Lecture: Fungal Keratitis

Fungal keratitis is a common sight threatening cause of ocular morbidity in hot tropical climates. They usually have a chronic presentation and predisposed to individuals who have sustained injury with vegetative matter, diabetics etc. The diagnosis is limited by lack of availability of microbiology laboratory services and corneal subspeciality care. Antifungal drugs are mainly fungistatic and limited by issues of bioavailability. The surgical modalities of treatment include intrastromal and intracameral injections, superficial keratectomy, tissue adhesives and penetrating keratoplasty. Our discussion on this topic will address clinical presentation, diagnosis, medical and surgical management options for fungal keratitis.

Lecturer: Dr. Aravind Roy, L V Prasad Eye Institute, India


DR ROY: Hello, everybody. In the hot, humid tropical climates or in the setting of a chronic debilitated patient, fungus is an important etiologic agent for infectious keratitis. So once again, I welcome everybody for this topic on fungal keratitis. I am Aravind Roy. I’m a consultant at the Cornea Institute, KVC Campus, Vijayawada, Hyderabad, from the LV Prasad Eye Institute. Before we start this series, could you please indicate the type of your practice? Okay. Thank you. So most of the attendees are in the comprehensive or in the training phase of their careers. And I’m sure you will find this disease very relevant in your clinical practice. So the objectives of our talk today are to understand the burden of fungal keratitis, the predisposing risk factors, common clinical presentations, the microbiology, and the management aspects, which includes both the medical and the surgical management options for the treatment of fungal keratitis. Infectious keratitis is very relevant in the developing world, especially in the Southeast Asia, as well as in all the hot, humid tropical, subtropical climes. The reported incidence ranges from 11,030 to 7,990, which translates to 840,000 new cases of corneal ulcers every year in India alone. Now, if we study the prevalence of fungal infections, both in terms of the smear positivity, as well as the culture positivity, from this publication, from Aravind Eye Hospital, in Madurai, you can note these curves in the fungal positivity in the smears to be progressively higher, and significantly higher than the bacterial smears. As well as the cultures, which show that there is a significant culture positivity too. And amongst the organisms that were positive, it was found that fusarium and aspergillus, which are basically the highly filamentous fungi, are much more prevalent, and fungus is a significant proportion of all the infectious keratitis that presents to the ophthalmic clinic. If we also study the trends in the pediatric keratitis, fungus also forms an important part of this practice. There are distinct seasonal trends in microbial keratitis, and so also for the fungi. And we see in a study which was conducted in South India, as well as in China, that there is a seasonal prevalence of more and more fungal keratitis seen in the winter, and also in the monsoons. When we look at the predisposing factors, which may predispose to a person getting fungal keratitis, trauma with vegetative matter or people who are more employed in agriculture, agriculture-based activities, or in manual labor, are more predisposed to fungal keratitis. That said, there are certain systemic conditions which also predispose the patient to fungal keratitis, and this includes having a history of uncontrolled blood sugar or diabetes, and any unsolicited corticosteroid topical usage. Usually these patients present within a week to two weeks of the manifestation of their symptoms. And this study on fungal keratitis conducted in-house at LV Prasad Eye Institute listed out all the important ocular and systemic factors for fungal keratitis. And it was found that prior surgery and lagophthalmos were some common ocular risk factors, and diabetes was the leading systemic risk factor for fungal keratitis. When the patients present, they present either with a topical antifungal, which was prescribed as well, or often with a combination with antibiotics. And the combination of antibiotics, antifungals, or over the counter medications which often have a steroid, often confuses the clinical picture. However, there are certain very classic pictures of the fungus, when we see it on the first visit. And typically, it’s a very dry-looking ulcer. It has an indistinct, feathery, fuzzy margin. There may be a ring infiltrate, and some satellites. As you can see in this case, there is a classic ring infiltrate that is building up, with a very dry appearing lesion with ill-defined margins. So often they tend to progress without response to the antifungals, and they may need a therapeutic graft, which is the final treatment for these cases, in case they do not respond to medical management alone. This is another case where there is a plaque-like presentation, and the convex shaped hypopyon in a pregnant lady. And it progresses very fast, leading to rapid perforation of the cornea, and that led to a decision of surgical management and a therapeutic penetrating keratoplasty was performed in this case. She