Abstract:[Background] The traditional method of bacterial culture is the gold standard for the diagnosis of lower respiratory tract infection. However, bacterial culture needs a long cycle and has low sensitivity. Loop-mediated isothermal amplification (LAMP) as a simple and rapid method can be used for rapid detection of common bacteria in clinical respiratory tract infection. [Objective] To evaluate the detection capability of LAMP for seven common bacterial species of respiratory tract infection. [Methods] Specific primers were designed for the LAMP of seven common pathogens of respiratory tract infections, including Klebsiella pneumoniae (KP), Acinetobacter baumannii (AB), Staphylococcus aureus (SA), Pseudomonas aeruginosa (PA), Streptococcus pneumoniae (SP), Moraxella catarrhalis (MC), and Haemophilus influenzae (HI). The sensitivity and specificity of this LAMP method were evaluated by gradient dilution method and cross-reaction experiment, respectively. We then retrospectively analyzed the residual tract specimens (sputum and bronchoalveolar lavage fluid) from 240 patients with suspected lower respiratory tract infection in Peking University Peoples Hospital from November 2019 to March 2021. After DNA extraction by the automatic nucleic acid extraction system, we employed LAMP to detect the seven common pathogens. Further, we compared the results obtained from LAMP and bacterial culture to evaluate the sensitivity and specificity of LAMP. The cut-off value (Ct value) of different species were optimized, and the application value of LAMP in clinical practice were discussed. [Results] The LAMP method had a good linear correlation with the template concentration. The LAMP system had good specificity as there was no cross reaction with other strains. A total of 218 sputum specimens and 22 bronchoalveolar lavage fluid specimens were detected in this study. The seven pathogens were detected in 178 specimens based on the bacterial culture method and in 176 specimens based on the LAMP method. The LAMP method generally consumed 2–3 h for detection and could simultaneously detect multiple specimens. The optimal Ct values of HI, KP, AB, MC, SA, PA, and SP were 18.5, 20, 20, 15, 25, 19, and 18, respectively. The sensitivity and specificity were as follows: KP (90.7%, 94.1%), AB (84.0%, 94.2%), PA (90.8%, 89.1%), MC (75.0%, 99.2%), SA (81.8%, 97.4%), HI (75.0%, 90.3%), and SP (33.3%, 95.4%). The LAMP results for KP, AB, and PA showed good coincidence with the semi-quantitative results of sputum culture, and the R2 values were 0.855 7, 0.804 4, and 0.924 3, respectively. [Conclusion] LAMP had low requirements for operating personnel, short cycle, and high specificity (89.1%–99.2%) compared with the culture method. It has high sensitivity (81.8%–90.8%) for the detection of K. pneumoniae, A. baumannii, S. aureus, and P. aeruginosa, and thus can be used for the rapid detection of these bacteria in clinical respiratory tract infection. The results of LAMP were in good agreement with the semi-quantitative results of sputum culture. The detection of H. influenzae and S. pneumoniae needs to be further evaluated due to the lack of positive culture specimens.