BOSTON – The sole risk factor common to patients with bloodstream infections from either drug-resistant extended-spectrum beta-lactamase–producing strains (ESBL) of Proteus mirabilis or ESBL-negative strains was treatment with an in-dwelling catheter, a difference that might help clinicians more easily identify patients with drug-resistant infections, Italian researchers reported Sept. 13 at the annual Interscience Conference on Antimicrobial Agents and Chemotherapy.
In a retrospective, hospital-based case-control study, significant risk factors for bloodstream infections with ESBL-positive P. mirabilis were admission to a hospital from a long-term care facility, previous antibiotic therapy, indwelling urinary catheter use, and previous hospitalizations.
In contrast, the only significant risk factors associated with ESBL-negative P. mirabilis infections were immunosuppressive therapy, age over 65 years, or indwelling urinary catheter use, said Dr. Mario Tumbarello from the Policlinico Universitario Agostino Gemelli in Rome.
In the study of 89 patients with P. mirabilis sepsis treated from 1999 through 2008, ESBL positivity was associated with significantly worse outcomes. For example, among all patients with P. mirabilis bloodstream infections, the 21-day mortality rate was 31.5%. Among patients with ESBL-positive infections only, the 21-day mortality rate was 52.9%, compared with 18.2% for patients who had infections with non-ESBL strains (P =.001).
Similarly, 30-day Kaplan-Meier survival estimates were significantly better among patients with ESBL-negative infections (P = less than.001), said Dr. Tumbarello at the meeting, which was sponsored by the American Society for Microbiology.
The investigators looked back over 10 years to see whether they could identify risk factors for the isolation of ESBL-producing P. mirabilis or non-ESBL–producing P. mirabilis in blood cultures. They considered variables such as patient demographics, comorbidities, medical/surgical history, catheter use, previous immunosuppressive therapy, type, dose and duration of antibiotics, length of stay, and APACHE II score.
They found, using multivariate logistic regression analysis, that risk factors associated independently with ESBL-producing P. mirabilis bloodstream infections were admission to the hospital from a long-term care facility (odds ratio [OR] 8.68, P =.008), previous antibiotic therapy (OR 4.14, P =.003), indwelling urinary catheter (OR 3.13, P = .002), and previous hospitalization (OR, 2.76, P = 0.03).
For non-ESBL–producing isolates, independent significant risk factors were immunosuppressive therapy (OR 4.58, P =.002), age greater than 65 years (OR 4.03, P less than 0.001), and indwelling urinary catheter (OR 2.57, P = 0.01).
The study was internally funded. Dr. Tumarello said that he has no conflicts of interest relevant to the study.