refused any further surgery, but gradually the graft had a neurotrophic keratitis, and it led to a graft melt, and this was the final picture, with a neurotrophic graft and pseudocornea and a perforation in the graft. She refused the tarsorrhaphy, which would have certainly avoided this kind of a clinical presentation. However, because she was in an advanced pregnancy, she refused further intervention, as far as surgery was concerned. This was another classic case, which presented with microbiology negative, diffuse margins, convex hypopyon, and a lesion which assured some amount of melting and thinning along the inferonasal cornea. There were no predisposing factors. At the outset, I knew we were dealing with fungal keratitis, especially because these cases are so very common in our clinical scenario. However, as a protocol, it is recommended that one should not diagnose fungal keratitis, unless there is microbiological evidence. This is important, because fungal keratitis tends to have a chronic, long-term course, and unless one does not have a definitive evidence of the fungus, it is important that you do not start antifungal drugs. And this patient was started on a broad spectrum empirical therapy, and this was the resolution. Over two weeks. This is a microbiology negative possibly bacterial keratitis. So these are some of the complexities, when we look at fungal keratitis. And another important factor that is a differential in fungal keratitis is stromal keratitis. That is one part of the clinical web. The picture is very confusing. However, points in favor of it is that there is a history of repeated recurrence, a much more chronic duration, florid vascularization, areas of scarring, and footprint scars, which help the clinician distinguish between fungus and virus. This was another case of a 71-year-old male, with a two-week history and no antecedent, history of injury with any vegetative matter and so forth. So a penetrating keratoplasty was performed, because this was a chronic ulcer. It was not responding to medical management alone. The grafts often fail, as is the case with a hot keratoplasty, meaning a keratoplasty that is performed in the setting of an infection. And they may require further surgery. And this was done in this case, where a cataract extraction and DSEK was performed, which led to clarity and resolution of vision. In addition to hyaline fungus, candida yeasts are also important etiologic agents in fungal keratitis. In a study from our institute, we found that candida albicans and candida parapsilosis are two other species that commonly infect cornea, and they may present as keratitis. These etiological agents are usually responsive to amphotericin B or caspofungin or voriconazole, as can be noted in the MIC 50 and 90 values. And when we do an E test with a Mueller Hinton agar with glucose and methylene blue, this is quite a long culture of candida, as can be seen with the amphotericin, voriconazole, and caspofungin. So a significant zone of inhibition around the strips placed in the center of the plates. The other important etiological agent in the setting of fungal keratitis are dematiaceous fungus. These are basically pigmented fungi, and in a dextrose agar, they grow as fluffy cottony blackish colonies. And when a slide is prepared with lactophenol wet mount, we can see filamentous hyaline and a brownish to blackish pigment. Typically these ulcers may also present with a fair amount of pigment around it, which hints that probably we are dealing with a dematiaceous fungus. Usually there are several species that come under dematiaceous fungus. However, this could be the commonest that have been isolated from several large series, including one large series in-house, from LV. This is an important slide, which shows the outcome of infection with dematiaceous fungus. And usually the commonest fungus that is recovered is curvularia. And the first line treatment for these infections is natamycin. Usually they heal well, with a fairly good visual recovery. A few percentage of patients are referred for penetrating keratoplasty or may have a poor outcome and need to be eviscerated. But by and large, they respond well to antifungals. The basic microbiology of corneal ulcers remain the same, irrespective of if we are dealing with bacteria or fungal keratitis. Now, this includes a smear using a CF wet mount, Gram stain, and Giemsa stain, solid media, and liquid media. It includes 5% sheep blood agar, chocolate agar, dextrose agar, Sabouraud dextrose agar, and brain/heart infusion broth. To report as fungus positive, either the corneal scrapings should reveal fungal elements in the smear, or there is growth of fungus from more than one medium in the presence of negative smears. Or if fungus grows in a confluent manner at the site of inoculum on a single solid media, or fungus grows in one media with fungus in the smears, these are the criteria in which a report of fungus-positive comes from the microbiology lab. And if we look at the sensitivity and specificity of fungi, when we compare gram, Giemsa, and KOH, we see it is a fairly sensitive and specific to detect fungus. Meaning that if you get fungus in the slides, that means that a fair amount — you can be fairly