Lecture: Update on the Medical Treatment of Infectious Keratitis

This webinar will review current evidence based recommendations for the medical treatment of infectious keratitis including bacterial, fungal, parasitic and viral infections.

Lecturer: Dr. Soledad Cortina, MD, Associate Professor of Ophthalmology and Visual Sciences, University of Illinois Eye & Ear Infirmary, Chicago, USA


DR CORTINA: All right. Good morning, everyone. Thank you for having me again at one of the Orbis… (inaudible) how to diagnose and treat what is infectious keratitis, but also going over what are the relatively new concepts that we’ve learned through the literature in the past few years, and what is the future, and what are the things that we need to be looking at, to incorporate into our practices in the future, to be better at diagnosing and treating infectious keratitis? Or more effective. So my disclosures are — I have no financial disclosures for this talk, but we are gonna talk about some treatments that are used off-label. So what is the global picture? Well, infectious keratitis is a significant burden for the world. And in fact, we think that corneal disease is the number one cause of monocular blindness in the world, and it typically affects marginal populations the most. Infectious keratitis is the main cause of corneal opacities, and the 4th cause of blindness in the world. And it’s responsible for 10% of other preventable causes. So if we did a better job at preventing infectious keratitis, we could significantly decrease the burden of corneal blindness. So this is the first poll question. We’re just diving right into it. And the question is: What is the gold standard for the diagnosis of bacterial keratitis? All right. I like it. The gold standard, I think, still, in this day and age, despite all the pitfalls, gram stain and cultures are the gold standard for diagnosing bacterial keratitis. But other modalities are gaining traction, and we’ll talk about what the future holds. Obviously we want a fast and accurate diagnosis so we know what we’re treating, we can treat it fast, and achieve resolution as soon as possible. Not just for bacterial keratitis, but in general, Gram and Giemsa stains are our standard. The problem is they’re low sensitivity. When they’re positive, the specificity is excellent, but they’re not always positive, and then we’re treating an infection and don’t always know what organism in general is causing it. But in general they have pretty good specificity. We can use different media for different bacteria. With blood and chocolate, more common for bacteria. For fungal, we can use DHI, or for acanthamoeba, we can use blood with overlay, and so different media, depending on what is the organism we’re suspecting. PCR — we typically use PCR mostly for viral infections. But we’re gonna talk a little bit more about broadening its application to perhaps fungal and bacterial keratitis. And for the diagnosis of viral keratitis, it is largely clinical. And most of us don’t typically do any confirmation. But if we were to do PCR on this, just because of the dendrites — they’re so characteristic that sometimes we don’t need a confirmation. But in those equivocal cases, we could do PCR, and it has a very high sensitivity and specificity. We can use confocal microscopy, like the image on the right. It typically is able to diagnose larger organisms. We’re able to see acanthamoeba, filamentary fungus, and nocardia. Anterior segment OCT is something that’s been proposed as well for the diagnosis of certain keratitis. More than for the diagnosis, it is to follow them up in a more objective measure. So we can objectively measure the size of the infiltrate, we can monitor for thinning during treatment, and such things. So let’s talk a little bit about PCR for infectious keratitis. What are the advantages? Well, it has a high sensitivity, compared to culture and staining, which is the main drawback of the culture and the staining. It has the ability to test for antimicrobial resistance. So right off with that, we would know… You know, if these microorganisms are resistant, we would know what to treat it with. It can be performed pretty quickly, within 4 to 8 hours. Whereas for the cultures you at least need to wait 24 to 48 hours. And if it’s a fungal organism, probably days or weeks. It is shown to be more cost effective with selective use than culture and staining. This was surprising to me. But it’s a lot more expensive than culture and staining. If it’s going to decrease the burden of treating all those patients that are misdiagnosed and end up with a severe keratitis, the cost of treating those patients is much higher than actually applying PCR. So the thought would be, for those areas where they have a high incidence of infectious keratitis, perhaps a center that performs PCR for all those patients might be cost effective. There is increased ability to detect less common organisms, such as viruses, for example. But it does have some disadvantages. And that’s why it is not yet applied everywhere. And we use it sporadically. It can have lower specificity compared to culture and staining. Because it basically will give you everything that’s there. Need for a narrow list of causative agents to use specific primers — we have to select and tell them: Okay. I want you to test for staph, or I want you to test for pseudomonas, or I’m looking for candida or aspergillus, or whatever. It becomes less cost effective when it’s performed with a multiorganism PCR approach. If you’re looking just for HSV, that’s easy and very cost effective. But