As frequent intraocular injection of anti-VEGF drugs has become increasingly common, the issue of optimizing this expensive and burdensome process has become a major concern. Over time, the protocol known as treat and extend—in which the macula is treated monthly until dry, followed by a gradual increase in between-treatment intervals until a maximum safe interval has been determined—has become accepted as the best option. However, it continues to evolve and be refined.

Here, three experts share their current experience and thoughts on the most effective ways to use the treat-and-extend approach.

Setting the Interval

“Managing macular degeneration is a balancing act between overtreating and undertreating,” says Carl Regillo, MD, FACS, a professor of ophthalmology at Thomas Jefferson University and director of the Retina Service at Wills Eye Hospital. “I’d say the greater of the two evils is undertreating, because if you’re not on top of the disease, the vision gains you get early on in treatment are likely to be lost to some degree.

“There’s a lot of real-world management data that shows a mean visual acuity decline in wet macular degeneration patients over years in the course of treatment,” he continues. “On the other hand, there’s also data showing that it’s possible to maintain those vision gains. In fact, there’s now some data from the prospective, controlled TREX AMD study that suggests that treat and extend used for two years compares favorably to the gold standard monthly anti-VEGF injection technique, in terms of maintaining vision gains. However, in order to achieve that level of effectiveness you have to have a low threshold for treating. You need to average a relatively high number of treatments per year, and compliance with follow-up has to be good.”

Dr. Regillo says he begins treatment with several monthly injections, regardless of which anti-VEGF drug he’s using. “I treat monthly until the macula is as good as I can get it,” he says. “Once I’ve reached that point, I’ll try to extend, usually by two-week intervals. As long as nothing is worsening and we’re able to maintain the vision gains, I’ll usually cap the extension at about 12 weeks. With today’s drugs, 25 or 30 percent of patients make it out to 12 weeks; the rest are maintained at a more frequent interval that’s specific to their needs. That might be as frequent as every four weeks, although most patients end up in the six- to eight-week range. It’s rare that a patient requires treatment more frequently than every four weeks.”

Dr. Regillo says he’d like the macula to remain completely dry, but he’ll sometimes tolerate small amounts of subretinal fluid. “Generally, the deeper the fluid, the better it’s tolerated, especially if it’s a small amount and it’s not changing over time,” he says. “If a small sliver of subretinal fluid doesn’t go away after three consecutive monthly treatments, I’ll think about extending. You can certainly tolerate some degree of pigment epithelial detachment, if that’s present. If I extend and the fluid gets worse, I’ll go back to a more frequent interval. Sometimes I’ll even bring patients in more frequently than every four weeks to see if they’re responding adequately, or if the drug is wearing off quickly.”

Hemorrhages are also a reason to adjust treatment. “A new hemorrhage is an indication of new activity, and it will prompt me to reduce the interval,” says Dr. Regillo. “If it’s just a small hemorrhage and vision isn’t affected much, I’ll go back to the previous disease-free interval. But if it’s a big hemorrhage—or for that matter, any major setback—I’ll go all the way back to a four-week interval and maintain the patient there until the macula gets back to being as good as possible. Unfortunately, with a large hemorrhage the patient may not regain all of the lost vision.

“In this situation I’m much more cautious about re-extending, especially if the setback was a large hemorrhage,” he adds. “I’ll also speak to the patient’s primary care provider to see if the patient is at risk for bleeding because of the use of anti-platelet or anti-coagulant agents such as aspirin, Plavix or Coumadin, and I’ll work with those doctors to balance the systemic and ocular risks. Some retinal specialists will keep a patient who has had a big bleed at frequent injection intervals. I’ve done that too, especially if the patient had a bad outcome in the fellow eye and has to stay on blood thinners, for example. After a large hemorrhage I may not extend that patient at all.”