confident that you are dealing with fungus as the only etiology in this kind of ulcers. The normal ocular flora has a biodiversity, and recently next generation sequencing has helped us understand that what proportion of this normal ocular biodiversity changes in health and disease. So this paper talks about how the ocular microbiome changes when there is an infection. Especially when there is a fungal keratitis. So these bands actually tell us which genera are present in the normal ocular microbiome in healthy individuals, where the swab has been taken, and there is a lot of unclassified fungi that is found. And when we take smears from patients who are suffering from fungal keratitis, we see there is more aspergillus, less of the unclassified fungi, and when we take corneal scrapings and subject them to the next generation sequencing, again, we see similar patterns, of more aspergillus and other fungi, as opposed to the unclassified fungi. So that means that the possible etiological agent is populating, and also it is pushing out the other fungi that are part of the normal ocular microbiome. And this is how the normal ocular flora responds to a disease. And this helps us understand the burden of the disease. When clinical studies were done on the type of fungal keratitis and the infective organism, it was found that hyaline filamentous fungus is very common in our part of the world. And they constituted predominantly of fusarium and aspergillus. And this was seen both in large studies from South India, as well as in North China. Confocal microscopy is another important tool, wherever available, that aids the physician to make a diagnosis of fungal keratitis. This is important, because often the corneal infiltrates may be deep stromal, and therefore are not amenable to scraping, or we may not get access to provide a sample to the microbiologist, as regards to the etiology. So the picture on your right is the confocal image of a normal cornea. So what we see in the figure A and B are the anterior and the posterior stroma, where the keratocytes are noted. The figure C shows the single monolayer of the normal human corneal endothelium. And the figure D shows the nerve plexuses, which are present at this level. Figure E and F shows the large stromal nerve fibers, which are often 15 microns, and these sizes actually help us to detect fungus, or any other foreign body or any other pathology that may be there in the diseased cornea. So as we can see in the subsequent slide, these are the confocal images from a patient of fungal keratitis. And here we see that the fungal filaments appear as highly reflective double walled septate filaments. They are 3 to 8 microns, which means they are the same dimension as the subbasal nerve plexus, and they are irregular branching of uniform width. When we see this kind of characteristic clinical picture, we are reasonably sure that we are dealing with fungus as the probable etiologic agent. There are a lot of antifungal drugs that have been recommended for usage in fungal keratitis. Broadly, they have been classified into polyenes, azoles, pyrimidines, and echinocandins. The polyenes and azoles inhibit cell wall synthesis. The pyrimidines and echinocandins lead to cell death. So it was found that caspofungin and the triazoles were much more effective, because they have much lower minimum inhibitory concentration. And as you can see in the MIC 50 and 90 values, they’re significantly lower, compared to, say, natamycin. And it was found that the newer drugs are much more sensitive in killing the fungus. However, does that actually translate to clinical benefit? And to study that, a very important trial was conducted. This is level 1 evidence, which found that the natamycin, which is one of the polyenes, was actually significantly better than voriconazole in managing filamentous hyaline fungi. The mycotic ulcer treatment trial is a landmark study that helps us understand which drugs are the best and are first line in the management of filamentous fungi. And we understood from this study that natamycin is the single best first line molecule in the management of filamentous fungus, and cases that are treated with natamycin have significantly better best corrected visual acuity than voriconazole, and they were less likely to have perforation or require a need for therapeutic penetrating keratoplasty. A follow-up of the mycotic ulcer treatment trial was the mycotic ulcer treatment trial II, or the MUTT II, as it is called. And the study question for this was: Whether additional oral voriconazole was beneficial. When we are adding that to a regimen of topical antifungals for treatment of hyaline filamentous fungus. And this study conclusively proved that there is no additional benefit of adding oral voriconazole to a patient who is being treated with topical antifungal drugs. Intracameral drugs have also been involved, and there have been several anecdotal reports about the use of intracameral agents in the management of deep seated corneal mycotic keratitis. The technique consists of injecting voriconazole into the mid to deep stroma. Usually a dose of 50 micrograms in 0.1 milliliter is administered in divided doses all across the lesion. It is important to see