Journal of Pediatric Ophthalmology and Strabismus

Eye to Eye 

Management of Recurrent Retinopathy of Prematurity Following Anti-VEGF Injection

Amit R. Bhatt, MD; R. V. Paul Chan, MD; James D. Reynolds, MD; Rudolph S. Wagner, MD

Abstract

Wagner: A premature male neonate born at 23 weeks of gestational age developed zone I acute posterior retinopathy of prematurity (APROP) at 32 weeks of age. He was treated with injections of 0.2 mg of ranibizumab. The neovascularization resolved within days and a 360-degree vascularization process began with what I describe as small, multiple, radial vessels progressing to the periphery 360 degrees. There was a recurrence at 42 weeks of gestational age. There now was a new onset of plus disease and neovascularization along the previously advancing edge, now into zone II nasally and zone III temporally. Is this an expected finding in similar cases, and how would you manage it?

Chan: I don't think it's unexpected to see recurrence. However, one of the issues that we struggle with as a community is how to define recurrence. The screening criteria, recommendations for when to treat, and treatment guidelines have not been well established by consensus.

Reynolds: That's an excellent point. Even the terminology varies from group to group.

Chan: When you look at the literature, the definition of recurrence after anti-VEGF treatment has not been well established. In this case, the questions are whether the examiner saw a true occurrence of plus disease after anti-VEGF treatment and whether our threshold for plus disease is different after anti-VEGF treatment. If there is definitely neovascularization, then I think that treatment or identifying this as recurrence may seem reasonable. The other issue is how are you identifying the neovascularization?

Wagner: It's observation.

Chan: Are you only seeing some neovascularization at the interface? Did you do a fluorescein angiogram? I will almost always recommend fluorescein angiography at some point for these children. This is especially true if I suspect recurrence, to identify the leakage and neovascularization. I've also encountered ischemia on fluorescein angiography that isn't always apparent on a clinical examination with indirect ophthalmoscopy.

Reynolds: The bottom line is that we do not know the natural history of ROP after anti-vascular endothelial growth factor (VEGF) treatment. No study has reported natural history data on these patients, and that's what we really need. We don't know that the current neovascularization is going to result in traction if untreated. Everybody thinks this is going to be just like untreated zone II disease, but we don't know that. We're all in treatment mode. We don't know that we have to be as aggressive in recurrences as we do in initial treatment. The location alone means that the total retinal detachment rate is going to be lower and we know the detachment rate is higher for zone I. In this case, the recurrence is in zone II and we know that the blindness rate is lower for these patients.

Bhatt: I agree that defining recurrence is important. Many infants treated with a lower dose have some type of recurrence, but not all of them require treatment. We need to decide when we should intervene and when we should not.

Wagner: In this particular case, what would be your treatment?

Bhatt: Our institution has treated approximately 100 eyes with injection and we've had to re-treat 10 to 12 of them at this point. The question in my mind was whether to try a repeat injection or laser treatment. In my experience, the repeat injection has not been that successful. Others have had success. One of the issues is that you don't really get further retinal growth after a second injection. The 4 eyes that I've attempted to re-inject all ended up progressing and needing laser treatment to eventually improve or resolve. I prefer laser treatment for recurrence.

Chan: The…

Wagner: A premature male neonate born at 23 weeks of gestational age developed zone I acute posterior retinopathy of prematurity (APROP) at 32 weeks of age. He was treated with injections of 0.2 mg of ranibizumab. The neovascularization resolved within days and a 360-degree vascularization process began with what I describe as small, multiple, radial vessels progressing to the periphery 360 degrees. There was a recurrence at 42 weeks of gestational age. There now was a new onset of plus disease and neovascularization along the previously advancing edge, now into zone II nasally and zone III temporally. Is this an expected finding in similar cases, and how would you manage it?

Chan: I don't think it's unexpected to see recurrence. However, one of the issues that we struggle with as a community is how to define recurrence. The screening criteria, recommendations for when to treat, and treatment guidelines have not been well established by consensus.

Reynolds: That's an excellent point. Even the terminology varies from group to group.