Peter K. Kaiser, MD, the Chaney Family endowed chair in Ophthalmology Research and professor of ophthalmology at the Cleveland Clinic Lerner College of Medicine, Cole Eye Institute, says he’ll extend the interval out to three or three-and-a-half months, depending on the type of lesion the patient has and the drug he’s using. “I might go to the longer end of the spectrum with a type 1 lesion,” he says. “The exception would be a patient with polypoidal choroidal vasculopathy who has had a major bleed in the fellow eye. For those patients, I use shorter intervals. I also keep the interval shorter with type 2 or type 3 lesions because they are more likely to produce a sudden decrease in vision from rapid activity or a bleed.

“Generally speaking, I’m wary of going beyond the amount of time that the drug has biologic activity,” he adds. “I generally can go longer with aflibercept than ranibizumab or bevacizumab.”


“Treating diabetic macular edema or retinal vein occlusion is a little different,” notes David M. Brown, MD, FACS, who practices at Retina Consultants of Houston and helped design many of the major trials involving anti-VEGF agents. “Diseases like DME and RVO are inner-retinal diseases, diseases predominantly of the inner capillary and inner plexiform layers. You can have recurrent edema in those areas without having catastrophic vision loss. So in a diabetic patient I do my best to eliminate fluid, but I’m much more tolerant of allowing a recurrence.

“With DME, I don’t necessarily treat with anti-VEGF if your fovea is dry with a nice foveal reflex and the only edema is outside the fovea,” he continues. “In that situation I treat until the foveal reflex is restored and extend when the fovea is dry. RVO and DME patients typically regain lost vision when the edema is resolved with the next injection. In contrast, macular degeneration is a disease of the retinal pigment epithelium and Bruch’s membrane and the photoreceptors. You can’t afford to repeatedly damage those areas, because photoreceptors don’t regenerate.”

“Diabetic macular edema is very different from macular degeneration,” agrees Dr. Regillo. “With DME, you’re treating abnormally leaking retinal blood vessels, not abnormal blood vessels that are growing and invading and destroying the RPE. So for DME, my approach is PRN. I treat until the macula is dry, and then I watch and wait. DME patients tolerate small recurrences relatively well, unlike wet macular degeneration.”

Dr. Regillo points out another difference between macular degeneration and DME: the time it takes to get the macula dry. “With wet macular degeneration, you can get the macula dry, or mostly dry, after three or four monthly injections in the majority of patients. DME tends to be slower to respond, so you tend to need a longer time frame to achieve a relatively dry macula—sometimes up to a year or more. However, after the first year DME patients tend to need a lot less treatment to keep the macula edema-free; the likelihood of having a recurrence goes down, so in most cases you can take them off treatment or switch them to very infrequent treatment. With wet macular degeneration, you usually need eight or nine injections in year one, and then six or seven in year two and beyond to maintain those vision gains. For DME it may be a similar number of treatments in the first year, but the average drops to just three or four in year two, and even less thereafter, according to several PRN treatment-based DME studies.”

Dr. Regillo adds that in many cases using treat and extend with DME patients wouldn’t reduce their treatment burden. “You’re going to be seeing diabetic retinopathy patients pretty frequently anyway, because you’re monitoring other aspects of their retinopathy,” he says. “So you’re not reducing the burden by using a treat-and-extend approach.”

Treatment Alternatives

Doctors agree that in some situations it may make sense to switch to PRN treatment or stop injections altogether.

Treating PRN instead. Dr. Brown says that in a few cases he may be willing to shift treatment to PRN. “If I’ve extended the interval past 10 weeks and the patient is binocular, or is really anxious to stop having injections, I may switch to PRN treatment,” he says. “However, you have to be careful because if you get rid of one weed in your garden, another weed is likely to pop up. You can calm neovascularization down, but a lot of the time those patients get recurrent leakage.”

Dr. Brown says he appreciates that many surgeons don’t believe that it’s ever safe to switch to PRN when treating macular degeneration, but he notes that there’s no data showing that Lucentis or Avastin lasts more than 10 weeks. “Eylea may or may not last longer than that,” he says. “We don’t have the data. But we have good data on Lucentis, and if the patient is coming in every 12 weeks, that patient is going naked for two weeks. And I’ve never had a patient who could go 12 weeks but not 13. On the other hand, we have a lot of patients who can go for eight weeks but not 10. That’s why, if I’m going to continue the drugs, 10 weeks is my maximum interval for Lucentis and eight weeks is my maximum for Avastin. Eylea seems to have a little more durability in the short run, but there’s not a lot of data to confirm that.”

