Lecture: Treatment of Type 1 Retinopathy of Prematurity

During this lecture, Dr. Harper reviews comprehensive evidence-based approaches for treating and following Type I ROP in order to obtain the best possible results. Discussion includes treatment options including laser and anti-VEGF therapies as well as fluorescein angiogram management of post-treatment retinas.

Lecturer: Dr. C. Armitage Harper, Austin Retina, Texas, USA


DR HARPER: Good morning, everybody. My name is Dr. Armie Harper, and I’m really grateful to Orbis for allowing me to talk to you all today. Everything in here is my opinion, and how I treat babies on a daily basis. I’m in Texas, and I’m associated with two University of Texas Universities. If you have any questions, we have a question and answer session at the end. In addition, my email will be at the end and you’re always welcome to email me. So we’ll get started. Let’s see. So when we talk about ROP, it’s the leading cause, now, of childhood blindness throughout the world. There are approximately 50,000 babies in the world currently blind as a result of ROP. Even in the US, we still have about 600 babies that go blind every year. This is a largely preventable problem, but the combination of properly monitored oxygen supplementation, and then systemic disease prevention and appropriate treatment, once the type I ROP begins. We’ve been doing this for quite a while. We also have a small non-profit, and we want to thank our global partners, and you can see those listed there. We’re also really grateful to Project Orbis for allowing us this opportunity to share this information with the world. We’re committed to being the voice for every premature infant, to give them the likelihood of experiencing a full and healthy life. I’m a consultant for Regeneron, and I will be talking about off-label drug use, bevacizumab and ranibizumab in the US. It was approved by the European equivalent of the FDA. So there’s a couple different types of treatment modalities for ROP worldwide. And as you all know, cryotherapy was originally used. Started in Japan. Pioneered by Earl Palmer at the University of Oregon, and subsequently was fairly effective. After that, laser kind of took over ROP, and it is still an excellent treatment for type I ROP. Now we do things a little differently. Again, laser can be used, and is super effective in the right hands. We also use anti-VEGF therapy now. Anti-VEGF therapy — there are many different drugs now, that can be used for anti-VEGF therapy. But the two that we’ll talk about today are bevacizumab and ranibizumab. In addition, some people use anti-VEGF therapy plus laser, either concurrently, which means at the same time, or later. And I’ll discuss that, because that’s basically what we do. So with diode laser photocoagulation, we use the red laser and treat the avascular retina, in a dense pattern. The success rate is excellent, again. In this study in 2000, 0% of zone 1 eyes progressed, and only 3.8% in zone II eyes had a recurrence. This is an example of a type I ROP, before laser, and you can see the vascular tortuosity, and then the ridge of stage 3, and avascular retina. I put the picture in the right in there, because the baby nutritionally was doing very poorly and had two broken femurs as well, secondary to nutrition. And of course that goes along with retinopathy of prematurity. Is the nutritional aspect of things. So for type I ROP, this is two weeks status post laser. You can see the beginning regression of the stage 3 ridges. And then you see the vascular tortuosity regressing. It is not unusual to have a little hemorrhage at the edge after laser, and that occurs quite commonly, also with the intravitreal anti-VEGF treatments. So the problem with laser is it’s very time-consuming. It’s difficult. Requires mentored training. In the future, after this is accomplished, you’re destroying avascular retina, so you do have some decreased peripheral vision. Decreased night vision. High myopia, because of the anterior segment being a little bit deeper and having a different shape, and most people use general anesthesia, because the babies move around a lot, and it’s a little safer, if you can control the airway, in my opinion. But it does work well. This is one of my patients, now 17 years old. At this point. That was injected at 35 weeks, postmenstrual age. Currently the vision is 20/20. So you can see they can do quite well with this type of laser for a type I retinopathy of prematurity. But this is what I always talk about with my residents: We want to visualize the perfect drug for type I ROP. We want a drug that’s immediately effective for the cessation or stopping of the type I ROP. We want rapid dissolution from the systemic circulation. We don’t want a drug that has systemic side effects. We want normal VEGF levels to occur. Because they’re essential for neurogenesis, and also development of almost any organ in the body. So we need normal levels of serum VEGF. We want the normal vasculature to grow. In other words, when you apply a dose of intravitreal anti-VEGF, regular vasculature, and the retina stops growing for a while too, and we want that to continue on, if possible. And we don’t want abnormalities of normal macular vasculature. There are some excellent articles now out of Duke, which talk about the evidence — they talk about OCT changes after intravitreal anti-VEGF, and something always to look at. So far, we don’t have this drug. There are two common drugs that we use in ROP. Bevacizumab and ranibizumab. Bevacizumab, Avastin, obviously, is much less expensive. It was developed as an anticancer drug, and has been used for years and years for macular degeneration, other vascular diseases of the retina. Ranibizumab is a smaller molecule, and we’ll talk about that. It was developed specifically for the eye. FDA approved for use in adults, in the US, and in Europe, for babies. So the basic biochemistry of angiogenesis — I know you all know this. But I thought I would go over it very quickly. Angiogenesis is the growth of new blood vessels. It plays a role in normal development and maintenance of the