[Bacterial keratitis. Current diagnostic aspects]

Deep learning-based classification system of bacterial keratitis and fungal keratitis using anterior segment images

Introduction

Infectious keratitis is a vision threatening disease. Bacterial and fungal keratitis are often confused in the early stages, so right diagnosis and optimized treatment for causative organisms is crucial. Antibacterial and antifungal medications are completely different, and the prognosis for fungal keratitis is even much worse. Since the identification of microorganisms takes a long time, empirical treatment must be started according to the appearance of the lesion before an accurate diagnosis. Thus, we developed an automated deep learning (DL) based diagnostic system of bacterial and fungal keratitis based on the anterior segment photographs using two proposed modules, Lesion Guiding Module (LGM) and Mask Adjusting Module (MAM).

Methods

We used 684 anterior segment photographs from 107 patients confirmed as bacterial or fungal keratitis by corneal scraping culture. Both broad- and slit-beam images were included in the analysis. We set baseline classifier as ResNet-50. The LGM was designed to learn the location information of lesions annotated by ophthalmologists and the slit-beam MAM was applied to extract the correct feature points from two different images (broad- and slit-beam) during the training phase. Our algorithm was then externally validated using 98 images from Google image search and ophthalmology textbooks.

Results

A total of 594 images from 88 patients were used for training, and 90 images from 19 patients were used for test. Compared to the diagnostic accuracy of baseline network ResNet-50, the proposed method with LGM and MAM showed significantly higher accuracy (81.1 vs. 87.8%). We further observed that the model achieved significant improvement on diagnostic performance using open-source dataset (64.2 vs. 71.4%). LGM and MAM module showed positive effect on an ablation study.

Discussion

This study demonstrated that the potential of a novel DL based diagnostic algorithm for bacterial and fungal keratitis using two types of anterior segment photographs. The proposed network containing LGM and slit-beam MAM is robust in improving the diagnostic accuracy and overcoming the limitations of small training data and multi type of images.

[Effectiveness of initial antibiotic therapy for treatment of contact lens-related bacterial keratitis]

BACKGROUND

Contact lens-related microbial keratitis is a cause of potentially sight-threatening corneal opacification. Effective initial antimicrobial therapy is crucial to prevent long-term complications. This investigation was undertaken to test the effectiveness of current routine empirical antibiotic treatment regimens.

METHODS/PATIENTS

All consecutive cases of contact lens-related keratitis presenting in the outpatient clinic of the Department of Ophthalmology at the Medical University of Innsbruck between January 2010 and April 2012 were retrospectively analyzed.

RESULTS

Cultures were positive in 69 out of the 123 cases included in the study. Culture results identified 59.4 % Gram positive strains, 50.7 % Gram negative strains and 7.2 % fungal strains. Mixed infections accounted for 29 % of cases. The combination of an aminoglycoside and a second generation quinolone antibiotic was the most common initial treatment regimen (87.8 %). In vitro this regimen was less effective compared to combinations of moxifloxacin and ciprofloxacin or moxifloxacin and gentamicin.

CONCLUSION

Empirical combined regimens remain an effective treatment of contact lens-related keratitis. Fluoroquinolones proved to be inadequate for monotherapy.

In-vitro analysis of the microbicidal activity of 6 contact lens care solutions

BACKGROUND

Contact lens-related infections are often associated with inadequate contact lens hygiene, and therefore, contact lens care products should be able to sufficiently minimise the amount of pathogens that are responsible for these infections. In 2001, the EN ISO 14729 was introduced to ensure adequate disinfection efficacy of contact lens care solutions, but this norm has recently been criticised.

METHODS

In this study, six frequently used contact lens care solutions were retested according to the Stand Alone Test of the EN ISO 14729 (2001). The Stand Alone Test is a quantitative suspension test. In addition, the products were tested in a modified setting adding an organic load. The load was a mixture of human blood serum, lysozyme, and mucine, which resembles tear fluid.

