Methods: Adult patients admitted to the ED of a tertiary care hospital in Colombia, between January and April 2016, treated with empiric antibiotics for cUTI were included and followed until completion of their antibiotic regimen. An antibiotic flowchart which considered risk factors for multidrug-resistant bacteria and patient stratification by severity was implemented to guide cUTI treatment. Patients who were treated empirically following the flowchart were considered the adherent group, and those not treated according to the flowchart were the non-adherent group. Clinical and economic outcomes were compared between groups. Data was analyzed using descriptive and inferential statistics.
Results: A total of 50 patients with cUTI were included; 39 (78%) were adherent and 11 (22%) were non-adherent. The mean age was 66 years old; 29 (58%) women and 21 (42%) men. Escherichia coli was the most common bacteria isolated (72%). Complications occurred more frequently in the non-adherent group (45% vs. 13%, p = 0.01). After 48 hours of antibiotic therapy, 78% of adherent patients improved their symptoms in contrast to 9% of non-adherent patients and at the end of treatment, the symptom’s resolution was more frequent in the adherent group (78% vs. 36%, p = 0.006). The mortality rate was also higher in the non-adherent group (20% vs. 2.5%, p = 0.06). Regarding clinical tests, the mean cost per patient in the non-adherent group was $282 USD vs. $113 USD in the adherent group (p = 0.01).
Conclusion: Our findings demonstrate that clinical and economic outcomes are significantly better for patients in the ED treated for cUTI according to an antibiotic flowchart. Flowcharts for other common infections should be implemented and measured as it seems to be a very efficient way to implement prompt and appropriate empiric antibiotic therapy in the ED.
C. Hernández-Gómez, Merck Sharp & Dohme: Consultant , Consulting fee
K. Escandón-Vargas, None
D. Aragon, None
M. V. Villegas, Merck Sharp & Dohme: Consultant , Consulting fee and Research support
See more of: Poster Abstract Session