vasculature. And an abnormal angiogenesis is indicated in a number of diseases, including proliferative retinopathies, which includes diabetes, branch retinal vein occlusions, and of course ROP. VEGF-A is one of the prime drivers. It’s a powerful stimulator of angiogenesis. It is secreted by a number of cells, including endothelial cells. It’s a homodimeric protein, and VEGF-A is a key mediator of angiogenesis. It binds and has signal transduction, gene expression, and you can get the vascular leakage in angiogenesis. So we want it blocking this particular protein. So when we talk about drug design, we want a really — a drug that fully penetrates all the retinal layers, from intravitreal injection. That has rapid systemic elimination. That has a low potential for cell or complement mediated cytotoxicity, and that binds to multiple VEGF isoforms. And so we want a smaller molecular weight, because we want to get into the eye, get out of the eye, and then get into the systemic circulation, and that’s super important. Also, we want a drug that has a high affinity for the VEGF isoforms that we’re looking to block. So when I thought about this, I wanted to design something, where we would have a road map, where we could stop at different waypoints, and talk about why we do what we do, why we think about the way we think, and how to get to the basic endpoint. And the endpoint that we want to get to, and our destination today, is the successful resolution of type I ROP. In other words, no progression to stage 4 or 5. With the caveat that we don’t want to cause any damage from what we’re doing systemically. So here’s a bunch of supporting articles. And you can pull this up later, because this is a recorded lecture. And you can have these. They’re excellent articles. Some of them are a little bit older. Some are brand-new. Just a few months ago. So something definitely to look at and pull up and use. So our first road map — and this is what I talked to my neonatologists about all the time — and this was pioneered by Shukla, and also the folks at Cleveland Clinic, talking about biphasic oxygen distribution. And we had STOP ROP before, which wasn’t really effective, but these more recent papers have shown great effectivity of controlling biphasic oxygen distribution. One of the things you have to have is an oxygen blender. And many of our other countries that don’t have the availability of blenders have high rates of ROP, or oxygen induced retinopathy, because the only way you can deliver oxygen to a baby is through the nasal cannula, and the wall oxygen. You can’t blend it with the air, and have a decreased sat rate. And I’ll talk about that a little bit as well. But Jonathan, Sears, and Shukla did a great job. I’ll talk about that here. It’s important to understand there are two phases to ROP. One is the hyperoxic phase, which leads to retinal vascular growth attenuation and vaso-obliteration. The vessels just stop growing. So early, when the baby is born to 33 or 34 weeks, we keep the saturation rate between 88 and 92%. That’s of course individualized for every baby, and this is really in control of our neonatologists, who I love to work with. And then the phase 2 is a hypoxic phase, where you have pathologic angiogenesis, in response to overexpression of the cytokines. In other words, the VEGF rises. And so we want a higher sat rate at that time, to downregulate everything. And what I do — the paper is complicated, but what I do, in order to make it easy for me, is that any time I see any stage 2, or beginning vascular tortuosity, I know the VEGF levels are up. I’ll ask my neonatologists to raise the sat rates to 96 to 99. You don’t want to go over 99, because you want — if you go over 100, you don’t really know how much saturated oxygen is in the baby’s system. And that way, we can downregulate. We’re just using oxygen, basically, as a drug. It works really, really well. And sometimes just like you would inject anti-VEGF, you’ll see regression of the ROP that you had seen previously, by using this type of model. They demonstrated in their parameters greater than 50% reduction in type I ROP. Treatable ROP. Using a biphasic model. Now, in their model, they actually turn up the oxygen saturation on all the babies they have, and we don’t do that. It’s a more selective method. And we are collecting data on that. But it’s my gut that this is doing a really good job, in helping us have less treatable type I ROP. Our second waypoint. We know that bevacizumab, the drug that’s most commonly used, at the dose most commonly used, significantly reduces plasma VEGF in premature infants after intravitreal injection. And this was shown in Retina in 2015, and there were 11 eyes from 6 infants who received intravitreal injections, and they looked at the plasma concentrations of VEGF. What’s measured in the serum. And it’s very difficult to measure in the serum accurately. But they did feel that the plasma concentration of VEGF didn’t return to original levels in the samples until eight weeks after the injection. So that’s something important to think about. If you only have Avastin to use as a drug, then that’s what you have to use as a drug, but it’s something to consider for the future, and something to push your people for, to be able to obtain other drugs as they come out in the near future. We want normal VEGF levels in these infants. So their other organs and other systems can grow. So the RAINBOW Trial was a trial that was done using Lucentis, the results of which were published just a few months ago. And 225 infants were enrolled. And there were 75 bilateral injections of 0.1 milligram of ranibizumab. And 73 of 0.2 milligram. And then there were 69 bilateral retinal laser photocoagulations. And basically they measured the VEGF levels and came to the conclusion that there was no significant difference in the plasma free concentrations of VEGF between any groups or over time. In other words, it didn’t matter if you lasered or used 0.2 or 0.1. There were no significant differences in plasma VEGF. So that may mean it’s really something important for the future of these kids. The