Lecture: Acanthamoeba Keratitis: A Case Study for Outbreak Investigations

During this live webinar, we describe an ongoing outbreak of Acanthamoeba keratitis in the United States and the process of investigating, identifying and modifying the likely responsible risk factors through an outbreak analysis conducted at a single treatment center.

Lecturer: Dr. Elmer Tu, University of Illinois Eye & Ear Infirmary, Chicago, USA


(To translate please select your language to the right of this page)

DR TU: My name is Elmer Tu, and I’ll be talking about acanthamoeba keratitis as a case study, as to how to manage outbreaks of atypical infections of the cornea. Let me go ahead and start the talk here. So what we’re gonna talk about today is basically looking at atypical infections, and how you as a physician or eyecare professional can look at these infections and judge whether or not there is an outbreak going on, and how to handle those, and how to get to the root cause, to try to prevent further injury to other patients, and hopefully come to some conclusions about how to better manage these conditions. So there is a lot of talk now about big data, and I think this talk, ironically, is probably more about small data than big data. Big data is often difficult to come by, it’s expensive, but there are some things you can learn from small data that you can’t learn from big data, that can give you some clues as to how to approach certain disease processes. If we go back, if we look at England, for example, in the 1800s, you’ll notice along the blue line below that the life expectancy in England changed very little from 1500 all the way to the 1800s. In fact, a physician, by the time you finished training, you would only have a few years of practice before you either retired or died. The interesting thing is this jagged appearance to the life expectancy varied from the mid-30s down to the 20s, from one year to the next. If you look historically, the reason for this was that in Europe, generally, but in London specifically, there were these outbreaks of pandemics, which would kill hundreds of thousands of people at a time. Often younger people. And this would really have a significant impact on their life expectancy. Now, many didn’t understand how these were occurring, and one of the popular theories of the time was something called a miasmatic theory. That the disease was transmitted from exposure to bad air, usually from foul or rotting organic matter. This was a theory that had been held throughout history, and specifically that disease was transmitted through breathing, and not direct contact or ingestion. And this was applied to things like bubonic plague, cholera, et cetera. If you look at London specifically, as I mentioned earlier, you can see that there were these cholera outbreaks that occurred, and these had a very significant impact on life expectancy in London specifically. And because of this, they decided to do something about it, since it continued to happen in cycles every few years or so. The city began a large project, basically directing human waste away from homes directly into the river Thames, which flows from North to South. You can see a diagram here, for example, of how human waste was put into a sewer and that led directly to the River Thames. Also, transport of water to homes was expensive, so most of London, as in many places, relied on central taps or water fountains, where people gathered their water and brought it back to their homes. The interesting thing about this was there was an obstetrician/anesthesiologist, named John Snow, who was a skeptic of this theory of disease transmission. He went in and looked at the pattern of how cholera occurred and some of the recent outbreaks that occurred there. He identified that there was a particular water company that supplied the South of London that was responsible for almost 50% of all the deaths in London from cholera, and from this, he felt that there was some other mode of transmission other than breathing in bad air. An opportunity came with the SoHo outbreak in August of 1854. The opportunity was that there was a severe outbreak that was pervasive in a community. There were 127 dead within three days of the initial case. 500 dead within the first ten days. And about 13% of all the people who contracted the disease actually died. You can see here in the diagram on the right — from the outbreak, they basically were looking at all types of temperatures, humidity, wind gusts, and everything else. In order to look for a pattern that might explain why this cholera outbreak occurred at that time. John Snow met Reverend Henry Whitehead, who was part of a commission to investigate this outbreak. He spoke with John Snow, gave an account of his theories and writings, and he was a skeptic that the miasmatic theory was correct. So he went out and interviewed the local residents, and after he interviewed them, he was much more in line with John Snow’s theory that bad air was not the cause. In fact, in the upper left hand side of your slide, you’ll see the central tap, which was called at that time, the Broad Street tap, where there was a water pump. And Reverend Henry Whitehead mapped out all the cases of cholera and realized that almost all the cases were centered around this particular tap. This is a mockup on the right here, which is not from the original investigation, but showing that almost all these cases radiated from the Broad Street tap and did not have any other relationships to the taps around it. This is a photograph of the tap. It still exists as a monument to this particular outbreak. As you remember, they were using the sewer system now in order to funnel waste away from homes. What happened in this particular situation was that there was actually a break in the pipes that led to the sewer, which led to direct contamination from the human waste, from this one particular home, directly into the Broad Street tap. So a baby actually contracted cholera in this home, and all the waste was going directly into the Broad Street tap. And this was discovered as the cause for the cholera outbreak. Particularly severe and pervasive in just this one community. What this taught us was that public health