(285) Risk Factors of Antimicrobial Resistance in Patients with Pneumococcal Infection in Hong Kong – a Population-based Unmatched Case-control Study

Background: Antimicrobial resistance (AMR) presents significant challenges for the effective treatment of pneumococcal diseases (PD) caused by Streptococcus pneumoniae (S. pneumoniae). Little has known about the clinical risk factors associated with AMR PD infection.

Objectives: Among the PD episodes identified from 2012-2021, we aimed to explore the risk factors associated with AMR among PD-related hospitalisation compared to those with isolate susceptible to antibiotics.

Methods: We retrieved data from the Hong Kong Clinical Data Analysis and Reporting System. PD-related hospitalisations were identified by diagnostic codes in International Classification of Diseases, Ninth Revision, Clinical Modification. Antimicrobial susceptibility tests (AST) were retrieved to define the AMR status. PD episodes are defined as cases with any isolate having intermediate susceptibility or resistance to any antibiotics in AST, and controls are defined as those whose isolates were susceptible to all antibiotics. We identified the risk factors for PD with AMR S. pneumoniae by comparing the characteristics of cases with controls using mixed multivariable logistic regression model. Such regression model also cconsidered the potential correlation among the same patients who had multiple PD episodes during the study period. Potential risk factors, including patient demographics variables and prior comorbidities, were added into several models in a stepwise manner to compute the odds ratio (OR) of the potential risk factors.

Results: A total of 2,255 patients had 2,286 PD episodes within the study period. 80% of the eligible episodes had S. pneumoniae isolates with non-susceptibility against at least one antibiotic tested. The results from several mixed multivariable logistic regression models consistently indicated significantly increased odds ratios of AMR infection among the age 2-17 when compared to age 18-64 (OR: 1.86 (95%CI: 1.34-2.61)). Proportion of antimicrobial non-susceptibility is lower in IPD episodes compared to non-invasive pneumococcal pneumonia. No prior comorbidities showed statistical significance in predicting the AMR status. Additional analysis conducted among adult patients revealed that prior cardiovascular disease is associated with a marginally increased risk of AMR with an OR of 1.43 (95%CI: 1.00, 2.05), while other conditions were not statistically significant.

Conclusions: Measures to reduce AMR S. pneumoniae should focus on children and adolescents at schooling age. Further research is warranted to pinpoint the disease pattern and risk factors associated with AMR PD infection among paediatric patients.