Abstract
# Purpose #
To explore the use of optical SmartProbes for the rapid evaluation of corneal scrapes from patients with suspected microbial keratitis, as a clinical alternative to Gram stain.
# Design #
Experimental study with evaluation of a diagnostic technology.
# Subjects #
Corneal scrapes from 267 patients presenting with suspected microbial keratitis at a referral cornea clinic within a specialist tertiary care hospital in South India were evaluated.
# Methods #
Corneal scrapes were flooded with SmartProbe (BAC One (all microbe, n = 107) or BAC Two (specific to gram-negative bacteria, n = 160)) and evaluated by fluorescence microscopy (without the need for sample washing or further processing). The microscopist was masked to the clinical result when scoring the SmartProbe sample.
# Main Outcome Measures #
Score the SmartProbe labelled sample as bacteria/fungi/none (BAC One) or gram-negative bacteria/none (BAC Two) and compare to Gram stain and culture reports, determining the sensitivity, specificity, positive and negative predictive values and accuracy.
# Results #
Compared to Gram stain, BAC One demonstrated sensitivity and specificity of 80.0 and 87.5 percent respectively, positive and negative predictive values of 93.8 and 65.1 percent and an accuracy score of 82.2.
BAC Two demonstrated sensitivity and specificity of 93.3 and 84.8 percent respectively, a negative predictive value of 99.2 percent and an accuracy of 85.6 percent. Twenty percent of fungal samples were recorded as positive for BAC Two, leading to a positive predictive value of 38.9, however, when morphology and positive fluorescent signal were considered together, the positive predictive value increased to 88.8.
When the corresponding culture results were compared to the direct microscopy (Gram stain) result, the sensitivity and specificity were 73.4 and 70.7 percent respectively, the positive and negative predictive values were 86.5 and 51.0 percent, and there was an overall accuracy of 72.6.
# Conclusions #
Fluorescent SmartProbes offer a comparative method to direct microscopy (Gram stain) for delineating gram-positive or gram-negative bacteria or fungi within corneal scrapes. We demonstrate equivalent or higher scores for sensitivity, specificity, positive and negative predictive values and accuracy than culture to Gram stain. Our approach has scope for point-of-care clinical application to aid in the diagnosis of microbial keratitis.
To explore the use of optical SmartProbes for the rapid evaluation of corneal scrapes from patients with suspected microbial keratitis, as a clinical alternative to Gram stain.
# Design #
Experimental study with evaluation of a diagnostic technology.
# Subjects #
Corneal scrapes from 267 patients presenting with suspected microbial keratitis at a referral cornea clinic within a specialist tertiary care hospital in South India were evaluated.
# Methods #
Corneal scrapes were flooded with SmartProbe (BAC One (all microbe, n = 107) or BAC Two (specific to gram-negative bacteria, n = 160)) and evaluated by fluorescence microscopy (without the need for sample washing or further processing). The microscopist was masked to the clinical result when scoring the SmartProbe sample.
# Main Outcome Measures #
Score the SmartProbe labelled sample as bacteria/fungi/none (BAC One) or gram-negative bacteria/none (BAC Two) and compare to Gram stain and culture reports, determining the sensitivity, specificity, positive and negative predictive values and accuracy.
# Results #
Compared to Gram stain, BAC One demonstrated sensitivity and specificity of 80.0 and 87.5 percent respectively, positive and negative predictive values of 93.8 and 65.1 percent and an accuracy score of 82.2.
BAC Two demonstrated sensitivity and specificity of 93.3 and 84.8 percent respectively, a negative predictive value of 99.2 percent and an accuracy of 85.6 percent. Twenty percent of fungal samples were recorded as positive for BAC Two, leading to a positive predictive value of 38.9, however, when morphology and positive fluorescent signal were considered together, the positive predictive value increased to 88.8.
When the corresponding culture results were compared to the direct microscopy (Gram stain) result, the sensitivity and specificity were 73.4 and 70.7 percent respectively, the positive and negative predictive values were 86.5 and 51.0 percent, and there was an overall accuracy of 72.6.
# Conclusions #
Fluorescent SmartProbes offer a comparative method to direct microscopy (Gram stain) for delineating gram-positive or gram-negative bacteria or fungi within corneal scrapes. We demonstrate equivalent or higher scores for sensitivity, specificity, positive and negative predictive values and accuracy than culture to Gram stain. Our approach has scope for point-of-care clinical application to aid in the diagnosis of microbial keratitis.
Data Citation
Mills, B. (2020). Exploratory use of fluorescent SmartProbes for the rapid detection of microbial isolates causing corneal ulcer, [dataset]. University of Edinburgh. College of Medicine & Veterinary Medicine. https://doi.org/10.7488/ds/2812.
| Date made available | 6 May 2020 |
|---|---|
| Publisher | Edinburgh DataShare |