disease — we have a limited understanding of transmission, mechanism of disease, and treatment. And if you have a disease that has episodic outbreaks, there are usually a large number of cases which respected geography and allowed study of the case. The good thing was that there was an established infrastructure, in that there were health care professionals who had thought about this and written about it, and they had theories they could test out based on this particular outbreak, and with local community help, Henry Whitehead specifically understanding the community, and being able to track the cases accurately. What happened in this case is they began to get a clinical understanding of disease. And because of that clinical understanding, they were able to ask appropriate questions about the outbreak and come to a conclusion at a fairly rapid pace. And really, this is a testament to the collaboration between these two investigators. What this was — was really the birth of epidemiology. And this is more than just statistics. This brings together expertise from multiple disciplines, between physicians, epidemiologists, and biostatisticians. And it really proved the power of what a small, well-designed study, based on an outbreak of a particular infection can lead to, in terms of altering the course of history. And if you think about this, this actually led to the eventual demise of the miasmatic theory. As you may recall from the original slide, the life expectancy of all of Europe really exponentially increased after this particular mode of transmission was finally discovered. And proven. In no small part secondary to this. This has really led to our most recent definition of an outbreak. And basically this is the currency And basically this is the occurrence of more cases of disease than expected in a given area or among a specific group of people over a particular period of time. You have to establish what the disease is. You have to establish what the case definition is. What type of infection, how do you prove it, how do you confirm it. You want to confirm that the cases are real cases, and you want to establish the background of the incidence of the disease. It turns out that atypical corneal infections are actually ideal for this in a number of ways. When we approach corneal infections, we know they could be bacterial, viral, fungal, or parasitic. They’re generally rare or uncommon diseases, they’re difficult to diagnose, unresponsive to commonly available medications, and these infections unfortunately are often unique to the eye, because of the unusual character of the cornea and its immune privilege. They really offer the greatest challenge to treatment, but they also offer the greatest opportunity for understanding mechanisms of disease. So let’s take a look at acanthamoeba specifically. It’s a free living protozoa that’s found in most sources of water and soil. There’s a seasonal variation based on temperature and humidity. And this was only identified as an eye pathogen back in 1973. This is a traditionally rare disease. The incidence we’ll talk about in a minute. And it’s easily traceable, because of special diagnostic modalities, and specialized compounded medications. Generally these cases are not gonna get better on their own. So they need to go to a specialty area to treat them. It’s also primarily a disease of the eye. There’s really little help from the systemic medical literature, because of its rarity in terms of epidemiology, and also treatment. Acanthamoeba outbreaks are actually quite similar to what you saw with cholera, and they’re generally outbreaks that occur in different periods, in different places, around the world. The most studied areas are probably in the UK, with John Dart. There have also been a number of outbreaks here in the US. Specifically across the US, in the 1980s, and then most recently in Iowa in 1992, and the current one we’ll be talking about today. So who is at greater risk for contracting acanthamoeba keratitis? Soft contact lens wearers who wear their lenses only during the day, soft contact lens wearers who wear their lenses only overnight, gas permeable contact lens wearers who wear their lenses only during the day, or gas permeable contact lens wearers who wear their lenses overnight? Interesting. So we’ll talk about in in just a moment. But in fact, the greatest risk for contracting acanthamoeba keratitis is actually gas permeable contact lens wearers who wear their lenses overnight. Here’s a second question. How often is contact lens-related acanthamoeba keratitis bilateral? Never, 1%, 10%, or 25% of cases. 10% of cases. That’s actually correct? Although there’s quite a spread here. Who gets acanthamoeba keratitis is a very important question. When you have patients who come in with unusual infections that are not diagnosed, when you go back to the history, it points to the possible of acanthamoeba keratitis as being present in their disease process. It turns out that contact lens wearers — almost 85% to 100% of cases are in contact lens wearers. The rates, however, are very similar between the two in general. Meaning soft contact lens wearers and rigid or gas permeable contact lens wearers have a similar incidence of acanthamoeba keratitis. But if you look at it, orthokeratology specifically in patients who wear gas permeable hard contact lenses overnight have a significantly higher rate of acanthamoeba keratitis than any of the other groups, as I mentioned in the question before, and this may have to do with tap water exposure of the contact lenses just prior to insertion at night as part of the care regimens. Interestingly, if you look at the series, bilateral disease occurs in anywhere between 7% to 11% of patients with acanthamoeba keratitis, probably by far the highest rate of any of the pathogens that can infect the cornea. This is probably because the contact lenses are kept in the same environment that’s responsible for acanthamoeba colonizing their contact lenses in the first place. The good news is that if you’re a non-contact lens wearer, the rate of infection is actually quite low. So moving on, historically, we know that there’s a geographic variation of acanthamoeba. In the Iowa outbreak, for example, they were able to identify those areas that had the highest rates of acanthamoeba keratitis as being those counties that were flooded in a recent upper Mississippi flood. And also the use, interestingly, of municipal water versus well water. Municipal water actually seemed to be a source or a risk factor for infections. This is not surprising. Much of the wonderful work that was done by John Dart in the UK showed that there was a significant geographic impact on rates there. And water quality had a significant impact on that. Where there was hard water, there was a greater risk of infection, probably because of biofilm in the water supply. And looking at several of their cases, 27 of which — they went into the patients’ homes, and were able to do multiple samples of the domestic water, including showers, kitchen water, et cetera, and they found that in 30% of those, that they were able to identify acanthamoeba in their home domestic water supply, and that six of those isolates were actually identical genetically to the keratitis specimens that were isolated from the patients’ eyes. These patients all had these water storage tanks or rooftop cisterns that kept water in times of shortage. And acanthamoeba tended to grow in the biofilm that accumulated at the bottom of those particular cisterns, as a source for their infection. So looking at our most recent outbreak, here at the University of Illinois, Chicago, where I am, we had some confocal microscopy since the 1970s, to the point where we became a regional center for confocal microscopy, most of which were for cases of acanthamoeba keratitis. This was entirely consistent with the previous incidence of cases in the US, estimated to be 2 cases per million contact lens wearers. That was outbreak incidence; the baseline incidence was probably far lower than that. We began seeing, however, in 2003, an increasing number of cases that really owed to the referral patterns we had in place. We published this paper in 2004, showing that the incidence prior to 2003 and after was significantly higher — almost 6 times higher — between 6 and 40 cases in the period before and after 2003. This was also noted by investigators at the Wills Eye Institute, who presented in Chicago, in 2005, where they identified 19 cases over the preceding couple of years. We eventually did also publish our cases in 2006, as I mentioned earlier. We had had 40 cases. You can see a patient here with an epithelial form of acanthamoeba keratitis in the photograph. The CDC became interested and did a survey of 10 centers around the country, and they were able to reproduce the increase that we had seen after 2003. Almost an exponential increase, as you can see here. Of both culture and non-culture-confirmed cases. Hearkening back to the cholera outbreaks, where you see that they were mapping all types of different conditions that were associated — previously associated with an outbreak of cholera — these were some of the factors that were looked at originally, including the introduction of no-rub solutions in the early 2000s, and the dominance of the market with silicone hydrogels, in and around 2003, but we also looked at AMO Complete Moisture Plus, which was a new solution introduced in 2004. And when we looked at this — and the CDC also looked at this at the same time — we found that the odds ratio was about 16 times greater if you had been exposed to AMO Complete Moisture Plus for contracting acanthamoeba keratitis as opposed to other solutions. There were other factors, including solution reuse, infrequent rubbing of lenses, and showering with lenses that contributed, and this went toward the recall of that particular solution from the market. But in our paper, we identified that even though it was independently associated with a risk of contracting acanthamoeba among contact lens wearers, we felt there were probably other factors involved. When the CDC published their paper a few years later, you could see they were remarkably similar in terms of the risk factors we had identified, which gave us some assurance that we were on the right track in terms of our dataset. We only found — and this was confirmed in other studies — that only 50% of the patients were using AMO Complete Moisture Plus. If you took all of those patients out, you would still have, in Chicago, at least, 5 times greater the number of cases than was estimated in the 1980s outbreak. So there was something else other than just AMO Complete Moisture Plus as a factor in this outbreak of acanthamoeba keratitis. So we were table to understand this and look into this more deeply. This was a testament to Charlotte Joslin, one of my collaborators on this study. Dr. Joslin saw that all the cases were peripheral when the population was less. There were a few cases where the population was higher not just per capita, but total, in the Chicagoland area. She started looking into reasons for this, and hearkening back to the Broad Street tap investigation, you can see here again the geographic variation was very important. And if you looked at the water distribution system in Chicago, there was actually a central tap. Meaning that water is tapped out of Lake Michigan, processed with two plants here on the coast of Lake Michigan, and then distributed through the rest of the Chicagoland area. So what we’re seeing is that at the end of these water runs is where those infections were occurring. Not so centrally. Looking into why this is, we had all of these other factors that we had looked at and identified as possible causes for the acanthamoeba outbreak. But interestingly, also in 2003, our environmental protection agency changed rules on the amount of disinfectant by-products there could be in the end supply of these water runs. And for this reason, different communities had taken different tactics of either changing their disinfectants completely or reducing the levels, to try to bring them to quote-unquote safe levels, to try to prevent cancer et cetera. It was actually not very well proven as causes for this. So you can see here: Dr. Joslin looked at the water treatment system, and she was able to identify and show that water that was entering the peripheral areas of Chicago actually had a significantly lower amount of chlorine than prior to 2003. And even though we don’t have specific confirmation of this, it appears that the reason for the increase in acanthamoeba keratitis was probably a decrease in the amount of disinfectant in the water supply. The disinfectant specifically does not kill acanthamoeba, but it does kill bacteria, and bacterial biofilm is actually a place where acanthamoeba lives, and also the bacteria is considered a food source for acanthamoeba. So any increase in biofilm, any increase in bacterial contamination of the water and pipes, would lead to an increase of acanthamoeba in the water system. And you can see here that over time, as that disinfectant continued to decrease in its efficacy, that the number of cases that were occurring in Chicago increased as you got closer and closer to the center, as time went on, which is not surprising, as the pipes basically started to deteriorate. And where this was confirmed was that even after the recall of Complete Moisture Plus, there was not a significant reduction in the number of cases of acanthamoeba, either in the Chicagoland area or nation-wide, just from that recall. There was a small decrease, but generally it did not return to baseline, as you can see here from the CDC study looking at the same 10 centers they interviewed before. And the University of Illinois, over the past several years, continued to have an incidence of 10 to 12, sometimes higher, cases here. In a study that we did of the water supply, we were also able to show — and this was a non-random sampling — that the water supply in the Chicagoland area had the highest reported rate of acanthamoeba found in the water taps here. Short of the specific case of the UK, where they were only looking at patients who had acanthamoeba. So in summary, for this portion, acanthamoeba keratitis is a rare condition. From this small study, we were able to identify environmental risk factors that probably are not modifiable at this time. And we started to recommend strongly that the tap water not be involved in the system of contact lens rinsing or hygiene, either for hard or gas permeable contact lenses. So in addition to this, this is also another opportunity that’s been taken in the past, when there are large outbreaks, really to look at refining the management of these particular atypical infections. Using the patients that we have to try to improve their therapy and improve their outcomes. So now we’re going from big data to small data. And then from small data, we’re actually going to really small data. So now we’re not looking at the large dataset. Now we’re looking at small portions of those patients as a guide to how to improve therapy, which is even more difficult. So atypical infections have the characteristic of being difficult to diagnose, being poorly responsive. When we have an infection that isn’t responsive to medication, we look at them can being do microbiological testing, imaging, really with the goal to identify the pathogen. We always start with empiric therapy, because antibacterials are usually the first line of therapy. They will be the most catastrophic if untreated, but this is always considered a best-guess therapy. If it fails that, then we have to try to see what the next step would be. One is, if you’ve identified the pathogen, then you can determine a specific therapy, based on historical studies of efficacy. We can treat based on that. If, however, the pathogen is unidentified, then we have to keep guessing, and that includes other empiric therapy, there may be further diagnostics, and if it remains unidentified at that point, usually people will default to it being a herpetic infection, acanthamoeba, or some sort of non-infectious process, and ultimately, if the eye continues to deteriorate, they may require surgery. So what have we done in terms of advances in clinical management? There are diagnostic techniques for acanthamoeba keratitis which vary widely and wildly between centers. In fact, in the UK at the time, they really did not employ confocal microscopy, which was heavily relied on here in the US. When we published the study, there was always some question as to whether our cases were consistent with cases of other large outbreaks, to determine whether our diagnostic capabilities were correct and also whether our treatment outcomes were really congruous with other centers and other previous studies. So the standard microbiologic techniques we use for acanthamoeba include culture, histologic smears, PCR in some centers, and in the US they relied a lot on confocal microscopy. Here’s an example of a Giemsa stain on the left, showing the edges of an acanthamoeba cyst. You can see the same patient with an acanthamoeba target cyst on the right. So we looked at our first 53 patients, doing all tests that were available to us at the time, excluding PCR, and we were able to show that the sensitivity and specificity were actually quite good, whether you use culture positivity only as a standard, or if you used a composite diagnosis, including clinical microbiology and smears. So we felt that in our hands, at least, the confocal microscopy was helpful, and was easily reproducible and sensitive. So from this, we were able to draw conclusions from patients that we had seen, that would, I think, be helpful in further diagnosis. Here’s the first question in this section. What is the most important presenting prognostic factor for a patient’s visual outcome who has acanthamoeba keratitis? So there had actually been quite a bit written about this, prior to our study. And I think it’s been adopted now that our study is probably correct in determining how outcomes are determined. Generally most people would consider the duration of symptoms was the hallmark of when a patient would do well or not do well. One month is usually pointed to as the break point. But in our study of prognostic factors, duration of symptoms, even though it led to deeper disease, was not independently tied to visual outcomes. And in fact, really the patients with a good prognosis had superficial disease. Meaning they had just primarily an epithelialitis, and radial neuritis was not a prognostic factor in our study, or anterior stromal