| dc.description.abstract | In the last two decades, intensive use of broad spectrum antibiotics has led
to the emerge of antibiotics-resistant strains of Enterobacteriaceae that produce
Extended-Spectrum (3-Lactamases (ESBLs). The mortality rate in
'susceptibility/treatment mismatched patients' has ranged from 42-100%. To
begin initial appropriate antibiotic therapy in a timely manner, a predictor will be
needed. One of the latest scoring system to predict ESBL infection is Duke Model
Score.
This study aims to assess the accuracy of the Duke Model Score as a
predictor ofESBL infection.
Cross-sectional study was conducted from January - August 2014 in RSUP
H Adam Malik Medan, in patients with infections. Italian Score was calculated
and bacterial cultures were taken from the source of infection.
By using a cut off point of2::8, Duke Model Score sensitivity of91.3% was
obtained, specificity was 82.6%, positive predictive value was 84% and negative
predictive value was 90.5%. From the analysis using ROC curves, the Duke
Model Score accuracy was 93.1% (95% CI: 88.1 to 98.1%, p = 0.0001).
Antibiotics that can be used against ESBL infections only amikacin, imipenem,
ertapenem, meropenem and tigeciclin.
Conclusion: Duke Model Score can be used as predictors of the presence
of ESBL infection | en_US |