if you’re gonna send a large list of potential organisms, that makes it less cost effective. And supply costs — of course machinery fees and training expenses need to be taken into account. So what’s the future of diagnostics? I still think PCR is gonna play a large role for the diagnosis of bacterial and fungal infections. But we need to get better at validating this procedure, at having better particles for the sample collection and diagnostic thresholds, and of course, we need to reduce the cost. Another interesting technology is next generation sequencing. These technologies develop to determine DNA sequencing. It’s associated with bioinformatics to detect pairs between the sample that we’ve collected and a gene sequenced database. It has the potential to determine the etiologic agent with a lot of precision, and very fast, and it does not require primers like PCR. So potentially the application could be wider. And then anterior segment OCT — I thought about including this, because it’s something interesting. Maybe we’re not using this to diagnose cytomegalovirus endotheliitis, but it’s a nice sort of confirmation or adjuvant test that we can do. In the areas where you see the typical coin lesions with the endotheliitis, you see a thickened portion of Descemet’s membrane, and you can see these cells at high resolution and high magnification. And confocal microscopy can also be used. We talked about it for the classic indications, which are acanthamoeba, nocardia, and filamentary fungal keratitis. But CMV can also actually be diagnosed with confocal microscopy, where you can see the owl’s eye cells, the large cells within the coin lesions right here. So our next poll question is: Which of the following organisms is most likely causing this corneal infection? All right, yeah. The answer is acanthamoeba. I like that not everybody answered acanthamoeba. That’s the thing about confocal microscopy. It depends on who’s doing it and who is interpreting those images. It’s not always very clear cut. Sometimes we’re going back and forth and looking at multiple images and trying to figure out… Is this something? Or is it not something? But in this image you can clearly see the cyst with the wall. There’s actually a lot of cysts in this image. To give you a hint that this is indeed acanthamoeba keratitis. What about the treatment for bacterial keratitis? What have we learned recently? We’ve largely treated it with antimicrobials. It is the first line. Topical antibiotics. We have a choice between a broad or reduced spectrum antibiotic. We take into account the cost of the medication we’re giving, the toxicity. Its availability in the country or area that we’re at. And specifically, we want to target specific pathogens according to the area that we live in. So there was this Cochrane review done by McDonald, where he looked at 16 studies that were comparing two or more antibiotics that were given for at least 7 days or more. And he showed that no matter what antibiotic was used, they all had similar efficacy, and the same risk of adverse events. It did show, though, that there was some antibiotics that had a relatively minor higher risk of toxic conjunctivitis or discomfort. And not surprisingly, aminoglycosides were the ones with the most adverse events, in terms of toxic conjunctivitis, followed by cephalosporins and lastly the one that most of us end up choosing, because it’s low side effects, is fluoroquinolones. This graph gives us an idea of the initial antibiotic choice for bacterial ulcers. In the US and in international areas. This was done by a survey. And we see that largely in the US for severe ulcers vancomycin and tobramycin is preferred, with 50% of cornea specialists using these antibiotics. Whereas internationally, the first choice would be monotherapy with a fluoroquinolone. What about resistant organisms? So not everything responds to our initial antibiotic. The CDC actually estimates there’s at least 2 million infections a year by resistant organisms today. 80% of methicillin resistant staphylococcus aureus in the United States is actually resistant to fluoroquinolones in ocular samples. When we look at all the ocular samples, all the MRSA that we recover, 80% will not respond to fluoroquinolones. In the survey for corneal ulcer trial, we saw that in vitro susceptibility correlates with clinical results. So for antibiotics, susceptibility — if it’s resistant to the antibiotic, most likely the infection will not respond. Despite the fact that we know that antibiotic sensitivities are based on blood levels of these antibiotics, when administered systemically, and we know that we do achieve much higher concentrations in the eye, by application of fortified antibiotic topically and more frequently. But I think it is still recommended to obtain cultures and sensitivities in all corneal ulcers, and particularly for this very reason, because we’re having a significant rise in resistant organisms. We of course are gonna evaluate the response to treatment, and if things are not going the right way, we may change perhaps an initial monotherapy with a fluoroquinolone to fortified antibiotics. But if the initial therapy actually already includes fortified antibiotics, then we have to consider other things. It is pretty unlikely that one of the organisms that we’re treating will be resistant to, say, fortified vancomycin and fortified tobramycin. So we need to consider toxicity as an impediment for reepithelialization of the cornea. And so perhaps instead of intensifying therapy, what we need to do is reduce the frequency of