone cameras that have been trialed and written about. But the ones that we use in the United States, the RetCam itself, is several hundred thousand dollars to purchase. You need speculums, you need multiple camera lenses to try to prevent infection, et cetera. Which can be hard. And then the larger and older babies are obviously much more hard to image because they don’t want to be held down. So, the image quality can degrade as they get older. When we’re doing tele-screening, it — one of the other negatives is that it often requires some additional exams for babies. So when I’m seeing a baby in the NICU and I see that, you know, even if it was a 24 weeker and I’m seeing them at 31 weeks and they’re immature in Zone II already and I got a great view, I will give them two weeks for follow-up when they’re being tele-screened, they’re getting extra exams. Typically high-risk babies or where retinopathy of prematurity is present on the image, seen weekly. Low risk babies. I use the DIGIROP screening algorithm, you can kind it on the internet. To help risk stratify. It’s nice, all you need is the birth wait and gestational age. There are other variants where you can take into account the degree of retinopathy prematurity, whether there was ROP in the last exam. If they’re at high risk for developing type 1 ROP, I screen them once a week to give them a break. These are babies that are older and larger. And older children higher risk without ROP, the doctor from Stanford who has been long a proponent of tele-screening has published a study in ophthalmology retina in the last two years that demonstrated babies over 40 weeks who did not have retinopathy of prematurity or severe retinopathy of prematurity did not progress after 40 weeks as long as they were stable and biweekly screening. Requiring these exams and they’re harder and the children are larger. Now we have defined the problem. The next question becomes, how do we treat it? These are the major advancements of the 20th — 21st century. And 20th century. But 21st century since that’s where we are now. This diagram, when to treat. This is where type 1 ROP comes into play. And talk about it in a little bit. But type 1 ROP meaning stage 3 in zone I without Plus disease, stage 2 in zone 1 with. Plus disease or stage 1 in Zone 3 with Plus disease. These are what we use to determine whether a baby needs to be treated or not because treatment has been shown to improve visual outcomes in those patients. And then which treatment modality we choose, fortunate enough to have both at Children’s. We try to tailor it to the baby. So, this is a picture what have threshold would look like from CRYO-ROP. As I mentioned, they define threshold at 8 contiguous clock hours or 5 contiguous clock hours with Plus disease. Something that’s treated aggressively now, they wait longer and they really only had CRYO at that point. That was the only treatment option. And in eyes with threshold disease, this leads to a 50% chance of retinal detachment if left untreated. It’s interesting, it was not 100%. So, some of these babies still did regress. With CRYO therapy, you’re ablating the avascular retina. Destroying the cells that create the VEGF signals that are causing the retinopathy of prematurity to get worse. And like the laser, there’s still VEGF signaling in the eye itself and it takes one to two weeks. Does so, we don’t really use CRYO in the United States anymore. Here is the — and pre-threshold type 1 and 2, demonstrated to be equivocal whether treated or not, zone I, stage 1 or 2 without plus disease, zone II stage 3 without plus disease. I like to put an important caveat, which is one specific type, the zone II, stage 3 without plus disease. There — if there is stage 3 temporally, right in line with the macula, 2 or 3 clock hours or more, that stage 3, it is a regresses spontaneously can still cause traction. Needs to be watched carefully and in some cases may warrant treatment that’s a little bit earlier depending on the condition of the baby. It’s just important to watch out for the potential for — to develop traction. With treatment, usually laser because these kids are older, it just helps the neovascularization resolve faster. And using the ETROP treatment paradigm, structural outcomes improve — the presence of bad structural outcomes decreased from 15.6% to under 10%. So, type I ROP is treatment-indicated. Type 2 it follow closely within one week. Often type 2 ROP patients depending how they look, I may see twice a week depending on, you know, where they are and how they’re progressing. And so, here is, again, that picture from ICROP 3, the patient with type 1 retinopathy of prematurity. As I mentioned, a couple different options. Not using cryotherapy anymore, but we are using laser and Anti-VEGF. And here’s why. These are the pictures from the BEATROP study. A baby in zone I with ROP, and treated with conventional laser therapy almost up to their fovea. And you can see the retina that was ablated. And this is only going to severely constrict their peripheral vision. This was fortunately a good structural outcome because the retina remains flat. The level of vascularization is never going to increase beyond where they — where the laser took place. Versus a child with — who received bevacizumab, and you can see the white arrows, the smaller arrows, that’s the old