Stopping because the patient seems to be “regressed.” Dr. Regillo says that when he was first using these drugs he would sometimes stop the injections when a patient was successfully extended to 12 weeks without incident. “I thought that if we got the patient out to 12 weeks the drug was probably not having any significant residual anti-VEGF effect,” he says. “I figured that the choroidal neovascularization was ‘regressed’ and didn’t need ongoing therapy. But published papers and personal experience have shown that most patients inevitably do recur, and every time you have a recurrence, you can have a setback that the patient may never fully recover from. So today I very rarely stop treatment, especially if a patient is doing well.”

Dr. Kaiser says he also rarely stops treatment. “If I can get patients out to three or three-and-a-half months and they remain dry for a year of extended intervals, then I’ll try getting them off the drug,” he says. “This is rare, but it does occasionally happen. However, if I do stop the injections, I’m going to bring those patients back in a month—i.e., four months after the previous injection—because I’ll want to watch them more closely.”

Stopping because treatment is futile. “In wet AMD, if the patient starts to develop disciform scars or subretinal fibrosis, that means there is permanent damage to the photoreceptors,” notes Dr. Kaiser. “At that point there’s minimal visual acuity left to save, so we usually don’t continue treatment unless it’s the patient’s only eye. We might also discontinue treatment if the patient has a large subretinal hemorrhage that’s caused damage to the photoreceptors. In that situation it’s also unlikely that continued anti-VEGF is going to be all that beneficial.”

Dr. Regillo says he’ll sometimes stop treatment because there’s some doubt as to whether the macular disease was ever truly neovascular. “There are a variety of conditions that can masquerade as neovascular macular degeneration, such as central serous chorioretinopathy and serous pigment epithelial detachments,” he notes.

Treatment Strategies

When deciding on a course of treatment for your patient:

Make sure you have the right diagnosis. “I think the biggest mistake clinicians make is not making sure they have the right diagnosis at baseline,” says Dr. Kaiser. “At the Cleveland clinic, as a tertiary-care center, we see a lot of patients who’ve been getting injected for diseases that are not macular degeneration and who don’t need the treatment. Thankfully, most of these patients weren’t harmed by the injections.”

Dr. Kaiser says he encounters patients with diseases that mimic macular degeneration fairly often. “For example, chronic central serous chorioretinopathy can mimic neovascularization in older patients,” he says. “That condition is different from macular degeneration in that the choroid is very thick; in macular degeneration the choroid is usually thin, except in polypoidal choroidal vasculopathy patients. In CSC, the fluid doesn’t respond at all to treatment with anti-VEGF agents.

“Another problem that mimics macular degeneration is vitelliform dystrophy,” he continues. “Those patients have what looks like neovascularization and subretinal fluid, but again, they won’t respond to anti-VEGF treatment. In these cases, having an OCT-A instrument is useful because it can definitely help you differentiate between true wet macular degeneration and masquerade syndromes.”

Don’t undertreat. Dr. Regillo says he believes the biggest mistake doctors make is occasionally undertreating. “If you look at real-world data, the mean number of treatments after year one tends to be lower than what studies with good outcomes would suggest it should be,” he points out. “Of course, this doesn’t imply that it’s the doctor’s or patient’s fault, per se. Often it’s other health problems that keep the patient from following up as recommended. DME patients are often younger, so compliance could be a bigger issue there.”

Dr. Brown agrees. “It’s human nature to empathize with the patient,” he says. “It’s easy to think, ‘Oh, the patient really doesn’t want to be coming back this often. Maybe I’ll allow a little fluid and extend a little longer.’ Sometimes it’s a reaction to having so many patients in the waiting room. The problem is, any time you undertreat, you leave vision on the table.”