application, to allow the epithelium to heal. So this is our next question. And it’s: What is the treatment of choice for methicillin resistant staphylococcus aureus keratitis? Perfect. Vancomycin. This is a little bit of a tricky question, because vancomycin has a wide spectrum of Gram positives and is the treatment of choice today for methicillin resistant staphylococcus aureus. We’re starting to see probably more systemic infections — not so much in the eye, though cases have been reported — of vancomycin resistant staph aureus. Linezolid might be appropriate in these cases. It is significantly more expensive, but can be used if there is allergy to vancomycin, for example, or toxicity in some way, to vancomycin. Moxifloxacin, like we talked about, at least in the US, high rate of resistance from MRSA to quinolones. So it won’t be a good enough choice, and tobramycin is usually Gram negative. While it can be adjuvant in some infections, it’s not treatment of choice as a monotherapy. If we talk about MRSA keratitis, we decided vancomycin is the treatment of choice. But how much vancomycin? What is the ideal concentration? Do we need to give vancomycin 5%? Do we need to give 2.5%? Do we need to give 1.5%? What exactly is the concentration that would be better? And this is a study that was recently published in Cornea. It is actually a study done in rabbits. It’s not a human study. But basically what they did is they infected the rabbit corneas with MRSA, and then they treated it, and they looked at the different concentrations. And they found that the most effective concentration is still the 5%, 50 milligrams per CC, of fortified vancomycin. However, this one is also the most toxic. They saw that lower concentrations appear to still be active, but they need extended therapy. So you need to use it more frequently and for a longer period of time. The main concern about the higher concentrations is that it burns when you put it in, and it has toxicity to the epithelium. So because it burns so much, sometimes non-compliance is an issue, and that may delay resolution, and also, if we have delayed epithelial healing because of epithelial toxicity, that could also delay the resolution of the keratitis. So how to reduce the inflammatory response? Because largely, you know, and unfortunately, infectious keratitis continues to be a significant burden in the world, because it will give residual opacity that permanently affects the vision, and can cause severe complications like necrosis and perforations, requiring therapeutic keratoplasty with permanent sequelae in these patients. So how can we reduce the inflammatory response and hopefully reduce the necrosis and residual opacity so that we can reduce the burden of disease? What are the factors that could reduce inflammatory response? Well, one of them — or at least one of the drugs that has been proposed — are anticollagenases. The goal would be to reduce the lysis of tissue. Therefore end up with less perforations and less rate of therapeutic keratoplasties. Tetracyclines are largely what we’re talking about, when we talk about anticollagenases. They do inhibit collagenases, and have an antimetalloproteinase activity in vitro. There are two study — again, these also have been done in rabbits — that showed a significant reduction in ulceration and risk of perforation. But unfortunately there are really no randomized clinical studies in humans that guide the use of tetracycline. Relatively safe drug. And I feel that we all use it, despite the lack of strong evidence for it. But we all, at the risk of melting and perforation for various corneal diseases, not just infectious keratitis, we may decide to use tetracyclines orally. And then steroids. This is the eternal debate. Most cornea specialists feel that steroids are largely beneficial for corneal ulcers, or for most corneal ulcers. There’s always the concern, though, about — why do we think that? We think it helps reduce scarring, neovascularization, and stromal lysis. The disadvantage is the delayed reepithelialization and potential worsening of an infection. That’s why oftentimes we’re scared to start steroids. So it’s always the eternal debate. What is the evidence? So there have been several studies, high quality studies, performed to try to answer this question. When I looked at this Cochrane review, actually, we found really four studies that were randomized studies. That used antibiotics, plus or minus topical steroids. Unfortunately, three were too small. Not necessarily powered to show a difference. But there was one study that had enough power to detect differences, and this was the steroid for corneal ulcer trials. So this review concluded that unfortunately there’s not enough evidence to recommend its use in clinical practice, but I think it’s the worthwhile to look at this SCUT study, and we’re gonna answer this question and then go into the SCUT study, and see what we learned from it. Our next question says… Let me see. I can’t quite read it. In which of the following organisms are topical steroids contraindicated? This is part of what we learned from the SCUT study, so that’s why this question is here. Right. So this is good, that we’re gonna go over the results of the SCUT study. The answer is nocardia. A bit of a trick question, because acanthamoeba was not tested. This was only for bacterial keratitis. Steroids are sometimes used in acanthamoeba, but it’s one of the organisms that we want to be cautious about. But let’s talk about what the SCUT study showed. This is the steroids for corneal ulcer trial, a multicenter randomized, double blind study, that compared topical steroids as adjuvant versus placebo. 