P30 – Antibiotic Treatment Outcomes in Complicated Intra-abdominal Infection (cIAI)—Results from a Propensity-Matched Cohort Study
Author(s):
Joseph Solomkin, C. Mullins, Alvaro Quintana, Christian Eckmann, Ahmed Shelbaya, Mingfang Zhao, Frank Ernst, Michelle Krukas, Arlene Reisman, University of Cincinnati College of Medicine
Background: Treatment outcomes for patients suffering from cIAI are known to depend on various severity- and treatment-related variables. To examine 3M™ APR™-DRG severity of illness the effectiveness of tigecycline and other antibiotic treatments, we constructed a propensity score using the Premier research database. We then matched patients with this score to examine outcomes of tigecycline vs. other antibiotic therapy.
Hypothesis: Severity of illness descriptors predict outcomes in patients receiving broad-spectrum antimicrobial therapy.
Methods: Retrospective, observational study using a large clinical database maintained by Premier. Success was defined as 1 operation, no additional antibiotics for the cIAI, and survival to 30 days.
Results: The final propensity score model included: APR-DRG severity of illness, APR-DRG risk of mortality, ICU admission, vasopressor use, mechanical ventilation, site of infection, and a range of demographic data for patients and hospitals. The c-statistic for this model was 0.7. A total of 2424 patients were matched (tigecycline [606]; other antibiotic therapy [1818]). Organ site of infection (combined treatment procedure groups): stomach (61), gall bladder (134), appendix (29), small intestine (670), large bowel (1040), and other (875). Treatment was successful in 426 (70.3%) tigecycline-treated patients and 1294 (71.2%) patients receiving other antibiotic therapy. Similar treatment success occurred across all infection sites. Of those survivors, all-cause rehospitalization within 180 days was 45.2% for those considered “cured”, and 49.6% for those considered “failed”. cIAI-related hospitalization was 6.4% for cures and 7.8% for failures.
Conclusions: These data demonstrate the association between severity of acute illness and outcome. Tigecycline and other antibiotic therapies were similar in effectiveness in the overall matched population, and among those in the extreme severity of illness category.