disease. If, however, they had deep stromal keratitis or ring infiltrate or any signs of extracorneal infection, this led to a 10 times more likely risk factor for ending up with a worse visual outcome, meaning 20/30 or worse. So this was really the basic tenet of that. Next question: Does having acanthamoeba keratitis confer immunity against future infections? So many patients — their first question is: When can I get back into my contact lenses again? The answer is complicated. Basically the answer that all of you have taken is no. And this came to light in this particular patient. We had a 16-year-old male who had two acanthamoeba keratitis episodes, one year apart, in the same eye. I worried that we had not fully treated the acanthamoeba in their first eye and that was a recurrence rather than a reinfection. But we had been genotyping all of these, and we were able to show in that publication that this patient contracted two separate isolates of acanthamoeba one year apart and had moved 900 miles away, so it wasn’t entirely his environment. There is some suggestion in many studies that patients who are at risk for acanthamoeba once may actually be at higher risk for acanthamoeba a second time. So it’s something you should be aware of. Next question: Which of the following have been identified as though pathogens with acanthamoeba keratitis? At co-pathogens. So it turns out that acanthamoeba is a very interesting organism. If you look at all of the literature, it’s basically been associated with infections from every single genus. That includes bacteria, herpes simplex, mold, and even other amoeba. This comes from a colleague of mine, where a patient was diagnosed with acanthamoeba keratitis and was improving dramatically over the period of a few weeks. From this slide to this slide was one day. The patient went from seeing better and feeling better to feeling much worse. The patient developed these deep infiltrates, which turned out to be a strep viridans. In looking back through the literature, we found that crystalline keratopathy was actually related to acanthamoeba keratitis quite strongly, as far as 20 or 30 years ago. First reported by Elizabeth Cohen at that time, at the Wills Eye Institute. And this spurred us to look into our cases of acanthamoeba keratitis, to see how many of them actually were polymicrobial. And we only were able to demonstrate about 4.5% that had co-isolates. And we feel that this is probably because there is an availability of broad spectrum antibiotics that sterilize these corneas prior to coming to us as a tertiary care center for diagnosis. But if you look through history, you can see that there are significant levels of co-pathogens noted, including fungus and bacteria, in all of the large studies, which suggests that acanthamoeba does coexist with other contaminants and pathogens, both as a food source and for other reasons. And you can see here: One of the characteristics of acanthamoeba that makes it ideal for this is that it can harbor endosymbionts. So it does eat bacteria, but often the bacteria are not digested by the acanthamoeba. And these can all exist inside of the acanthamoeba, untouched by its own immune system. And so we were able to look at — this was actually from Alfonso Iovieno, who was at Miami with Darlene Miller, who looked at their acanthamoeba isolates. 60% of them actually were harboring endosymbionts. And this was reflective of their environmental prey. Eric Perlman, who was at the time in Cleveland, did a study in mice which showed that in those patients with acanthamoeba that had an endosymbiont, that those caused a great deal more inflammation and scarring when treated, versus those that did not. So these confer pathogenicity, and also alter the outcome of these patients, as a possible reason for why some patients do better and some patients do worse. And this is not unheard of with Onchocerca specifically. We know this can be a corneal infection that comes from the systemic circulation. But interestingly, most of the significant scarring occurs after killing the onchocerca. It turns out that in these particular pathogens that they have an endosymbiont called Wolbachia. And treating the Wolbachia will often reduce the amount of inflammation that occurs, because it’s not liberated after killing the onchocerca. So the question of whether patients should be on concomitant antimicrobial therapy. Bacterial co-pathogens are the most common. We know that others are relatively rare. In the first paper, we didn’t include this, because we weren’t sure what to make of it. But we noticed in the prior therapy that those patients who used a non-benzalkonium chloride-containing antibiotic prior to presentation resulted in a poor outcome. This did not stay in the significant category once the multivariate analysis was done. So we looked at this as benzalkonium chloride. We know this is a hotly debated additive in ophthalmic drugs. It’s basically in everything, including most antiinfectives. The interesting thing is if you look at benzalkonium chloride by itself, it has an equal efficacy to hydrogen peroxide for certain species of acanthamoeba, while the fluoroquinolones have none. So our supposition is if you do treat early a patient with acanthamoeba keratitis with a benzalkonium chloride antibiotic, you’re probably getting some therapeutic effect from the benzalkonium chloride and not from the moxifloxacin. This is for many different species of acanthamoeba. It holds true for almost all of them. It’s interesting to think that moxifloxacin was also introduced around the same time. Finally, if you look at medical therapy, and how we’ve been able to improve the treatment of acanthamoeba keratitis, based, again, on this small group of patients, the standard medical therapy that we usually deployed prior was mechanical debridement. This helps debulk the organism load. It also helps the diagnosis. Acanthamoeba is an infection that takes a long time to treat and may have many ups and downs, and being unsure of the original diagnosis can lead you astray and make you try to treat something else or treat it differently. Some of the milder cases, we’ll start with a biguanide, chlorhexidine. If you have available to you hexamidine