500 participants with culture positive bacterial keratitis. We already knew this was bacterial. We didn’t include anything else. The drug used was prednisolone sodic phosphate, so it’s not milky and white, and it looked just the same as the placebo. It was started 48 hours after starting topical moxifloxacin, and the dosing was four times a day. The results show that in general, the study doesn’t show a significant difference in opacity or the rate of perforation in these patients. So whether they used steroids or they didn’t use steroids, these two endpoints did not change for all patients, on average. So this was surprising. We all could have bet that opacities were gonna be less, that perforation was gonna be less. This was not the case. But we did learn that it did benefit in certain subgroups. For example, patients with low visual acuity, less than count fingers at presentation, patients with central ulcers, the central 4 millimeters, and patients with deep ulcers, we saw that those that were randomized to the steroid group had better visual acuity at 3 months. We also saw that the type of organism was important. We saw that nocardia, which was 10% of all the patients included, had worse visual acuity if they were given steroids. That’s why I was asking the question about: Which one is the organism in which steroids are contraindicated? And it’s for this reason. We saw that largely if you look at all pseudomonas cases, steroids didn’t make a difference. But if you look at pseudomonas of the invasive subtype, we saw they ended up with better visual acuity at 3 months if steroids were used. And we also learned that timing is important. We saw that the earlier the steroids were started, within the first two to three days, then the better the outcomes. So what are the take-home points on steroids? That I can gather from the evidence that we have? Well, I think that as a result of this subgroup analysis, we could recommend that the use of topical steroids in bacterial ulcers — we could recommend the use of topical steroids in bacterial ulcers, except on patients that have nocardia, and we should start them relatively quickly, 48 hours after initiating appropriate antibiotic therapy. This is the recommendation. Some other data — and there’s a few studies out there now that are showing that we didn’t see a significant result in the SCUT study overall, because the dosing wasn’t enough, and if they compare low dosing, twice a day, of topical steroids, with high dosing, more than six times a day, they saw that those that have a higher dosing of the steroids had a significantly better visual acuity. And again, these are not large studies, like the SCUT study. But it’s a concept that’s beginning to gain traction, and we are gonna need to look at it in further studies, to confirm exactly what’s the optimal dosing for those topical steroids in bacterial keratitis. What about fungal keratitis? Well, we know that it definitely has worse outcomes than bacterial keratitis. There’s less evidence to guide treatment. And it is largely more common in tropical climates, where it represents almost 50% of all the ulcers. Here in the US, contact lens use is a significant risk factor, and we’ve had a couple of our contact lens solutions removed from the market for this reason. So what are the options for topical treatment? Well, unfortunately, at least here in the US, natamycin was approved in the 1960s and we haven’t had another topical antifungal approved since. So it is effective, but it has limited penetration. Amphotericin B is effective but needs to be compounded. But it is in general effective. Voriconazole has good ocular penetration and has very good antifungal effect in vitro, but we’re gonna learn that sometimes in vivo that is not the case. So our next question is: What is the topical treatment of choice for fusarium keratitis? Good. So the majority, 60%, picked natamycin. And I think that’s what our latest studies have shown. And that unfortunately we thought the voriconazole was gonna be a great drug, but it didn’t seem to have such clinical effect as we expected. So let’s go over the MUTT trial, the mycotic ulcer treatment trial. Again, it’s a randomized, double blinded, multicenter study, so it hits all the points you want to see on a well designed study to answer this question. And basically looked at natamycin versus topical voriconazole for the treatment of mycotic ulcers. Mostly filamentary fungus, especially fusarium. So there were 323 patients included, but the study control had to stop the study, because they showed that the group treated with voriconazole showed a higher rate of perforations and need for therapeutic keratoplasty. So actually the study had to be stopped. So this was pretty interesting. We did not expect this. What did the results show? Natamycin had better visual acuity at 3 months, mostly in fusarium ulcers. There was less corneal opacity at 3 months. This was only for fusarium ulcers. And there was a reduced number of positive cultures, six days after treatment, with natamycin. So we concluded that natamycin is superior to voriconazole in all filamentous species. There was a second randomized study and a Cochrane review that confirmed these findings. So we can say that natamycin is the initial treatment of choice for fusarium keratitis. Now, the problem is penetration into the cornea. If we have a very deep ulcer, we have to look at alternative therapies, because natamycin might not get deep enough to treat this infection. So