median elimination time — and this is calculated on our computer model — is 5.6 days from the eye and 0.3 days from the serum. So our third waypoint. The neurodevelopmental effects of anti-VEGF treatment are still in question. And that’s super important to recognize also. It’s very difficult to assess neurodevelopmental issues in children that are born premature, because they have so many other things which they have challenges for. But this was an article published in Pediatrics. And they compared the babies that had intravitreal bevacizumab, versus laser treatments. And they looked at motor, language, and cognitive. And they said that the odds of the severe motor developmental skills — neurodevelopmental skills — were 3.1 times higher in the Avastin group. It’s a small chart review. This is limited, the study. Some data was obtained from parental interviews, which is always not the best choice. More infants in the bevacizumab group had zone 1 disease. In other words, were probably sicker babies. Scores were not calculated if the children were severely developmentally delayed. The neurodevelopmental outcomes — this was another study, recently published in OSLI, by Dr. Grieve, where again, they looked at the Bailey scores at two years. The average Bailey scores for cognition, language, and motor did not differ between the two groups. So we don’t really know at this point what neurodevelopmental changes can occur with any of these drugs. There’s a lot of literature. The RAINBOW Trial actually is doing a prospective trial, up to five years, to look at this question also, and we won’t have that data for a few years yet. We’re at year four in that RAINBOW Trial at this point, waiting for that data to come in. Waypoint number four. This is super important. This was something that was initially not recognized, when a previous trial was done, but most infants, when injected with anti-VEGF, are peripherally avascular, and require careful monitoring in perpetuity, or in our case, laser to the avascular area. The issue is: When we first injected Avastin, and you looked in there, it’s very difficult to tell, especially when babies are highly pigmented, where the demarcation area is, between vascular and avascular retina. So when you started looking with fluorescein angiogram, you can actually see the demarcation lines. So that’s something that we started to think about, and have done some work on, and published some papers. We looked at — we did a retrospective evaluation. Including RetCam Fundus and fluorescein angiograms of 16 neonates who received ranibizumab for type I disease, between April 13 — April 2013 and January of 2015. We looked for maturity to zone 3, based on a standard ETROP definition, vascular blunting, vascular loops, vascular dilatation, vascular dropout, and leakage. This was a 25 week old male, 690 grams. We treated with 0.15 milligrams ranibizumab for type I ROP at about 37. We performed a fluorescein angiogram at 60 weeks. This is what it looked like. And what you can see here is you can see vascular fluorescein leakage at the edge. You can see the big swath of avascular retinas in both eyes, and babies unfortunately at this point — their retina really, the blood vessels don’t advance very much past this point. They have basically stopped growing. But that vascular leakage does indicate some activity. And we don’t know what that means for the future. But we’ll talk about how to mitigate risk, in my opinion. So you also see only growth at that anterior edge of zone 2. Not into zone 3. Case study number two. This is a former 23 week old male twin, 320 grams, who was treated with 0.15 milligrams of ranibizumab at 37. FA was performed at 99. This was when I was just starting to perform FAs, so we went back and picked up all these kids. That’s why some are a little bit different, of when we performed the FAs. Now I perform almost all FAs at 60 weeks, if possible. So this is — remember, this is at 99 weeks. And you can see there’s only growth to the anterior edge of zone 2. You see some vascular loops as well. And then you can see some vascular fluorescein leakage, indicating activity. And our results — we only had 50% of the eyes that actually reached zone 3. In other words, the blood vessels only grew to zone 3. 90% of the eyes had vascular blunting, 93.4% of eyes had vascular loops. 40% had that vascular leakage. In other words, that little bit of activity. 90% of eyes had vascular dilatation. And 93.4% of eyes had capillary dropout. And so our conclusion was that ranibizumab is effective for the initial cessation of type I ROP, but full vascularization was only achieved in 50% of these babies. And the majority of eyes had some vascular anomalies. What this says is necessitate long-term observation of the avascular peripheral retina. And if observation is not possible, consideration should be given to photocoagulation in these avascular areas. We know from looking at different studies — and my friend Manny Cheng is looking at this, where you have children who have been injected with anti-VEGF, and if you look at them as teenagers, many of them have some abnormalities, such as lash degeneration, retinal tears, and even retinal detachments. So waypoint number five is: The premature infants who develop ROP are more likely to have an FZD4 variant which may predispose them to peripheral vascular anomalies. You’re all familiar with FZD4 and other genetic anomalies like TSPAN12, which are associated with FEVR. This is a pediatric retinal disease characterized by the appearance of an ROP-like syndrome in the absence of premature birth that can cause devastating blindness and can be active throughout life. Wnt signaling is an ubiquitous pathway that modulates cellular and tissue differentiation. In regards to eye development, a particular Wnt pathway, Norrin-FZD4 has been identified as playing a role. And a disproportionate percentage of infants and children with a diagnosis of ROP or FEVR have this variant, 7.5% in the ROP population compared with the general population. And this study was performed by Kim Drenser at Royal Oak. More recently, in the