the fluid wave in the corneal stroma, once you inject the drug. And that is the guide to the completion of the treatment. So as it is being shown in this representative video, voriconazole is being injected all around the lesion, and the hydration of the cornea lets us know that we have uniformly spread the drug into the cornea. And usually one to three injections are repeated at 72-hour intervals. And before we declare whether this is an effective treatment, or we should progress to other surgical modalities, such as lamellar keratoplasty or therapeutic keratoplasty. As regards to the evidence for or against intrastromal voriconazole injections, these two papers provide a lot of insight. Initially, the thought was that intrastromal voriconazole tends to help the resolution of recalcitrant and deep mycotic keratitis. However, a later study from the same group found that there is no additional benefit of intrastromal voriconazole when topical voriconazole is given to a regimen of topical natamycin for deep mycotic keratitis. Therefore the verdict is still out, and probably it is recommended to use these modalities of treatment on a case per case basis, and not really as a standardized uniform protocol. A further study that was recently published also found that it is better to have a global approach, or a holistic approach in managing fungal keratitis. And this was labeled as the TST protocol, which is the acronym for topical, systemic, and targeted therapy. So basically it means that the first line treatment is natamycin. Undoubtedly in the view of the evidence that has been accumulating over the years. And in addition, one needs to use systemic antifungals and add topical voriconazole, or continue them, or add intrastromal or intracameral injections. And if there is a poor response, then proceed for therapeutic keratoplasty. The authors reported an 80% success, after following this protocol. We also recently reviewed the surgical aspects of managing fungal keratitis. Lamellar surgeries have been involved for quite some time. And this is important. Because lamellar surgeries are extraocular procedures. There is no harm to the host endothelium. One can easily take off the sutures in a short postoperative course, and wean the patient off the corticosteroids. And it’s also important in a resource-limited setting, where the tissues which are of poorer quality can also be utilized in these cases. So basically the lamellar surgeries or the deep anterior lamellar keratoplasty involves removing the superficial layers of the corneal stroma, and trying to dissect off all the diseased tissue. So that can be achieved by a layer by layer dissection, or by a big bubble technique. So as you can see in this video, DALK is being performed for an infective keratitis. And after partially dissecting into the corneal stroma, we try to dissect off all the unhealthy appearing stroma. And a big bubble DALK is attempted. Once the big bubble is achieved, it is important to dissect off all the corneal stromal layers, before putting in the graft. So there have been quite a bit of publications in this regard, and a large study from China found that out of 55 eyes that were part of this study, 51 eyes resolved with lamellar surgery alone. The histopathology showed that there were fungal hyphae aligned with the corneal stroma. A study showed that about 15% of these cases had recurrences. They were not limited only to fungal keratitis, because this study also talked about advanced infectious keratitis, and they included fungus, bacteria, acanthamoeba, et cetera. And they found that typically all those cases where the Descemet’s were not completely bare, in those cases, there were some lamella where the organisms might have remained, and that led to recurrences. We evaluated in an in-house study 167 corneal buttons of fungal keratitis, and we found that patients who have a heavy fungal load, or who have a poor host response, a poor response to the fungus, usually went in for a surgical option, and therapeutic penetrating keratoplasty was performed in these cases. The histopathology of the buttons showed a lot of fungal hyphae in the deeper stromal layers, and they were also noted to cross the Descemet’s into the anterior chamber. Therefore it is important that one needs to dissect off and bare the Descemet’s, before attempting any lamellar surgery. And in case we suspect a full thickness infiltrate, it is important to completely excise the unhealthy cornea and perform full thickness procedures, rather than go for lamellar procedures, which may predispose to a recurrence. The other important surgery in the diagnosis of a deep seated fungal infection is a corneal biopsy. This is recommended for those lesions which are off the visual axis. And this is a representative video, where we have a deep seated corneal infiltrate, which is away from the visual axis. So it is first important to dissect off the superficial healthy appearing cornea. And then from the deeper layers, the scraping is being taken, and a small block of tissue is excised off from the deeper cornea. And this is subjected to histopathology. Once that has been taken, the flap is secured into position with 10-0 nylon. And when we perform the histopathology, we