Microbial Keratitis in Nepal: Predicting the Microbial Aetiology from Clinical Features

Abstract

Fungal corneal infection (keratitis) is a common clinical problem in South Asia. However, it is often challenging to distinguish this from other aetiologies, such as bacteria or acanthamoeba. In this prospective study, we investigated clinical and epidemiological features that can predict the microbial aetiology of microbial keratitis in Nepal. We recruited patients presenting with keratitis to a tertiary eye hospital in lowland eastern Nepal between June 2019 and November 2020. A structured assessment, including demographics, history, and clinical signs, was carried out. The aetiology was investigated with in vivo confocal microscopy and corneal scrape for microscopy and culture. A predictor score was developed using odds ratios calculated to predict aetiology from features.A fungal cause was identified in 482/642 (75.1%) of cases, which increased to 532/642 (82.9%) when including mixed infections. Unusually, dematiaceous fungi accounted for half of the culture-positive cases (50.6%). Serrated infiltrate margins, patent nasolacrimal duct, raised corneal slough, and organic trauma were independently associated with fungal keratitis (p < 0.01). These four features were combined in a predictor score. The probability of fungal keratitis was 30.1% if one feature was present, increasing to 96.3% if all four were present. Whilst microbiological diagnosis is the “gold standard” to determine the aetiology of an infection, certain clinical signs can help direct the clinician to find a presumptive infectious cause, allowing appropriate treatment to be started without delay. Additionally, this study identified dematiaceous fungi, specifically Curvularia spp., as the main causative agent for fungal keratitis in this region. This novel finding warrants further research to understand potential implications and any trends over time.

Diagnosis of Fungal Keratitis in Low-Income Countries: Evaluation of Smear Microscopy, Culture, and In Vivo Confocal Microscopy in Nepal

“…The full protocol for this study and results have already been published [20,21]. We have previously described the methodology relating to clinical findings, microbiological diagnosis, and in vivo confocal microscopy in our earlier work [22]; we therefore describe these briefly here.…”

“…We collected data on demographic details, ophthalmic clinical history and clinical examination findings using a structured case record form, as previously described. [22]. Clinical examination included the best spectacle-corrected visual acuity (BSCVA) in Log-MAR and slit-lamp examination, which followed a structured approach: eyelid assessment, corneal ulcer features, anterior chamber characteristics (flare, cells, hypopyon shape, and size), and perforation status.…”

“…Laboratory diagnosis was determined using smear microscopy and culture, as previously described [22]. Following the application of preservative-free topical anaesthesia (proxymetacaine), corneal scrape specimens were collected from the base and edge of the ulcer using a slit lamp and 21 G needles.…”

Application Progress of High-Throughput Sequencing in Ocular Diseases

“…The common solid medium includes various types of agar medium, such as brain-heart infusion and soybean-casein digestive agar medium, while the liquid medium includes cooked meat medium [ 17 ]. Depending on the localization of the eye disease, the samples used for pathogen culture can be acquired from the conjunctiva, cornea, aqueous humor, or vitreous humor [ 18 , 19 , 20 , 21 ]. Gram-positive bacteria account for a large proportion of culture-positive extraocular and intraocular pathogens [ 20 , 21 ].…”

“…Depending on the localization of the eye disease, the samples used for pathogen culture can be acquired from the conjunctiva, cornea, aqueous humor, or vitreous humor [ 18 , 19 , 20 , 21 ]. Gram-positive bacteria account for a large proportion of culture-positive extraocular and intraocular pathogens [ 20 , 21 ]. However, the microbial culture method is limited by the low positive rate of pathogen identification [ 17 ].…”