Background The Antimicrobial Stewardship Program (ASP) at King Abdulaziz Medical City - Western Region (KAMC-WR), initiated in November 2016, aims to improve the use of antimicrobials at our facility. It is internationally recognized that one of the cornerstones of ASP is prospective audit and feedback (PAF). In an effort to improve antimicrobial use in areas at high risk for antimicrobial resistance, the ASP initiated a PAF service in February 2018 at the pediatric intensive care unit (ICU), pediatric cardiac ICU, adult ICU, and adult hematology-oncology ward. This retrospective study describes PAF interventions and antimicrobial appropriateness based on PAF.

Methods An automatically generated daily antimicrobial report built by the ASP and ISD team is used by the PAF to assess patients not under infectious disease (ID) consultation on prespecified antimicrobials for appropriateness of indication, dose, route, and frequency. The PAF service is run by the PICU or ASP/ID clinical pharmacist, ID fellow, and ID consultant. The PAF teams meet with the respective departments for discussion and communication of recommendations twice weekly for pediatric patients and daily for adults. A customized PAF electronic note is filled for each order. The audited antimicrobials are imipenem, meropenem, colistin, tigecycline, linezolid, vancomycin IV, anidulafungin, caspofungin, and voriconazole. An electronic report of the PAF note is generated regularly by the ISD team to assess PAF. Simple descriptive statistics were used for the analysis.

Results A total of 747 PAF consultations were documented from February 2018 to February 2019, 92% of which were in adults (686 of 747). 35% of all consultations were inappropriate. Of the 259 inappropriate medication orders, most were in adults (97%) with pneumonia (40%) and sepsis (27%). 68% of inappropriate orders were empiric, and the rest had no clear indication or were therapeutic. The most common inappropriate element of adult orders was choice of antimicrobial, followed by dose, frequency, and then route (83%, 6%, 4%, and 1%, respectively). All inappropriate pediatric orders were inappropriate choices. For adults, meropenem and vancomycin consisted of 67% of inappropriate choices of antimicrobial (140 of 208). Of all antimicrobials, tigecycline had the highest rate of inappropriate choice (10 of 21).

Conclusion This study shows that antimicrobial use guidelines for meropenem, vancomycin, and tigecycline need to be re-evaluated and reinforced through continuous PAF and the creation of clinical practice guidelines with electronic order sets for pneumonia and sepsis. The results of this workflow embedded electronic assessment will help the ASP at KAMC-WR tailor future interventions that promote safe and effective antimicrobial use.