A global perspective of pediatric non-viral keratitis: literature review

Incidence and clinical characteristics of paediatric keratitis

BACKGROUND/AIMS

To report the incidence and clinical characteristics of paediatric keratitis diagnosed over a 10-year period in a well-defined population.

DESIGN

Retrospective, population-based study.

METHODS

Setting: multicentre.

POPULATION

patients (<19 years) diagnosed with keratitis as residents of Olmsted County from 1 January 2000, through 31 December 2009.

MAIN OUTCOME MEASURES

calculated annual age-specific and gender-specific incidence rates, demographic information and initial and final visual acuity.

RESULTS

A total of 294 diagnoses of keratitis occurred in 285 children during the 10-year period, yielding an incidence of 78.0 per 100 000 younger than 19 years (95% CI 69.0 to 87.1) or approximately 1 in 1282 children. The incidence increased throughout the 10-year study period (p<0.001). The mean age at diagnosis was 15.3 years (range, 0.2-18.9) and 172 (60.4%) were women. The observed forms included keratitis due to contact lens wear in 134 (45.6%), infectious keratitis in 72 (24.5%), keratitis not otherwise specified in 65 (22.1%) and keratitis sicca in 23 (7.8%). The visual acuity was reduced to ≤20/40 in 61 (21.4) of the 285 patients at the initial examination and in 24 (8.4%) at the final examination. Children with infectious keratitis had the poorest presenting vision and the best final vision, whereas the reverse was true for those with keratitis sicca.

CONCLUSIONS

Keratitis, regardless of aetiology, was observed in approximately 1 in 1300 children by 19 years of age in this population-based cohort. Nearly half were related to contact lens wear and a decrease in vision to ≤ 20/40 occurred in 1 in 12 patients.

Genomic and Phenotypic Trait Variation of the Opportunistic Human Pathogen Aspergillus flavus and Its Close Relatives

Fungal diseases affect millions of humans annually, yet fungal pathogens remain understudied. The mold Aspergillus flavus can cause both aspergillosis and fungal keratitis infections, but closely related species are not considered clinically relevant. To study the evolution of A. flavus pathogenicity, we examined genomic and phenotypic traits of two strains of A. flavus and three closely related species, Aspergillus arachidicola (two strains), Aspergillus parasiticus (two strains), and Aspergillus nomiae (one strain). We identified >3,000 orthologous proteins unique to A. flavus, including seven biosynthetic gene clusters present in A. flavus strains and absent in the three nonpathogens. We characterized secondary metabolite production for all seven strains under two clinically relevant conditions, temperature and salt concentration. Temperature impacted metabolite production in all species, whereas salinity did not affect production of any species. Strains of the same species produced different metabolites. Growth under stress conditions revealed additional heterogeneity within species. Using the invertebrate fungal disease model Galleria mellonella, we found virulence of strains of the same species varied widely; A. flavus strains were not more virulent than strains of the nonpathogens. In a murine model of fungal keratitis, we observed significantly lower disease severity and corneal thickness for A. arachidicola compared to other species at 48 h postinfection, but not at 72 h. Our work identifies variations in key phenotypic, chemical, and genomic attributes between A. flavus and its nonpathogenic relatives and reveals extensive strain heterogeneity in virulence that does not correspond to the currently established clinical relevance of these species. IMPORTANCE Aspergillus flavus is a filamentous fungus that causes opportunistic human infections, such as aspergillosis and fungal keratitis, but its close relatives are considered nonpathogenic. To begin understanding how this difference in pathogenicity evolved, we characterized variation in infection-relevant genomic, chemical, and phenotypic traits between strains of A. flavus and its relatives. We found extensive variation (or strain heterogeneity) within the pathogenic A. flavus as well as within its close relatives, suggesting that strain-level differences may play a major role in the ability of these fungi to cause disease. Surprisingly, we also found that the virulence of strains from species not considered to be pathogens was similar to that of A. flavus in both invertebrate and murine models of disease. These results contrast with previous studies on Aspergillus fumigatus, another major pathogen in the genus, for which significant differences in infection-relevant chemical and phenotypic traits are observed between closely related pathogenic and nonpathogenic species.