Journal of Ophthalmology, in 2020, they looked at many children. They found 9 infants with FEVR-related eye disease, causing gene mutations, that had ROP. In the 18 eyes of these 9 patients, 9 of the eyes exhibited severe ROP. So in conclusion, the LRP5, FZD4, NDP, and TSPAN12 may play a role in the pathogenesis of ROP and can cause atypical ROP or preterm FEVR. The fundus lesions of ROP patients with disease causing gene mutations are more serious and may be FEVR combined with ROP. ROPER, as defined by Berrocal. So it’s important to recognize that even though you may not know, and you may not have the genetic ability to test for this, if you inject a baby with ROP and they have one of these genes that may put them at risk in the future for neovascularization, again, consider lasering the periphery. And laser the periphery in these babies, and we can talk about this a little bit. It’s not like confluent laser as we would treat a type I ROP. Paul Kuram has recently published a paper, showing that we don’t have to have confluent laser. It’s more non-confluent, more PRP laser, but I would advocate fairly heavy laser up against the vascular/avascular zone, because we know that’s where VEGF is produced. And again, these FZD4 and ROP and other genes — if an infant has this genetic anomaly and has been treated for type I ROP, is he or she at risk for reactivation after anti-VEGF treatment? At this point, nobody really knows, but it’s something to keep in the back of your mind always. So waypoint number 6: Premature infants injected with anti-VEGF drugs, bevacizumab or ranibizumab, may reactivate and develop retinal detachments. Super important to know. It is not a one and done phenomenon. It doesn’t work like that. Just like in your diabetic patients, or your macular degeneration patients, these drugs will bind to the protein VEGF, but then they fall off. More VEGF is produced. And babies can reactivate. We know from 2018, Dr. Gunicolis’s patient, that progressive retinal detachments can occur in this condition. Fiber vascular contraction and transactional retinal detachment are recognized complications associated with the use of these drugs. So I’ll show you some pictures. This was an international multicenter interventional retrospective case series. 35 eyes with retinal detachment from 23 infants were included. Inclusion required the antivascular endothelial growth factor treatment for type I ROP, with progression to tractional retinal detachments. These are pictures from the paper. And as you all know, these are very, very difficult to fix, once these do occur. So my case — this is a former 23 week old and 3/7 week old infant injected three times elsewhere with bilateral intravitreal Avastin. First seen by me at age 59 weeks of age, and you can see what’s happened since the injection. So in other words, it can recur. You need to pay attention. And if it does recur or you start to see you have the options of treating with another anti-VEGF, but laser is definitive, and it works well. And once the babies have grown out of zone I, then laser is probably the most appropriate treatment, in my hands, at this point. This one, unfortunately, required vitrectomy. This was the other eye. Which had a stage 5 retinal detachment at this point. So case number two is a former 24 week old infant, injected four times with bilateral intravitreal bevacizumab elsewhere, and was first seen by me at 18 months of age. And the right eye was already phthisical, with a funnel 5b retinal detachment. And the left eye fortunately still had an attached retina, and you can see the abnormalities of the vasculature, centrally, in this left macula, and you can see the activity on the edge, and I lasered this extensively, and you can see the regression, which occurred in the color photograph on the right. And we fortunately were able to help this child. Waypoint number 7. The combined treatment with ranibizumab and laser or laser alone is effective in treating micropremature infants, less than 750 grams with type I ROP. We really wanted to look at small babies. In other words, less than 750 grams. Not everyone has these babies. But this is something we were really interested to see. What our results were, when we went back and looked at all the babies that we had seen, who were super small. And as you know, the smaller the babies are, the higher the percentage of folks that get type I ROP, and as you see from the photographs on the bottom, this was previously anti-VEGF treated eye, by my partner on the right with beautiful laser. So this was a 23 week old and 4/7 male, 550 grams, 35 weeks and 42 weeks had intravitreal ranibizumab, and the dose at that time was 0.15 milligrams. And a lot of times, you will always see this, and Tony Capone will always talk about flat neovascularization occurring in zone I. Super important to pay attention to that. It’s not gonna look like a stage 3 normally. So if you see that, remember to put in your mind that flat neovascularization can be awful as well, and will cause really severe problems if you don’t laser it or treat it appropriately. This was the FA at 50 weeks, and again, this is that same laser on the right, that my partner Dr. Ryan Young did, to save this child. And really, this is still in zone I, and this was after a couple injections of anti-VEGF. Some babies are so sick, and if they continue to be sick, they’re gonna have sick retinas as well, and you have to treat that appropriately. So we did a retrospective review. And we looked at 100 babies. They were less than 750 grams. The time to regression was defined clinically. And at that time, we used 0.15 milligrams of ranibizumab, subsequently have gone to 0.2 milligrams, based on the results of the RAINBOW Trial. Anyway, we had 100 neonates, 63% were male, 597 grams was the mean birth weight, from 310 to 750, and the mean gestational age was 24.2 weeks. Our results — and remember, I was lasering more at this time — we had 100 neonates, 40 neonates developed type I ROP, 40%, so pretty high at that time. We did 21 just regular laser, and 19 we did ranibizumab, followed by 19 lasers, at 60 weeks of age, if possible. I choose 60 weeks, because at that time, the lungs have matured really well. The anesthesiologist usually will let them go home from the hospital on the same day. Which is super important. And we feel like the retinas stop growing at that point, so the laser can take care of the rest, where the vessels haven’t grown out to, in the avascular area. So our comorbidities that we looked at — we wanted to look at that as well. There’s no correlation between the instance of systemic comorbidities and ROP, but there were a lot of comorbidities. And these are the ones that we usually expect. And by the way, many of these things can be partially resolved by using the appropriate oxygen levels. In the NICU. None of our patients progressed to stage 4 or 5. 