NEW ORLEANS – Patients seen in the emergency department present with a wider range of risk for resistant infections than do hospitalized patients overall, and as a result, hospital-wide antibiograms that dictate empiric therapy may not translate well to the ED setting, a study showed.
Instead, investigators suggested that clinicians choose empiric antimicrobial therapy in the emergency department based on factors associated with differences in susceptibility.
The study researchers compared common antibiotics to treat Escherichia coli between adults treated at their institution overall versus the emergency department, and by sex, patient age, and whether people came to the ED from home versus a long-term care setting. They found some significant differences that could guide empiric treatment in the ED setting.
“Our ED pharmacist observed a lot of broad-spectrum antibiotic prescribing for UTIs,” lead investigator Sarah Jorgensen, PharmD, said at the annual meeting of the American Society for Microbiology. “Also, we’ve had a culture follow-up program in place for the last 5 years, and they had to intervene on a lot of postdischarge antibiotic mismatches.”
E. coli was the most common urinary pathogen detected in this study of 500 randomly selected ED patients with ICD-9/ICD-10 diagnostic codes for urinary tract infection. Investigators found E. coli in 64% of the 226 culture-positive patients presenting to the ED at Huntington Hospital in Pasadena, Calif., between July 2015 and June 2016.
“What was surprising for us, because our enrollment was based on ICD codes for a UTI, is that only about 50% had a positive urine culture,” said Dr. Jorgensen, a pharmacy resident at the University of Southern California in Los Angeles. “Urinalysis was positive in about 99% of the population.”
Dr. Jorgensen and her colleagues found overall low susceptibilities of 71% for ciprofloxacin, 66% for trimethoprim/sulfamethoxazole, and 67% for cefazolin susceptibilities in the ED. They also found that 8% of isolates were positive for extended-spectrum beta-lactamase isolates, as were 1% of E. coli isolates.
The 67% cefazolin susceptibility was significantly lower in the ED, compared with the 86% susceptibility in the institutional antibiogram (P less than .001).
The investigators found E. coli susceptibility to ciprofloxacin was lower in men, at 55%, compared with 74% among women, but the difference was not statistically significant (P = .14). A similar pattern emerged with cefazolin – 55% susceptibility among men and 69% among women (P = .26). In contrast, trimethoprim/sulfamethoxazole susceptibility trended higher in men, at 73%, vs. 64% among women in the ED (P = .63).
When they divided patients by age 50 years and younger versus those older than 50, the investigators found ampicillin susceptibilities were lower in the younger group, at 30%, compared with 51% among those in the older cohort (P = .03). Similarly, gentamicin susceptibilities were 80% in the younger group, compared with 92% in the older group (P = .04).
Ciprofloxacin susceptibility was significantly lower among people coming to the ED from a long-term care facility than among those coming from home – 35% vs. 77% (P less than .001). Differences in ciprofloxacin susceptibility between admitted and discharged patients was less striking – 63% vs. 78% (P = .04).
Nitrofurantoin was the only oral agent with susceptibility greater than 80% in all patient groups, with susceptibility ranging from 88% to 100%.
Because it typically takes 2-3 days to get the susceptibility results back at Huntington Hospital, many patients are discharged on empiric therapy, noted Mira Zurayk, PharmD, a resident at Huntington Hospital. That can present multiple challenges, particularly with homeless patients who are difficult to find and provide follow-up for, Dr. Jorgensen added.
Based partly on the study findings, the investigators developed a clinical algorithm specifically to address UTI antimicrobial prescriptions in the ED. The algorithm incorporates different recommendations for different groups of patients because of their different resistance trends.
“I think that is a good way to tailor empiric therapy when you don’t have culture results up front,” Dr. Jorgensen said. “We just implemented the algorithm, and I’m now analyzing the outcomes.”
Having more data on outcomes will help the clinicians target lowering the rate of “drug-bug mismatches,” as well as UTI-related revisits to the ED. In addition, the work could help expand the antibiotic stewardship program in the hospital to the ED for the first time, Dr. Jorgensen said.
Dr. Jorgensen and Dr. Zurayk had no relevant financial disclosures. One of the study coauthors, Annie Wong-Beringer, PharmD, receives grant funding from Merck.