An Emerging Global Threat of Mycotic Keratitis Caused by Uncommon Fungal Species: A Systematic Review and Meta-Analysis

Abstract

Purpose: The purpose of this study was to analyze epidemiological characteristics, clinical spectrum, and treatment outcome of mycotic keratitis (MK) caused by uncommon species.

Methods: The systematic review in compliance with Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines has been registered on “PROSPERO” (CRD42023410825), whereas the systematic literature search was conducted on PubMed, Cochrane, Google Scholar, and Semantic Scholar on uncommon MK from January 1963 to March 2023. The main keywords for the literature search comprised: “mycotic keratitis,” “fungal keratitis,” “keratomycosis,” “oculomycosis,” “uncommon,” “rare,” “emerging,” “atypical,” “unusual,” and various combinations of it.

Results: The study identified a pool of 13,662 articles. Five hundred sixty-six studies were deemed suitable, and 186 studies met the inclusion criteria to ascertain the pooled prevalence. A total of 154 uncommon fungal species/genera were identified among 61 countries. Australia exhibited the highest pooled prevalence, whereas India reported the maximum number of cases, genera, and species. Clinical presentation varied from mild to severe disease with unequivocal response to standard therapeutic regimes. Microbiologically proven species with reported sensitivity had better visual and structural outcomes.

Conclusions: The study provides the first-ever global prevalence estimate of MK caused by uncommon species, estimating 154 fungal genera/species with varying severity, assuming that several culture-negative cases with specific predispositions would also qualify as cases of MK. Microbiologically proven species with reported sensitivity have better visual and structural outcomes. Cases that fail to respond to standard therapy should be re-evaluated for uncommon species with a high index of suspicion. Prompt diagnosis with culture and sensitivity analysis, PCR or in vivo confocal microscopy (IVCM)-based test and timely treatment remain the most important factors in salvaging visual and structural function.

Translational Relevance: This review catalogues the epidemiological, clinical, and morphological traits of rare fungi implicated in atypical MK and also provides a global prevalence estimate. Further, it emphasizes the role of implementation of specialized diagnostic techniques and collaborative efforts to combat the visual disability stemming from afflictions due to rare or atypical fungal species. Information on continent and country wise prevalence of atypical species would be helpful in appropriate management of such cases, in event of inconclusive diagnosis and consequent suboptimal response to treatment.

Introduction

Infectious keratitis is a major cause of corneal blindness and one of the leading causes of preventable blindness worldwide. Out of all the etiological causes of infectious keratitis, mycotic keratitis (MK) poses a substantial global health concern with incidence ranging between 1% and 44%.^1–3 

The species that commonly cause infection of the cornea include Fusarium, Aspergillus, and Candida.^4–6 However, the emergence of uncommon fungal species posing significant challenges in disease management necessitates in-depth evaluation, discussion, and meaningful review. These lesser-known fungal pathogens have been reported in sporadic cases, which warrant prompt diagnosis through microbiological analysis and subsequent intervention.^7–12 Any inadvertent delay in initiating appropriate treatment can cause unfavorable visual and structural outcomes.^13–15 

The present study aims to appraise the existing literature on a comprehensive analysis of epidemiological characteristics, diagnosis, and treatment outcomes of uncommon MK to efficiently manage such cases and prevent end-stage blindness. The critical insight and clinical relevance of this systematic review (SR) lie in the detailed perspective of rare fungal species presenting as a global threat. Further, the recommendations based on the analysis of various interventions can aid clinicians in making informed treatment decisions to optimize treatment outcomes. 

Materials and Methods

Search Strategy

This study was conducted in accordance with Preferred Reporting Items for Systematic Reviews & Meta-Analyses (PRISMA) guidelines^2,16 and has been registered on the international prospective register of systematic reviews “PROSPERO” (registration number: CRD42023410825). A comprehensive literature search on uncommon MK was carried out across multiple databases, including PubMed, Cochrane, Google Scholar, and Semantic Scholar. The primary keywords used in this extensive literature search included “mycotic keratitis,” “fungal keratitis,” “keratomycosis,” and “oculomycosis,” in conjunction with “uncommon,” “rare,” “emerging,” “atypical,” “unusual,” and rare fungal species. The search was confined to articles published in the English language from January 1, 1963, to March 10, 2023 (Supplementary Table S1, search strategy). 