demarcation line and the level of vascularization doubled at 13 months. What I tell parent when I’m recommending Anti-VEGF injections, one of the major benefits is the retina will still have a chance to grow and it will help improve the amount of retina they have to see. I also talk about how the Anti-VEGF agents when we inject them immediately bind to the circulating VEGF and therefore tend to start working within 24 to 48 hours. What I’ve noticed more recently with aflibercept injections, babies who are active, type 1 ROP in Zone I, high burden of VEGF, the Anti-VEGF works much faster. And the older baby where there is less VEGF is circulating, it tends to be a less dramatic effect. You can see on the picture on the bottom in frame C, where if up to laser their patient, you would then have to wait two weeks to see it take effect. It’s a harrowing two weeks. It may progress, versus the injection where it takes it away immediately. So, it tends to be a very good indication for Anti-VEGF agents, although not everyone has them. So, with laser treatment, obviously the pros are that it’s one treatment. So, if it’s done correctly, and it does take a lot of skill and practice to become good at ROP lasers, they’re the hardest laser that you can do. There’s a decreased need for follow-up. And it’s more permanent. Meaning the retinopathy of prematurity regressed after one, sometimes two sessions within and then you are able to not follow the infants as closely and they — depending on where the parents live, if they live close to your center or far question from your center, that can obviously play a role. Social situations can as well. If you don’t think the infant are brought to follow-up, then lazier is a much more kind of durable and permanent option, obviously. The cons that it can take two beaks to demonstrate effect. You can have cataract formation, glaucoma, retinal tear or detachments. Treatment failures. Not all centers do this, at Children’s we tend to intubate and sedate these babies. And it’s a fairly difficult procedure, especially in situations like in the past slide where it was in Zone I and there was very active ROP to cover the entire retina. But it can become very important because it cuts down on the need for follow-up, depending on that. And obviously we want the babies to be as safe as possible, even if we can’t see them. Anti-VEGF injection, the way I think about, it is much easier up front. So whereas a laser is much easier after the lazier is done and the ROP is regressed, an Anti-VEGF injection is much easier up front. So, you — an injection, obviously, takes a minute. It generally does not need sedation. What we’ve found is in the older babies who are bigger, sometimes they do need a little bit of intranasal vera sed or oral lorazepam just to relax a little bit so they don’t bear down as much when we’re doing the injection and push the medicine out of they eye. It obviously takes rapid effect. You don’t have to wait for the two weeks for the laser to take effect. And it continues to make the retina grow afterwards. The cons, similar to laser, in the sense that you could nick the lens and cause cataract formation, you can have increased pressure afterwards. This tends to be associated with the volume of liquid that you inject. So, most of our injections are either in .01 or .02 — ccs. And .05 is higher risk of intraocular pressure. You can get retinal tear. One of the things we’ve adopted is the safer protocol for injections. We use a 32-gauge needle. We sterilize the eye with 5% Betadine. We have close follow up. After injection, see them within 24 to 36 hours to assess for infection. And then if it’s around 36 hours, a lot of times we can see improvement already. And then can follow them a couple of days later just to make sure that no late injection develops. Or if the — if it’s 24 hours and we haven’t seen improvement yet, then continue to follow them until there is improvement. From the ROP 1 treatment study failures on the initial injection were called at approximately post-op day 4. So, that’s kind of you really want to see Pi post-op day 4. This is an unknown effect on development. We are injecting Anti-VEGF agents into very small children and we are obviously injecting less quantities of it now than we have in the past. But it’s a really hard patient population to kind of assess for development because they obviously have a lot of developmental issues related to their other comorbidities. And there is a level of systemic Anti-VEGF suppression which is thought to be less in our newer agents. And then there can be late reoccurrences. This is where I was kind of referring to in that you will — the work up front is easier for an injection, technically. But then you’re following these babies for long periods of time to see if they will reoccur. The most recent data from the FIREFLY next study published in 2026 in neonatology is babies that didn’t have recurrence by 50 week it is not have a recurrence up to 3 year, which is why the study data is. But there are case reports of 6-year-olds and older patients developing kind of reoccurrences and not fully vascularizing. And so, at Children’s, we take all of these children — this is also part of the safer protocol, for an exam under anesthesia with fluoro angiography around a year of life, or earlier with another procedure to look if the