then it was thought… Well, perhaps the MUTT1 trial didn’t do so well because the concentration of voriconazole in the cornea is not as stable. And perhaps oral voriconazole would be better and provide us better therapeutic levels to the cornea. So the MUTT II study was created, to compare voriconazole levels in aqueous humor. I’m sorry. Not to compare, but the basis of this study was that there were studies showing that the levels inside the eye, particularly in aqueous humor, were better if voriconazole was administered orally than if it was administered topically. The study design was: Patients were treated with a topical antifungal, and then voriconazole versus placebo orally was given to all of these patients. And they were randomized for each arm. And it did not show a difference in visual results, rate of perforation, or need for therapeutic keratoplasty, whether they had oral voriconazole or placebo. But we did see significantly more adverse events in the group that was randomized to oral voriconazole, and that’s to be expected. Increased liver enzymes and visual disturbances. Subgroup analysis showed, though, that there might be a possible benefit of oral voriconazole, particularly for fusarium ulcers. What about adjuvant treatments? Okay. We’re not able to resolve this infection with our topical or oral treatment. What can we do? Intracameral and intrastromal amphotericin and intrastromal voriconazole had been proposed, and are pretty widely used for the treatment of fungal ulcers. For intrastromal voriconazole in particular, some studies showed a possible benefit in certain cases, but this effect really needs to be proven in randomized controlled trials. And as an aside, I don’t think that we have a significant number of well designed studies, but more published series and comparative series looking at intrastromal voriconazole versus amphotericin. And I think it is that feeling of… I would say a large number of cornea specialists that treat fungal ulcers frequently that the rate of therapeutic keratoplasty actually has decreased since they’ve switched — after the MUTT trial results, they switched to intrastromal and intracameral amphotericin, instead of voriconazole, and their rates of therapeutic keratoplasty have gone down. So it may be even that when injected locally, amphotericin may be more effective. Other oral antifungals that may be considered are posaconazole. So in summary, natamycin is the treatment with the best support, and you may consider voriconazole if the organism is fusarium. This study is derived from the MUTT trial, from the same data. It’s just another way of analyzing the data. And what they looked at is the predictors for corneal perforations or need for therapeutic keratoplasty in severe fungal keratitis. So it’s a secondary analysis. What they showed is that the depth of the infiltrate, the size of the infiltrate, and the presence of a hypopyon are the three most important risk factors for perforation and need for therapeutic keratoplasty. So we consider that — if we consider these high risk features, we see the rate of perforation is almost 60%, compared to much less in the lower risk population. The size — when we consider a large infiltrate size, the size was more than 6.6 millimeters, and then the depth of the infiltrate was posterior 1/3 of the corneal stroma. So something to keep in mind. Just one quick thought about therapeutic keratoplasty in fungal keratitis. Today we’re not talking about surgical treatment of keratitis, but mostly medical treatment. If we have to do a therapeutic keratoplasty in fungal keratitis, I would advise caution with postoperative steroids, because it’s the number one risk factor for recurrence. What can we use if we don’t use steroids? We may have to use nothing, and let this transplant fail if it has to fail. There are some small series looking at cyclosporine A as an immunosuppressant. And cyclosporine A also has some expressive effect in fungal growth, so it’s something to consider. All right. We’ll switch gears now into viral keratitis. The diagnosis, like we talked about, for viral keratitis — it’s largely clinically. We can use PCR. Topical antivirals. We have trifluridine. It’s effective, but toxic to the epithelium. Acyclovir is just as effective, but less toxic. Unfortunately in the US it’s not available topically. Ganciclovir has a broader spectrum against HSV, VZV, and CMV — that is the most important, the CMV. And there’s a randomized controlled trial that is right now investigating ganciclovir for herpes zoster keratitis. Adjuvants for the treatment of viral keratitis: Topical steroids. So the HEDS study, which is a study done several years ago — there were two of these studies looking at the treatment of viral keratitis or herpetic keratitis. And what we learned from this first study: The most important thing that we learned is that actually we do need to use topical steroids to help resolve stromal herpetic keratitis. Oral antivirals, we have three options. Acyclovir, valacyclovir, which is the prodrug for acyclovir, with better bioavailability, and requires less frequent dosing. And valganciclovir causes… It’s the drug of choice for the treatment of cytomegalovirus keratitis, but it can have some adverse effects, including aplastic anemia. So just comparing the topical antiviral treatment, so if we look at ganciclovir, trifluridine, and oral acyclovir, they all have almost the same therapeutic effect on epithelial keratitis. There are new antivirals with novel mechanisms of action, other than nucleoside analogs, that are developing, and we’re