100% success in that way. And this is what we’re effectively looking for. And this is why we do it the way we do it. There was no correlation of systemic comorbidities. They did have a high rate of ROP at this time. And the use of peripheral ablative laser still, with or without ranibizumab, was able to prevent progression past stage 3 in all cases. No infants progressed to stage 4 or 5. And on your right of your screen, this is Waylon, a 320 gram baby that we treated. So waypoint number 8. This is a safer technique, and I’m gonna talk about this. We published this in a paper in OSLI. The SAFER technique for the treatment of type I ROP minimizes risk and really optimizes the outcome, and we were trying to develop a technique where we could allow everyone to take care of these babies super safely. So the injection of intravitreal anti-VEGF necessitates a practical technique that is dependable and standardized, in order to help ensure the safest outcome possible. And so we put it all together. The short needle, 32 gauge, 4 millimeter needle. We found this needle because the original needles that were used were very long, 7.9, 8 millimeter needles. If you take a needle like that, you can put it all the way across a baby’s eye, you can put it through the lens, and you can cause a lot of damage, if you sink it all the way to the hub. So there had to be a better way to do this. So we found a 4 millimeter needle that we’ve subsequently used. And I’ll show you the manufacturer of that needle. And we looked — Linda Cernicchiaro wrote a paper, a multicenter retrospective trial. When you look at where to inject the needle, you can’t inject at 3 to 4 millimeters as you would in an adult eye. The baby’s anatomy is completely different. Looking at a nomogram from Manny Cheng, we have to inject 0.75 to 1 millimeter from the limbus. At this point, it’s difficult to have any complicated, and we talk about that. We use an antiseptic or antibiotic. There have been reports in the literature, even in the RAINBOW Trial, where they had endophthalmitis from injection, for type I ROP. We used Betadine 10%. I leave it on the eye for about five minutes, and then after the injection, I apply Betadine again. I don’t use antibiotics for the antibiotic prophylactic part. I will use an antibiotic-steroid combination afterwards, just because the Betadine itself is so caustic. But Betadine has been around for a long time, and is a great drug, and works well for us in preventing any cases of endophthalmitis. The F in SAFER is super important. If you inject an adult, we know that adults will tell us if they have pain in their eye, or they can’t see, or anything like that. They’re gonna call our office, or come back into your office, and say: Hey, this is not working for me. Something is going on. So if they get an infection, they know. Babies can’t tell you that. About the only thing that can happen is they can develop a fever or they’ll get fussy, and they won’t eat, but some of them are on parenteral feeds in the NICU. So it’s very important that you go back to the NICU in 48 to 72 hours to make sure no infection has occurred. So E stands for extra attention to detail. We use sterile instruments. We use the ora nomogram that I described earlier, and look for the presence of conjunctivitis. If the baby has conjunctivitis, you should not be treating at that time. Need to start them on some antibiotics, and make that clear, or alternatively, you would perform laser at that time. And then we recheck every one to two weeks, following anti-VEGF, and then we don’t do fluorescein angiogram every one to two weeks, but we do it between 60 to 65 weeks on every baby I treat with intravitreal anti-VEGF. Rechecking every one to two weeks can ensure that you don’t miss anything and have a recurrence. If there’s anything active, in other words, if I see anything that is just not plain avascular, with stage zero ROP, I come back every week and see these kids. And they come to my office every week. Until they’re ready for surgery. At 60 weeks. 60 week old baby — I have that also, because babies at that point are very difficult to examine, so it’s super important that those babies are looked at. Here’s the little needles from TSK, and you can see the size of that needle. I put this box on there, so if anybody wants to order them, they can do that. So the background of the SAFER technique, and we talked about this a little bit, the BEAT-ROP 2011. It was a 7.93 millimeter. Even in the RAINBOW trial, it was a 7 millimeter needle. Those are huge needles to use in these babies. The gauges are a little different. We used a 32. When we look at the literature, we don’t find any evidence that there’s any less endophthalmitis in adult patients with a smaller gauge needle. Whether it’s a 30, 31, or 32. I do think that a 32-gauge needle in an adult causes less discomfort, and I kind of use that for my technique, for babies as well, and I want to think about that. And the RAINBOW study trial is published at this point. So in 2017, Lauren Wright described a technique for intravitreal injection with a 32 gauge 4 millimeter needle with concomitant pathology. She took an autopsy infant eye, 56 week old, and we looked at it with a 12 millimeter needle, and you could penetrate the lens or retina, go all the way across the other side of the eye, so it really gave us an idea