avascular retina still exists. And with we have to do a third the amount of laser as in the primary situation. And this, again, is — and I usually use when I can no longer examine them in the office as the time to start considering it because I really want to know what’s going on in the back of their eye and I don’t want to miss anything. But you can imagine how these fairly sick children also have other doctors appointments and they’re going to lots of doctor appointments and need to come to their eye appointment every one to two weeks or a little bit longer, depending how old they are, and that can be very challenging. And so, current) with Anti-VEGF agents. Following the PEDIG ROP1 studies, looked at lower doses of bevacizumab, it was patients from 2017 to 2019. Included patients with type 1 ROP in at least one eye. Low dose bevacizumab, .013, or very low dose, .004, and .002, with a fellow dose in both eyes were treated with Anti-VEGF agents. This was specifically with bevacizumab. Both received treatment at similar rates. Even though it was lower and lower doses, both were seen to be effective. Poor structural outcomes were low in both eyes and there was no difference in rate of reactivation. Though there was a tend towards shorter time to reactivation in the very low dose intravitreal bevacizumab arm. And so, this is usually between 10 to 12 weeks with avastin or bevacizumab. With ranibizumab, the RAINBOW trial, shown to be effective with laser. And what we have noticed in the personal experience and studies, is that ranibizumab has a typically higher rate of reoccurrence because it tends to leave the system faster. It’s a good thing and a bad thing. It’s good because it leaves their system faster, but bad because we don’t wanted to be dealing with ROP reoccurrences. And not every ROP reoccurrence requires treatment. Sometimes they reoccur and watch them and it resolves. It doesn’t become type 1 again. And finally, aflibercept, FIREFLEYE in Europe and Russia led to the approval of the eyes nut United States for the retinopathy of prematurity. And so all of these agents are used now in the United States. And I think with varying kind of — depending on what there center you’re at, they can all be effective. It’s important to know the half-life of the drug that you’re use, because you need to know when the reoccurrence might happen and how closely to watch. Finally on the horizon there was recently a study published in the Journal of ophthalmology about using dexamethasone 0.1% eye drops in the prevention of type 1 retinopathy of prematurity. Look at them, if you administer prior to the peak flee owe vascularization, around type 2 retinopathy, it significantly decreased the burden and progression of it. The thought was this is a minimally invasive way to prevent retinopathy regression. This is a study in ophthalmology, during two time periods, 2015 to 2018 and the 2020 to 2021, where they were using dexamethasone drops and the initial 2015 to 2018 one, they were not. So, as I mentioned, type 2 ROP got one drop every day when there was stage 3 in Zone II. It was every zero day if it was stage 1 to 2 in Zone I with no plus disease, but eventually require one drop every day, it would progress. In rare cases, sometimes twice day drops were used. But the goal was to decrease that as soon as possible. And then to taper to off when there was no longer type 2 ROP present. And if you look at their graphs, it’s actually quite dramatic. In terms of interventions that they’ve had tad. And in terms of the number of screened infants that required treatment. They both really went down significantly. Obviously there are some risks that haven’t really necessarily been well elucidated because this was not like a clinical trial. This was a retrospective study. And so, the systemic exposure is theorized to be very small. The infant eye can only hold about a fifth of a drop. Punctal occlusion was used. No patients developed adrenal suppression. Although it’s unclear how about were commonly given systemic steroids for this problem and for BPD, sorry. And ocular side effects, cataract formation, steroid induced glaucoma were not observed, but again important to consider. And then finally something that has been a problem — developing problem for our institution is now these nano premature infants, 24 weekers, and under 600 gram infants being born and published a paper in the New England, you know, of ophthalmology about long-term outcomes. What you expect, more significant number required treatment. Look at the national studies, nano premature infants, typically about 50 to 60% of them will require retinopathy of prematurity treatment. The treatment age is younger in these kids. They’re more at risk for having more posterior disease. Interestingly enough, long-term effects, there was no difference between them and micro premature infants, which were 24 to 26 weekers. Both groups, though, had spherical equivalence of one or less at four years for those we industrial follow-up for. And obviously in these kids, you are going to expect them to be more myopic. In a 4-year-old, you u you expect them to be more myopic than a plus one spherical equivalent. There is definitely a trend towards myopia. That was my talk. A whirlwind tour through retinopathy of prematurity. And I hope you guys enjoyed it. And I’m gonna take questions now.