find that there are fungal filaments in the GMS stain and the Gram stain showed fungal filaments as well. Fungus tends to grow as a plaque on the cornea. This acts as a debulking procedure, which allows the drugs to penetrate and decreases fungal load. So this also helps in fungal keratitis. This is especially important in the setting of a small perforation, where we do not have the need or requirement for a much more aggressive procedure, such as a full thickness penetrating keratoplasty. So it is important to dry off the area, scrape off any overlaying epithelium that is there, and then apply a thin film of cyanoacrylate. This is important to achieve tectonic stability of the globe. Once it is achieved, a small air bubble may be placed into the anterior chamber. There may be adhesions in the back of the cornea, so it is important to sweep the cornea with a blunt instrument. Once you have swept off all the adhesions out of the back of the area of perforations, a small air bubble is placed into the cornea, and it is allowed to heal on its own and a bandage contact lens is placed on top of it. Sometimes this was a case of a patient who had a corneal trauma, and a corneal trauma led to a secondary infection. So the air bubble that you see was the air bubble that was presented at the initial surgery. And there is a lot of tissue necrosis, and there are some broken sutures, where the infiltrate has allowed the sutures to cheesewire through them. So these cases tend to be quite bad, with a lot of corneal melting, and large perforations. So it’s important to scrape off all the necrotic debris, and then apply a thin film of tissue over it. Once that is done, one may make a small port or refresh the previous one, and just augment the amount of air that is there in the anterior chamber. Place a bandage contact lens on top of it, and continue with the drugs and allow the eye to heal before a definitive — or another surgical procedure is contemplated. In the setting of non-healing corneal ulcers, because often fungi leads to chronic ulcers, it is important to scrape off this area and place an amniotic membrane, which assists in reepithelializations and healing. Sometimes they need multiple amniotic membrane grafts, which need to be rolled or set on top of each other and allowed to fuse with fibrin glue. This helps in the healing of these lesions. The other aspect in the management of fungal keratitis is performing a full thickness penetrating keratoplasty, when all other modalities have exhausted themselves. This is required when there is a worsening of the fungal keratitis, or there is a non-response to medical management, a corneal perforation, involvement of the corneal limbus. It is important to first identify the adequate trephine, and then trephine the area that appears to be infected. Once that is done, it is important to make a controlled entry into the anterior chamber. This is an inflamed eye. Therefore it is important to perform more than one peripheral iridectomy. Once that is done, the host bed is prepared. It is important to place the adequately sized donor and secure it into position with 16 interrupted sutures. So this is the suturing that is being performed. And at the end, the chamber is well formed. And the surgery is completed. So sometimes that might be an old perforation, or a patient who has had a tissue adhesive that has been in place, because of a prior perforation. And the lesion has progressed underneath. It is very important to make a very controlled entry in these cases. This is required because there might be some adhesions of the lens-iris diaphragm to the back of the cornea. Therefore, a full thickness penetrating keratoplasty may easily cut off the iris too. So it’s extremely important to do a controlled entry into the anterior chamber. Once it is done, all the unhealthy appearing adhesions have to be removed. It is important to size the trephines accordingly. It is important to oversize the donor by 0.5 millimeters up to 9 millimeters. And from 10 millimeters and beyond, it is important to oversize the donor by 1 millimeter. All the grafts which use a donor which is more than 10 millimeters — one should use 24 sutures to close the graft. After that, any membranes or inflammatory membranes that are formed in the pupillary axis or underneath the iris needs to be carefully excised. These eyes tend to bleed a lot, so it is important to ask the assistant to gently irrigate the surgical field, so that there is a view of the surgical field and we can quickly continue with the dissection before completing the surgery. Once that is in place, it’s important to quickly perform the peripheral iridectomies, before putting the host corneal button. Sometimes there might be a total corneal infiltrate. It is important to wash the exudates and assess exactly the amount of cornea that is actually infiltrated. So when we see a lesion, something like this, it’s important to first make a controlled entry into the chamber. And then gently wash off all the exudates from the anterior chamber. So as the wash is completed, we realized that the infiltrate is somewhere within the cornea itself. And that helps us to size the donor and host trephines accordingly. So this is the wash that is being