Comparing Intravitreal Injection Protocols and Practices

An expert roundtable on clinical decision-making and the nuances of technique

Comparing Intravitreal Injection Protocols and Practices

An expert roundtable on clinical decision-making and the nuances of technique.


In the five years since the introduction of Macugen by Eyetech Pharmaceuticals ushered in a revolutionary new era of AMD therapy, intravitreal injections have become the mainstay therapy for this disease and an everyday occurrence in most retina practices. Although the clinical performance of several anti-VEGF therapies has been studied in exhaustive detail, less attention has been paid to the day-to-day issues involved in how to perform the procedure in a safe and efficient manner that minimizes discomfort for patients.

To meet this need, Eyetech recently convened a roundtable of experts well-versed in anti-VEGF use to discuss and debate the practical details such as injection location and method, preparation of the syringe, lid speculum use, the utility of sterile or clean gloves, pre- and postop medication use, and many other decisions that are largely driven by physician preference and anecdotal experience rather than hard science. The result was a lively discussion that generated a multitude of clinical pearls that can help even the most seasoned retinal physicians improve their intravitreal injection technique. We're pleased to present the following edited transcript of their discussion.


Kirk Packo, MD, Professor and Chairman of Ophthalmology, Rush University Medical Center, Chicago; also in practice at Illinois Retina Associates.


∙ David S. Dyer, MD, private practitioner at Retina Associates PA of Kansas City.
∙ Thomas R. Friberg, MD, Professor of Ophthalmology and Bioengineering, University of Pittsburgh; UPMC Eye Center, Pittsburgh.
∙ Baruch Kuppermann, MD, PhD, Professor of Ophthalmology and biomedical engineering, University of California, Irvine; Chief, Retina Service; Vice-Chair, Clinical Research, Gavin Herbert Eye Institute.
∙ Michael Tolentino, MD, Director of Clinical Research, Center for Retina and Macular Disease, Florida.

Dr. Packo: Retinal surgeons have been comfortable putting needles through the pars plana into the vitreous cavity for a long time, but in January 2005 our day-to-day practice radically changed when Eyetech began sales of Macugen, the first anti-VEGF therapy, followed in short order by Lucentis and Avastin. With these advances, we went from being surgeons to being technicians delivering drugs. Now that we give injections all day long, each of us has developed our own methodology in handling the patients and our own favorite ways of doing it.

The only guidance in the literature has been a single paper published in Retina1 that presented the best advice of thought leaders in our specialty. Good advice, to be sure, but it did not represent a real conglomerate of thought nationally. What we'd like to do with this panel is to get your thoughts as experienced clinicians on the nitty-gritty of giving an intravitreal injection. How is it that you make this a safe experience, a positive experience for the patient, and one that allows you to get the drug where it belongs?

1. Aiello LP, Brucker AJ, Chang S, Cunningham ET Jr, D'Amico DJ, et al. Evolving guidelines for intravitreous injections. Retina. 2004 Oct;24(5 Suppl):S3-19.


Dr. Packo: Let's start with a patient presenting to our office for a scheduled injection. Typically, we check the vision and intraocular pressure, dilate the patient and obtain an OCT scan. Do we routinely need to do all those metrics? Can you streamline the visit by perhaps omitting some of these things?

Dr. Tolentino: It depends on your injection schedule. When injecting every four or six weeks, I don't think you need a dilated exam each time. But if you use an induction and maintenance strategy like “treat and extend,” you do need to examine the patient to assess response to treatment and determine the viability of extension.

Personally, I always examine patients. I don't want to just be an “injector” per se; I want to exercise clinical judgment. Side effect profiles of the medications we use have to be considered. In addition, there could be asymptomatic uveitis, inflammation or increased intraocular pressure. If you don't do an exam, the patient could harbor that for a long time.

Dr. Kuppermann: I examine each time. I tend to be an aggressive PRN injector, so I examine, talk with the patient, look for blood or subretinal fluid, get an OCT and make a decision at that point about injection. I always do the full battery of tests. If you don't examine them, you don't know if they've developed new subretinal hemorrhage, which could influence your decision to inject or not.

Dr. Dyer: I completely agree. If the patient's visual acuity is dropping off, I want to know what's changed since the last visit. If their pressure is high, I may want to manage that first and then do the injection. It's important to look each time for new subretinal fluid, blood or inflammation. Perhaps there's been a reaction to the last injection.

Dr. Packo: I would just like to acknowledge that there are certainly physicians who do excellent work who give an injection whenever there's evidence of fluid, without sitting the patient down at the slit lamp — and I don't know that that's bad medicine at all. Even if there were a new hemorrhage present, it wouldn't prompt a change in the care they provide, although I do think knowledge in the chart is a good thing too.

Dr. Friberg: When you say “scheduled injection,” that might be a different scenario than a scheduled visit and an injection. Suppose on a previous visit a patient has been assessed and gives approval, and now they're coming in just to be injected. Even on that day, I think at a minimum it is important to measure visual acuity. But I don't think you need to measure IOP and perform a dilated exam unless there's been a lot of time between the date of the exam and the visit for the injection.

Dr. Dyer: That's a good point. I typically do my injections the same day I see the patient. If a patient comes back to just have an injection done, I don't do an exam as long as they've had a recent exam, but I still will check the vision and a pressure on that day.

Dr. Packo: Previously, we instructed our staff to ask the patient how they're doing and if anything has changed since the last visit. We now routinely instruct staff to specifically ask the patient if they have had a TIA, stroke, or been hospitalized. Given the still unknown true systemic risk of these drugs, I feel it's important to inquire about thromboembolic events. If an event were mild, the patient may not volunteer this information. It may not change what you do on that day, but we will never know the risks unless we document what's going on with our patients.

