Comparative Study of Clinical and Microbiological Characteristics of ST8 and ST72 Methicillin-Resistant Staphylococcus aureus Bacteremia in Korea = 한국에서 분리된 ST8형과 ST72형 메티실린내성 황색포도알균 균혈증의 임상적 및 미생물학적 비교연구

다국어 초록 (Multilingual Abstract)

Introduction: Staphylococcus aureus is a major global pathogen causing both community- and healthcare-associated infections. Following the introduction of antibiotics, methicillin- resistant S. aureus (MRSA) emerged as a major healthcare problem. While the USA300 clone (ST8-SCCmec IV, PVL-positive) has dominated community infections in the United States and spread globally, Korea has been characterized by the PVL-negative ST72-SCCmec IV clone. However, ST8 lineages, including USA300, have recently emerged in Korea, with increasing reports. Surveillance studies have shown a gradual rise in ST8 prevalence, mirroring patterns seen in Japan and Taiwan. Despite this trend, comparative clinical data between ST8 and ST72 MRSA bacteremia remain limited, prompting this study to investigate their clinical and microbiological differences.
Methods: This retrospective cohort study analyzed adult patients (≥18 years) with S. aureus bacteremia (SAB) admitted to Asan Medical Center, Seoul, between July 2008 and December 2020. Data were derived from a prospectively maintained SAB registry, and only first-episode cases were included after excluding polymicrobial, insignificant, or outpatient-only infections and prior or pretreated cases. Clinical data, infection focus, and outcomes up to 90 days were reviewed by infectious disease specialists using standardized definitions. S. aureus isolates were identified according to CLSI methods or MALDI-TOF, and antimicrobial susceptibility was tested using the MicroScan system. Vancomycin MICs were determined by both broth microdilution and E-test, and molecular typing (MLST, agr function, spa type) was performed for all MRSA isolates. Statistical analyses were performed using R, employing nonparametric tests for group comparisons and logistic regression for mortality risk factors, with Firth and exact methods applied to address small-sample bias. Temporal trends in ST distribution and vancomycin MIC were analyzed using ANOVA and logistic regression by year, with sensitivity analyses excluding incomplete 2020 data. Additional comparisons assessed clinical outcomes by vancomycin MIC levels (≤1 vs. >1 mg/L), and significance was defined as p < 0.05.

Results: A total of 1,975 cases of SAB were identified during the study period, including 998 cases of MRSA (50.5%). ST72 and ST8 accounted for 327 (32.7%) and 23 (2.3%) of the MRSA isolates, respectively. Patients infected with ST8 and ST72 MRSA were similar in age, sex, acquisition mode, and comorbidity profile. Pneumonia was more frequent among ST8 cases, but other infection sources and infection severity did not differ significantly. Antimicrobial susceptibility testing showed that ST8 isolates exhibited significantly higher resistance to ciprofloxacin and erythromycin, while susceptibilities to other agents were comparable. All isolates were susceptible to vancomycin by CLSI criteria, though E-test results revealed higher MIC distributions among ST8 isolates. Agr dysfunction rates were similar between groups. Longitudinal analysis demonstrated a significant increase in the proportion of ST8 MRSA among SAB cases from 1.3% in 2008 to 3.3% in 2020, while ST72 remained stable. In contrast, vancomycin MIC >1 mg/L proportions decreased significantly over time by both BMD and E-test methods, and these MIC variations did not correlate with adverse outcomes. Clinical management, including vancomycin use, trough levels, time to therapy, and duration of antibiotics, was comparable between groups. Mortality and recurrence rates were also similar; 30-day mortality was 13% in ST72 and 0% in ST8 infections, without a statistically significant difference. Univariate analysis identified liver cirrhosis, solid cancer, and high comorbidity burden (CCI > 4) as predictors of 30-day mortality, whereas infectious focus removal was protective. In multivariable models, liver cirrhosis and CCI > 4 remained independent risk factors. Firth and exact logistic regression confirmed no significant association between MRSA genotype (ST8 vs. ST72) and mortality. Post hoc power analysis indicated sufficient statistical power to detect clinically relevant differences, supporting that the non-significant findings reflected true similarity in outcomes. Overall, ST8 MRSA showed distinct resistance and clonal profiles but no difference in clinical severity or prognosis compared with the long-dominant ST72 clone in Korea.

Conclusions: ST8 MRSA displayed distinct resistance and clonal profiles but showed no difference in clinical severity or prognosis compared with the long-dominant ST72 clone in Korea. Mortality was mainly influenced by host comorbidities rather than genotype, emphasizing the need for ongoing surveillance to monitor the clinical impact of emerging ST8 clones.

목차 (Table of Contents)

• Summary i
• Table of Contents iv
• List of Tables v
• List of Figures vi
• Introduction 1
• Methods 3
• 1. Study design and population 3
• 2. Laboratory and microbiological data 4
• 3. Data collection and definitions· 4
• 4. Statistical analysis 5
• Results 8
• 1. Baseline and clinical characteristics 8
• 2. Antimicrobial susceptibility and genotypic features 11
• 3. Temporal trends and vancomycin MIC distribution 14
• 4. Clinical management and outcomes 23
• 5. Predictors of mortality and recurrence 25
• Discussion 32
• Conclusions 37
• References 38
• Korean summary 44