Impact of removing ESBL testing report from cultures on the selection of antibiotics for the treatment and 30-day mortality of patients infected with ESBL-producing organisms

Abstract

Abstract Background The increasing incidence of carbapenem-resistant organisms is well recognized as a global crisis. Several studies have identified previous carbapenem use as a risk factor for the subsequent development of infections with carbapenem-resistant Gram-negative organisms. We characterized the impact of the removal of the ESBL designation from microbiology reports on inpatient antibiotic prescribing and mortality. Methods This was a historical control and interventional analysis performed in a 1300-bed university hospital to compare 1 year before (1 August 2019 to 31 July 2020) and 1 year after removal of the ESBL designation (1 August 2020 to 31 July 2021), with a 1-month washout period (1 August 2020 to 31 August 2020). All statistical tests were two-tailed, and P < 0.05 was considered statistically significant. Results A total of 213 and 207 patients were included before and after elimination of the ESBL designation in the Hospital Information System. The most commonly isolated organism was Escherichia coli. The most common specimen in which these organisms were detected was urine (57.8% and 62.2%, before and after periods). Definitive prescribing of carbapenems, as the primary outcome, decreased from 56.5–41.3% (P = 0.01); cefepime prescribing decreased from 13.6–3.5% (P < 0.05); and piperacillin-tazobactam use increased significantly from 10.4–28.7% (P < 0.05). TMP-SMX was prescribed more frequently, increasing from 0 to 2.8%; however, this change was not statistically significant. The 30-day mortality rate from any cause did not differ between the before and after periods, with 22/154 patients (14.3%) before and 24/143 (16.8%) after (P = 0.55). There was a non-statistically significant decrease in carbapenem prescriptions for definitive treatment of bacteremia caused by ESBL-producing organisms (84.4% and 68.6%; P = 0.155), while definitive piperacillin-tazobactam prescribing increased from 0 to 6.0% (P = 0.025), comparing the before and after periods, respectively. Carbapenem prescriptions for urinary tract infections decreased from 49.4–28.1% (P = 0.005), while piperacillin-tazobactam was prescribed more often (11.2% versus 32.6%; P = 0.001) (before and after, respectively); carbapenem use for pneumonia did not decrease. Conclusions Definitive prescribing of carbapenems and cefepime decreased and piperacillin-tazobactam use increased after removal of the ESBL report. Our findings confirm the importance of collaboration between microbiology and antimicrobial stewardship programs.