looking forward to having something different for the treatment of herpetic keratitis. So which of the following antivirals is most likely to cause aplastic anemia? Let’s see if you guys were paying attention. All right, yeah. And it is ganciclovir, when it’s given systemically, which is the way that it’s given for the treatment of CMV endotheliitis. The side effects — when this drug is given systemically, there are topicals you know, for example, herpetic keratitis, and these did not cause aplastic anemia or these severe side effects at the same rate. So what about prophylaxis? This is what we learned with our HEDS II study. Basically this classic teaching, very important, is that the use of oral acyclovir reduces the recurrence of stromal keratitis by 45%, so by a lot. Very important. But there are a lot of questions that have not been answered with this study. For example, we know it reduces the recurrence of stromal keratitis. Does this still apply to epithelial keratitis? Or to uveitis? We pretty much think yes, but it was not assessed in this study. And what happens when the prophylaxis is discontinued? For how long do we have to use this prophylaxis? The study only used prophylaxis for one year, and then they tested recurrence at that. So recurrence at one year was reduced for 45%, but if we had followed these patients for longer than one year, perhaps their reduction in recurrence would have been even longer. So we don’t know for how long we have to use it. Most of us use it sort of long-term, for life, or until they haven’t had an episode for several years. But this is a question that we don’t have an evidence-based answer yet. We now have the ZEDS study, which is the study looking at the treatment of herpes zoster ophthalmicus. And basically what this study wants to assess is the effect of valacyclovir as a suppressive agent for zoster keratitis. I think the sense from the cornea community is that patients with zoster keratitis require high doses of acyclovir or valacyclovir, to keep these eyes quiet. And this is what this study is actually looking at. So the patients that are included are those that have herpes zoster ophthalmicus, and then they get a recurrence. So these are the patients included in the study. So it’s a very small subset of patients. It has been difficult to recruit. But this is an ongoing study. And the third thing that is important for viral keratitis is vaccination. Again, this is something relatively new. Not that new, but relatively new. And the idea is to prevent HZO and other zoster complications. It is recommended for all adults over 50. It does give — particularly the recombinant vaccine — these flu-like symptoms that are not mild. So most of us recommend if you’re gonna get the vaccine just get it when you know you can take a day or two off, just to be on the safe side, or on a Friday before the weekend. Something like that. And I think in recommending vaccination, I think we as ophthalmologists play a large role and I think it’s really important, when it affects patients, it really gives a lot of morbidity, and we hear story after story, after a patient got herpes zoster, how their quality of life declined significantly. So we have two FDA approved vaccines here in the US. The live attenuated vaccine and now the newer adjuvant subunit vaccine that seems to be more effective than the live attenuated virus vaccine. And the CDC again recommends for all individuals over 50 with no contraindications. The study looking at the live attenuated vaccine — it showed approximately a 50% decrease in the incidence of HZO with vaccination, and that vaccination decreases the severity of the disease in patients who contracted HZO, even if they still got it after being vaccinated. We don’t have any data for the attenuated subunit vaccine yet, but we know it’s more effective than the live attenuated, so we’re expecting even better results. What about antiviral resistance? Does this exist? Well, actually, it happens relatively rare in immunocompetent patients. It’s more a problem for immunocompromised patients, in which it can be as high as 36%. Rousseau looked in a prospective study and revealed that herpes virus I resistance to acyclovir is a significant cause of failure of prophylaxis in patients with HSV keratitis, and that it is associated with longer disease duration. And so a lot of us, when we have a patient that is on appropriate prophylaxis, and they have a recurrence with HSV, we may decide to keep them on a higher dose of prophylaxis from that point on. So this is something that could be done. However, I think due to acyclovir resistance, and cross resistance between the different antivirals that are now on the market, I think an alternative drug that has a different mechanism of action for HSV is largely needed. What about prophylaxis after corneal transplantation? So there’s been some studies, particularly this Cochrane database systematic review that looked at oral antivirals for preventing recurrent HSV keratitis after corneal grafting. And unfortunately the evidence is not conclusive, but it does strongly suggest a decreased rate of recurrence and rejection rate in graft failure. They state that the effect of the prophylaxis may actually last longer after cessation of treatment, and this again is one of the things I was discussing before, that we don’t have the definite answers to how long we need to use the prophylaxis and what happens once we stop it. But there’s no consensus regarding the most effective dosage and the duration of treatment in this and other study that have been published. But