Dr. Tolentino: I've had several patients with a recent history of heart attack or stroke that had been scheduled for an injection, and I delay the injection as long as possible. When you look at some basic science models of stroke, VEGF is very critical in the first six weeks post-recovery. Inhibiting VEGF at that time could be detrimental. That's why clinical trials of anti-VEGF therapies exclude patients who have had a stroke within six months. There's a scientific rationale based on this class of drug in cancer patients, and also the understanding of stroke and involvement of VEGF in its pathogenesis. However, that concern has to be weighed against the level of visual impairment. One must consider the severity of the stroke and also the visual prognosis, and make a clinical decision.

“I always examine patients. I don't want to just be an ‘injector’ per se; I want to exercise clinical judgment.”

—Dr. Tolentino

Dr. Kuppermann: It's dose-dependent — systemic anti-VEGF use would pose a greater risk than intravitreal injection — but I too would exercise caution. There is a hypothesis that Macugen may be less of a concern because it's a VEGF-165 specific blocker as opposed to a pan-VEGF blocker. That's not borne out in the literature because no study has been powered enough to show a difference, but there is a scientific hypothesis that is rational.

Dr. Friberg: When you say inhibiting VEGF is bad, which VEGF are you talking about?

Dr. Tolentino: VEGF-121 is probably more critical in the stroke literature, but I'm sure VEGF-165 also has a role. I'm hesitant to use any anti-VEGFs for a certain period after a stroke, and I think we should all be wary about it.

Dr. Kuppermann: I agree with that. I try to wait three months; a six-month delay can be difficult for the patient. Also bear in mind that Avastin clearly has a longer half-life systemically than other anti-VEGF agents, and therefore is perhaps a greater risk in a stroke patient than Lucentis, for example.

Dr. Packo: True, but on the other hand, Lucentis penetrates from the serum into the CSF much better than Avastin. Thus, we really don't know which of these two compounds has the greater stroke risk.


Dr. Packo: In the general surgery literature, it's been very well documented that the use of antibiotics is effective as prophylaxis, but only when given prior to the procedure in a defined time that allows the drug to get to a reasonable level in the tissue, and of course it must have an appropriate efficacy profile. That's science that guides the general surgeon.

What do we have that's science in ophthalmology, and what's perhaps more akin to cultural habit or conventional wisdom? What's the role of pre-op antibiotics in getting a patient ready?

Dr. Friberg: I don't use pre-injection antibiotics. For me, they don't have much of a role, particularly if I'm going to use betadine, as I virtually always do. I don't pre-treat the patient unless a part of the study protocol calls for that.

Dr. Packo: Does anybody use pre-op antibiotics?

Dr. Dyer: I have a partner that uses pre-op antibiotics for three days. I use one drop of antibiotic the day of injection, and also a drop of betadine. But, in my mind, that antibiotic is really given as a head start for their post-injection antibiotic regimen. I'm just trying to get some antibiotic in the eye as soon as possible. If the patient goes home and doesn't start postop drops for 12 or 18 hours, they've at least had one.

Dr. Kuppermann: But it's not even clear that the postinjection use of antibiotics is needed. We've got a herd mentality: most of us are doing it because everybody else is.

Dr. Friberg: In certain countries, particularly Japan, they really are fairly adamant about not treating anybody unless they have had pre-operative antibiotics. They feel that it's more dangerous to treat them right away than to wait a few days or longer, because of this issue.

Dr. Kuppermann: Has anyone on the panel had a case of endophthalmitis post-injection?

Dr. Dyer: I've had one in five years. It's rare.

Dr. Kuppermann: That low incidence rate has to be compared against the imminent loss of sight from delaying injection.

Dr. Friberg: This is fascinating. Have we ever seen any data regarding a delay of X number of days or weeks causing less visual improvement?

Dr. Kuppermann: You can't design a study that way, but we know that over time it happens. I have had situations where I've had to choose between the eyes in bilateral cases, and I have had bleeds happen in that week in the untreated eye.

Dr. Packo: Without knowing the data, my guess is the Japanese data is very similar to what we've got here, and a delay of a few days probably wouldn't make a difference.


Dr. Packo: Let's talk about antiseptics. Chlorhexidine is toxic to the cornea. No one uses silver anymore. So what we're left with is povidone iodine. How do you prepare it?

Dr. Friberg: Just break the top off the bottle and drop it in the eye.

Dr. Packo: Undiluted commercial betadine is 10%, which has 1% free iodine. That does cause a little irritation and stippling to the corneal epithelium. There is a commercially available ophthalmic version that is a buffered 5% (0.5% iodine) concentration. We take a brown glass bottle, fill it halfway with betadine and the other half with BSS, and just use that. It's a stoppered bottle and every month we dump it out and start over again.

“The data on preop betadine use says that 19 seconds should provide complete bactericidal effect, so I wait about half a minute before injecting.”

—Dr. Kuppermann

Dr. Dyer: We add betadine to a bottle of artificial tears to achieve a 5% concentration.

Dr. Packo: What about the fact that betadine is light sensitive? Do you think it needs to be protected in a brown bottle?

Dr. Dyer: We keep everything in cabinets or drawers so we think that it's fine that way.

Dr. Kuppermann: I don't use betadine or povidone drops. We have packets of swabs. I open one up and swab the lids, and then use a separate one for the conjunctiva.

Dr. Packo: Let's talk about that: what's important to treat here? Do you treat the conjunctival surface? Do you routinely treat the eyelashes or the eyelid skin around it? And is that even necessary?