Chan: When you look at the literature, the definition of recurrence after anti-VEGF treatment has not been well established. In this case, the questions are whether the examiner saw a true occurrence of plus disease after anti-VEGF treatment and whether our threshold for plus disease is different after anti-VEGF treatment. If there is definitely neovascularization, then I think that treatment or identifying this as recurrence may seem reasonable. The other issue is how are you identifying the neovascularization?

Wagner: It's observation.

Chan: Are you only seeing some neovascularization at the interface? Did you do a fluorescein angiogram? I will almost always recommend fluorescein angiography at some point for these children. This is especially true if I suspect recurrence, to identify the leakage and neovascularization. I've also encountered ischemia on fluorescein angiography that isn't always apparent on a clinical examination with indirect ophthalmoscopy.

Reynolds: The bottom line is that we do not know the natural history of ROP after anti-vascular endothelial growth factor (VEGF) treatment. No study has reported natural history data on these patients, and that's what we really need. We don't know that the current neovascularization is going to result in traction if untreated. Everybody thinks this is going to be just like untreated zone II disease, but we don't know that. We're all in treatment mode. We don't know that we have to be as aggressive in recurrences as we do in initial treatment. The location alone means that the total retinal detachment rate is going to be lower and we know the detachment rate is higher for zone I. In this case, the recurrence is in zone II and we know that the blindness rate is lower for these patients.

Bhatt: I agree that defining recurrence is important. Many infants treated with a lower dose have some type of recurrence, but not all of them require treatment. We need to decide when we should intervene and when we should not.

Wagner: In this particular case, what would be your treatment?

Bhatt: Our institution has treated approximately 100 eyes with injection and we've had to re-treat 10 to 12 of them at this point. The question in my mind was whether to try a repeat injection or laser treatment. In my experience, the repeat injection has not been that successful. Others have had success. One of the issues is that you don't really get further retinal growth after a second injection. The 4 eyes that I've attempted to re-inject all ended up progressing and needing laser treatment to eventually improve or resolve. I prefer laser treatment for recurrence.

Chan: The BEAT ROP study indicated that APROP may be a risk factor for recurrence of disease. However, it's been shown that there can be diagnostic variability of APROP between different examiners. I've also seen cases of children who have had multiple injections and end up having persistent avascular retina. We recently published a case that had persistent avascular retina 4 years after injection. Personally, I now prefer to use laser treatment for these cases, but we still don't have reliable data supporting lasers versus no laser for these avascular areas. And we don't have good prospective studies that support combination treatment, laser after injection, or repeat injection.

Reynolds: I would say that people are re-treating in recurrences based on anecdotal evidence and convenience. I've come to understand that retina physicians are approaching recurrences differently than pediatric ophthalmologists are. Pediatric ophthalmologists are much more likely to re-inject and retina physicians are much more likely to use the laser. I think the difference is that there has not been a significant conversation about persistent avascular retina and what that means. My impression is that the presence of persistent avascular retina is more disturbing to retina physicians than to pediatric ophthalmologists. What is the significance of 4 disc diameters of persistent avascular retina? I don't know the answer to that.

Wagner: In the study that you mentioned, did you use bevacizumab only?

Bhatt: Yes. It was all at the 0.25 mg dose of bevacizumab, which was a low dose at the time. The Pediatric Eye Investigator Group (PEDIG) study is using even lower doses but we don't have all the data yet. How long do you observe these patients with attempts at in-office examination? When do you consider putting them under anesthesia to get a really thorough look at the peripheral retina? Is it normal? Is it abnormal? Can you do fluorescein angiography? If you find some abnormalities, do you laser it or do you watch it? Is it going to have any clinical significance?

Reynolds: The case that we are discussing is the perfect case of recurrence. Because this happened, it was treated at 32 weeks. The neonate is still in the hospital at 42 weeks. That makes your decision easier and I think that you have an opportunity to put further questions about recurrences, persistent avascular retina, and examination under anesthesia in the future to rest if you laser this patient at this point in time. Despite the fact that I would be a strong proponent of not using the laser for this patient initially because of the APROP and the zone I disease and the fact that this is 32 weeks, this is a sick infant and a sick retina that is going to do poorly. I would inject this patient, but when that recurrence came in the hospital at 42 weeks, I would use laser treatment.

