Abstract
Identifying pathogens in patients with lower respiratory tract infections (LRTIs) has always been a major challenge. Metagenomic next-generation sequencing (mNGS) technology is a new diagnostic tool that can assist clinicians in the etiological diagnosis of LRTIs. This study focuses on the clinical value of mNGS in the diagnosis of suspected LRTIs. A total of 400 patients with suspected LRTIs admitted at the First Affiliated Hospital of Nanchang University from July 2020 to February 2023 were enrolled in this retrospective study. Bronchoalveolar lavage fluid (BALF) samples were analyzed using both mNGS and culture methods. The diagnostic accuracy of two approaches was systematically compared against the final clinical diagnosis, which served as the gold-standard reference. Of the 400 enrolled cases, 82.3% (329/400) were diagnosed with LRTIs. From these cases, mNGS identified 76.8% (307/400) truepositive cases, 8.0% (32/400) falsepositive cases, 9.8% (39/400) truenegative cases, and 5.5% (22/400) falsenegative cases. mNGS demonstrated significantly higher sensitivity [93.3% (307/329) vs. 55.6(183/329)%] alongside greater negative predictive values [63.9% (39/61) vs.25.9%(51/197)], whereas culture offered higher specificity [54.9%(39/71) vs. 71.8%(51/71)]. The area under the receiver-operating curve (AUC) of mNGS[0.744(95%CI: 0.67-0.82)]was significantly higher than that of cultures[0.636(95%CI: 0.57-0.71)]. Specifically, mNGS detected more Streptococcus pneumoniae (7.0% vs. 0%), Haemophilus influenzae (6.7% vs. 0%), Aspergillus (9.4% vs. 3.5%), Pneumocystis jirovecii (11.9% vs. 0%) and other intracellular pathogens. Of the 329 patients with LRTIs, antibiotic treatment was modified based on the mNGS results in more than half of the patients(50.5%,166/329), including 20 cases of adjusted antimicrobial regimens, 70 cases de-escalated the empirical antibiotic treatment, and 76 patients escalated the treatment by increasing dosage or medication. 60.8%(101/166) of patients responded to modified antibiotic treatment. Significant benefits of mNGS have been shown in pathogen identification and antimicrobial treatment stewardship in patients with LRTIs. For those with suboptimal therapeutic responses, physicians should be alert to some emerging intracellular pathogens, including Chlamydia psittaci, Mycobacterium tuberculosis, and Pneumocystis jirovecii.