Study Selection and Quality Assessment

The published articles were initially screened via titles and abstracts and followed by full-text reviews. We included human studies related to MK caused by atypical/uncommon fungal pathogens. Inclusion criteria covered randomized controlled trials, prospective (Ps), retrospective (Rs), ambispective (As), case series (Cs), cross-sectional studies (CSSs), and case reports (CRs). Excluded from our analysis were studies related to keratitis caused by non-fungal (bacteria/protozoa/virus) or common fungal pathogens, abstract-only papers, review articles, comments, editorials, books, articles without full-text availability, cell culture, and animal studies. Studies reporting mixed infections, from non-fungal/common fungi and uncommon fungi, were excluded because such cases often involve the clinical course and outcomes confounded with cross-infection. However, isolated cases where bacterial infection ensued secondary to atypical fungal infection (superinfection) were included. 

The Joanna Briggs Institute (JBI) Critical Appraisal tool was used to assess the risk of publication bias.¹⁷ Two reviewers independently screened and assessed the quality of studies, achieving consensus through group discussion. 

Data Extraction

Data collection involved the extraction of various variables, including the first author’s name, publication year, study design, country, demographic data, fungal species, number of atypical MK cases, total MK cases, frequency of various fungal genera/species, microbiological investigations, management including medical or surgical interventions, and clinical outcomes related to different treatment strategies. 

For inclusion in our analysis, cases needed confirmation through microbiological diagnosis. In studies on infectious keratitis, we exclusively extracted data from rare fungal species. 

Data Synthesis

The studies were categorized into three groups: Cr/Cs, original studies, and prevalence studies. The primary objective of this review was to determine the epidemiological characteristics, distribution of uncommon MK, and treatment outcome. We compiled a list of rare fungal genera/species from all studies, categorizing them by countries, and documenting implicated genera/species. Additionally, the pooled prevalence of these genera/species across different continents through a meta-analysis (MA) of prevalence studies was ascertained. 

A species-specific comprehensive analysis was conducted for cases where both management and outcomes were fully documented. Individual case data were collected, however, when not feasible, summary estimates were incorporated to evaluate the treatment response of various genera/species, determining the most effective approach – whether medical, targeted, or surgical. Response to treatment was assessed in terms of visual or structural outcome (corneal scar or structural disfigurement due to evisceration/enucleation). Visual outcome was categorized into good and poor based on the World Health Organization (WHO) classification of blindness, whereas visual acuity of LogMAR 1.3 or better was categorized as good and less than 1.3 as poor.¹⁸ 

Statistical Analysis

The MA was conducted using prevalence estimates within 95% confidence intervals (CIs). Heterogeneity impact was assessed using the I² statistic: 0% = none, 25% = low, 50% = moderate, and 75% = high heterogeneity. 

We used the Freeman-Tukey double arcsine transformation to stabilize variance and mitigate prevalence estimates.¹⁹ Random-effect models were used to aggregate prevalence estimates, considering study variability from individual studies. Forest plots visually summarize the results. 

Subgroup analysis examined fungal keratitis prevalence across continents and countries. Publication bias was assessed using Egger’s linear regression test and metabias command, with potential bias corrected using the “metatrim” command’s trim-and-fill procedure. Sensitivity analysis was done to assess the robustness of the MA. STATA version 16 and “metaprop” packages were utilized for statistical analysis, with a significance level of 0.05 for 2-tailed P values. 

Results

The search strategy initially identified a pool of 13,662 articles. Finally, 566 studies were deemed suitable for the Supplemental References. From this set, 186 studies met the criteria for inclusion in the MA to ascertain the pooled prevalence of uncommon MK (Fig. 1; Supplementary Table S2). 

Figure 1.

 

The PRISMA flow diagram.

The majority of studies primarily originated from India (n = 200, 35.3%), followed by the United States (n = 86, 15.2%), Australia (n = 24, 4.2%), and China (n = 23, 4.1%). Within 566 studies are 320 Cr, 51 Cs, 100 Rs, 85 Ps, 8 CSSs, and 2 randomized control trials (RCTs; Supplementary Tables S3A, S3B). All studies have not provided detailed data on study variables like demographics, investigations, and management. Cr and Cs have described the aforementioned parameters concerning particular patients, whereas the majority of Ps and Rs studies have described a common pool of MK cases, including both common and uncommon species. Thus, complete information on uncommon species could not be obtained from all 566 studies. To address this variability, subsequent sections present data in units of “n” and “%” as available from a specific number of studies and their reported cases, varying over different variables. 

Demography and Risk Factors

Data on gender distribution was available in 392 studies, including 921 cases. Men (n = 618) were affected more often than women (n = 303). Mean age was 50.48 years (range = 5–87 years). Risk factors mentioned in 350 studies revealed trauma as the most common predisposition (n = 598 cases). Trauma with vegetative matter was identified as the cause in 229 instances (38.3%). Postoperative infection, contact lens usage, steroid administration, and systemic conditions like diabetes mellitus were other causative factors in the development of keratitis (Table 1). 