done. And once the wash is complete, it’s important to irrigate the field. Make the peripheral iridectomies. And then proceed with suturing the host — donor button into the host. This is another such scenario, where there is a longstanding corneal ulceration. So sometimes there might be dense adhesions at the back of the cornea. So it’s important to very gently irrigate them, and excise them, before proceeding with the surgery. As you can see in this case, there is a dense and thick membrane in the pupillary area. Therefore, it is important to carefully dissect off the corneal lamellae, and then make the opening into the pupillary axis. Sometimes the PI may also lead to vitreous prolapse, as you can see in this case. There is a lot of vitreous here. So this is because there was an old standing perforation, and often the lens-iris diaphragm is prolapsed forward. In that case, it is important to wait, realize what has happened, and do a thorough vitrectomy. Once the vitrectomy is done, it is important to identify that there is no loose vitreous wicks, and then proceed with the rest of the surgery. Place a bit of viscoelastic to form the angles of the chamber, and then proceed with the keratoplasty, by putting a donor corneal button in place. Sometimes there are longstanding, large perforations. So as you can see in this case, there is a large inferonasal perforation. And the moment we lift off the diseased cornea, we see that there is a lot of underlying infiltrates and exudates at the pupillary axis. Often there would have been a perforation, and sometimes we see that it’s just the vitreous that is sitting there. It’s important to identify this, and also perform this procedure quickly, because the eye is open, and the expulsive hemorrhage can happen at any moment. A thorough, quick, automated vitrectomy is performed. And after that, the button is quickly secured into position. It’s also important that once we excise off the diseased-appearing cornea, a ledge of tissue remains at the margins of the host cornea. So in this video, what has been demonstrated is that this ledge needs to be excised off very gently and carefully. This serves two purposes. Number one is that it does not allow the donor button to override the host cornea. Number two is that sometimes there might be some infiltrates at the periphery. And it is important that one should excise all the unhealthy-appearing cornea before placing the donor button. So that will minimize the chance of recurrence. Once this has been done, it is important to gently wash the chamber. Also wash the angles of the eye. Quickly perform the peripheral iridectomies before proceeding for the surgery. In some cases, especially those where there has been a perforation, it is important that we very carefully excise the host cornea. This is to avoid injury to the underlying iris, which may be tightly adherent, as you can see in this case. Once that is done, excise off any diseased tissue and the ledge that may be there. After performing this procedure, it is important to check whether there is a complicated cataract that might be there in this case. In such a case, it is important to very carefully remove the cataract. This is important, because if there is an intumescent cataract, your chamber will not form. And in that case, it is important that we take out the complicated cataract and allow the chamber to form, before proceeding into the surgery. In addition to the different modalities that are present, collagen crosslinking in the management of infectious keratitis has been in vogue. This is so because there have been several isolated reports which have found that in end stage infectious keratitis, collagen crosslinking may help in resolution of infectious keratitis. But when we come to fungal keratitis, the body of evidence has been mixed. There have been several reports which have suggested that probably in the setting of mycotic keratitis, the collagen crosslinking procedure is not of much help. However, a randomized controlled trial, which was recently published in Ophthalmology, found that there is very minimal benefit of collagen crosslinking in the setting of fungal keratitis. They used four subgroups for this study, where natamycin and amphotericin B were given alone, as well as with collagen crosslinking, and the effect was studied. And they found there was no improvement in microbiological cure, infiltrate, scar size, percentage epithelialization, or adverse events in those groups subjected to collagen crosslinking, versus those groups where collagen crosslinking was not done. Therefore this gives reasonable evidence that in the setting of fungal keratitis, collagen crosslinking is of not much help. So with this, I would like to thank my colleagues and fellows, and we can take a few questions, before we end the topic for the day. Thank you very much.

>> Thank you, Dr. Roy. So you can go ahead and stop your screen share. And we have about 13 questions open.

DR ROY: So let’s go with the questions. One after the other. So how important is the role of scraping and staining? By the way, are you also seeing the questions on your screen too, as I am seeing them?

>> So everyone can open them on their own screen. Yeah. Everyone can see them.