Dr. Kuppermann: I'm not sure it's necessary to treat the skin. We do a lashes prep and then put in a closed-blade speculum, and then treat the area of the injection. Some patients have a betadine sensitivity, not a frank allergy, and complain of irritation. If so, I just do a quick dab at the site of the injection.

Dr. Dyer: How long do you wait before injecting?

Dr. Kuppermann: The data says 19 seconds should provide complete bactericidal effect, so I wait about half a minute.

Dr. Packo: I have a woman with bilateral neovascularization who has a history of anaphylactic shock when given intravenous iodine. What would you do with her?

Dr. Tolentino: That's one of the few patients for which I use the pre-treatment antibiotic drops, a quinolone for three days before the injection.

Dr. Packo: I do exactly that. This lady is using antibiotics for three days prior to every visit she comes in.

Dr. Friberg: Would you use any betadine at all then?

Dr. Packo: No. Not if it's true anaphylaxis.

Dr. Dyer: Before we started doing all the intravitreal injections for anti-VEGF and this became a more common routine for us, let's say back in the mid-'90s when you were doing ganciclovir injection or an occasional triamcinolone injection, were you using betadine and antibiotics, or just the antibiotics back then?

Dr. Kuppermann: Just betadine. We would frequently use a drop of antibiotic at the end of the procedure, but no prolonged antibiotic. It would tend to just be betadine to the surface.

Dr. Dyer: I didn't use betadine until five years ago, just antibiotics, and I never had a problem with that. Perhaps I just didn't do enough to have a problem.

Dr. Kuppermann: It's a low-risk procedure.

Dr. Packo: You have to keep in mind that an antiseptic does not sterilize the eye, although it certainly lowers the flora by well over 95%. In the EVS, there were patients who developed endophthalmitis who had betadine preparation ahead of time.



Dr. Packo: What if the patient has severe blepharitis? We've all seen people that have rosacea, a rip-roaring blepharitis and gunky lashes. Do you delay that patient and put them on antibiotics?

Dr. Dyer: In the past five years, I've only had one or two patients that had such severe blepharitis that I felt uncomfortable about injecting and had to delay. As a rule, I don't treat blepharitis patients any differently, but I think using a lid speculum solves a lot of that.

Dr. Packo: Retinal surgeons by and large are fairly poor at looking for it and recognizing it, and there may be some patients at risk that we're letting slip through.

Dr. Kuppermann: We tend to apply some betadine to the lashes anyway, so I would apply more in the presence of blepharitis. But the flip side is that I try not to be too aggressive, because you might release more bacteria-laden oil or debris.

Dr. Friberg: If they had what I would consider level 7, 8 or 9+ blepharitis, I would not treat them that day.

Dr. Packo: If the patient had a documented MRSA infection, would you do anything differently? If there is MRSA in their nasal cavity, it's likely in their conjunctival flora as well.

Dr. Tolentino: Are they being actively treated for the infection? I've injected patients with MRSA.

Dr. Dyer: I don't treat them any differently.

Dr. Kuppermann: When I'm aware of it, I do have a tendency to treat locally more and consider pre-treatment antibiotics.


Dr. Packo: If a patient is taking an anticoagulant such as Coumadin, do you do anything differently?

Dr. Tolentino: I recently had my first case of a bleed in a patient on Plavix, Coumadin and aspirin. We use a 30-gauge needle, and he bled from the injection site for seven hours. I tried numerous methods of coagulation, including cautery and a patch, and consulted his internist to inquire about a platelet transfusion, which he advised against.

Dr. Friberg: Did you think you hit a conjunctival vessel?

Dr. Tolentino: That seems likely. He'd had injections for the last two and a half years and this was the first time it ever happened. It was just a fluke. I do so many injections that I'm not going to change my strategy for anticoagulant patients. We just need to be aware of what may happen and how to manage it.


Dr. Packo: What if a patient has severe glaucoma preop? The hypothesis might be neovascular glaucoma with rubeosis from a central vein occlusion. Pressure is 45 mm Hg. The referring physician says, “Why don't you give the patient some Avastin?”

Dr. Tolentino: I wouldn't initially. I 'd get the pressure down first.

Dr. Friberg: I would give them the Avastin and maybe do a paracentesis.

Dr. Kuppermann: But then you're apt to be removing some of the Avastin that you inject.

Dr. Dyer: If you're injecting into the vitreous cavity and then you take some fluid out of the anterior chamber, if you're removing any Avastin at all it's a very small amount.

Dr. Kuppermann: The patient may be pseudophakic. I think that makes a difference.

Dr. Friberg: If you do the tap first, I think you're more likely to have a hemorrhage in the anterior chamber.

Dr. Kuppermann: There's no question that a pressure of 45 needs to be managed, but in my experience you don't necessarily need to pre-manage it. I've injected eyes with high pressure, and after five to 10 minutes they're back to their baseline pressure.

Dr. Packo: Checking the pressure immediately after the intravitreal injection, you may find it went down from some release of liquid vitreous.

Dr. Kuppermann: Exactly. So I normally do not modify my procedure.

Dr. Dyer: If they started with a relatively normal optic nerve, I don't think a short-term pressure spike is a big issue. Rather, I worry about the end-stage open angle glaucoma patient with extensive cupping and who has split fixation. What do you do with that? Do you give them an injection once a month, increase their pressure for 15 to 30 minutes, and just hope you're not causing damage?

Dr. Packo: I have a patient who had a large bleed in his only good eye, and he has 100% occlusion of the carotid on that side. When I give him the injection, I close the artery just from light touch alone. He's probably the only patient I've ever done a paracentesis on following a routine anti-VEGF injection.