“Every study we’ve seen to date has shown that the lower the number of injections—especially during the first year—the worse the outcomes are,” adds Dr. Kaiser. “Being aggressive with treatment to get the neovascularization under control quickly is important.”

Avoid extending too far if the patient has a lot at stake should vision loss occur. “If some individuals get a hemorrhage or lose a line of vision, they could end up having to move into a nursing home and lose their independence,” says Dr. Brown. “In that situation, I’m much more likely to treat and extend out to six weeks and see if we can keep it there.”

Remember that it takes more injections to achieve results when managing DME. “A lot of doctors give up too soon, especially with DME,” says Dr. Brown. “When treating macular degeneration, if you give four or five shots and the patient is still ‘count fingers,’ that patient is probably never going to improve. In contrast, with DME it can take eight or 10 or 12 shots to get the edema under control and improve the patient’s vision. I tell my diabetic patients, ‘It took you years to get into this shape; it’s going to take me a while to dig you out.’ They understand that.”

Be cautious about re-extending the interval. Dr. Regillo says that one result of his many years of experience with treat and extend is that he’s much less inclined to frequently rechallenge the patient. “If a patient has fluid when I extend her to 10 weeks and I’ve brought her back to eight weeks, I’ll probably keep her at eight weeks for a while,” he says. “A lot of studies suggest that that the disease-free interval remains pretty constant for a given patient.1 Nevertheless, if everything is going well after six months or so and the patient is maintaining good vision, I will try extending again.”

Expect an occasional chronic IOP increase. “A consistent increase in IOP after anti-VEGF injections is relatively rare, but it does happen,” says Dr. Kaiser. “I’d say it happens in fewer than 5 percent of our patients, and the IOP isn’t usually super-high; it’s manageable with IOP-lowering drops. If it starts to happen, we’ll reduce the number of injections as much as we can and try to use a longer-acting anti-VEGF agent.”

Dr. Brown notes that any injection adds volume to the eye. “Older eyes are less distensible,” he points out. “A harder, stiffer eye will be more likely to experience a pressure increase. As a result, a couple of phenomena may occur. Some patients will develop glaucoma. Other patients just get decreased compliance of the eyewall, causing their vision to gray out as the pressure rises. You do the injection and they say, ‘Doc, I can’t see.’ If that happens even once, we typically do an anterior chamber tap before each subsequent injection, taking out about the same amount of fluid I’m going to put in. I do the same thing with my glaucoma patients, and for patients with suspicious optic nerve heads.”

Clinic Strategies

To maximize your efficiency when dealing with a large number of patients needing injections:

Consider using alternating treatment-only visits. “For wet macular degeneration patients who can’t be extended much beyond six weeks, if everything has been constant in their pattern of treatment, I’ll schedule alternating treatment-only visits,” says Dr. Regillo. “That means that when they come in for their next visit, they don’t get an exam. They get their vision tested and get the OCT for me to look at to make sure the macula is still in an optimal state, and then they go right to the treatment room. They don’t get a formal examination unless they’ve had a change in their vision or the OCT shows something new. On any given day I may have five or six of these treatment-only visits. It’s a much faster encounter, and it eases the burden on both sides.”

Dr. Kaiser says he also moves to this type of format, usually after a few visits with a full exam. “It depends where the patient is in the process,” he says. “Early on, we’re much less likely to do it, but as we continue to see the patient, we’re more likely to include those types of visits. The history we take is very abbreviated; we mostly make sure there are no adverse effects. Also, there’s usually no reason to do extensive visual acuity testing or refraction at those visits.”

“The clinical exam is overutilized by some doctors,” agrees Dr. Brown. “They do it every time. It’s pretty unusual for that kind of exam to cause a change in your management. However, I do think you should examine these patients at least once every three months because you can get a hemorrhage or some other complication that you won’t notice if you don’t look.”

Don’t try to run your clinic yourself. “The doctor should just be doing what he knows how to do, which is seeing the patient, talking to the patient, documenting and injecting,” says Dr. Brown. “Have somebody else manage your flow. Get either a flow leader and a team manager or a clinic leader who can tell you what to do, and then do what they say.”

Don’t schedule all of your injections at once. “Some doctors have injection clinics,” Dr. Brown notes. “That makes no sense to me. What’s taught in lean manufacturing is to have multiple varied things scheduled each hour. That way, while I’m seeing one or two new patients, who may take a long time, the technicians can set up two or three injections. I can then pop out and do those injections. Then, while I see the next new patients, they can set up two or three more injections.”

If you’re running behind, tell your patients the reason. “Sometimes you can’t help falling behind schedule because of an emergency,” notes Dr. Brown. “In that situation, everyone in the office needs to be telling the patients the reason, and they all need to say the same thing. Patients get very annoyed if they’ve waited for two hours and no one has said anything to them. When that happens, the patient comes into my room and I have to defuse the patient’s anger for the first three or four minutes. That’s a waste of my time.”

Injection Strategies

These suggestions will help ensure that injections are done safely and quickly:

To improve your injection technique, observe other surgeons who have extensive experience. “There are a million ways to do an injection, and there are dogmas all the way around,” says Dr. Brown. “Some surgeons use a drape, some don’t. Some use a speculum, others don’t. Some people do it one-handed; some do it two-handed. I think it’s really helpful to watch someone you trust who’s doing 50 or 60 injections a day and see how he or she does it. Their technique may not be perfect for you, but you’ll get some ideas from watching. Most experienced surgeons will be happy to let you observe them. The idea is to learn different techniques and then experiment to find the technique that works best for you.”

When performing the injection, minimize talking, manipulation and the number of people present. “In this situation, less is more,” says Dr. Regillo. “We always follow a standard protocol. I’m notified when the patient has been prepped; I come into the room and the technician assists me with holding the lids. We have a no-talk zone over the patient that helps minimize the risk of infection, so any counseling is done ahead of time. We’ve already explained to the patient why we don’t talk during the injection.

“It also helps to not have family in the procedure room,” he notes. “They tend to talk to each other, or the family member asks the technician questions, and the technician shouldn’t be speaking once the prep has started. If the patient is well-versed in treatments and knows what to expect, it’s a fairly smooth, quick and safe process.”

Minimize the use of subconjunctival anesthetic. “I don’t use subconjunctival, although some of my partners do,” says Dr. Brown. “Some of them hate the way patients flinch. To me, the increase in subconjunctival hemorrhages isn’t worth it, except for the 5 or 10 percent of patients that you’ll have to scrape off the ceiling if you don’t do it.”

Don’t inject until the eye is numb. “The big mistake is injecting too soon, when the eye is still sensitive,” says Dr. Kaiser. “That will cause a lot of discomfort.”

If the patient refuses the use of Betadine, do the injection anyway. “Betadine is important, but if the patient refuses the injection with Betadine, you’re better off doing the injection without Betadine than foregoing the injection and letting the patient go blind from the disease,” says Dr. Brown. “You just have to know that the patient will have an increased risk of infection. Even in this situation, the infection rate is still really low.”

Make sure you don’t leave residual disinfectant in the eye. “Make sure you really clean the eye up with something like Betadine, but then be sure to flush all of the particulate matter out of the eye so that there’s no residual cleaner left,” says Dr. Kaiser. “The best injections are those in which the patient has no discomfort at home, and discomfort at home is almost always related to the cleansing, not the actual injection. I also tell my patients to use artificial tears when they go home; the flushing process helps to minimize discomfort.”  REVIEW

Dr. Regillo does investigative work in the clinical trials of anti-VEGF drugs from Regeneron/Bayer, Genentech/Roche, Allergan, Alcon, and multiple other companies. Dr. Brown is a consultant for Regeneron/Bayer, Genentech/Roche, Allergan, Alimera, Alcon/Novartis, and Thrombogenics. Dr. Kaiser is a consultant to Alcon/Novartis, Regeneron/Bayer, Kanghong and Allergan.

1. Freund KB, Korobelnik JF, Devenyi R, et al. Treat-and-extend regimens with anti-VEGF agents in retinal diseases: A literature review and consensus recommendations. Retina 2015;35:8:1489-506.