So the first question: Is gene therapy possible? I think the answer is in cases where there are something called ROPers, or ROP kind of extended spectrum. Meaning some of these children may have worse ROP because they have an underlying fever or some other genetic condition. I think that those could be targeted for this. But unfortunately, there’s at least not yet has there been elucidated some genes that we can target for this and I think a lot of it has to do with how these infants are cared for and how their clinical course goes. And that can be very hard to determine going further. The next question is: If a child already has a stage 5 retinal detachment, is it possible to get treatment? The answer is yes. You can treat stage 5 retinal detachments. They are extremely hard to treat, and the visual outcomes are generally very poor. But there are some very skilled retina surgeons who specialize in pediatric retina and retinopathy or prematurity who can try, but the good outcome chances are very small. The role of the optometrist in screening was one question. And so, I think that it is — it’s? Hard. I think it depends on your area and level of expertise. In the United States, optometrists typically do not screen for retinopathy of prematurity because the level of acuity. But in — and that may be different in other resource — more resource-poor settings. I think this really is the role of the ophthalmologist. In your experience, how can clinicians optimize examinations of older, uncooperative infants while minimizing the need for examine under anesthesia? I think this is a great question. Thing all comes down to just like in surgery, your assistant. I typically when seeing outpatients have one of our nurses whose — who is a NICU nurse as well. We have two on our staff. Obviously that’s not available for everyone. But I think if you have a very good technician who is used to holding babies, I think having someone who is skilled and experienced in holding the children is very helpful. Swaddling is also very important. So, for the larger babies we will — even the bigger ones, we’ll just get a bigger blanket and we will swaddle them, very helpful. And sugar water can be helpful as well and a pacifier. Those are my big tips as to what I do. Also, as you get older and more experienced yourself, the other maneuver that I use that I was taught was the doll’s eye. So the less you have to use your depressor, the better. And if you doll’s eye, you can kind of get the eye into a different position and oftentimes they’ll bells or look in the direction that you want and you just kind of have to use their moving to help you out. Another question we have is for older babies, why not IV eye before laser? That’s a great question. It all depends on the baby itself. If a baby is sicker and will not tolerate a laser, that’s another reason to do an intravitreal injection. If the baby is older and has grown further out to a narrow degree, had they seen we had been younger and injected them, then if we have the ability to do the laser and cut down on the need for follow-ups and the outcomes would be similar, then we would consider laser. As I mentioned, I mentioned it’s very baby-specific. But if we can do a laser and kind of save the degree of follow-ups and there really won’t be much different, than doing an intravitreal injection before then, you might be kicking the can down the road. Also, depending on the level of neovascularization, there is always the risk of the Anti-VEGF crunch you can see in diabetics who have detachment. What it’s describing, when you give the Anti-VEGF agents, the stage 3 regresses so quickly that it can cause traction. It has been seen. It’s definitely been seen in diabetics more so. But it’s just one of the things we think about in these older babies. Also, as I mentioned, older babies can be a little bit harder to inject and bear down. And I’ve unfortunately seen them squirt the medicine out of the tract just by bearing down. There are things to consider, but those are the big ones. The next question is how many intravitreal injections? And it very much depends. So, babies who are these nano premature infants that we have seen often have been the ones who have required more than one intravitreal injection. And it’s — we theorize that it’s because they’re so young and they just have such active VEGF levels for so long. Once we treat them once, the second time we treat them again, we go through this idea of, you know, would a laser be more feasible? Would a laser cut down on follow-ups depending on where they are? Some of these nano premature infants, it’s after the reoccurrence of ROP after still in zone I. And obviously, we treat them with an injection. It becomes an area where you have to be very personalized to the patient. Our choice — so the next question is: What is the choice of intravitreal drug agent and what is the dosage and site of injection from the limbus? So our current choice aflibercept. Because it’s approved by the Federal drug Administration in the United States. That’s why we’re using it. We’ve had good success with it. Its dosage is .4 milligrams per on the 01mLs, which is about a fifth, I believe, of the adult’s dose. We like it because it comes in a pre-packaged syringe. And so therefore theoretically reduces the risk of infection. And then our site of injection is usually 1 millimeter from the limbus. So, you have to remember when you’re injecting in