249. Extensively Drug-Resistant (XDR) and Pandrug-Resistant (PDR) Bacteria in the Post-Antibiotic Era: A Study on Epidemiology and Outcomes in a High-Prevalence Setting
Session: Poster Abstract Session: Antimicrobial Susceptibility and Resistance
Friday, October 21, 2011
Room: Poster Hall B1
Handouts
  • poster_idsa_Stardeli.pdf (199.5 kB)
  • Background: Gram-negative bacteria (GNB) resistant to all antibiotics have appeared in Greece. Limited data exist on the outcome of such infections. We assessed the resistant patterns and prevalence of multiresistant GNB in our 767-bed Hospital, and the impact of infections to hospitalized patients.

    Methods: All clinical samples (October-December 2009) that yielded Enterobacteriaceae, A. baumannii and P. aeruginosa were analyzed. Susceptibility patterns (EUCAST criteria) to b-lactams, carbapenems, quinolones, aminoglycosides, cotrimoxazole, tigecycline and polymixins were recorded. Isolates were categorized as: Pandrug-resistant (PDR), Extensively drug-resistant (XDR), Mutlidrug-resistant (MDR) and non-multiresistant (non-MDR), if found non-susceptible to all, all but 1 or 2, all but 3, and <3 antibiotic classes, respectively. Next, a case-control study compared the clinical characteristics and outcomes of infections from PDR and XDR strains with those from non-MDR counterparts, in patients matched by age and sex.

    Results: 838 isolates comprised of 562 (67%) non-MDR, 195 (23.2%) MDR, and 81 (9.8%) XDR, 20 of which were resistant to all antibiotics tested. 27.2% (34/125) of K. pneumoniae and 17.5% (20/114) of A. baumannii were susceptible to no more than one agent. 38 infections (cases) from XDR/PDR GNB that were evaluated, occurred on average 34 days after admission, as opposed to 17 days until infection from a non-MDR pathogen (p=0.012). Comorbidities and icu history did not significantly differ between cases and controls. All-cause and infection-related mortality were much higher among cases than controls (55.3% vs 28.9%,  and 23.7% vs 5.3%, respectively; odds-ratio for death = 3). The isolate was empirically covered successfully in 32.4% of cases and 80% of controls, whereas the regimen after isolation contained an active agent in 48.5% and 95.8%, respectively (p<0.01, both). De-escalation was done in only 10.5% of controls.

    Conclusion: The high rates of very resistant pathogens are alarming and associated with increased mortality; to this contribute the inability to successfully cover the pathogen during the initial crucial hours of empiric regimen, as well as the limited therapeutic options available for targeted therapy.


    Subject Category: A. Antimicrobial agents and Resistance

    Thomai Stardeli1, Olympia Samoglou1, Victoria Gouma1, Eirini Triantafyllaki1, Kostoula Arvaniti2, Eleni Siskou2, Vasiliki Koulourida2 and Spiros Miyakis3, (1)Medical School, Aristotle Univeristy of Thessaloniki, Thessaloniki, Greece, (2)Papageorgiou General Hospital, Thessaloniki, Greece, (3)Medical School, Aristotle University of Thessaloniki, Papageorgiou General Hospital, Thessaloniki, Greece

    Disclosures:

    T. Stardeli, None

    O. Samoglou, None

    V. Gouma, None

    E. Triantafyllaki, None

    K. Arvaniti, None

    E. Siskou, None

    V. Koulourida, None

    S. Miyakis, None

    Findings in the abstracts are embargoed until 12:01 a.m. EST Thursday, Oct. 20 with the exception of research findings presented at IDSA press conferences.