Pediatric Microbial Keratitis: Identification of Clinical Biomarkers for Prognosis and Outcome of 218 Cases From 2009 to 2019

PURPOSE

The aim of this study was to analyze the risk factors, microbiological profile, and treatment efficacy in pediatric microbial keratitis (MK) and to identify clinical biomarkers prognosticating outcome.

METHODS

A retrospective analysis was conducted from patients younger than 16 years with MK-excluding viral, marginal, or interstitial keratitis. Data pertaining to predisposing factors, symptom duration, prior treatment, ulcer characteristics, microbiological profile, time to resolution, and final outcome were recorded. Statistical analysis was performed. The mixed-effects linear regression model with random intercept was used to evaluate factors affecting time to resolution.

RESULTS

Among 218 episodes of 215 pediatric patients with MK, the geometric mean of central [median 3 mm, interquartile range (IQR) 1-4.3 mm] and peripheral ulcers (median 1 mm, IQR 1-2.5 mm) was significantly different ( P < 0.0001). Organisms identified were bacteria (56.9%), fungi (31.5%), and acanthamoebae (2.3%). Of 172 cases (78.8%), which resolved in a median resolution time of 22 days (IQR, 11-44 days), 107 (81.6%) with absent/negative microbiology healed on empirical therapy. On multivariate analysis, peripheral ulcers and geometric mean ulcer size affected time to resolution. Significantly higher percentage of eyes, which worsened or perforated, were on topical steroids compared with those which healed (31.8% vs. 9.2%, P = 0.0061).

CONCLUSIONS

Good outcome even in culture negative cases suggests empirical therapy may be instituted for nonsevere peripheral pediatric MK; however, the importance of a microbiological workup cannot be underscored enough. Ulcer location and geometric mean size of ulcer may be used as clinical prognostic markers for resolution.

Pediatric Non-Viral Microbial Keratitis: Predisposing Factors, Microbiological Profile, Treatment Modalities, and Visual Outcome

Purpose

To describe the predisposing factors, causative organisms, treatment modalities, and visual outcomes of childhood non-viral microbial keratitis in our region.

Patients and Methods

All cases with the clinical or microbiological diagnosis of non-viral microbial keratitis in patients ≤18 years presenting to Dhahran Eye specialist Hospital, a tertiary eye care hospital in Dhahran, Eastern Province, Saudi Arabia, between 2010 and 2020 were included. A retrospective chart review was conducted. Demographic data, predisposing factors, clinical characteristics, isolated microorganisms, and visual outcomes were recorded.

Results

Fifty-nine patients were included in this study, of which three cases were bilateral. The mean age was 9.3 ± 6.3 years (range: 14 days - 18 years). Predisposing factors were identified in 89.8% (n=53) of cases. Contact lens wear was the leading cause 35.6% (n=21), followed by trauma 27.1% (n=16), ocular diseases 11.9% (n=7), systemic diseases 10.2% (n=6), and ocular surgery 5.1% (n=3). Out of all cases, 66.1% (n=39) have undergone corneal scraping, out of which 43.6% (n=17) showed positive growth. Gram-negative organisms accounted for 47.1% (n=8) isolates of all culture-positive cases. Pseudomonas aeruginosa was the most common pure isolate, which accounted for 41.2% (n=7) of culture-positive cases, followed by Staphylococcus aureus 11.8% (n=2). The most common complication was corneal scar in 71.2% (n=42). Nineteen (32.2%) patients had poor outcome. Seven patients (11.9%) required further intervention, these included penetrating keratoplasty (n=1), deep lamellar keratoplasty (n=3), Photorefractive keratectomy (n=2), and Phototherapeutic keratectomy (n=1).