or propamidine, every hour — after two or three weeks of therapy, patients will have significant discomfort just from the diamidine itself. The backbone of therapy is the biguanide 0.02%. We will add systemic meds, which I’ll explain in a moment. And steroids are eliminated or reduced at diagnosis. We know the prior use of steroids prior to effective acanthamoebal therapy will worsen their prognosis, but it’s unclear what to do with them if they’re already on them. We know if they discontinue them acutely, the patients may have significant rebound inflammation, which may result in more morbidity. Than reducing and limiting steroids at the time of diagnosis. So as I was intimating earlier, we were using 20/30 as a visual result. And if you look at past studies, you can see actually the visual results in patients who were successfully treated for acanthamoeba are actually pretty good. You can see 20/40 or better in most of the studies, constituted somewhere between 60 to 80% of patients. In ours, it was significantly higher. But if you look at all of these studies, however, there are some patients who don’t do well. Meaning that they end up with less than 20/100 vision, or they end up with a corneal transplant. Or even loss of the eye. So we know there can be clinical resistance to acanthamoeba keratitis. Up to 5% may remain persistently culture positive, even with intensive and appropriate therapy for these patients. And this was published in 2002, and still remains true today. Voriconazole was an agent that we looked at, specifically from a paper from the CDC, and we know from other ophthalmic uses that it has high bioavailability. There’s an oral and an IV form. It’s good that it has equal tissue levels regardless if it’s given intravenously or orally. It’s been extensively studied for use in endophthalmitis locally, when it’s administered intravitreally. Based on the paper I was mentioning, which I’ll describe in just a moment — this is actually the first paper listed in the reference on the right here. It does have clinical activity against acanthamoeba, and was used in a cocktail of other drugs, to combat acanthamoeba encephalitis successfully in a few cases. There are significantly mixed results, however, in in vitro sensitivity testing, which is inaccurate and unstandardized at this time. So in vitro sensitivity testing is often unhelpful in identifying whether a compound is effective or not, but may be helpful in identifying candidate compounds, if you choose to use them to treat acanthamoeba in the future. So as far as voriconazole and acanthamoeba keratitis, this was described as both a topical adjunctive therapy as having some benefit. We published a paper using oral voriconazole, and found in three eyes of two patients that had chronic, culture-proven stromal keratitis, that there was resolution with its sole use after several months of therapy. In fact, both of these patients that we treated — we treated for a month, they got significantly better, we discontinued treatment, and they all recurred 6 to 8 weeks later. We put them back on therapy for 3 to 7 months, and the patients finally resolved their acanthamoeba infections. I’ll find that publication below. The other agent that was looked at, in that particular paper by Dr. Visvesara, was miltefosine. This is an antileishmaniasis drug, with low toxicity, and used widely in developing countries with leishmaniasis. It’s much more effective than voriconazole. This is the first patient we treated with miltefosine. A patient with multiple recurrences of acanthamoeba, each one proven with pathology. You can see after treatment the patient had significant inflammation, but eventually resolved, and with a new transplant, the patient kept this transplant clear now, free of disease, for about 7 or 8 years. So we feel this is another alternative therapy. Other alternative therapies include caspofungin, which has been used in the past, but there’s much less experience with at this point. I should say a brief word about crosslinking and surgery. Corneal crosslinking has been used in many centers as an adjunctive therapy. It’s not a primary therapy for many infections, and specifically for acanthamoeba. We know the riboflavin and wavelengths of energy that are delivered for standard crosslinking really don’t have significant effect against acanthamoeba cysts. And it may have more of a collagen stabilizing effect. It may reduce pain by killing the surface nerves, and may reduce inflammation by killing white blood cells in the area. While the acanthamoeba is being killed by chemical and medical therapy. Surgery — generally we reserve that for patients who have failed all other therapy. Prognosis is not great. Recurrence rates can be high, but they’re significantly better than 20 or 30 years ago, primarily because we have better medications now to take care of any residual infection that might be there. So in summary, we’re going from big data to small data to really small data, but the important thing is that it be good, clean data. This is probably more important than having a large dataset that’s sort of messy and unverified. You want to have a team approach to planning, gathering, and analyzing data, and you want to take the time to think about that data, and how to generate new hypotheses, based on your own knowledge of the disease processes, which is something that biostatisticians generally don’t have. And working as a team, you’ll come up with likely explanations for things, to minimize the amount of work you have to do to come to a hypothesis and conclusion. We know also that medical treatment of recalcitrant keratitides relies on accurate identification and isolation. We have the added benefit of being able to use compounded antiinfectives, where the safety and efficacy can be better proven. Acanthamoeba keratitis, as I mentioned earlier, is a rare infection. We know that contact lens wear in developed countries is a primary risk factor, but water exposure in Third World or underdeveloped countries is the primary risk factor there. Deviations in local water quality will amplify the number of infections and the amount of risk that patients are exposed to, and the current risk factors in the US are probably not