an infant’s eye with the lens takes up a significant portion of the eye itself. So, you’re injecting straight down rather than injecting towards the nerve like you would do an adult. And you’re very close to the limbus with the goal of not hitting the lens. The next question is about tele-screening, how to get the improved care by tele-screening as opposed to retinal findings that lead to treatment or follow-up, especially in older infants who are harder to image? This is a great question. I think the answer is twofold. One is the NICU teams that are doing our tele-screening — so the nurses and nurse practitioners doing this in the NICUs, they tend to improve over time. Their ability to get further and further out is really quite impressive. And then the other thing I would say is most peripheral findings that indicate a need for treatment you can see more centrally that there is something going on that would indicate the need for an in-person exam. So, if the vessels start to change in an older child and start to look a little bit more like plus disease, starts to, you know, just — you can kind of see as they go out if they’re getting more tortuous and dilated to suspect that there is active retinopathy whether you go further out. One of the things I tell our fellows when they’re doing in-person exams. Follow the vessels, if they start to look weird or funny or dilated and you don’t see ROP, you’re missing it. What I have seen on tele-screening, you can do the same. I think that some care in a lot of instances is better than none. I think our experience with tele-screening has been great because of our ability to get good pictures. I think the chances of missing peripheral findings are much smaller than you would think. So the next question was from a provider in Colombia who is in a rural practice and wanted to know how often premature and pediatric patients should be referred to an ophthalmologist after NICU discharge? They’re usually less sick in this patient setting. So, for pediatric ophthalmology follow-up, our usual protocol is to see them until their retinopathy of prematurity has resolved and do a repeat exam around 9 to 12 months looking at refraction, need for classes, development of amblyopia. If they look good on that exam and have a normal reaction, discharge them to the regular screening that goes on. The next question is about Vabysmo, I’m not aware of any Vabysmo trials for retinopathy of prematurity in particular. It’s a high dose inhibitor, I’m sorry, I’m not an adult retina provider. But with we don’t use that, we use the regular aflibercept, they’re babies. The trend is towards less-concentrated and lower dose agents. How long do you follow babies who have received only an Anti-VEGF injection? As I mentioned, our protocol at Children’s, when they’ve had Anti-VEGF injection, we follow them until they go for their exam under anesthesia and then have laser. And I will continue to follow them after the laser for the par or peripheral avascular retina. But this is very specific to Anti-VEGF-injected babies and usually see them once or twice a year from then on to monitor for changes. I think it’s really important that any baby who has been injected with an Anti-VEGF agent then have lifelong follow-up in retinopathy of prematurity. There was another question about Vabysmo, it does look like it’s a different antibody. I know it’s big in the adult retina world. So, I think — so I guess does not like it’s aflibercept. It’s something else. I would say we don’t use it yet. Do you always perform laser even if they are stabilized prophylactically for older babies treated with Anti-VEGF? Yes, but at much later. A year of life. Post 60 to 70 weeks gestation or later. It very much becomes when I can no longer examine them in the office. It’s part of that SAFER protocol, written by a number of ROP experts. And I highly recommend people checking out that they recommend in the protocol after 60 to 65 weeks taking them for an exam under anesthesia. Whether you do angiogram, you can see leakage, and active retinopathy of prematurity. And the case of later activations are enough to make us do as best as we can to prevent that from happening. The question on dexamethasone. What are the side effects? In the treatment trial, I should say retrospective cohort study published in ophthalmology, there were no side effects that were noted. But any steroid eye drop, steroid hypertension or glaucoma. Cataract formation if they’re on it for too long. The theorization of the systemic cortisol suppression is another one that’s important. Although with punctal occlusion, the risks of it are very, very low. So, those are the major side effects. We have another question about managing late recurrence after Anti-VEGF and ROP. Monitoring them, we continue to monitor them as any other ROP patient. Watch until they’re completely vascularized or for an exam under anesthesia. A reoccurrence is one thing. You can develop a reoccurrence, it can be stage 1 retinopathy. A reoccurrence, type 1 ROP, a repeat treatment, laser or injection. If it’s late, it’s usually laser because they have grown out so much. So, in terms of monitoring, I usually follow these patients every one to two weeks until you get through a time frame when they can reoccur, which, again, is depending on the agent you use around 8 to 12 weeks or so. I think 10 is kind of the sweet spot for bevacizumab and aflibercept. And it can — and then