of what this really looked like inside the eye. However, the 4 millimeter needle, as you can see on your right, super small. It’s very difficult to damage the eye with this needle. And it effectively penetrates into the vitreous cavity. And as you well know, these needles — the sclera at this point is about a millimeter, 0.75 millimeters thick. So it’s easy for the needle to get in there. So Linda Cernicchiaro wrote a great paper, when she looked at a combination of all the results, along with Bascom Palmer. And we’ve used a short needle since 2014. These are all the centers that I see on the right. We see in Austin 550 new babies a year. And Bascom Palmer sees a huge amount of babies as well. And the type I ROP we treated with the 32 gauge, 4 millimeter needle, and you can see how small that needle is. It’s a single use hypodermic sterile, and 0.5 to 1.5 millimeters from the limbus, depending on the patient’s chronological age. And the results, the demographics, Bascom Palmer, they had — the gestational age and our gestational age was almost exactly the same. The birth weight in grams, they’re a little smaller in Bascom Palmer and inject a little bit later. They injected Avastin and we injected ranibizumab. We had no complications in any of the groups. Cataract, endophthalmitis, vitreous hemorrhage, or anything else. Retinal detachments. So this was the first step demonstrating the safety with the 32 gauge needle. The benefit it offers to small eyes — it’s lower risk of iatrogenic breaks, and a lower risk of lenticular injury. So we talk about antibiotics. You can use 5 to 10% Betadine. Most of the time the Betadine comes as a 5%. I use 10%. Just because I worry. And you can consider the antibiotic-steroid, again, to decrease the inflammation, secondary to antiseptic. Prevention of infection during this procedure is paramount. It’s super important. Topical Betadine has been used for almost 40 years. It remains the most effective bactericidal agent. Following the 48 to 72 hours, you’ve got to look. Babies, again, can’t tell you whether they’ve gotten an infection. So it’s super important to know if they do. And then here’s our setup. Here’s what it looks like. We don’t use this light to assess where the ora is. I’m just giving you an example of this on the right. But here’s the setup that we use in general. We use masks as well. Everybody is using masks for everything. And then we recheck them one to two weeks, following treatment, and we perform our fluorescein angiogram at 60 to 65 weeks in the operating room under general anesthesia. And again, recurrence can happen up to 5 years later, so it’s super important to pay attention. Here’s our ROP injection video. And now what we also do with this… We use this technique — I put a little towel over the nose and over the mouth, to prevent particulate from heading up that way. And then we also place another drop of Betadine in the baby’s eye. And you can see that one more time. You can see that little short needle there in this little syringe. So the destination is the successful resolution of type I ROP, no progression to stage 4 or 5. There’s a call to action, and this is not the end of the story. This is how I do it. You have to use what you do, and if you’re successful with the technique that you use, then by all means, continue that way. But there are many, many babies in the world that are going blind from ROP, and our goal is to cause resolution, and make that happen for people. We want that to go away. We are — through my non-profit, Small World Vision, working on oxygen blenders, in order to provide a more reasonable cost basis for a blender that can help these NICUs and help these babies survive and thrive. Here’s more references, if you’re interested to getting to the end. And then I really want to thank all of you for listening this morning. Here’s my email, if you would like. I would also like to thank Jessica Goldstein, my assistant, for really helping me put this whole thing together, and really collecting all this data. So we have a few questions. I don’t use bevacizumab. The first question is: Do you use bevacizumab and what dose? Most people use 0.625 bevacizumab. And that has been super successful for almost all people. There recently has been a de-escalation trial, using very microdoses of bevacizumab, but it has a 40% failure rate, and they had several retinal detachments, and even a cataract, so I would caution the use of very microdoses of bevacizumab. And again, I don’t use bevacizumab, so I don’t prepare it by myself. Also, with microdosing, it’s very difficult. Because you would have to develop it under the hood, and the risk of infection is very high, so I certainly would not do that. Next question is: For how long do you follow the babies treated with anti-VEGF? We follow the babies, again, to 60 weeks, and then I take them to the OR and do that. And treat them with a laser. Following that, I’m a retina surgeon, so I’m certainly — my refraction is lights on and lights off. So I send them to the pediatric ophthalmologist, my great colleagues, to take care of them, for their refractions, if they have any. Recently we’ve done a trial and submitted a paper where we actually looked at the refractive errors for treatment with Lucentis, followed by laser, and we found that we had no refractive errors, compared to the average, so it does work well in allowing the anterior segment to develop. One question is: What is the recurrence rate after bevacizumab versus ranibizumab. The recurrence rate may be higher. We don’t really know. I personally don’t have a higher recurrence rate with ranibizumab, especially at the 0.2 dose. Bevacizumab, again, we’re still trying to figure out exactly what the best dose is. Next question is: Is there any development of macular edema, as a complication of laser treated babies? The answer is yes, they do develop macular edema, but it almost always resolves spontaneously. If you treat too heavily with laser, and that’s an important thing to talk about, you sometimes will get exudative detachments, and we have to give them systemic steroids, to allow them to