The prediction values of carbapenemase detection methods and carbapenem susceptibility testing for clinical outcomes of patients with Acinetobacter bacteremia under carbapenem treatment

Background: Carbapenem-resistant Acinetobacter species have emerged as notorious pathogens causing nosocomial infections. Several phenotypic methods have been developed for detecting carbapenemase production in Enterobacteriaceae. The accuracy of these methods in the prediction of carbapenemase production in Acinetobacter species has not been studied well. Methods: This retrospective study enrolled adult patients with Acinetobacter bacteremia from four medical centers in Taiwan between 2012 and 2016. Their demographics and clinical outcomes were recorded. The carbapenem susceptibility of the Acinetobacter species was determined using the agar diffusion method. The carbapenemase genes were detected by PCR. Four phenotypic methods, including the modified Hodge test (MHT), modified carbapenem inactivation method (mCIM), Carba NP test, and CarbAcineto NP test were carried out to determine the production of carbapenemase. Results: We analyzed 257 adults who received initial carbapenem monotherapy for the treatment of Acinetobacter bacteremia. Shock within three days of bacteremia and acquisition of carbapenem non-susceptible isolates were independently associated with a higher 14-day and 30-day mortality in patients with Acinetobacter bacteremia. Among the four phenotypic tests for carbapenemase detection, MHT using the imipenem disc displayed the greatest sensitivity (94%; 95% confidence interval [CI], 89–97%) and specificity (81%; 95% CI, 73–88%) for predicting imipenem non-susceptibility. Conclusion: Carbapenem non-susceptibility and shock were independent risk factors for mortality in patients with Acinetobacter bacteremia. The MHT could predict the carbapenem susceptibility of Acinetobacter isolates. It is a cheap and quick assay, which could be applied in clinical practice.