Dr. Friberg: Some data suggests these drugs cause sustained intraocular pressure for an hour or so, rather than just the pressure effect of the volume. If you have a susceptible patient, I think you might want to modify your technique in the way you've just explained.

Dr. Packo: It's something that by and large the retina community really doesn't know about or look for. We're not accustomed to looking at the cup-to-disc ratio really well, and there may be some advanced glaucoma patients that are getting choroidal neovascularization. That is something we need to be aware of, as most surgeons would say they almost never do a paracentesis, but there still might be a role for it in a patient with advanced glaucoma.

Dr. Tolentino: Does anybody pre-treat the high pressure, such as in a patient who has an obstructed outflow tract? I've given aqueous suppressants for three days prior, and sometimes I've given Diamox if they have a prolonged high intraocular pressure.

Dr. Packo: All these are good suggestions, and I think that we have to start with awareness and get a message out to the community. I think it's a subset of patients that are being defined now as we're learning more about these drugs that we need to be cautious about.


Lid Speculum:

Dr. Packo: Dr. Kuppermann, you were frequently giving injections prior to January 2005 when the anti-VEGF era began. You were involved in multiple studies with intravitreal injections for the AIDS population, and the VITRASE study, for which you were the PI. Did you use a speculum? How valuable is that?

Dr. Kuppermann: It varied from trial to trial, but frequently a speculum was employed. In the '90s, taking care of patients with AIDS and cytomegalovirus retinitis often required weekly injections in profoundly immunosuppressed patients. We would typically inject superotemporally back then, and we tended not to use specula at the time. We shifted mostly to inferior location sites when we converted to injecting triamcinolone on a regular basis, to keep the opaque drug out of the visual axis. Now, out of consistency, we tend to inject all the drugs inferiorly and I use specula routinely.

Dr. Dyer: The VITRASE study initially didn't require a lid speculum. Those patients that didn't have a speculum used had a higher rate of endophthalmitis. When the VISION trial for Macugen required use of a speculum, there were only two cases of endophthalmitis reported.

Dr. Packo: I think endophthalmitis rate in the VISION trial didn't prove anything. There was no power to show cause and effect there. It's exactly the way the company should have acted: take any potential variable out of the picture. I had one of the endophthalmitis cases in that trial and I did not use a speculum. Although I switched to using a speculum for the VISION trial, subsequently in my routine practice I never use a metallic speculum to give my anti-VEGF injections. I define a speculum as something that holds the lids open and keeps lashes pointed out of the way. Two fingers can do that. Plus, two fingers is much more comfortable for the patient than a metal speculum. Finally, the release and removal of the speculum occasionally is a painful maneuver as well, and can actually release oil from the lid margins. Gently letting go of your fingers avoids all of these concerns.

“I define a speculum as something that holds the lids open and keeps lashes pointed out of the way. Two fingers can do that.”

—Dr. Packo

I put another drop of betadine in and have the patient look down and over. I inject superotemporally and let go very gently. I'm always concerned about the occasional patient that will fight you taking the speculum out. I've put a speculum in myself to see what it feels like, and it hurts. The anesthesia we give typically doesn't neutralize that reflex, so the speculum can add to an uncomfortable experience. I'm the only person in a 12-physician group that does it this way and I have patients all the time that come back complaining, “Why is your partner hurting me so much by putting that metal thing in my eye?”

Dr. Tolentino: I have the exact same experience. I don't use a speculum. I consider my fingers better specula because I have more control over them. I have many patients who want me to do their injection because I don't use a speculum.

Dr. Kuppermann: That is exactly the technique I employed in the AIDS and CMV retinitis era — before there was a lot of comment about specula — and there weren't any complications. Later, I adopted some of the “community standards” like specula and more aggressive antibiotics and I've adhered to those.

Dr. Packo: If someone asks me, “Do you use a speculum?” I'm going to answer “yes.” I define a speculum as my first and third fingers and they do exactly what a metallic speculum does.

Dr. Friberg: If a patient gets endophthalmitis and you're questioned about whether you used a speculum and you said you did but you really used your fingers, I'm not sure that would be very compelling. I'm not disagreeing that using your fingers and being extremely careful in injecting might not be better than a speculum, but I don't think we should parse what is a speculum and what isn't.

I use a speculum essentially to keep the lid from touching the needle. If the patient is cooperative and is not going to move their eye, even with a speculum, they can easily do that, so a speculum doesn't completely protect you from complications. In fact, as you mentioned, if it's not a comfortable speculum, it could even be more problematic.

Dr. Packo: A wire speculum, rather than a bladed speculum, can actually point the lashes inward, and even create some more expression of the Meibomian glands.

Dr. Kuppermann: I never use wire specula because I'm worried about Meibomian gland expression, so I use a bladed speculum.

Dr. Dyer: I use a plastic bladed speculum, and the plastic is more comfortable than metal. Metal is cold, which I feel is more noticeable than a plastic lid speculum that has no sensation of being cold.

“When people go to the delicatessen to get a sandwich, the workers have gloves on. If we're not going to put gloves on when we're doing injections, we have to ask ourselves, why aren't we?”

—Dr. Friberg

Dr. Packo: There was a plastic product on the market for a while that had two pieces of tape that attached to the upper and lower lids and pulled them apart as it sprang open. It didn't wrap around the lash margins, but I don't know that it's available anymore for intravitreal injections.