Table 1.

 

Consolidated Characteristics, Demographic, and Predisposing Factors of 566 Studies of Uncommon Mycotic Keratitis

 

Geographic Distribution

Investigating 566 studies across 61 countries spanning 6 continents, revealed intriguing insights into the geographic distribution of atypical MK (see Supplementary Tables S3C, S4). A total of 10,436 cases were documented, comprising 10,452 fungal isolates, uncovering 154 rare genera/species. Curvularia (n = 2652), Alternaria (n = 1753), and Acremonium (n = 1001) were the leading causes of keratitis among atypical fungal isolates followed by Penicillium, Bipolaris, and others (Table 2). The highest number of cases are reported from India and China, closely followed by the United States (Fig. 2A). Notably, a similar trend was observed in these countries concerning the diversity of rare genera/species (Fig. 2B), further highlighting the consistent pattern across the regions studied. European countries like Belgium, Portland, Poland, and Sweden exhibited significantly fewer genera/species, emphasizing the distinct disparity in both the prevalence and diversity of atypical MK on a global scale. 

Table 2.

 

Frequency Distribution of Uncommon Fungal Genus/Species

 

Figure 2.

 

(A) Map showing worldwide distribution of uncommon mycotic keratitis cases. (B) Map showing worldwide distribution of uncommon fungal species causing keratitis.

Genera/Species

One hundred fifty-four uncommon fungal genera/species have been classified into two broad categories and subcategories based on mycelium morphology and spore generation (Supplementary Fig. S1). The majority of included genera/species (n = 126, 81.81%) belong to the division Ascomycota, often referred to as sac-fungi, typically seen in multicellular form and branched/septate mycelium. 

Clinical Features

Clinical features seen in atypical MK are similar to those of common species, however, the severity grade has been reported to be more intense owing to delayed diagnosis and virulence of the species. Reported clinical manifestations comprised of corneal ulcer, corneal abscess, hypopyon, satellite lesions, corneal melt and sloughing, infectious uveitis, endophthalmitis and panophthalmitis with varying severity. Epidermophyton floccosum, Malassezia restricta, and Microspaeropsis have feathery, indistinct margins with deep yellow infiltrates. Curvularia keratitis is more likely to have a corneal ulcer with a raised surface and the presence of hypopyon. Species such as Acrophialophora, Hormographiella aspergillata, Podospora austroamericana, Scedosporium, Apiospermum, Schizophyllum commune, Trichophyton Nattrassia mangiferae, Arthrographiskalrae, Purpureocillium lilacinum, Rhodotorula, and Chrysosporium were characterized by deep infiltrates presence of endothelial plaque, hypopyon, and rapid progression to corneal melt. 

Management

Management data (investigative modalities, treatment, and clinical outcomes) has been provided in a subset of 413 studies, including 778 cases, which is detailed in Supplementary Table S5. 

Diagnosis

Microscopy and culture were done in 96.53% (n = 751) and 100% (n = 778) cases, whereas culture-based methods exclusively identified genera/species in 77.89% (n = 606). Polymerase chain reaction (PCR)-based DNA identification was used in 20.05% (n = 156), matrix-assisted laser desorption ionization time of flight mass spectrometry (MALDI-TOF MS) in 2.31% (n = 18) and in vivo confocal microscopy in 3.47% (n = 27; Fig. 3A). 

Figure 3.

 

(A) Summary of various microbiological assays reported in 778 uncommon mycotic keratitis cases. (B) The overall visual and structural outcomes in 778 cases.

Treatment

There were 98.58% of the patients (n = 767) who received medical interventions, primarily with topical antifungal agents like Natamycin, Voriconazole, and Amphotericin B (AMB). Oral antifungal agents included Voriconazole, Ketoconazole, Itraconazole, and Fluconazole. Voriconazole, topically or orally, was effective for refractory keratitis. Combination therapy with Voriconazole and other agents like Terbinafine, Posaconazole, and Caspofungin was effective in Tintelnotia, Purpureocillium, and Alternaria species. Sphaeropsis subglobosa and Subramaniula asteroids responded well to Clotrimazole and Isavuconazole, respectively. Fifteen percent (n = 117) of the patients who were started on medical therapy later required targeted therapy with AMB and Voriconazole. Out of which, 20.51% (n = 24) resolved without further intervention. 

