To investigate the significance of genotype identification and drug resistance analysis of children’s respiratory drug-resistant Haemophilus influenzae (HI) for antibiotic treatment.

Total of 92 children with respiratory HI infection admitted to the Fifth People’s Hospital of Hainan Province from March 2018 to March 2022 were selected. Serotype, biotype, genotyping (TEM-1 and ROB-1), antibiotic sensitivity and β-lactamase expression were collected. The influence factors of positive HI β-lactamase in children’s respiratory tract were analyzed by Logistic regression analysis; A line map model of children’s respiratory tract with HI β-lactamase positive was constructed. The model was evaluated by receiver operating characteristic (ROC) curve, calibration curve and clinical decision curve, respectively.

The 92 strains of HI were classified into 8 biotypes, which were Ⅰ, Ⅱ, Ⅲ, Ⅳ, Ⅴ, Ⅵ, Ⅶ and Ⅷ. Among these, types Ⅱ [35.87% (33/92)] and Ⅲ [28.26% (26/92)] were the most prevalent. The serological typing revealed that HIa accounted for the highest proportion [28.26% (26/92)]. The detection rates of TEM-1 and ROB-1 genes were 57.61% (53/92) and 21.74% (20/92), respectively. Sensitivity testing on the 8 biotypes of HI showed that types Ⅴ, Ⅵ, Ⅶ and Ⅷ exhibited no antibiotic resistance. Type Ⅳ HI demonstrated a relatively high resistance rate to ampicillin [57.14% (4/7)], while type Ⅱ HI exhibited a higher resistance rate to cefuroxime [39.39% (13/33)]. Among the 92 strains, 37 (40.22%) were β-lactamase positive and 55 (59.78%) were negative. β-lactamase positive strains showed significantly higher resistance rates to cefpodoxime and ampicillin [33 (89.19%) and 37 (100.00%), respectively] compared to β-lactamase negative strains [4 (6.35%) and 7 (11.11%), respectively], with statistically significant differences (χ² = 68.628, 74.747, both P < 0.001). Multifactor Logistics regression analysis revealed that using antibiotics continuously for ≥ 5 days (OR = 163.464, 95%CI: 8.420-3 173.439, P < 0.001), having ≥ 2 courses of continuous medication (OR = 19.890, 95%CI: 2.300-171.977, P = 0.007), combining ≥ 2 types of drugs (OR = 32.468, 95%CI: 2.792-377.616, P = 0.005) and frequently changing medication ≥ 2 times (OR = 30.769, 95%CI: 3.358-281.921, P = 0.002) were independent risk factors for β-lactamase-positive HI in children’s respiratory tracts. Constructing a column chart model based on these risk factors exhibited high discrimination, accuracy and effectiveness in predicting β-lactamase-positive HI in children’s respiratory tracts without altering the structure.

Clinically, the separation and identification of respiratory HI and the monitoring of drug resistance should be strengthened; the resistance mechanism of HI should be analyzed correctly, the antimicrobial drugs should be rationally used.