Dr. Tolentino: One advantage to using your hand as a speculum is that you can modulate for the patient. If you have someone who is a very forceful blinker, you put on a little bit more pressure to keep that eyelid open. If they're not, you can be gentler. You can also use your fingers to stop their eye from moving just by putting pressure one way or the other. I think it's just a more refined way. With a metal speculum, you can't control how they move their eye, but with your hand, I've been able to do that.


Dr. Packo: Let's talk about other things one can do under the banner of sterile technique. Anyone use gloves?

Dr. Kuppermann: There are three options: no gloves, clean gloves that you pull out of the box, and individually-wrapped sterile gloves.

Dr. Dyer: I don't use any gloves. I don't see the point if the needle is sterile.

Dr. Kuppermann: I don't use sterile individually-wrapped gloves, but I do use clean gloves. Even though I wash my hands regularly all day, it's a way to keep a clean procedure going.

Dr. Friberg: I use clean gloves. When people go to the delicatessen to get a sandwich, the workers have gloves on. If we're not going to put gloves on when we're doing injections, we have to ask ourselves, why aren't we? Do you think anything can fall off of your hand into the eye?

Dr. Dyer: Then you should have a cap, gown and face cover. You can have skin falling off or possibly aerosolize body fluid, since you're talking to them. You have to draw the line somewhere. It looks good to the patient to have gloves on, but it's not doing anything medically. I do not use gloves.

Dr. Kuppermann: I don't think there's any right or wrong in this. If I were a medical-legal expert, I wouldn't condemn anyone for not using gloves. It reminds me that in the days of AIDS and CMV retinitis, I didn't use gloves either.

Dr. Packo: According to the PAT survey data, about 30% of doctors do use gloves and about 70% don't.

Dr. Kuppermann: Among our panel, three do not use gloves and two use clean gloves rather than sterile ones. That's somewhat in line with the PAT findings. Interestingly, outside the United States many of these injections are done in the operating room under completely sterile conditions. A number of cultural and regulatory factors likely come into play when accounting for that distinction.

Fixating Forceps:

Dr. Packo: How about fixating the eye when you give the injection? Do you use fixation forceps?

Dr. Dyer: I don't use anything for fixation. I like to have the patient look up, as I inject inferotemporally. A lot of patients don't want to see what's going on. Also, if somebody is nervous, their natural tendency is to look up and away. I would rather have them look where they want to look and treat them inferotemporally.

Dr. Packo: I would offer that the concept of looking up sometimes is confusing to the patient, particularly if they're lying on their back. They will look at the ceiling straight ahead rather than looking up behind them.

Dr. Dyer: That's a great point. I don't say, “Look up.” I say, “Look at the wall behind you.”

Dr. Packo: Looking at your shoulder is also a very easily recognized landmark for patients to go to. Looking down and over I've found is extremely easy for the patients.

Dr. Friberg: There might be some rationale for not injecting inferiorly. You have a tear lake down there.

Dr. Kuppermann: It might be more contaminated down there.

Dr. Packo: There is actually excellent data to show that the conjunctival flora is on the order of about three to four times greater in the area where the palpebral fissure is open, and it's much less under the upper lid. So if you want to go for conjunctiva that has significantly less flora, it's in the superotemporal quadrant. That was suggested to me by a uveitis specialist; when I heard that, I switched to superotemporal injections.

Dr. Kuppermann: It also hides bleeds a little better.

Dr. Friberg: And with respect to the needle, you can come from the side of the patient. I think it's important that they don't see what you're doing.

Dr. Packo: Do you use loupes or an indirect ophthalmoscope to see what you're doing?

Dr. Dyer: No. I don't mark either. I've done so many, I know where I'm going to go.

Dr. Packo: I think most people don't use a caliper either. They can use the Luer Lok.

Dr. Kuppermann: I use the tip of the syringe.

Dr. Friberg: I use nothing.

Dr. Tolentino: I use nothing.

Dr. Packo: I think a lot of people are comfortable knowing where 4 mm is. As a presbyope and a hyperope, I found that the +2 within an indirect ophthalmoscope, plus reading glasses underneath it, makes a marvelous telescope when it's illuminated, so I really can see exactly where I'm going with the light of an indirect on.


Dr. Packo: One of the biggest variables among retinal specialists is how we anesthetize the eye prior to injection. Everyone wants to make this as comfortable a procedure as possible, and yet there's wide variability in how we do it and what we're willing to accept from our patients as a positive experience.

Dr. Kuppermann: Initially I used subconjunctival injection, then switched over to pledgets, but I still have some patients who prefer the sub-conj injection once they've gotten used to it. I tried to convert them over to pledgets but they were used to the complete lack of sensation from the sub-conj. The newer patients who haven't experienced sub-conj seem to be quite tolerant of pledgets.

Dr. Tolentino: I use sub-conj injections almost exclusively. I tried to switch to a pledget of lidocaine gel, but my patients find it so comfortable the way I do it that when I've tried to change it, they've all “boycotted” me.

Dr. Dyer: We take the Alcon foam instrument wipe, cut those into little squares and put them into a bottle of xylocaine 4%, then fold one of those in half like a taco and put it down inside the inferior fornix (or underneath the upper lid for a superotemporal injection) for 10 minutes and then take it out. Patients don't feel anything at all during the injection. If you leave it in for five minutes, you can still do the injection, but they'll feel a small, brief pinprick. It's still tolerable. Either way, it avoids hemorrhages that are more likely to occur from a sub-conj injection and the abrasions and irritation that may occur from a cotton swab.

Dr. Friberg: I use sub-conj and I put a little xylocaine on a Q-tip and I move away the conj a little bit, and give sub-conj xylocaine and then virtually immediately, as soon as I grab the other syringe, I'm injecting them.