Surgical interventions were needed in 40.87% (n = 318) of the cases, including corneal transplants in 81.13% (n = 258; see Supplementary Table S5). Notable cases include a Phomopsis case requiring repeat optical keratoplasty and eight cases needing repeat therapeutic keratoplasty involving Paecilomyces (3 cases), Arthrographis kalrae, Rhodotorula, Chrysosporium, Schizophyllum commune, and Phialophora species. Arthrographis Kalrae cases necessitated three repeat transplants and still showed recurrence. Destructive procedures were performed in 11.63% (n = 37) of the surgical cases. 

Treatment Outcomes

Visual outcomes were reported in 488 cases, with 83.40% (n = 407) achieving visual acuity > LogMAR 1.3. Regarding structural outcomes among 290 cases, 84.14% (n = 244) healed as corneal scarring, whereas 14.83% (n = 43) showed structural disfigurement from destructive procedures or phthisis bulbi. Additionally, there were two cases (0.69%) of clear corneal grafts and one case (0.34%) with recurrent infection (Fig. 3B). 

Risk of Bias

The application of the JBI tool revealed that single-arm studies were associated with a low risk of bias (75%), whereas cross-sectional studies had a moderate risk of bias (56.25%; Supplementary Table S6). 

Meta-Analysis

Pooled prevalence with 95% CI across 186 studies was 23% (21%–25%; Table 3). Due to high heterogeneity, a subgroup analysis was performed, grouping data by continents and countries. Australia had the highest uncommon species prevalence at 37% (20% to 54%), followed by Asia = 24% (21%–26%), South America = 24% (10%–39%), North America = 21% (15%–27%), Africa = 20% (12%–28%), and Europe = 17% (10%–23%; see Supplementary Figs. S1A–S1F). 

Table 3.

 

Countrywise Pooled Prevalence of Uncommon Fungal Keratitis

 

Country-specific analysis showed the highest prevalence in Pakistan = 52% (32%–72%), followed by Taiwan = 43% (33%–53%). The United Kingdom had the lowest prevalence at 9% (5%–13%). Whereas considering the reported studies on uncommon species, India was found to have maximum studies (99) with a pooled prevalence of 18% (17%–18%) as opposed to 16 countries reporting single studies with varying prevalence rates. Few countries exhibited non-significant heterogeneity (I²  =  0.00%, P  =  0.64), indicating diverse factors contributing to heterogeneity (Supplementary Fig. S2). Sensitivity analysis based on risk of bias showed low-risk studies have an overall prevalence of 23% (21%–25%), which also aligns with the overall prevalence, whereas high-risk studies showed a prevalence of 20% (9%–31%) with wider CIs. Sensitivity analysis based on the number of countries with 3 or more studies showed that Taiwan (4 studies) had the highest prevalence of 45% (29%–61%), followed by Australia (4 studies) 37% (20%–54%; Supplementary Figs. S3, S5). 

Publication bias was detected through the Egger’s test (P = 0.03), primarily stemming from methodological differences and chance variations. Trim and fill analysis, addressing this bias, removed 19 studies, resulting in a symmetrical funnel, indicating no publication bias. Despite this, the new pooled effect remained close to the original calculation at 16% (13%–24%), suggesting the publication bias did not significantly impact the robustness of the MA (Supplementary Fig. S4). 

Discussion

Given that MK is increasingly reported as a disease with varied distribution, challenging course, and unpredictable outcome, it is pertinent to identify major determinants of visual and structural outcomes. One of the important factors remains the misdiagnosis or delayed diagnosis due to uncommon fungi. There is an absolute lacuna of studies quantifying the extent of rare MK. The principal insight that prompted us to undertake this review is the invariably grave visual and structural prognosis in cases of undiagnosed MK and consequent inapt treatment.^2,20,21 It is particularly worth mentioning here that during the conceptualization of this study, we contemplated 60 to 70 odd species to be the probable search result. However, to our surprise, a total of 154 uncommon genera/species were identified with worldwide distribution (see Table 2). 

A detailed systematic perusal of 566 articles revealed 10,452 uncommon fungal isolates implicated in the causation of MK. The epidemiological and geographic variation in the prevalence pattern of these genera/species, along with response to treatment and outcome, is reflective of a unique effect. 