Conclusion

Childhood non-viral microbial keratitis is uncommon; however, it carries significant risks. Most cases were associated with preventable risk factors, with contact lens wear being the leading cause. Early detection and management are mandatory to reduce the risk of vision-threatening complications. The difficulty in assessment should not jeopardize proper evaluation and management of suspected cases.

Determination of probability of causative pathogen in infectious keratitis using deep learning algorithm of slit-lamp images

Corneal opacities are important causes of blindness, and their major etiology is infectious keratitis. Slit-lamp examinations are commonly used to determine the causative pathogen; however, their diagnostic accuracy is low even for experienced ophthalmologists. To characterize the “face” of an infected cornea, we have adapted a deep learning architecture used for facial recognition and applied it to determine a probability score for a specific pathogen causing keratitis. To record the diverse features and mitigate the uncertainty, batches of probability scores of 4 serial images taken from many angles or fluorescence staining were learned for score and decision level fusion using a gradient boosting decision tree. A total of 4306 slit-lamp images including 312 images obtained by internet publications on keratitis by bacteria, fungi, acanthamoeba, and herpes simplex virus (HSV) were studied. The created algorithm had a high overall accuracy of diagnosis, e.g., the accuracy/area under the curve for acanthamoeba was 97.9%/0.995, bacteria was 90.7%/0.963, fungi was 95.0%/0.975, and HSV was 92.3%/0.946, by group K-fold validation, and it was robust to even the low resolution web images. We suggest that our hybrid deep learning-based algorithm be used as a simple and accurate method for computer-assisted diagnosis of infectious keratitis.

Microbial Warfare on Three Fronts: Mixed Biofilm of Aspergillus fumigatus and Staphylococcus aureus on Primary Cultures of Human Limbo-Corneal Fibroblasts

Background

Coinfections with fungi and bacteria in ocular pathologies are increasing at an alarming rate. Two of the main etiologic agents of infections on the corneal surface, such as Aspergillus fumigatus and Staphylococcus aureus, can form a biofilm. However, mixed fungal–bacterial biofilms are rarely reported in ocular infections. The implementation of cell cultures as a study model related to biofilm microbial keratitis will allow understanding the pathogenesis in the cornea. The cornea maintains a pathogen-free ocular surface in which human limbo-corneal fibroblast cells are part of its cell regeneration process. There are no reports of biofilm formation assays on limbo-corneal fibroblasts, as well as their behavior with a polymicrobial infection.

Objective

To determine the capacity of biofilm formation during this fungal–bacterial interaction on primary limbo-corneal fibroblast monolayers.

Results

The biofilm on the limbo-corneal fibroblast culture was analyzed by assessing biomass production and determining metabolic activity. Furthermore, the mixed biofilm effect on this cell culture was observed with several microscopy techniques. The single and mixed biofilm was higher on the limbo-corneal fibroblast monolayer than on abiotic surfaces. The A. fumigatus biofilm on the human limbo-corneal fibroblast culture showed a considerable decrease compared to the S. aureus biofilm on the limbo-corneal fibroblast monolayer. Moreover, the mixed biofilm had a lower density than that of the single biofilm. Antibiosis between A. fumigatus and S. aureus persisted during the challenge to limbo-corneal fibroblasts, but it seems that the fungus was more effectively inhibited.

Conclusion

This is the first report of mixed fungal–bacterial biofilm production and morphological characterization on the limbo-corneal fibroblast monolayer. Three antibiosis behaviors were observed between fungi, bacteria, and limbo-corneal fibroblasts. The mycophagy effect over A. fumigatus by S. aureus was exacerbated on the limbo-corneal fibroblast monolayer. During fungal–bacterial interactions, it appears that limbo-corneal fibroblasts showed some phagocytic activity, demonstrating tripartite relationships during coinfection.