modifiable, since we continue to see acanthamoeba keratitis at basically the same levels we had previously. And then finally, in cases — the increase of cases in acanthamoeba keratitis have led to innovation in treatments, beyond just discovering what the base risk factors are, and then we’re able to identify alternative topical and systemic medications, judge better the role of surgical intervention, and also the role of immunosuppression in preventing immune-related complications of acanthamoeba keratitis, while you’re treating it. There are also advanced diagnostic methods and future contact lens disinfection systems will be in play. I will refer you to the particular publication that’s on this page, looking at outbreaks of acanthamoeba microsporidia as models of what you can do locally, so please feel free to take a look at that. It’s a complex tree here, in terms of investigating something, but it’s important that you have some infrastructure, that you have clinical expertise, and have good collaborators. And from that, you can really do a lot of good with a small outbreak, in terms of, for example, we see about 250 to 300 patients, and most of these conclusions we came to within the first 100 or so. So if you have something like that, there’s a lot you can do with it, with relatively small resources. So manpower and thought is probably your best ally here. And I do want to acknowledge that this did involve a lot of different people. Without question, Charlotte Joslin was the main driver of this investigation. My partner and mentor Joel Sugar, here at the University of Illinois, and then members of the Ohio State University, and Case Western University as well. All of whom were involved in many of these investigative studies, and I thank you for your attention. I would be happy to take questions. I’ll take a look here. Okay, so… First question is: Do the new contact lens materials or designs reduce the incidence of acanthamoeba keratitis? So the answer is no. The silicone hydrogels were the ones that were introduced in 2003, 2004. And the vast majority of acanthamoeba keratitis infections we see today are with those materials. Daily wear lenses, daily disposable lenses, I’m sorry, do probably have a lower risk of acanthamoeba keratitis, but that’s not from the material. That’s most likely because they’re not stored in solution, and they’re not handled as much. Now, you would think that it would almost be zero, but it’s not. If you look at the two studies that John Dart and Fiona Stapleton did, there was some small incidence of acanthamoeba keratitis in even daily disposable lens wearers. The question is: Corneal decompression rate? I’m not exactly sure what that is. Perhaps you could retype that question. Because I think it’s half in. So here’s a question. We don’t have biguanide chlorhexidine here. How can we treat this acanthamoeba keratitis? The answer is it’s difficult. But you do have chlorhexidine. Chlorhexidine is a common compound. The question is whether your government or your hospital will allow you to compound it in the safer concentrations, again, 0.02%, that are required for treatment. You can go up to 0.06%, but we always start with 0.02%. Chlorhexidine is a common compound, and the barrier is not its availability. The barrier is whether or not your local government or facility, compounding facility, will allow you to compound it for topical use. If you don’t have that, really the only commonly available drug that’s been used in the past is neomycin. And to be honest with you, neomycin… I don’t even include in our treatment regimen, as you saw earlier, because I don’t think it’s very effective. If you have access to propamidine, that has some effect on trophozoites, but not cysts. So I think chlorhexidine is the most effective, and you’ll have to do work with local facilities to learn how to compound it. Another compound is polyhexamethylene biguanide, which is actually a pool cleaner, which is available in every country around the world, and relatively inexpensive. But has to be compounded. Chlorhexidine is easier, because it’s in the pharmacopeia. It has information used as a human drug, because it’s used in hand disinfectants et cetera. So it’s more likely to be allowed to be compounded than PHMB. But either one should be available. The question is whether or not your local government will allow you to use it. How do we know when to add fortified voriconazole? If you don’t have anything available, you can try… I haven’t found the fortified topicals to be as effective as biguanides. I’ve used them as an adjunct and haven’t found them to be the answer, necessarily, in those patients. If you have nothing else, I think it’s worthwhile to use that. But my experience is really more with systemic voriconazole than it is topical. But I think I would add it, if whatever treatment you have is not — after a couple of weeks or so is not showing an effect. Neomycin I answered. There have been some case reports of neomycin — actually maybe one case report in the past — of neomycin being effective. But generally in comparison to the current agents we have, including the biguanides and the diamidines, I think neomycin has very low efficacy in this particular case. Worthwhile using if you have nothing else, or as an additive if you can’t get something. But not terribly effective. So anonymous attendee is asking about unresponsive to conventional treatment. And that is where I was talking more about the voriconazole, the miltefosine, collagen crosslinking might be of some benefit in that particular case as well. The prognosis, I think, for corneal transplantation in these cases is poor if they’re not responding to medical therapy, but continued therapy and allowing the acanthamoeba to cause other problems, either in the adnexa or in the eye itself… Corneal transplantation may be an option at some point. Before the patient starts to develop glaucoma and other problems, including intumescent cataract. But those would be the things I talked about in the talk, as far as other potential treatments. Caspofungin, miltefosine, voriconazole, and collagen crosslinking. You would have to look at the chlorhexidine mouthwash to see what’s in it. They often have other agents