Prevalence

The prevalence pattern of particular species is largely dependent on the geographic and sociodemographic settings. Multiple analyses from India have reported MK prevalence ranging between 21.5% and 39.8%.^22–25 Whereas China, the United States, Australia, and Nigeria are at 33.18%, 6.06%, 1.83%, and 15.79%, respectively.² The reported incidence is higher in countries within tropical and subtropical regions with warm and humid climates conducive to fungal growth. In developing countries, it is largely a disease prevalent in rural settings consequent to injury by vegetative matter, animal residues, and soil-contaminated objects, particularly common in agricultural workers, laborers, and animal handlers. Furthermore, individuals with aforementioned occupations, being in lower socioeconomic strata, have limited access to medical resources, cost constraints, and treatment dropouts, leading to increased severity and potential complications.^21–25 

The pooled prevalence of uncommon MK is notably high in Australia (37%), indicating that such affections are not infrequent, especially in its tropical regions²⁶ with conducive warm, humid climatic conditions. A similar pattern is observed in Asia, with a prevalence of 24%. Here, India and China emerge as the major contributors to reported studies. This increased incidence can be attributed to climatic conditions, occupational hazards among agrarian populations, and risk of ocular injury, particularly with vegetative matter. Australia's higher prevalence appears to be equivocal on account of more uncommon species compared to overall prevalence. Whereas India and China have an overall increased prevalence of MK and thereby proportionately lesser prevalence of uncommon species. Europe, on the other hand, exhibits the lowest prevalence (17%), likely due to environmental and occupational factors, resulting in fewer cases of MK.^27,28 Environmental conditions favoring a cold climate are detrimental to fungal growth with a resultant decrease in prevalence rates. In these geographic areas, the common predispositions are contact lens usage and systemic diseases like diabetes mellitus. Given the availability of adequate medical resources, awareness about the disease, timely treatment, and compliance in developed countries, the disease severity and treatment outcomes are reportedly better.^29,30 Country-specific analysis suggests that Pakistan exhibited higher pooled prevalence (52%), likely due to limited studies and smaller sample sizes compared with India and China (see Table 3). Whereas considering three or more studies from a country, Taiwan is found to have the highest pooled prevalence (46%), which might be attributed to genetics, environmental factors, and lifestyle choices. 

Demography and Risk Factors

Age and gender incidence has been variably reported with some predisposition toward male subjects, particularly in younger age groups of 20 to 50 years,^27–30 the male:female ratio ranging from 70.1%:29.9%³¹ to 67.3%:32.7%.²¹ This can mainly be attributed to productive age groups being exposed to trauma, particularly with vegetative matter owing to occupational risks.^27,32 

The present review shows that atypical MK predominantly affects men between 30 and 70 years, whereas they are found to impact only a small percentage (1.09%) of children. Children exhibit a lower susceptibility to MK due to their limited exposure to objects of injury and parental supervision, with prevalence rates ranging from 4% to 14.9%.^33,34 However, Panda et al., in pediatric patients, observed a significant number of uncommon strains (46%) to be the causative agents.³⁵ 

Major risk factors implicated worldwide in the causation of MK are ocular trauma, particularly with vegetative matter, steroid administration, postoperative infection, contact lens usage, and systemic disorders. Our review also ascertains ocular trauma as the main predisposition to atypical MK. Further, a direct correlation between atypical MK and contact lens usage has been seen. Certain species, such as Metarrhizium anisopliae, Arthrographis kalrae, and Tintelnotia destructans, have been specifically associated with contact lens use.^36–42 Another implicated risk factor is inappropriate use of steroids in cases either misdiagnosed as viral keratitis or where the diagnosis was delayed. Similar observations are consistent across various common fungal species as well.^43–47 

Geographic Distribution

The prevalence pattern and epidemiological distribution of MK varies widely over different geographic areas. The geographic distribution of uncommon MK is reflective of a pattern consistent across common fungi as well.²⁹ Certain areas like India and China experience a higher frequency of cases due to the complex interplay of environmental, occupational, and sociodemographic factors. This trend is evident not only in an increased number of cases but also in the notable diversity of fungal species, with India reporting a total of 83 uncommon genera/species (see Supplementary Table S4). However, regardless of the reported frequency of uncommon fungi in different geographic locations, it is imperative to understand that these numbers may not accurately represent the actual magnitude owing to improper sampling techniques or inconclusive cultures in patients already started on antifungals. Additionally, lower socioeconomic strata and residents of rural areas might not visit long-distance ophthalmic centers and financial constraints with consequent under-reporting of cases and, thereby, the prevalence pattern. Further, most studies conducted at tertiary healthcare facilities receive patients in advanced stages, sometimes superimposed with other microorganisms, resulting in underestimation of species-specific prevalence and consequent overestimation of poor outcomes. 

