
Methods: A prospectively collected and retrospectively reviewed cohort study was conducted at 10 hospitals in South Korea. Adults who suffered SAB over a 7-year period were reviewed, and pMRSAB was defined as an MRSAB longer than 3 days despite of adequate antimicrobial therapy. Among 1,588 cases of SAB, 866 (54.5%) were MRSAB, and 241 (27.8%) of those were pMRSAB. Blood isolates were analyzable in 172 (71.4%) cases. Functionality of agr locus, spa type, SCCmec type, presence of genes for PVL and PSM-mec, and vancomycin minimum inhibitory concentration (MIC) were determined. Those microbiological factors were analyzed together with clinical factors to reveal the risk factors of in-hospital mortality.
Results: Among 172 cases of analyzable pMRSAB, 57 (33.1%) were in-hospital mortality. The number of agr dysfunction was significantly higher in in-hospital mortality than in survival (75.4%, 43/57 vs. 52.2%, 60/115, P=0.008). SCCmec type, positivity of PVL and PSM-mec genes, and vancomycin MIC were not significantly different between those groups. In multivariable analysis with clinical factors including Charlson’s comorbidity weighted index score, Pitt bacteremia score, and pneumonia as a primary infection foci, agr dysfunction was revealed as an independent risk factor for in-hospital mortality in pMRSAB (adjusted odds ratio, 2.38; 95% confidence interval, 1.12-5.04, P=0.024) (Table).
Conclusion: agr dysfunction is an independent risk factor for in-hospital mortality in pMRSAB.
Survival (n=115) |
In-hospital Mortality (n=57) |
Univariate OR (95% CI) |
Multivariate OR (95% CI) |
|
Charlson’s comorbidity index (±SD) |
4.6 (± 2.7) |
5.8 (± 2.7) |
1.18 (1.05-1.33)a |
1.16 (1.02-1.32)b |
Pitt bacteremia score, median (IQR) |
1.0 (0.0-2.0) |
2.0 (1.0-4.0) |
1.28 (1.09-1.50)a |
1.25 (1.05-1.48)b |
Pneumonia (%) |
2 (1.7) |
9 (15.8) |
10.59 (2.21-50.87)a |
5.70 (1.12-28.95)b |
agr dysfunction (%) |
60 (52.2) |
43 (75.4) |
2.82 (1.39-5.70)a |
2.38 (1.12-5.04)b |
OR, odds ratio; CI, confidence interval; SD, standard deviation; IQR, interquartile range
a P < 0.01; b P < 0.05

C. K. Kang,
None
C. J. Kim, None
K. H. Song, None
E. S. Kim, None
S. J. Kang, None
N. H. Kim, None
W. B. Park, None
Y. K. Kim, None
H. C. Jang, None
S. Lee, None
Y. S. Kim, None
Y. G. Kwak, None
K. T. Kwon, None
S. Kiem, None
H. B. Kim, None
See more of: Poster Abstract Session