in there, and I am not familiar with anyone using that. It needs to be in the proper concentration. Because as you know, 3% chlorhexidine will rapidly cause corneal death, within a few seconds of its application. So at full strength, chlorhexidine cannot be used in the eye. So you have to be very sure about the concentration that you’re using. It has to be 0.02%, 0.04%, or 0.06%. Anything approaching the concentration that’s used specifically for hand disinfection or wound disinfection will rapidly cause the cornea to decompensate irreversibly. We see that every now and then, at least, reports of it, from people who use it in the eye, and Betadine-allergic patients, that is very, very toxic. You need to be sure of the concentration and also make sure that the additives that are in there, that enable it to be used as a mouthwash, are not toxic to the eye. So here’s a question about how to diagnose acanthamoeba keratitis clinically at an early stage. And actually, that is the most important aspect of diagnosis, is really to suspect the possibility of acanthamoeba keratitis. So in a patient who has herpetic keratitis, for example, or something that looks like herpetic keratitis, that doesn’t respond appropriately to antivirals, or persists beyond a week or ten days that it normally would persist as an epithelial disease, or if they have a stromal keratitis that’s not resolving properly, either with the use of topical steroids or the use of antivirals, or if they’re a contact lens wearer… So basically my answer is that if the patient has high risk factors, for acanthamoeba keratitis, including a history of water exposure, or contact lens wearers with persistent infection that will not resolve with standard treatment, those patients immediately — the first thing that you should think about is acanthamoeba keratitis. Whether it be an epithelialopathy or a stromal keratitis. And try to think of it early and try to do the appropriate diagnosis or appropriate referrals, to get the question of whether it is acanthamoeba answered. So really, I think the best advice I would have is basically to have the possibility of acanthamoeba in your mind as a possible diagnosis, in every single patient, but not go chasing it necessarily, in every single patient, but make sure that’s not at the back of your mind, where you think of it two or three months later. And then finally, how can we prevent — not finally — but waterborne acanthamoeba keratitis? Specifically, there’s not a lot you can do in terms of environmental water. It can be found in pond water, it can be found in domestic water, meaning water in your home. So if you’re a contact lens wearer, which increases your risk, you want to make sure patients minimize their exposure to water when they’re handling contact lenses. Their hands should be dry after washing them. Preferably if you can use daily disposables and get rid of those lenses daily, so you don’t have to reuse them or store them near water, and also not to shower in lenses. So really exposure, limiting exposure, is your best bet. If, however, you have acanthamoeba in your home water supply, that’s a problem. It may be in the municipal supply. It may be in your home. If it’s because it exists in a biofilm, often getting biofilm out of pipes is very, very difficult. Replacing them, flushing them with formaldehyde, are industrial ways of getting rid of it. But once it’s colonized, it’s very difficult. So your best bet is really to try to avoid exposure to water. And bottled water, distilled water, those are not sterile. Those may also harbor acanthamoeba. So you really need to have boiled water or filtered water that filter out these parasites, for use in those particular scenarios. So what sensitivity does confocal microscopy have? Yeah, there was a recent paper. I can tell you… I think I was on that paper, actually. It varies from center to center. And so for me, I am the one… I am the technician, and I also read the confocals, and then also a take a look at corneal scrapings right after I do the confocal, to try to get an idea. And that’s really my experienced. And I think it’s unparalleled. So if you have the ability to actually do the scans yourself and actually look at the slides yourself, afterwards, that will give you the best experience in determining what specific structures you’re seeing on the confocal lens. If you saw my slide earlier, in our center, the sensitivity and specificity are quite high. In the 80% to 90% range. In other centers, it’s much lower. And I think it has to do — many of these centers don’t use… They use a technician who may not be imaging the correct area. So there are many different things you can do to increase your sensitivity and specificity of confocal microscopy. And then finally, could this disease link to the swimming pool? In my country, there’s more likely water contamination than contact lens wear. So the answer to that is yes. Chlorine, as I intimated earlier, which is a disinfectant that’s basically been reduced, we think, here in the Chicagoland area, that’s led to the increase, does not directly kill acanthamoeba. Acanthamoeba is usually screened from the water supply by filtration. Depending on the disinfection of your pool water, yes, acanthamoeba could potentially live there very easily. If the water is not properly filtered, it’s also gonna get in there from rain washing into it, and anything else. So it could be linked to swimming pools, it could be linked to swimming in ponds. It could be linked to swimming in the ocean, even. The ocean has different types of acanthamoeba that are found there, that are slightly different from fresh water. That’s absolutely correct. In different countries, the risk factors are different. So if you have a patient who has a significant history of water exposure, swims regularly, or something like that, who has an unusual-appearing keratitis, which may have some of the features of acanthamoeba, whether it be an epithelial disease or something deeper, that may be a risk factor that you’ll need to look at as well. As leading to a possibility of acanthamoeba keratitis. Okay. I think that’s it for the questions.

Download Slides


July 16, 2018

Last Updated: October 31, 2022

Leave a Comment