Species

Sac fungi with septate mycelium are considered to be more invasive and virulent as compared with other divisions. Species belonging to Ascomycota division, by their morphology, dense infiltration, and tissue lysis, can cause severe ulceration, whereas yeast species cause surface plaque formation. The most common species implicated in the causation of MK belongs to the genus Fusarium, Aspergillus, and Candida.^4,15 However, other uncommon genera/species and their related risk factors, clinical features, and spectrum of etiological agents have also been increasingly reported. In the context of most predominant fungal genera, Curvularia, Alternaria, Acremonium, Penicillium, and Bipolaris, each documented in over 500 instances, can now be justifiably classified as common genera due to their frequent occurrence, followed by Mucor, Scedosporium, and others (see Table 2). 

When considering the virulence of studied species, it was seen that Acrophialophora, Hormographiella aspergillata, Podospora austroamericana, Scedosporium, Apiospermum, Schizophyllum commune, and trichophyton are generally invasive with poor visual and structural outcomes.^48–54 Certain species like Nattrassia mangiferae, Arthrographiskalrae, Purpureocillium lilacinum, Rhodotorula, and Chrysosporium, required multiple therapeutic keratoplasty (TPK) with equivocal response indicating a relentless course.^41,55–58 It is thus imperative to be apprised of the potentially severe attributes of these species for close follow-ups and timely interventions in such cases. 

In tropical and subtropical regions, MK is generally caused by filamentous fungi from the genera Aspergillus and Fusarium. Conversely, yeast-associated MK due to the genus Candida has been observed from temperate regions.^15,29,59 Consistent with this observation, common species implicated in the causation of MK are Aspergillus, Fusarium, and Candida, whereas the most common species implicated in the causation of uncommon MK are Curvularia. 

Diagnosis

Identification of uncommon or rare fungal species warrants a high index of suspicion, culture/sensitivity, and appropriate molecular-based methods. Various stain-aided microscopy techniques like Gram stain, KOH, or lactophenol cotton blue offer preliminary differentiation among bacterial, yeast, and filamentous fungal pathogens. However, it may be challenging to identify specific genera and species of fungus.^28,60 Therefore, a comprehensive approach involves combining microscopy with culture and sensitivity analysis. Culture-based identification, relying on colony morphology and color, remains the most widely used investigative modality.^29,59 However, it is not without limitations, such as inadequate samples, prior antimicrobial use, and incorrect media inoculation leading to false negative results. Uncommon species often require extended inoculation periods and specialized plant-based agars like potato dextrose or cornmeal agars for accurate identification.⁶⁰ 

For the convenience of ophthalmologists and microbiologists, we have created an image plate that displays the anatomic structure and colony morphology of all 154 fungal species Figure 4. 

Figure 4.

 

Diagrammatic representation of morphology of reported uncommon fungal genera/species.

Figure 4.

 

Continued.

In 77.89% of the cases, the implicated species could be isolated by culture/sensitivity and morphological identification, whereas molecular assays like PCR provide rapid identification with an accuracy rate of 92.6% for specific genera/species within 4 to 8 hours.⁶¹ Additionally, MALDI-TOF has emerged as a promising tool for early diagnosis, particularly for rare fungal species.^62,63 However, the widespread adoption of these molecular diagnostic modalities is impeded by their high cost. In vivo confocal microscopy (IVCM) has the added advantage of offering real-time corneal imaging at the cellular level, bridging diagnostic gaps^64,65 with the limitation that it may be available at a select few centers. Nonetheless, this study highlights the primary importance of Culture sensitivity for accurate diagnosis and appropriate treatment selection. 

Treatment

MK is increasingly gaining global attention, leading to its inclusion in the WHO list of neglected tropical diseases.⁶⁶ Given that MK caused by uncommon species can potentially have an epidemiological variance, the choice of treatment may also vary. Variations in antifungal susceptibility patterns render it imperative to ascertain the definitive species and, thereby, their sensitivity-guided treatment. 

This review shows that 98.6% (n = 767) of patients received medical intervention. Topical antifungal agents remained the cornerstone of treatment. Natamycin, voriconazole, and AMB have shown effectiveness against a wide range of rare fungal species, supporting their continued consideration as primary choices in antifungal treatment with supplementation of systemic antifungals in severe or deep-seated keratitis. In refractory cases or resistant fungal species, combination therapy should be initiated. A combination of Voriconazole with Terbinafine, Posaconazole, or Caspofungin has shown promise and should be considered when appropriate.^67–72 