I think there is enough evidence — and this is common clinical practice — that we all use prophylaxis after we graft patients with HSV. What about alternative treatments for infectious keratitis? We talked about — that obviously the gold standard, the treatment of choice, is topical antimicrobials. But what happens when these don’t work? Some people are really interested in collagen crosslinking. For the treatment of infectious keratitis. Because it has an antimicrobial effect. And because it increases the resistance to tissue necrosis. But what is the evidence? What do we know? We know that for bacterial infections efficacy has been proven in in vitro studies. There have been several case reports now that showed benefit in patients with recalcitrant corneal ulcers. There was a study that showed — that reported on 16 patients, treated only with collagen crosslinking, and they showed that 14 out of 16 resolved just with the crosslinking, and 2 out of 16 required antibiotic treatment. Another study looked at 32 patients, and these were randomized to antibiotic and crosslinking that was one arm, versus antibiotic alone that was the second arm, and this study also showed benefit of crosslinking with a reduction in treatment time. So it seems that for bacterial infections, there might be a benefit here. What about fungal infections? In vitro, it has not shown efficacy. In vitro, as a monotherapy. There is a randomized controlled trial evaluating collagen crosslinking as an adjuvant in the treatment of fungal infections, and actually the early results suggested a possible increase in the perforation risk. When collagen crosslinking was used. So what are the take-home points about collagen crosslinking? I would say studies on the effect of collagen crosslinking in infectious keratitis are limited. I don’t think we can say one way or the other for sure yet. But there may be a benefit in bacterial keratitis. And I think larger randomized controlled trials are needed to determine the real efficacy of this therapeutic modality. And this is something new. It’s kind of hot off the press, published about a month ago in the American Journal of Ophthalmology. And it’s a concept similar to crosslinking, but with a different technique. It’s called rose bengal photodynamic antimicrobial therapy. Basically this case series was published where they looked at patients with progressive infectious keratitis that was unresponsive to standard medical therapy. They used rose bengal to the deepithelialized cornea for 30 minutes, and then it was followed by irradiation with custom made green LED source for 15 minutes. And they showed a pretty good effect, with resolution of several of their patients. So again, something to be considered, to be looking out in the future, as perhaps a new modality or alternative for recalcitrant patients who don’t respond to topical antimicrobials. So in summary: I think despite adequate antimicrobial treatment, infectious keratitis outcomes are often not favorable. Strategies focused on reducing associated morbidity should include: Prevention, better techniques for early diagnosis, better antimicrobials for resistant organisms, and therapies architected towards immune response modulation, very important, so we can reduce corneal opacity and corneal perforation. I thank everyone for listening, and we can answer some questions.

>> So Dr. Cortina, we have about ten questions.

DR CORTINA: I’m seeing them now. Is vaccination recommended for people who already had varicella when they were younger? How about people who already had HZO elsewhere? I think for people who already had varicella, the answer is yes, because the immunity can be reduced over time. So yes, it is still recommended for those patients. For those that already had HZO, one could argue that their immune system was already boosted by that episode, and I don’t think we have a definite answer for that. Any studies for oral fluconazole in fungal keratitis? So no recent studies comparing fluconazole, and we do it use it orally. At least, I do. And I do think it’s effective sometimes. It’s probably not the drug of choice. But sometimes because of insurance purposes not covering other drugs, we have to use it. And I think it is effective, or has some effect. But I don’t think we have as good studies as the ones that we have in the MUTT trial. Next question: It seems like there is still a lot of… To be defined when it comes to treatment and prevention of infectious keratitis. Are there any studies done currently that you think are promising? No, I think there’s a study looking at the use of new things. I think the study looking at the use of steroids in acanthamoeba that we should have results soon. But not… I can’t think of anything else right now. How about collagen crosslinking in acanthamoeba keratitis? I think there have been some reports, like I said, most of the studies — most of the evidence for crosslinking is just case reports or very small case series. There’s no series looking at acanthamoeba keratitis or crosslinking in acanthamoeba keratitis, but there is some case reports. Any vaccine for HSV? Great question. Unfortunately, no. That’s a big deficit that we have right now. It would be amazing if we could get a vaccine for herpes simplex. So natamycin is not available in my country. What would you recommend instead? I’m not sure what else is available, but compounded amphotericin and of course compounded voriconazole can also still be used. Compounded posaconazole or oral posaconazole are also options. These are even more expensive drugs. But caspofungin