resolve, and so it’s important to treat where you have — if you’re treating a baby with type I ROP, you want to use confluent laser and you don’t want a bright white spot. You want a gray spot, because those will light up laser and become more confluent. But it needs to be near confluent. Kind of a gray spot. Not white. And you’ll avoid the amount of macular edema you’re talking about. I think the macular edema resolves very quickly. I don’t think it’s gonna result in any amblyopia. And again, there have been some good studies out of Duke, looking at the OCT changes after laser. The ROP patients — why are they myopic? Premature babies in general tend to be more myopic than the average term babies. That’s one thing. Second thing: If you laser heavily and posteriorly for type I ROP, the anterior segment in these babies is not the right shape, and it causes a myopia. Do you treat type I ROP? That’s an excellent question. We treat all babies with type I ROP. That’s the definition of ETROP. And if you see it, you need to treat it. There are some cases of type II ROP that’s persistent. And Edward Wood from Stanford is looking at this question, trying to figure out which of these we need to treat. But there are some with persistent neovascularization that I will treat with type II. It’s pretty rare that we do that, though, but there are some. I have a feeling that some of these babies may have — may be a ROPER baby, in other words, have the genes for FEVR as well. Not sure about that. But I have a feeling that is the case. So a little bit different. Also, there can be asymmetric babies with that FEVR gene that require treatment. What is the safe measurements for giving anti-VEGF in ROP in neonates? There’s two things. One is where do you inject. And that’s about a millimeter from the limbus. And then the dosages — one of the things that’s difficult is the dosages are very difficult to see in a normal 1cc syringe. And that’s always been a source of frustration for everyone. So we’re trying to develop a new syringe that has a much smaller barrel. But that’s been very difficult so far. But we’re trying to do that as well. For type I, for stage 3, zone 3 ROP, when do you treat it? Well, it depends. If you have the definition of type I ROP, which is any vascular tortuosity, with stage 3, then you treat it. If you can’t follow this baby anymore, and you have stage 3, zone 3, then it probably should be treated. I would caution you that treating with anti-VEGF late — because you can develop a crunch phenomenon. So at that point, laser may be your best option. You’re not going to induce myopia when it’s stage 3. The vast majority of ROP, remember, goes away. So it does resolve by itself. Almost all of it resolves. Out of all the babies we see, the 550 new babies a year, we only treat about 3% of those babies. And those are all the babies less than 1500 grams. So it’s something to consider as well. This is rare, that we end up treating them, but when we do, this is the information I’m giving you. Do you ever inject anti-VEGF and laser at the same time? So I don’t usually do that. I think that if you inject anti-VEGF… Well, there are two reasons. One is it works really well to get rid of that initial ROP. And then secondarily, I worry, because if you place laser at the same time, it’s going to egress into the systemic circulation much faster, because you’ve disrupted that barrier, which allows it to stay in the vitreous cavity. So I don’t usually do that. Have you ever injected more than once in babies? Yes. It is rare, but babies sometimes can be so sick that there’s no way that they can undergo general anesthesia. And that one baby that I showed you the picture of, he was injected twice with Lucentis. But he had severe disease. He was an IUGR kid, and was super small, at 320 grams. So when you see a baby like that, you need to really confer with your neonatologist, and figure out what’s going on with the baby systemically. Because if they have lung problems, if they have gut problems, if they’ve had NEC, if any of those things are happening, you should be on alert for development of ROP. In addition, you all know the WIN-ROP study. If any of your babies are falling off the growth curve, they’re going to develop ROP, so it’s important to look at what’s going on with all the babies. And any time a baby is in the NICU, you need to sit down with the neonatologist and have a conference so they can communicate to the parents what’s going on with the baby every time. Is there any evidence of systemic effects of anti-VEGF? The answer is the jury is still out. We still don’t really know. We know that, for instance, we have seen evidence that the dose of bevacizumab that’s currently used by most people does cause some systemic suppression of serum anti-VEGF. And we know that from the RAINBOW trial, they did not see any serum anti-VEGF suppression. Or serum VEGF suppression. So it’s super important to pay attention to that. And I think we’ll know in the future, but we’re not really sure quite yet. And the question is: Do you give laser treatments as first line therapy at all? I will do that sometimes. It depends. Sometimes some of our kids… I know that I may lose them. They may leave the country. They may have a social situation, where I’m worried. And I have to make a decision, whether to give anti-VEGF, or give laser, and I will do laser. In addition, if a baby is older, and I already know the baby’s retinas have grown out pretty far in zone 2, almost to zone 3, but still need treatment, then laser is a good option for this baby, and if they’re healthy enough for the general anesthetic. You can do laser without general anesthetic, but I like airway control. I don’t like to have to rapidly intubate somebody, or have the fellows do that, because it is very distressing to the baby and to everybody around. So I do like airway control. It is doable. There was a study that was shown that if you do intubate and have control, you’ll do a little bit better job with the laser. And there are other anti-VEGFs that are coming on the market. There is — Regeneron has a trial using Eylea, in a very