Targeted delivery of antifungal agents, like AMB or Voriconazole, via subconjunctival, intrastromal, or intracameral routes, provides higher drug concentrations at the site of infection. Intrastromal Voriconazole is effective in fungal keratitis not responding to standard topical therapy, particularly against Alternaria and Acremonium.^73–76 Intracameral /intrastromal AMB is effective against certain filamentous fungi, especially in cases with deep mycosis.^77–79 Further studies have shown the effectiveness of targeted therapy, particularly for species like Fonsecaea, Lasiodiplodia, Cylindrocarponlichenicola, Scedosporium, Metarrhizium, and Paecilomyces.^80–85 

Surgical interventions should be considered depending on the severity and progression of keratitis with individualized type and timing of the procedure. Keratoplasty, as the main vision restoration procedure, is primarily indicated in medical treatment failure. However, in advanced keratitis, severe corneal thinning, and impending perforations, the decision to perform keratoplasty must be taken early. 

This review revealed that surgical procedures were required in over a third of the cases, varying from 11.20% to 100% depending on the genera/species. The choice of treatment was TPK, followed by destructive procedures, surgical debridement, and glue glue-BCL (BCL). Genera/species like Alternaria, Acremonium, Beauveria bassiana, and Colletotrichum required TPK consequent to suboptimal response to medical therapy. Certain genera/species like Nattrassia mangiferae, Arthrographis kalrae, Purpureocillium lilacinum, Rhodotorula, and Chrysosporium, required multiple TPKs with an equivocal response indicating a relentless course (see Supplementary Table S5). Studies on common fungal species have reported that 19.7% of the patients require TPK and 18.9% require evisceration, particularly when treatment is delayed by 10 days or more, indicating that common species are largely responsive to instated treatment as opposed to uncommon ones, which may have a primarily progressive course.³¹ 

Treatment decisions should take into account the specific fungal pathogen, its susceptibility profile, and the patient’s clinical condition. Empirical treatment should be avoided and antifungal sensitivity testing should mandatorily be conducted to guide treatment choices. Continuous monitoring of treatment response is essential, necessitating prompt adjustments in case of inadequate response. 

The early intervention led to good visual acuity in many species. However, 17 virulent species required surgical procedures with modest visual/structural outcomes. This is in contrast to visual acuity, as reported for common fungal keratitis. Studies from Pakistan, East Africa, Germany, and the United Kingdom have reported final visual acuity < LogMAR 1.3 in 59%, 66%, 57%, and 56% cases, respectively. Given these figures, it is expected that approximately 94,753 to 115,810 of eyes are surgically removed each year. In countries where eye care is suboptimal, the loss of eyes will probably be greater.^86–89 

This review shows that medical interventions were effective for genera such as Thielavia, Subramaniula, Plectosphaerella, Papulaspora, Malbranchea, Epidermophyton floccosum, Malassezia restricta, and Microspaeropsis olivacea, often achieving good visual acuity.^90–92 Whereas keratitis caused by Pyrenochaeta, Phaeoisaria, and Nigrospora resulted in corneal scarring and subsequent visual impairment.^48,93–100 Highly virulent genera/species like Scedosporium, Schizophyllum commune, and Trichophyton present significant challenges even after species identification and failure in salvaging the affected eye despite all measures.^52–54 Hormographiella aspergillata and Podospora austroamericana are invasive fungal genera/species associated with poor visual and structural outcomes.^49,50,54,101 Such emerging fungal pathogens and their resistance to certain antifungals have contributed to poor outcomes, as is being increasingly reported.^1,28 

Limitations and Strengths

This study suffers from a deficiency of synchronized data in reviewed studies about prevalence, demography, diagnosis, treatment, and outcome. Reviewed P/R studies comprised of prevalence pattern and clinical description of MK and not exclusively of uncommon keratitis, thereby, the data regarding diagnosis and management of uncommon MK were mainly extracted from Cr/Cs, rendering a restricted homogeneity toward cases reflecting the study parameters in various subsections. 

The major strength of this MA lies in its extensive data search, synthesis, and analysis of encyclopedic rare fungal species implicated in the causation of keratitis. The geographic distribution and magnitude of studied species can serve as a diagnostic differentiator in challenging scenarios. Additionally, we have showcased the morphological features of all uncommon fungi with the intent to provide valuable diagnostic tools in cases of ambiguous identity (see Fig. 4). Another major strength relates to the sensitivity analysis, which shows that low-risk studies’ prevalence and overall prevalence remain the same, nullifying the effects of high-risk studies. 