DR ROY: Okay, great. So scraping and staining is extremely important. As we discussed in the slide on microbiology, the sensitivity and specificity of almost all the scraping procedures are in their 90s. Which mean that if we get a fungal filament on the scraping, then we are reasonably sure that we are dealing with fungus as one of the etiologic agents. And because fungus tend to have a chronic course, it is important that we have evidence of the disease before we embark on a treatment. Therefore, this is extremely important, to have a simple test, as a 10% KOH wet mount, which has a reasonable degree of sensitivity and specificity, before proceeding with treating fungal keratitis. Steroids are not indicated in the treatment of fungal keratitis. In fact, unsolicited steroid use is one of the risk factors for poor prognosis in fungal keratitis. The third question is that of a patient who had undergone amniotic membrane transplantation, and has a descemetocele. If the patient is a fungal keratitis and if the incubation is so long as 6 months, it is probably useful to provide the options of a therapeutic penetrating keratoplasty to this patient. Keratectomy as a treatment of fungal keratitis has been fairly useful. We have reports where there have been successful resolution. Because sometimes the fungus spreads as a plaque on the superficial cornea. Therefore superficial keratectomy, when performed in a controlled manner, actually debulks the infection, and allows the drugs to penetrate into the remaining cornea, and assists in the resolution of fungal keratitis. So the other question is on a rural setting. So in a rural setting, I would say that if you can arrange a basic microscope, which most labs or most pathology/microbiology labs, in the laboratory settings, have them, and if you can use such an inexpensive thing as a 10% KOH wet mount, there is a 90% to 94% probability that you will pick up fungal keratitis. And this is more important, because most of the patients of fungal keratitis are either agricultural workers or patients who have sustained a vegetative trauma. Which is so much more common in a rural setting. And therefore you will actually increase your probability of treating them adequately, if you perform a simple test of a 10% KOH. Therefore it is strongly recommended that it is doable, and it should be done in the rural setting especially. So the clinical difference from bacterial keratitis is that fungus appears as a dry, feathery plaque-like appearance, with a presence of a ring infiltrate, and sometimes there might be satellite lesions. So this helps us distinguish fungus from bacteria. Other than keratoplasty, we discussed all the options. Pythium is an emerging disease. Pythium looks very similar to fungus. And unless carefully sought after, they may appear as hyaline tubules in the wet mount. However, they are not septate. They have been described as dot-like infiltrates, and they typically do not respond to antifungals. If you suspect Pythium, it is best at this point in time to refer early to a cornea service. As these patients need specialized treatment, and sometimes they also do require a penetrating keratoplasty. The primary tests that we should do is the 10% KOH wet mount, and when you have access to the microbiology lab, subject the scraping to both solid and liquid media, to allow for any growth. So if there is no availability of corneal scraping and staining, a basic lab microscope, which is available in any lab or any medical school, with a 10% KOH tablet, which is usually available in any chemist, you can make your own 10% KOH, and keep it in a capped bottle. Usually this remains fresh for the week. So you need to make it once a week. Put a little bit of the scraping directly in the center of the slide. Put one drop of the 10% KOH. Put a cover slip. And see this under a low powered microscope. Any low powered microscope can help you look into a corneal 10% KOH wet mount, and presence of filamentous fungi, which are very common in our setting — it can help you diagnose that and start adequate treatment. So it is really very simple to get it done. I’m sure if you tried once and have seen the pictures on the web or relevant lectures or your studies, I’m sure you’ll be able to catch and have a more definite treatment for your patients. 24 suture graft. So this is a good question. Actually, up to 10 millimeters of the donor cornea, we use 16 sutures. But whenever the grafts are larger than 10 millimeters, 24 sutures is recommended. This is because the larger size will not hold with 16 sutures, and very often you will see a dehiscence of the graft postjunction or there will be a prolapse of iris in the shallow chamber through the lesions. This is because a large graft needs 24 or sometimes 32 sutures for stability. Corneal tattooing. I don’t believe this is relevant to the current topic that we are discussing. Therefore it will be beyond the scope of this talk. Preserve the lens and keep them phakic? Absolutely one should try to preserve the lens whenever possible. In fact, if you have noticed the previous few videos that we were discussing, the lens is actually kept intact, and it is important not to lose the lens, because often lens expulsion is associated with vitreous loss, which actually worsens the prognosis in a penetrating keratoplasty. However, sometimes because in the setting of a perforated corneal ulcer there might be a complicated intumescent cataract, so in those cases, it is important that the intumescent lens is removed. Because if it is not done, it is almost impossible to form the chamber. And if you somehow do form the chamber, we also see these patients come back with the lens sticking to the back of the graft, and a very raised intraocular pressure. So it’s really important to