Diagnostic Value of Metagenomic Next-Generation Sequencing for the Detection of Pathogens in Bronchoalveolar Lavage Fluid in Ventilator-Associated Pneumonia Patients.

Xiaowei Fang,Tianjun Yang,Aijun Pan,Lei Zhang,Chunyan Zhu,Qing Mei,Shike Geng,Xiaoqin Fan

doi:10.3389/fmicb.2020.599756

Xiaowei Fang, Tianjun Yang + Show 6 more

Open Access

https://doi.org/10.3389/fmicb.2020.599756

Copy DOI

Abstract

Objective: To evaluate the diagnostic performance of metagenomic next-generation sequencing (mNGS) using bronchoalveolar lavage fluid (BALF) in patients with ventilator-associated pneumonia (VAP). Methods: BALF samples of 72 patients with VAP were collected from August 2018 to May 2020. The diagnostic performance of conventional testing (CT) and mNGS methods were compared based on bacterial and fungal examinations. The diagnostic value of mNGS for viral and mixed infections was also analyzed. Results: The percentage of mNGS positive samples was significantly higher than that estimated by the CT method [odds ratio (OR), 4.33; 95% confidence interval (CI), 1.78–10.53; p < 0.001]. The sensitivity and specificity of mNGS for bacterial detection were 97.1% (95% CI, 93.2–101.0%) and 42.1% (95 CI, 30.7–53.5%), respectively, whereas the positive predictive value (PPV) and the negative predictive value (NPV) were 60.0% (95% CI, 48.7–71.3%) and 94.1% (95% CI, 88.7–99.6%), respectively. A total of 38 samples were negative for bacterial detection as determined by the CT method, while 22 samples were positive as shown by the mNGS method. Conflicting results were obtained for three samples between the two methods of bacterial detection. However, no significant differences were noted between the mNGS and CT methods (OR, 1.42; 95% CI, 0.68–2.97; p = 0.46) with regard to fungal infections. The sensitivity and specificity of mNGS were 71.9% (95% CI, 61.5–82.3%) and 77.5% (95% CI, 67.9–87.1%), respectively. mNGS exhibited a PPV of 71.9% (95% CI, 61.5–82.3%) and an NPV of 77.5% (95% CI, 67.9–87.1%). A total of 9 out of 40 samples were found positive for fungi according to mNGS, whereas the CT method failed to present positive results in these samples. The mNGS and CT methods produced conflicting results with regard to fungal detection of the two samples. A total of 30 patients were virus-positive using mNGS. Furthermore, 42 patients (58.3%) were identified as pulmonary mixed infection cases. Conclusions: mNGS detection using BALF improved the sensitivity and specificity of bacterial identification in patients who developed VAP. In addition, mNGS exhibited apparent advantages in detecting viruses and identifying mixed infections.

Highlights

Ventilator-associated pneumonia (VAP) induced following mechanical ventilation remains a major health condition, which has to be treated by intensive care unit (ICU) physicians
The largest study is that conducted by Qing et al (Miao et al, 2018), which indicated that the Metagenomic next-generation sequencing (mNGS) was not superior to the Conventional testing (CT) methods for detecting common bacteria, whereas it exhibited a better performance than the CT methods with regard to fungal detection
All the patients enrolled in the present study developed VAP in the ICU and all samples were derived from bronchoalveolar lavage fluid (BALF)

Summary

Introduction

Ventilator-associated pneumonia (VAP) induced following mechanical ventilation remains a major health condition, which has to be treated by intensive care unit (ICU) physicians. It has been reported that the incidence of VAP in ICU is 9–27%, whereas the mortality rate ranges between 30 and 70% (Chastre and Fagon, 2002; Valles et al, 2007; Rosenthal et al, 2020). In the majority of the cases, VAP patients undergo endotracheal intubation and bronchoalveolar lavage fluid (BALF) becomes a more accessible clinical specimen with a higher diagnostic value than other specimens. Conventional testing (CT) methods of BALF, including bacterial and fungal smear and culture have been routinely performed to assess the microbiological information. These methods of assessment are time-consuming and exhibit a low positive rate, which cannot meet the diagnostic needs of a critically ill patient. Due to the limited detection range of the PCR kits, the combination of several diagnostic methods that can complement the viral detection is essential

Methods

Results

Conclusion