Dr. Packo: I give sub-conj injections as well. Here's a little trick: as I withdraw the sheath from the needle, I'll bend the needle 90° so it's beveled down. This accomplishes two things: (1) I can bring the needle tangentially across the eye and just squeak it under the conjunctiva to raise a little wheal of lidocaine, and (2) it's always away from their line of sight, so it's another way to hide the needle from them. I'm standing almost behind them with the syringe and since they're looking down and in they can't see the syringe at all due to the 90° angle of the needle. If you're bringing a straight, non-bent needle tangentially up to to the eye, they can usually see your hand and the needle. Again, doing it with the magnification of an indirect loupe, I can try to avoid vessels. Occasionally, I still get subconjunctival hemorrhages which sometimes can be severe and cosmetically unpleasant for the patient.

There are two commercially available topical viscous preparations: a viscous tetracaine called Tetravisc from Ocusoft, and a viscous lidocaine drop called Akten from Akorn. Do you have any experience with these, or do any of you use them?

Dr. Dyer: I have one partner who uses either one of those, and when I'm on call I'm almost guaranteed a phone call that day about somebody who's got corneal pain or an abrasion from the viscous tetracaine.

Dr. Packo: In terms of efficacy between the two, I've been impressed that the viscous lidocaine, Akten, actually gives significantly better anesthesia than the viscous tetra-caine, which surprised me a little bit from the science of those two drugs — but patients with the Akten rarely feel anything if you leave it on there for five minutes or so.

Dr. Kuppermann: There's been concern, and I don't know that the literature supports it, that the use of the viscous substances may inhibit the sepsis component of the procedure. When you then apply betadine or povidone iodine, the residual gel may inhibit the immediate contact of the povidone with the conjunctiva.

Dr. Packo: I always make a point of putting the betadine on the eye first, just after a proparacaine drop. And it still stings a little bit. I'll leave that there for a minute, and then put the viscous preparation on.

Dr. Kuppermann: I agree with that procedure — apply the betadine first.

Dr. Packo: For the sub-conj users here, when you draw up the subconjunctival lidocaine, I would assume you put it in a TB syringe, and if you're using Lucentis you're going to draw that up in a TB syringe as well, and they can look very similar. How do you mark them, and have you ever injected lidocaine by accident intravitreally? I'll admit I've done that twice.

Dr. Kuppermann: An additional complication is that I'm training fellows so there's more people involved. We do use a similar-looking TB syringe. It's got a 30-gauge needle for the sub-conj injection. When that is done, we routinely expel all remaining contents so there's never two syringes. Then and only then do we prepare the Lucentis syringe.

Dr. Packo: My bent needle is another visual clue that that was lidocaine and not Lucentis as well.

Dr. Tolentino: We have all our lidocaine syringes drawn up in the beginning of the day by our technicians in a sterile fashion, and then I'll label them with tape. Any lidocaine syringe will actually say “lidocaine” on it. The physicians draw up the anti-VEGF agent ourselves, so that sort of separation gives us several cues not to mix them up.

Dr. Packo: Another pearl is that if you inject subconjunctivally and you've left 0.2 cc in the syringe, throw the syringe out or expel it. Don't put it back on the table or else it may look like Lucentis.

Dr. Kuppermann: I don't throw it away because I use that sub-conj syringe as my marker or caliper. I want to show my fellows the position, so I use that impression ring. I don't throw it away, but we expel the contents for the same reason you indicated.

Dr. Packo: An additional comment on the accidental injection of intravitreal lidocaine — if this ever does occur, there is no need to panic. Lidocaine is well tolerated in the eye and causes no long-term damage. The eye temporarily goes NLP since the retina is anesthetized directly. The only issue is the embarrassment in telling the patient what happened.

One last point: There is now a new commercially available device that provides a novel way to create an anesthetic effect that works pretty well. It's a little cup called InVitria with a side-port that guides the intravitreal injection needle to standardize the position and depth you obtain. The margins of the device act as a lid speculum and the inner cup is equivalent to the limbal diameters. When you look down the throat of this cup, you see the eye. The device was designed by Arnaldo Gonçalves, MD, of the Netherlands.

His technique is to push on the eye and give the cup a little quarter turn to displace the conjunctiva. When he pushes down into the orbit and holds it in place, it does two things. It stabilizes the eye and raises the intraocular pressure — and that rise in pressure creates the anesthesia. If you raise the pressure sufficiently high, the patient won't feel the needle going through. I've used this device on several patients, and most like it, although most all admit that they still feel something. Rarely, it hurts. The cost is in the range of about $20 as a single-use device. Given the large number of injections that you may do in a day, this can add up and needs to be considered since you have no way of recouping this expense.


Dr. Packo: When you are drawing up the Lucentis, you can do so bottle-up or bottle-down. You get more of the drug out if you leave the stopper pointed up and go to the bottom of the glass with the needle. Do you, or your staff, draw up the Lucentis?

Dr. Dyer: Our staff draws it up.

Dr. Tolentino: I draw it up.

Dr. Kuppermann: I draw it up. We tend to use the filter needle and I swab the top of the stopper with alcohol.

Dr. Packo: I was told by an infectious disease specialist when I was in medical school that swabbing a bottle with alcohol is about as effective as praying over the bottle.

Dr. Kuppermann: Swabbing the bottle just a reflex to remind myself of the need for sterility. The stopper underneath the cap is said to be sterile.

Dr. Packo: Let's talk about dosages. We inject 0.05 mg as our standard recommended dose. Do you ever start injecting 0.1 mg or 0.15 mg?