I will space it every two weeks and until they get older, sometimes even more until I can’t them anymore. Which Anti-VEGF agent, is currently recommended? You know, I think that there are differences across regions even in the United States. I don’t think there is one that has necessarily been deemed to be better than another. I think that it depends on what is the easiest for you to get access to. Obviously ranibizumab and aflibercept can be much more expensive. And there are risks and benefits, pre-filled syringes versus not versus compounding. The terms of bevacizumab, what dose can you give? Are they comfortable giving lower dose? And there are institutions where they’re using half dose bevacizumab from the BEATROP study. One is not necessarily superior. With ranibizumab, there’s a fairly high rate of recurrence. As long as you’re aware that it happens, watch out for. Has the benefits, too. And leaves the system so quickly. I think that it’s important to just know the agent that you have. But I don’t think one is necessarily recommended above the others right now. Another question: When to consider failure of Anti-VEGF and when to reinject? So again I go back to the ROP-1 study and some of the other studies that it’s around four days. You know, I think if you have not seen a significant improvement by four days then you need to consider something else. One of the things, one of my mentors taught me, you also have to remember kind of how quickly ROP is not gonna get worse in the sense that back before Anti-VEGF agents existed, these babies would be lasered and they would have to wait two weeks to see a result. Anti-VEGF agents have followed along with the rest of us in the world and made us a little bit more impatient to see results. So, at four days if you haven’t seen improvement, that’s when I would consider reinjecting. It’s extremely rare. I think in the last five years aye had one baby require a reinjection like this. And that child, interestingly enough, got injections in both eyes, was a 23 weeker. With one eye failed after four days or never really succeeded and so required re-injection with bevacizumab. And the other eye failed within about a month. So, there is something I think also about that patient. These children are very sick and very young and obviously very different. This is a great question. What is the role of the pediatrician in ROP prevention? It’s such a team effort. Some of my favorite colleagues are our NICU providers because of the amazing care that they take care — that they provide the babies. These children are so vulnerable and so sick. I think it’s such a important thing to discuss with the NICU in terms of how concerned we are about their eyes. You know? What level of retinopathy do they have? Is it progress something getting worse? Talk about their oxygen levels, their lung status. The pediatricians can really inform us on how sick these children are. And we can inform them on how sick their eyes are. I think the role is that it’s a team effort and that we couldn’t do what we without the excellent care that the neonatologists provide. Can babies who have been intubated for a long time develop this disease? The answer is, unfortunately, yes. It’s often those who are intubated for a long time have worse lung function, more bronchopulmonary dysplasia and there’s a link between those two. Obviously the inability to oxygenate the retina is a problem. These are babies who we need to watch very, very carefully. How long — another question about dexamethasone drops. How long would the dexamethasone drops be given for? Usually under four weeks. A lot of times it’s two weeks. I highly recommend checking out that paper, I believe it’s free on PubMed. And it can give you a protocol for using it. But the goal is for a short amount of time. Next question is: If bilateral Type-I ROP, whether do you inject the other eye? Do you wait a week or not? We generally do simultaneous injections for the — for both eyes. We prep the field as a sterile field. It’s essentially an operating room in the NICU. And the — so we have two separate sets of equipment and two separate sets of gloves. So, we treat them as two separate procedures. We prep both eyes individually when we do them. I would generally not wait a week for type 1 ROP in any eye. The treatment protocol in the United States is that it needs to be treated within 24 to 48 hours of discovery. If you wanted to wait a day and inject the next day, that could be a reasonable decision depending on your rates of injection. Fortunately in my practice they are quite low. But they can happen. And also depends on what you thought caused the infection. Another point-to-point out is one of the major drivers of endophthalmitis in adults, oral flora. Babies on C PAP, we try to have the nasal prongs in if they’re on positive pressure, because we don’t want the dirty air from the nose and mouth to be blown up with towards the eye. That can cause injection. There have been case reports of that as well. With that, I hope they helped and I think we’re gonna wrap up for today. I really appreciate everyone coming and would love talking about ROP.
Last Updated: March 25, 2026

16 thoughts on “Lecture: Retinopathy of Prematurity in 2026 – Evaluation, Management and Disease Burden”

  3. What an insightful lecture, clear and precise. I was attentive till the end! Dr. Gise is a wonderful, respectful speaker.

    Reply

Leave a Comment