1238. Evaluating the Impact of an Automatic Antibiotic De-escalation Protocol at an Acute Care Hospital

Abstract

Abstract Background Previously, pharmacist-led stewardship programs have been shown to result in more timely de-escalation of antibiotics and lack of adverse outcomes in patients treated for pneumonia in the intensive care unit (ICU). This study evaluated the effects of an automatic pharmacist-led antimicrobial de-escalation policy outside of the ICU. Methods This was a retrospective pre- and post-implementation study at a 425-bed tertiary care hospital in Las Vegas, Nevada. Inclusion criteria consisted of patients admitted from October 2017 to October 2021 who had Escherichia sp., Proteus sp., Klebsiella sp., Pseudomonas sp., or Enterococcus sp. isolated from specified sites that were susceptible to certain antibiotics included in the de-escalation policy. Exclusion criteria included concomitant infections, polymicrobial infections with organisms outside of those mentioned in the inclusion criteria, deep-seated infections, or hemodynamic instability. The primary outcome was time to de-escalation, which was calculated by determining elapsed time between culture susceptibility results and start of the new antibiotic if de-escalated. If de-escalation did not occur, then time to de-escalation was calculated by comparing time of susceptibility results to antibiotic discontinuation. Secondary outcomes evaluated included rates of antibiotic de-escalation, occurrence of readmission at 30- and 90-days, incidence of all-cause inpatient mortality, length of stay, incidence of Clostridioides difficile infection, and time to clinical resolution. Time to clinical resolution was defined by when patient met pre-specified laboratory values following the first antibiotic dose. Results A total of 174 patients were included in the pre-group and 52 in the post-group. The results are shown in Table 1; there was a statistically significant decrease in time to de-escalation, rates of de-escalation, and 90-day readmission rates favoring the post-intervention group, with no differences seen in the other outcomes. Conclusion An automatic de-escalation protocol can significantly shorten time to de-escalation in addition to improving rates of de-escalation outside of the ICU without negatively impacting patients. Disclosures All Authors: No reported disclosures