Dr. Tolentino: Lucentis's HARBOR trial is currently researching if there needs to be a higher dose than 0.05 mg. Some patients respond to 0.3 mg of Lucentis.

Dr. Packo: There was a dose response in all the Lucentis trials; the 0.05 cc did better than 0.2 cc and 0.3 cc in every study. Genentech changed the formulation after uveitis was reported with the 0.1 cc dose. I know there are doctors who double the dosage for patients responding poorly to therapy. If you're going to treat bilaterally, you can split the bottle's dose between two eyes so you can charge for the second injection but not the drug. Most people use two separate bottles, though.

Dr. Kuppermann: When I do bilateral injections, I use two separate bottles.

Dr. Tolentino: I do, too.

Dr. Dyer: I do the same thing.

Dr. Kuppermann: For patients who need more therapy, I consider bringing them back sooner, which raises the issue of using Lucentis more often than every four weeks. In that case, I'll consider alternating Lucentis and Avastin. I have also increased the Lucentis dose for a patient who persistently leaks every time. There is some flexibility, even though the standard dose is 0.05 cc.

Dr. Packo: For patients with bilateral disease, are you comfortable treating both eyes on the same day?

Dr. Kuppermann: My standard is to do one-eye injection only, but I'm beginning to feel more comfortable with bilateral same-day injections. I treat it as two separate procedures and do each eye with a different set of everything.

Dr. Dyer: I've been doing bilateral injections for four years and I treat them as individual procedures. I've never had a problem and patients love it because it saves them a trip to the office for the injection in the second eye. It also saves you a slot in your clinic the next week.

Dr. Packo: The downside is that they may have bilateral irritation from the betadine. Also, you double the amount of the drug that goes into the bloodstream.

Dr. Tolentino: I do a lot of bilateral injections and I believe that Macugen is safer in terms of stroke risk. I choose one eye to get the Lucentis and the other eye to get Macugen, depending on how healthy the patient is and if there's any history of stroke or cardiovascular events.

Dr. Packo: When using Avastin for intravitreal injections, are you getting the drug from compounding pharmacies or from your own institution?

Dr. Friberg: We get it at a compounding pharmacy.

Dr. Kuppermann: I presented the compounding pharmacy price to the head of our pharmacy at the university. Within a day he called and said, “Get it from the outside compounding pharmacy. They can do it much cheaper.”

Dr. Packo: I'm at a university that refuses to trust any compounding pharmacy, so they prepare it onsite. It's significantly more expensive and they classify it as chemotherapy. Accordingly, there's a lot more paperwork before and after an Avastin injection.

Dr. Kuppermann: Before I was allowed to give my first Avastin injection in 2005, I had to go before the Pharmacy Committee because it is viewed as an oncology drug. I had to present the rationale and show them Phil Rosenfeld's original paper before I could proceed. Fortunately, it was a one-time visit that was collegial and relatively pleasant.


Dr. Kuppermann: There are two different approaches to hand positioning when injecting. Some fellows I train like to come in with their thumb already on the plunger. I favor the two-hand approach to stabilize the injection — going in halfway: mid-pupil, mid-vitreous — and get halfway down on the needle and depress the plunger with my other hand.

Dr. Dyer: I use one hand. I'll extend my little finger onto the cheek for stabilization and then inject. It's relatively quick.

Dr. Tolentino: I only do one-hand techniques. I've gotten good at that.

Dr. Packo: In addition to holding the syringe I'm going to inject with, I hold a dropperette of the betadine. I'll give a drop of betadine on the eye right before I inject, park the dropperette between my two outside fingers of my speculum hand, and then give the injection. On the off chance that some of the bacterial flora enter the eye along with the needle, they'll be bathed with iodine.

Immediately after withdrawing the needle, there commonly is a release of fluid. Is that a problem?

Dr. Dyer: I like to see a release of fluid because it lowers the eye pressure and may flush out any bacteria.

Dr. Packo: Do you consciously do anything in terms of duration of injection? Immediately after the injection, how long do you stay in the eye before you pull the needle out?

Dr. Dyer: I come out as quickly as possible.

Dr. Kuppermann: I inject slowly. It takes me about one second to depress the plunger. I use a tamponade with the cotton swab at the end of the injection.

Dr. Tolentino: I like to see the outflow, so I inject and pull out slowly.

Dr. Packo: I inject slowly, too. I'm worried about a jet stream hitting the opposite side of the eye. I use an indirect ophthalmoscope for visualization during the injection, and can often see the red reflex and the bolus of fluid going into the vitreous cavity, which takes about two seconds to stop. I'll wait that one or two seconds and then come out until the torrent of motion stops in the eye.

Dr. Friberg: I push out the conjunctiva to bring up a little tissue and then inject it with subconjunctival xylocaine. I use a diabetic syringe perpendicularly. However, sometimes I find the insulin syringe blocked and I can't inject it.

Dr. Kuppermann: With the TB syringe, if you're using kenalog or triamcinolone, occasionally it can block. I leave more space and do a quick blow down to the desired dose immediately before I give the injection to stop any crystals from getting lodged in the 30-gauge syringe. The 27-gauge syringe doesn't have that problem.

Dr. Packo: Any tricks in removing the speculum so you're not pressing on the eye?

Dr. Dyer: If you have the patient look down when you put the upper part of the lid speculum in, there's less pressure. I'll have them look up to put the lower part in. I tell them to keep looking up and pull the lower lid down and rock it out.

Dr. Packo: Afterwards, do you send the patient out of the room instantly, or do you do you check anything?

Dr. Kuppermann: I wait and check the pressure five minutes later with a Tonopen with a sterile tip, but I haven't tapped a single eye. All pressure eventually comes down. When the pressure is around 30, we send them out and document that in the chart.