Conclusions

The study fills a crucial knowledge gap by providing an estimate of the global prevalence of MK caused by uncommon fungi. Contrary to conventional beliefs, our research reveals that uncommon fungal genera/species are more prevalent than previously thought. We have estimated 154 fungal genera/species which can cause MK with varying severity, assuming that several culture-negative cases with specific predispositions would also qualify as cases of MK. Curvularia, Alternaria, Acremonium, Penicillium, and Bipolaris can now be classified as common genera due to their frequent occurrence. Australia has the highest pooled prevalence of atypical MK, whereas India reports a maximum number of cases, genera, and species. 

The burden of fungal keratitis is predominantly in developing countries, with limited access to advanced diagnostic techniques and treatment resources. Primary diagnosis of fungal genera can be made by culture and sensitivity analysis, being cost-effective and having wider availability. In cases with inconclusive results, other advanced techniques like MALDI-TOF and PCR have better sensitivity and can be utilized. However, due to cost constraints and limited availability even at higher centers, these techniques have been reported to be used less frequently, particularly in South Asian developing nations, ironically where the prevalence is higher. Guided by the insights from published literature, we have designed an algorithm for a step-by-step diagnostic approach in suspected rare fungal keratitis. This shall aid in standard clinical practice and better treatment outcomes (Fig. 5). 

Figure 5.

 

Diagnostic algorithm for uncommon fungal keratitis.

In most studies, microscopy and culture were the primary diagnostic tools for atypical MK, whereas molecular diagnostic techniques were used for subspecies identification. Microbiologically proven species with reported sensitivity have been shown to have better visual and structural outcomes as opposed to those with inconclusive microbiological identification. Thus, cases that fail to respond to standard therapy should be carefully re-evaluated for implicative uncommon species with a high index of suspicion. 

The mainstay of treatment continues to be topical and systemic antifungal medications. The efficacy of Natamycin, Voriconazole, and AMB against a variety of uncommon fungal species supports their continued use as first-line antifungal treatments. Systemic antifungals should be added in severe or deep-seated keratitis. Combination therapy of Voriconazole with Terbinafine, Posaconazole, or Caspofungin should be instated in refractory cases. Targeted delivery of antifungal agents, such as AMB or Voriconazole, via subconjunctival, intrastromal, or intracameral routes should be considered in cases with deep stromal /endothelial involvement and sloughing keratitis. At the same time, corneal transplantation remains the main vision restoration procedure in medical treatment failure. 

Given the wide diversity of rare fungal species causing keratitis, collaborative research efforts across institutions and countries should be encouraged. It seems logical to engage a multidisciplinary team consisting of ophthalmologists, microbiologists, and infectious disease specialists to ensure comprehensive care and optimal outcomes for patients with rare MK. 

The article not only provides valuable treatment recommendations but also equips clinicians and researchers with essential insights to improve MK management and contribute to evidence-based guidelines. In essence, this article is valuable for its contribution to understanding and managing uncommon MK, offering recommendations to optimize treatment outcomes for this challenging eye condition. 

Synopsis

The study emphasizes the prevalence of atypical MK as a developing epidemic. Further, the authors believe prompt diagnosis and early treatment initiation to be the most important factors in effecting better visual outcomes. 

Translational Relevance

MK caused by atypical organisms poses a significant global health problem with increased rates of sight-threatening complications. The review provides a comprehensive analysis of the different types of rarer etiologies implicated in MK across 61 countries, thereby filling the lacunae in literature about the epidemiological pattern of such infections. The results emphasize the need to combine culture-based methods with advanced diagnostics to improve accuracy and hence facilitate effective treatment strategies. Information on continent and country wise prevalence of atypical species would be helpful in appropriate management of such cases, in the event of inconclusive diagnosis and consequent suboptimal response to treatment. 

Further, such cataloguing of the clinical and morphological traits of these rare fungi helps provide a global prevalence estimate. In doing so, the study advocates for the wider implementation of specialized diagnostic techniques and collaborative efforts to combat the visual disability stemming from atypical MK, especially in resource-limited countries. 

Acknowledgments

The authors thank Muhammad Noufan CM, Athulya TH, and Mithali Rathi for the images and Sabir Mohammed for his assistance in the submission process. Authors acknowledge support of the Ruth M. Kraeuchi Missouri Endowed Chair Ophthalmology Fund, University of Missouri Columbia, MO, USA to RRM. 

The contents in the article do not represent the views of the United States Department of Veterans Affairs or the United States Government. 

Disclosure: M. Gautam, None; B. Lal, None; S. Patel, None; R. R. Mohan, None; A. Barathi, None; N. Yadav, None; S. K. Verma, None; R. Nyodu, None; A. Sampath, None; D. Koshti, None; B. Sharma, None