identify this condition and wherever possible, very carefully remove the lens and try to form the chamber. Intracameral and intrastromal injections. So as I discussed during my talk, there is a lot of evidence which goes both in favor, as well as against the use of intracameral drugs. Probably it is important to understand that this is one modality of treating fungal keratitis, and you may want to use them in deep mycotic keratitis, but not as a protocol. Because studies have found that it is as good as giving topical voriconazole, as opposed to giving intrastromal voriconazole. So the evidence is still out. The verdict is still out. And probably one needs to use it on a case to case basis. How do we treat after keratoplasty? In fungal keratitis, the protocol is to use the antifungal drugs for at least two weeks after the surgery. To check the margins and check whether there is any residual or recurrence. And once you are sure there is no residual or recurrence, start corticosteroids as topical treatment. This is important, because when we treat medically, we see that the complete resolution takes two weeks until scarring, before the fungi is eliminated from the area of infection. Therefore two weeks is the ballpark figure on which we decide when to start topical corticosteroids, postkeratoplasty, for fungal keratitis. Confocal is not really a diagnostic tool that is available to a lot of people. Therefore confocal, where available, can be used, and it also needs a fair bit of expertise to perform confocal. Because it is — it touches the eye. Patients tend to be apprehensive, and often there is a lot of noise. Meaning there is a lot of high reflectivity from the inflamed tissues. It is useful in deep mycotic keratitis, where the lesions are not available for scraping. The next question is: Should we scrape under slit lamp? Yes, we do scrape under slit lamp. And in this, we place a drop of topical anesthetic, such as proparacaine, 0.5%, allow the patient to be comfortable, place the patient in the slit lamp, and use a Bard Parker handle with a number 15 surgical blade, or a spatula, and take a little bit of sample for inoculating into the different media. The next question is: Do we prefer to start one medication, or should we combine them? The best molecule that we have at hand right now is natamycin. Therefore natamycin is the first line treatment in the management of fungal keratitis. And then as subsequent studies have also shown, one can also combine different modalities, and observe the response to that. So steroid is absolutely contraindicated in fungal keratitis. Therefore if you are dealing with a case of fungal keratitis, you should not use topical corticosteroids. As this worsens the prognosis. And allows the fungus to spread. The prototypical for corticosteroids after keratoplasty is just as we discussed. After we wait for two weeks, and we give antifungals for two weeks, after therapeutic keratoplasty, before starting topical steroids. There is no role of topical and systemic antibiotics in the management of hyaline filamentous keratitis. This question also we just discussed, that in addition to the antifungals, we may give cycloplegics, but apart from that, we do not add any other medications. Antifungal medications are fairly commonly available in those areas of the world where fungal keratitis is an important public health problem. Therefore it should be good to discuss with your authority where fungal keratitis is important to make availability of antifungal drugs. Especially topical antifungal drugs. These are fairly inexpensive, and probably it should be available in the pharmacy where you practice. In case you do have a lot of fungal keratitis in your practice. So topical treatment… So natamycin, as we understood from the evidence, from the mycotic ulcer treatment trial, natamycin is currently the single best molecule for the treatment of hyaline filamentous fungus. Corticosteroids are absolutely contraindicated, and therefore they should not be used in the management of fungal keratitis. Only when one has performed keratoplasty and after two weeks of natamycin when there is reasonable probability that there is no residual and recurrence, one can start with topical prednisolone. Once you are sure that the fungus is eliminated. But until that time, there is no role of corticosteroids in the management of fungal keratitis. The next question is: Is there any research on antiinflammatory topical, which prevents melting? No, unfortunately, none. We do use oral doxycycline, because they inhibit the tissue matrix metalloproteinases. And they can be advised in patients who have corneal melt. Topical antibiotics are not required in fungal keratitis. The empirical treatment when you get a microbiology negative is a broad spectrum antibiotic. So we use a combination of fortified cefazolin with ciprofloxacin or fortified vancomycin with ciprofloxacin, or some people also use moxifloxacin, as this is a newer generation fluoroquinolone with a broad spectrum Gram positive and Gram negative effect. We do not use 2% povidone-iodine as a topical treatment. Therefore I do not have experience with this line of management.

>> All right. Thank you, Dr. Roy. I think this might be a good place to stop. Okay.

DR ROY: Yes. So thank you, everybody, for joining in. It was great interacting with all of you. And we really appreciate your interest in this topic. In case you do have further questions, do write in. And we’ll try to address your questions. Thanks a lot for your great support, Lawrence. And the team at Cybersight. It’s always great to feature on this forum and speak to attendees all across the globe. Thank you very much.

Download Slides


Last Updated: October 31, 2022

1 thought on “Lecture: Fungal Keratitis”

Leave a Comment