Clinical and Microbiological Outcomes of Serious Infections with Multidrug-Resistant Gram-Negative Organisms Treated with Tigecycline

Kara B. Anthony () 2 4 Neil O. Fishman 2 4 Darren R. Linkin 0 2 3 4 Leanne B. Gasink 2 3 4 Paul H. Edelstein 1 Ebbing Lautenbach 0 2 3 4 0 Biostatistics and Epidemiology 1 Pathology and Laboratory Medicine 2 Division of Infectious Diseases, Department of Medicine, and Departments of 3 Center for Clinical Epidemiology and Biostatistics, University of Pennsylvania School of Medicine , Philadelphia 4 Center for Education and Research on Therapeutics Eighteen patients received tigecycline as treatment for infection due to multidrug-resistant gram-negative bacilli, including Acinetobacter baumannii and Klebsiella pneumoniae carbapenemase- and extended-spectrum b-lactamase-producing Enterobacteriaceae. Pretherapy minimum inhibitory concentration values for tigecycline predicted clinical success. Observed evolution of resistance during therapy raises concern about routine use of tigecycline in treatment of such infections when other therapies are available. - Infectious organism and patient Primary infection Comorbid condition(s) CHF, DM, HD Initial source of culture specimen Antimicrobial susceptibilitya a Active agent(s) listed. b Related or unrelated indicates relationship of death to tigecycline-treated infection. c Given to treat Pseudomonas aeruginosa, which was also isolated in the bronchial washing. d Given to treat Staphylococcus aureus, which was also isolated in the tracheal aspirate. e Given to treat P. aeruginosa, which was also isolated from the wound. initiation of antibiotic treatment), or not documented [9]. For Tigecycline-susceptibility testing was performed on bacterial microbiological response, if any criteria for positivity were met, isolates from 16 of the 18 patients before initiation of tigecycline the response was considered to be positive. Final disposition therapy (table 1). Of 9 A. baumannii isolates tested, 5 demwas defined as death related to infection (death in the setting onstrated intermediate resistance. Four (80%) of these 5 paof clinical evidence of active infection or within 5 days after tients with pretherapy isolates only intermediately susceptible the last positive culture result), death unrelated to infection to tigecycline died (all deaths were related to infection), whereas (death after an episode of infection but due to causes independent of the infectious process), or survival [9]. Categorical variables were compared using the Fishers exact test. 0 of 4 patients with pretherapy isolates susceptible to tigecycline died (P p .048). Among the 8 patients with nonA. baumannii isolates, pretherapy MIC appeared unrelated to survival. Twenty-one patients received tigecycline therapy There were 8 patients who had persistently positive culture 48 h within our facility during the defined study period results after initiation of tigecycline therapy; repeat testing for for treatment of a documented MDR gram-negative infection. susceptibility to tigecycline was performed for 6 of them. Of Of these, 18 received a full course ( 7 days) of therapy and these, patient 3 had an A. baumannii isolate that remained were included in this study (table 1). Of the 3 patients excluded intermediately susceptible (MIC, 3.004.00 mg/mL), and tigefrom the study; 1 was transferred to another facility during cycline treatment was discontinued after 28 days, because of treatment; 1 received 5 days of empirical therapy and, once clinical and microbiological failure. Patient 5 had an A. baumicrobiological data were available, was switched to treatment mannii tracheal aspirate isolate that was initially susceptible but with a targeted antibiotic; and 1 had an initial isolate found to that developed resistance during therapy (the MIC increased be resistant to tigecycline, so treatment was changed to another from 2.00 to 12.00 mg/mL after 14 days). The 4 other patients for whom susceptibility testing was Causative organism Resistance mechanism repeated had bacterial isolates that remained susceptible to tigecycline during therapy. Patient 13 had persistent K. pneumoniae tracheal and pleural isolates, despite 7 days of tigecycline treatment (MIC, 1.50 mg/mL), and ultimately died, after 16 days of treatment, of sepsis and respiratory failure. Patient 7 had a primary A. baumannii mediastinitis and secondary bacteremia, with blood culture results that remained positive after 5 days of therapy, despite repeated MIC values of 2.00 mg/mL. The patient died, on day 8 of therapy, of causes related to this infection. Patient 17, a recent heart transplant recipient, developed postoperative mediastinitis and an aortic pseudoaneurysm at the allograft anastomosis, both due to K. pneumoniae. His blood culture results were also positive (MIC, 1.0 mg/mL), and he remained persistently bacteremic, despite 140 days of therapy. The patient ultimately died of aortic rupture. Patient 18 had multiple recurrences of E. coli bacteremia (MIC, !0.75 mg/mL) in the setting of a retained venous catheter and septic thrombophlebitis, despite 1100 days of inpatient therapy. He was transferred to another facility, where he later died, on day 133 of therapy. Discussion. We describe 18 patients who received tigecycline for treatment of serious infections caused by MDR gramnegative bacilli. Tigecycline was used to treat a variety of infections, many not indicated in official FDA labeling for tigecycline. Most patients were critically ill at the time of tigecycline administration, and overall clinical outcomes were poor. Although tigecyclines potent in vitro activity against MDR Coadministered antibiotics gram-negative bacilli has suggested clinical success in the treatment of infections due to these organisms [3], there are few data from the clinical setting to support this. In our study, almost one-half of initial A. baumannii isolates showed intermediate susceptibility to tigecycline; this was associated with a higher mortality rate. This finding complements other recent reports of pre-existing reduced tigecycline susceptibilities [10] and suggests that pretherapy tigecycline MIC values may predict clinical outcome in these infections. It is important to note that no adjustment for the effect of potential confounders could be made in our analysis on the basis of our small sample size. In addition to pre-existing tigecycline resistance among A. baumannii, we observed that 1 isolate acquired full resistance to tigecycline during treatment. Recent reports have similarly described the emergence of resistance among MDR gram-negative organisms during therapy and question the durability of antimicrobial activity of tigecycline as its use becomes more widespread [1114]. Additionally, in our study, we observed persistent A. baumannii, E. coli, and K. pneumoniae bacteremia in patients receiving tigecycline treatment, despite isolates that maintained MIC values below the susceptibility breakpoint. Although at least 2 of these patients had potential sources for their persistent infection, recent reports b (...truncated)