has shown good antifungal effect. Role of povidone-iodine as antifungal? Very interesting. I don’t have any experience with this, but if you do, please share. Very interesting. Can microbial infection cause corneal melting in 72 hours, following uneventful small incision cataract surgery? I think the answer is yes, yeah. Sometimes microbial infections can progress pretty quickly. Particularly pseudomonas of the invasive type. So I think yeah, that is possible. Do you advise outpatient or inpatient care for microbial keratitis? I think that’s a very good question. And in the past, we used to do inpatient care, just for the frequency of dosing. Now we’ve realized at least here in the US that even if we put the patient in the hospital, the nurses don’t have the time to give the drops every 30 minutes or every hour, and we still largely depend on the patient or the family members self-administering the drops, so we don’t think there’s a huge benefit in having them — except when obtaining the medications or for some reason they’re not able to apply them themselves, and they don’t have any family members that can reliably help, and then we will admit them into the hospital. This is a good comment. It says that an eye study in Bhutan suggested effective primary prevention of fungal corneal ulcer, especially injuries acquired during harvest, by distributing ointment to village health workers. This is a good comment, thank you. Which treatment do you advise in bacterial keratitis that is resistant? We talked about how vancomycin 5% is most effective. So if you’re going for fast effect, you should use 5%. If you’re worried about toxicity, then you can decrease the concentration and perhaps more frequent dosing. What is the best option for treating corneal involvement of pseudomonas endophthalmitis? I think we always put them on oral moxifloxacin and treat them with topical tobramycin plus moxifloxacin, and then of course we’ll do either a vitrectomy or inject intravitreal antibiotics, ceftazidime, and… I think that’s the way. That’s our initial line of treatment. With bacterial infections, keratitis, we use subconjunctival injection of… I can’t quite understand that question, I’m sorry. In early stages, how can we make the difference between viral or fungal keratitis — which analysis can we use for this? So I think that PCR is something that we can use for this. And sometimes it is true. If you have a stromal keratitis, it’s difficult to know exactly what is causing it. And of course at the large medical centers like this one the patient gets to us after being treated for a month with therapies that didn’t work, and so for us, it’s like… Okay, it’s not this, it’s not this, we know it’s this. But at the beginning, you see a patient, and sometimes you don’t know. Is this herpetic keratitis? Or is this acanthamoeba? Or is this early fungal keratitis? And I think PCR in the future will be very beneficial. So the sample taken for PCR — there’s a special little probe that you use. And it comes actually with a tube where you put it into. That’s how you take the sample. Unless you’re taking a sample from the aqueous humor, and then there’s a little vial where you will put it in. Role of subconjunctival antibiotic in microbial keratitis? I think we’ve used it, but I think intrastromal has better effect. So if we inject, we typically end up injecting intrastromally. Is specular microscopy useful in diagnosing CMV endotheliitis? We talked about confocal microscopy and anterior segment OCT, and that’s what’s been published, but I think you could see our cells in specular microscopy. So I would say it should have a use. I haven’t tried it myself, but it sounds like an interesting idea. Okay. So… The next question is: Could we prepare… Oh, when you begin corticosteroid for the treatment of deep herpetic keratitis after beginning oral acyclovir — or when? Pretty quickly. I think that I have no qualms using topical steroids if the patient has adequate antiviral coverage. So if I think it’s stromal keratitis, I start them both the same day. And I think that… Can we use chlorhexidine for additional therapy in viral or fungal keratitis? So again, no studies to support this, but it’s an antiseptic, and it’s reasonable to think that it will have an effect. So if there is something that is not responding to your usual therapy, or this is all you have, I think it’s worth a try. How can we determine patient with bacterial keratitis that needs collagen crosslinking for good treatment? So when do we decide to do collagen crosslinking? There are two schools of thought. There are people who think that maybe collagen crosslinking could be the first line of therapy, so that then you don’t have to worry about patient compliance with medication or the patient not being able to come back daily for checkups and things like that. And then — but most of the evidence this we have right now, as we discussed, is more the use when it’s the last resort. Well, medical treatment didn’t work. Let’s try crosslinking. But those two randomized studies that I talked about, that are small, only 16 patients, but that used collagen crosslinking alone, and the one that used collagen crosslinking plus antibiotic both together, and then they showed that crosslinking may have a benefit. So I don’t think we have the right answer to that yet. All right. So I think with this I will end the webinar. I thank you very much for listening, and for tuning in on Cybersight today with Orbis, and come join us next time.

Download Slides


November 1, 2019

Last Updated: October 31, 2022

Leave a Comment