small dose, and that one is ongoing, and there’s others, one of which is used in other countries, such as China, though not in the US, and there are other drugs in the pipeline as well. So if ROP does not regress, how many times can you give intravitreal anti-VEGF? Super important question. The problem is: If you give more than one, you give two — you don’t know what the systemic effect is, and that’s a good follow-up question to the one we had. So if I give one, and it recurs, usually it’ll regress, but if it recurs — remember, for instance, in the one de-escalation trial, they only looked at the recurrence rate at 4 weeks. Any anti-VEGF, if you inject, won’t recur usually at 4 weeks, unless you’ve missed the window of opportunity. It usually recurs at 6 to 8 weeks, which is when we know it wears off in adult eyes, so it’s super important to pay attention. If you have laser at that time, you have recurrence, and the baby’s retinas have grown, then it’s time to do the laser. And that’s kind of where we are. I have had no experience with Eylea injection. Eylea is something, again, which is investigational. And that will be probably looking at something for the future. It’s a very small dose of Eylea. That is being used. So follow up these patients. Basically if you have an ROP kid, you need to follow up the patients forever. Because they’re going to have risks of retinal detachment. Sometimes they have vitreous hemorrhages. That occur at the avascular-vascular junction. So it’s important to have you or the pediatric ophthalmologist follow up with them for the rest of their lives. They’re gonna need glasses. They’re gonna need contact lenses sometimes, when they’re older. So it’s super important. How many high risk babies develop ROP? In the sense of anybody less than 1500 grams — our rate is 3 or 4%. Some other countries we know sometimes — in other countries, we can have rates of 40, 50, or 60% in babies that are much larger, because of the oxygen delivery systems that they have. Anti-VEGF injection performed in the NICU, or at the bedside? I usually perform them at the bedside in our institutions. But whatever people are comfortable with I would be comfortable with. Our neonatologists are excellent. And we have a very good working relationship. And it is a topical procedure. You’re using topical anesthesia, for injection, and pretty simple. I use a lot of Betadine. And the topical anesthesia — sometimes they give them a little Versed, but usually we don’t give them anything, just like you would do in your clinic. ROP occurs — I would encourage you to look into early treatment for retinopathy of prematurity study, that’s the classification we use, and it’s an excellent study and an excellent thing, but basically if you have plus disease and any stage 2 or stage 3, then you need to treat. The plus disease also can look like — if you have a baby with pulmonary hypertension, you can get some vascular tortuosity. So you need to differentiate that from that. Is there any reason to measure IOP in anti-VEGF, and what is the reason? I don’t measure intraocular pressure. We do look at the optic nerve to see if there’s pulsation. I think it’s very difficult, because the baby is squeezing, and you’ll have the lid speculum in there, so to get an accurate reading after injection is very, very difficult. And I answered this question. Do you perform anti-VEGF with local anesthesia or sedation? It’s just topical. And again, if the neonatologist wants to give them Versed, they can do that. I always have the charge nurse — and usually will have the respirator right there to resuscitate these babies if necessary, but I can’t remember having that happen before. Do babies have vessels in zone 3? Almost ’til the ora with leakage — need treatment? Very good question. I think if you’re within a disc diameter, if you have leakage, and you can see that on the fluorescein angiogram, I think you need to treat it. If it’s just avascular and you’re within a disc diameter of the ora, you’re probably gonna be okay. You can place some light lasers into there anyway, and I think these babies later will develop some thinning of the retina, and possibly lattice, so you need to consider that. We don’t really know yet. But this is what we do to be super safe. So according to your experience, which causes more myopic eyes? Babies treated with anti-VEGF only or laser only? That’s really been answered in the literature. And laser definitely causes more myopia than the anti-VEGF injections. And again, when I looked at my data, looking at anti-VEGF, plus laser later, we don’t have any induced myopia. Stage 4 — do you use anti-VEGF? No. If you develop stage 4 and you have traction, you do vitrectomy. That’s the only way you’re going to rescue that. And I worry about the traction in that case, again, causing crunch, pulling that traction off. Please share the laser parameters. This is difficult, because all lasers are a little bit different. The lasers — you have a newer laser, you have to use less power. If you have a highly pigmented baby’s retina, use less power, because the laser spot is gonna be a little hotter. So you have to be careful. And again, it’s titrated to a gray spot. I use the 28 diopter lens. We use a 100 millisecond interval. And then you probably need to put at least 1500 spots. In the baby. I don’t avoid the horizontal raphe, because that is the area of most activity in general. And if you avoid that, you are setting yourself up for a little bit of a disaster, because that’s where the recurrence rate can occur. I don’t have any experience using biosimilars in babies. Only in adults. In investigational trials. Anyway, thank you all so much, and if you have any other questions, please don’t hesitate in contacting us. I’m happy to answer. I’m happy to be of help to anybody in the community, and thank you so much to my staff, and thank you to Lawrence, who is our tech extraordinaire for Cybersight. Thank you so much. Appreciate it.

Download Slides


October 16, 2020

Last Updated: October 31, 2022

2 thoughts on “Lecture: Treatment of Type 1 Retinopathy of Prematurity”

Leave a Comment