RESULTS

The criteria of the Stand Alone Test recommended in EN ISO 14729 were only met by Aosept Plus. This 3% hydrogen-peroxide-based contact lens care solution attained a reduction factor of > 5 log units for bacteria and > 4 for fungi in all cases. Two further contact lens care solutions, Blue Vision and Optifree Replenish, met the criteria of a reduction factor of > 3 log units for bacteria and > 1 log unit for fungi, but only in the presence of artificial tear fluid. The three remaining products did not exhibit adequate disinfecting efficacy, at least against one of the tested microorganisms.

CONCLUSIONS

Through the observation that the artificial tear fluid used in this study influences the disinfecting efficacy of contact lens care solutions, especially that of multi-purpose solutions, in a different way than does albumin, mucine, or even the organic load suggested in EN ISO 14729, it becomes obvious that the test conditions in the EN ISO 14729 should be revised in order to create more realistic conditions, e.g., by using a more realistic artificial tear fluid. Furthermore, we suggest adapting the EN ISO 14729 to the European test hierarchy for chemical disinfectants and antiseptics, which consists of three test phases and also requests meeting stricter criteria in order to pass the test. Unless the test conditions guarantee a sufficient reduction of potential pathogens, the risk of contact lens-related microbial keratitis and other infections will remain for the users.

[Persistent contact lens-associated keratitis]

A 43-year-old woman presented with a 2-month history of contact lens-associated keratitis and corneal ulceration on the left eye. Without having isolated the pathogenic agent, a calculated anti-infective and anti-inflammatory therapy had already been administered with no signs of improvement. Filamentous fungi (Acremonium sp) were found in the scraping material and the contact lens solution. A keratoplasty à chaud was necessary despite immediate initiation of pathogen-specific therapy. Early identification of the pathogen via corneal debridement and culturing are standard procedures in order to initiate a specific therapy and prevent ulceration which can threaten the visual outcome.

A three-phase in-vitro system for studying Pseudomonas aeruginosa adhesion and biofilm formation upon hydrogel contact lenses

Background

Pseudomonas aeruginosa is commonly associated with contact lens (CL) -related eye infections, for which bacterial adhesion and biofilm formation upon hydrogel CLs is a specific risk factor. Whilst P. aeruginosa has been widely used as a model organism for initial biofilm formation on CLs, in-vitro models that closely reproduce in-vivo conditions have rarely been presented.

Results

In the current investigation, a novel in-vitro biofilm model for studying the adherence of P. aeruginosa to hydrogel CLs was established. Nutritional and interfacial conditions similar to those in the eye of a CL wearer were created through the involvement of a solid:liquid and a solid:air interface, shear forces and a complex artificial tear fluid. Bioburdens varied depending on the CL material and biofilm maturation occurred after 72 h incubation. Whilst a range of biofilm morphologies were visualised including dispersed and adherent bacterial cells, aggregates and colonies embedded in extracellular polymer substances (EPS), EPS fibres, mushroom-like formations, and crystalline structures, a compact and heterogeneous biofilm morphology predominated on all CL materials.

Conclusions

In order to better understand the process of biofilm formation on CLs and to test the efficacy of CL care solutions, representative in-vitro biofilm models are required. Here, we present a three-phase biofilm model that simulates the environment in the eye of a CL wearer and thus generates biofilms which resemble those commonly observed in-situ.

Two cases of keratitis and corneal ulcers caused by Burkholderia gladioli

We report two cases of protracted keratitis complicated by corneal ulcer. Burkholderia gladioli, primarily known as a plant pathogen, was cultured from corneal swabs. The epithelial defects healed after extended adequate antibiotic therapy. Despite identical patterns of both strains in restriction fragment length polymorphism analysis, a common infection source was not detectable.

Therapie und Prognose der bakteriellen Keratitis

This review discusses the current therapy for bacterial keratitis, especially that using quinolones of the fourth generation. In addition, the importance of additive steroids is discussed. Steroids should be administered at a low dose and tapered off slowly, whereas antibiotics should be applied at a high dosage and discontinued abruptly.