Dr. Tolentino: I make sure they can see my hand wave but I don't check pressure.

Dr. Dyer: As long as they can see, I don't check the pressure. If they go completely NLP, I'll wait five minutes; if they're still NLP, I'll check the pressure at that point. I've only had to tap one person because they were still NLP after five minutes.

Dr. Kuppermann: It's probably worth doing a thorough review of every step and think about why we're doing it and whether we can streamline the process. In many ways, the less we do, the better off we are.


Dr. Packo: What is your postop routine? Do you give instructions to patients and tell them what to watch for?

Dr. Tolentino: I prescribe or give a sample of postop antibiotics to use four times a day for two days. If there is extensive blepharitis, I'll prescribe a fourth-generation fluo-roquinolone. I'm re-evaluating this, though.

Dr. Kuppermann: At a university, you have to bill the patient for anything you open; we aren't allowed to give samples. Since we use antibiotics during the procedure, we used to give the bottle to the patient, because they've already paid for it. We have since been prohibited from that practice and now give them a prescription. It tends to be a fourth-generation fluoroquinolone, either Zymar or Vigamox. I'm also reconsidering whether this is a component of my technique that I want to discard.

Dr. Friberg: We aren't allowed to give samples, but we prescribe postop antibiotics and call them afterwards.

Dr. Dyer: I give them either a Vigamox or Zymar prescription with a regimen of three drops a day for three days. Then we go over the signs and symptoms of endophthalmitis, vitreous hemorrhage and retinal tear or detachment. We give them a sheet that lists the symptoms and tell them to call us if they experience any.

“It would be nice to get rid of postop drops, but what if a patient gets endophthalmitis and you get sued?”

—Dr. Dyer

Dr. Packo: I have partners who provide the patient with Polytrim at our cost, $3 a bottle. It's very inexpensive compared to a $90 bottle of Vigamox. Corneal specialist Terry O'Brien at Bascom Palmer published data showing that if you give Vigamox hourly, you quickly get high levels in the anterior chamber. I usually ask my patient to instill it hourly — it soothes their eye a little bit, too.

Dr. Dyer: It would be nice to get rid of postop drops, but what if a patient gets endophthalmitis and you get sued? There's evidence that shows that Vigamox and Zymar are penetrating at least the anterior chamber.

Dr. Packo: If the patient complains that the post-injection course was irritating, is there anything you can do for that? The complaint of irritation is most likely from betadine, I would guess. Alternatively, the anesthesia could dampen their blink reflex and allow the eye to dry out, giving them a secondary keratitis. I have found that using an NSAID drop post-injection can cut down of some of the irritation complaints. Any thoughts?

Dr. Dyer: My first step is to give them a sample of artificial tears to take once an hour after injections. If that doesn't work, I'll put them on a topical steroid and antibiotics at the same time. In patients that are more sensitive to the betadine drops, the steroid drop for a day or two tends to work well.

Dr. Packo: I'd worry the steroid might blunt their endophthalmitis prevention reserve whereas an NSAID might do the same without it.

Dr. Tolentino: I used to get a lot of calls for irritation after the injection. At that time, we didn't tell them to keep their eyes closed after we numbed their eye, which can predispose them to micro-abrasions. Also, the anesthesia, which is profound, inhibits the blink response. When I started telling my patients to close their eyes, it cut half the calls. I've told people who have complained to patch their eyes if they can't keep them closed.

Dr. Kuppermann: I use the antibiotic drops to rinse out the betadine.

Dr. Dyer: At the end of the procedure, I always lavage the anterior surface of the eye with an eyewash solution to rinse out excess betadine. The patients say it feels good.


Dr. Packo: Do you obtain a fresh consent each time, or is one consent on the record enough for your series of injections?

Dr. Dyer: We get consent every time. We also have a flowsheet that documents the drops and injections we used.

Dr. Kuppermann: I receive consent every time as well, but I thought I was doing something extreme because the university made me do it.

Dr. Packo: We get consent once a year, unless we need to go to the other eye. On our flowsheet, we put whether the consent was “not indicated” or “yes, within a year.” We verify that we have consent on the day of the procedure.

Dr. Kuppermann: I have a standardized form where I circle each element. We also started doing a “Time-Out,” where a technician and fellow come in to check things off before I come in to verify and do the injection.

Dr. Tolentino: The patient gets a consent form, but a big issue in Florida has been wrong-site surgery, so we incorporated a sticker system. The technician puts a sticker — “Left,” “Right,” “Prepped” — on the correct side of the forehead. There's also an OU sticker for bilateral simultaneous injections. When the doctor comes in, he asks both the technician and the patient for site verification.

Dr. Packo: How do you handle follow up? My staff calls patients the next day.

Dr. Friberg: I don't see them back, unless on an extra visit, because we call them. However, I'm not certain these calls are important.

Dr. Dyer: When should you call? The endophthalmitis cases we've seen occur anywhere from Day 1 to Day 7. If you call on Day 1 and everything's fine, you could miss a Staphylococcus epidermidis infection that might take three or seven days to erupt. Still, one nice thing about calling is the ability to reinforce that the patient should call if they have any problems or symptoms.

Dr. Tolentino: I don't see the patient until their next injection. I give them our phone number to call if there are any problems.

Dr. Kuppermann: We have the patient come back after the first injection. There's debate of the one-week follow-up. If the patient experiences symptoms at Day 3, they may say, “Well, I'll see the doctor in four days, I might as well wait,” even though they're instructed to come in right away. Again, it's the unexpected consequences of good intentions. RP