Wagner: Laser treatment was used in this case and it was considered successful.

Chan: We complicate this discussion by saying we don't have a lot of prospective evidence. But we also have to clarify that we're not saying there's no role for anti-VEGF treatment. With patients with APROP who do not have anti-VEGF treatment, we know that laser could also fail. I think the option of having anti-VEGF is good and there is some utility to it.I have a question for Dr. Bhatt about the dosage. How do you know you're giving that concentration of bevacizumab in the first place just by volume measurement?

Bhatt: That's a good question to ask the PEDIG planning committee that's currently looking at this.

Chan: Those of us who get bevacizumab from our compounding pharmacy assume that we're giving a smaller dose based on smaller volume, correct?

Bhatt: Correct.

Chan: And is that how you approached your study?

Bhatt: Our dose doesn't have to be diluted. The bevacizumab comes in a 25 mg/mL concentration, so our 0.25 mg dose is easy because it's an exact 0.01 mL. But in the PEDIG study, we're having to dilute it and that can lead to questions about concentrating problems in your local pharmacies. What is the exact dose that's given? And you're getting such small volumes. Our dose was 0.01 mL, which is the very smallest hash mark of a tuberculin syringe. How much of that volume gets in? How much stays in the needle? In the PEDIG study, the physicians are actually the ones pushing out the volume to get to that 0.01 mL. We're not trained to prepare syringes for an exact volume, so how accurate is the dose?

Reynolds: I think that we have to get away from this concept of specific dosage and start thinking about a range of dosages because that's the reality of what we're giving. To think that your 100 patients all got 0.25 mL is simply not true. It's not possible because of human error, the syringe, and the volume you're dealing with. I assume that the pharmacy gets it right.

Bhatt: But even that's an assumption, right?

Reynolds: That's an assumption, but that's a lot better assumption than those other three elements.

Bhatt: I think that's absolutely true. I'll be the first to admit I don't think every patient is getting the same volume. Syringes have to be packaged as sterilely as possible, so they're wrapped multiple times. Even that wrapping could slightly depress the plunger and cause a loss in volume. When you're talking about such a small volume to begin with, that's a big deal. I've put the needle into the eye and plunged and didn't really feel like much plunging happened, so I wonder if a lot of the volume wasn't already pushed out and there was air in there. It's difficult to know how much volume there is. I guess when you see results and the eyes improve, it gives you some positivity to the whole process, but there are a lot of questions to resolve.

Reynolds: I don't think that that basic flaw necessarily has to be fixed. I think that we could move into this concept of a range of dose. That's acceptable. You want to get the technology as good as we can. A whole new syringe should be invented for this with a smaller volume than a tuberculin syringe. But it's still acceptable, even when you're giving a range.

Wagner: I think the natural history of recurrence is important. There are some recent publications on the full developmental pattern following laser therapy versus injection and also the peripheral and foveal findings with intravenous fluorescein angiography 4 years after injection, which will hopefully provide more concrete recommendations. I'm not sure what it means yet as far as the ultimate disease process, but I think that is another topic for another time.

This Eye to Eye session was conducted on Sunday, March 18, 2018, during the annual meeting of the American Association for Pediatric Ophthalmology and Strabismus in Washington, DC.

Authors

Amit R. Bhatt, MD, is from Baylor College of Medicine/Texas Children's Hospital, Houston, Texas.

R. V. Paul Chan, MD, is from UIC Department of Ophthalmology & Visual Sciences, Chicago, Illinois.

James D. Reynolds, MD, is from University at Buffalo, Ross Eye Institute, Buffalo, New York.

Moderator: Rudolph S. Wagner, MD

The authors have no financial or proprietary interest in the materials presented herein.

10.3928/01913913-20181220-01

Sign up to receive

Journal E-contents