Objective To understand the characteristics of drug susceptibility of Streptococcus pneumoniae induced lower respiratory tract infection, identify the factors influencing the prognosis of the infection in preschool children in Nanyang First People's Hospital, Henan province, construct a prognosis prediction model and provide evidence for the prevention and control of S. pneumoniae infection.

Methods A total of 1425 preschool children with S. pneumoniae induced lower respiratory tract infection in Nanyang First People's Hospital were enrolled as study participants between January 2019 and October 2024, and drug susceptibility test was performed with disk diffusion method. According to the prognosis of lower respiratory tract infection, the children were divided into poor prognosis group (n=70) and good prognosis group (n=1355). The factors influencing the prognosis were identified by multivariate Logistic regression analysis. The prediction model was constructed, and the effect of the model was evaluated.

Results In 1425 strains of S. pneumoniae, the sensitivity rates to linezolid, ertapenem and vancomycin were all 100.00%, and the resistance rates to tetracycline, erythromycin and clindamycin were high (90.04%, 91.09%, 83.79%). There were significant differences in the incidence of premature birth, asthma history, roommate smoking, family history of respiratory system diseases, the prevalence of malnutrition and daily outdoor activity time between the two groups (P<0.05). Logistic regression analysis showed that premature birth odds ratio (OR)=3.955, 95% confidence interval (CI): 1.838−8.511, asthma history (OR=2.614, 95%CI: 1.531−4.464), roommate smoking (OR=2.234, 95%CI: 1.407−3.549), malnutrition (OR=2.006, 95%CI: 1.342−2.998), daily outdoor activity time <1 h (OR=1.921, 95%CI: 1.256−2.940) were the factors influencing the prognosis of S. pneumoniae induced lower respiratory tract infection in preschool children. The Hosmer-Lemeshow test indicated that χ² was 4.572, P was 0.802, and the fitting of the regression equation was good. The receiver operating characteristic (ROC) curve analysis showed that the area under the curve (AUC) of the prognosis prediction model was 0.941 (95%CI: 0.927−0.952) (P<0.05), showing good prediction effect.

Conclusion S. pneumoniae was sensitive to linezolid, ertapenem and vancomycin, but highly resistant to tetracycline, erythromycin and clindamycin. There were many factors associated with the prognosis of S. pneumoniae induced lower respiratory tract infection in preschool children. It is suggested to use highly sensitive drugs (such as linezolid) as the first choice for empirical treatment in clinical practice, and strengthen the monitoring of children with high risks (premature birth, asthma history). The application of the prognosis prediction model constructed in this study in the early risk stratification in primary medical settings can improve the targeted treatment strategies and reduce the incidence of poor prognosis.