Dataset Information

Accuracy of a commercial multiplex PCR for the diagnosis of bacterial vaginosis.

ABSTRACT:

Purpose

Bacterial vaginosis (BV) is a common clinical condition characterized by odorous vaginal discharge, vaginal itching and/or burning. BV can occur when vaginal lactobacilli are depleted and replaced by diverse anaerobic bacteria. We evaluated a commercial multiplex PCR (ATRiDA) for the diagnosis of BV.

Methods

Cervicovaginal samples were included from women reporting urogenital symptoms and from women notified for sexually transmitted infections (STI) - who were not (necessarily) symptomatic. Clinical BV diagnoses were obtained from electronic patient files. The ATRiDA test measures the loads of Gardnerella vaginalis, Atopobium vaginae and Lactobacillus species in relation to overall bacterial load. The ATRiDA test outcome was compared to the clinical BV diagnosis and to vaginal microbiota composition, determined by 16SrRNA gene sequencing.

Results

We included samples from 185 women reporting urogenital symptoms, of whom 81 had BV and 93 women who were notified for an STI, of whom 16 had BV. Overall, compared to the clinical BV diagnosis, the ATRiDA test demonstrated high sensitivity (96.9?%) and moderate specificity (70.2?%). The negative predictive value was high (>97.3). The positive predictive value differed by study group and was highest in women reporting urogenital symptoms (78.2?%). Sequencing showed that 54?% of women who had an ATRiDA BV-positive test outcome, but who were not clinically diagnosed with BV, had diverse anaerobic vaginal microbiota (asymptomatic vaginal dysbiosis).

Conclusion

The ATRiDA test is a sensitive method for the detection of BV but, given the high occurrence of asymptomatic vaginal dysbiosis, a positive test outcome should be interpreted together with clinical symptoms.

SUBMITTER: van der Veer C

PROVIDER: S-EPMC6230723 | biostudies-literature | 2018 Sep

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Accuracy of a commercial multiplex PCR for the diagnosis of bacterial vaginosis.

van der Veer Charlotte C van Houdt Robin R van Dam Alje A de Vries Henry H Bruisten Sylvia S

Journal of medical microbiology 20180709 9

<h4>Purpose</h4>Bacterial vaginosis (BV) is a common clinical condition characterized by odorous vaginal discharge, vaginal itching and/or burning. BV can occur when vaginal lactobacilli are depleted and replaced by diverse anaerobic bacteria. We evaluated a commercial multiplex PCR (ATRiDA) for the diagnosis of BV.<h4>Methods</h4>Cervicovaginal samples were included from women reporting urogenital symptoms and from women notified for sexually transmitted infections (STI) - who were not (necessa ...[more]

PMID: 29985123

Similar Datasets

Project description:Bacterial vaginosis (BV) is caused by the excessive and imbalanced growth of bacteria in vagina, affecting 30 to 50% of women. Gram staining followed by Nugent scoring based on bacterial morphotypes under the microscope is considered the gold standard for BV diagnosis; this method is often labor-intensive and time-consuming, and results vary from person to person. We developed and optimized a convolutional neural network (CNN) model and evaluated its ability to automatically identify and classify three categories of Nugent scores from microscope images. The CNN model was first established with a panel of microscopic images with Nugent scores determined by experts. The model was trained by minimizing the cross-entropy loss function and optimized by using a momentum optimizer. The separate test sets of images collected from three hospitals were evaluated by the CNN model. The CNN model consisted of 25 convolutional layers, 2 pooling layers, and a fully connected layer. The model obtained 82.4% sensitivity and 96.6% specificity with the 5,815 validation images when altered vaginal flora and BV were considered the positive samples, which was better than the rates achieved by top-level technologists and obstetricians in China. The capability of our model for generalization was so strong that it exhibited 75.1% accuracy in three categories of Nugent scores on the independent test set of 1,082 images, which was 6.6% higher than the average of three technologists, who are hold bachelor's degrees in medicine and are qualified to make diagnostic decisions. When three technologists ran one specimen in triplicate, the precision of three categories of Nugent scores was 54.0%. One hundred three samples diagnosed by two technologists on different days showed a repeatability of 90.3%. The CNN model outperformed human health care practitioners in terms of accuracy and stability for three categories of Nugent score diagnosis. The deep learning model may offer translational applications in automating diagnosis of bacterial vaginosis with proper supporting hardware.

Dataset Information

Accuracy of a commercial multiplex PCR for the diagnosis of bacterial vaginosis.

Purpose

Methods

Results

Conclusion

Publications

Accuracy of a commercial multiplex PCR for the diagnosis of bacterial vaginosis.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets