RESUMEN
Background: A reported history of penicillin allergy frequently leads to the prescription of carbapenems as a substitute for penicillin to avoid allergic reactions. Such self-reported allergies need to be accurately characterized to identify targeted antibiotic stewardship interventions that potentially minimize unnecessary carbapenem use. Design: Retrospective cohort study. Method: The proportion of hospitalized patients with penicillin allergy history receiving carbapenem prescriptions was evaluated between January 1st, 2017 and December 31st, 2018 at the University Hospital Basel, Switzerland. The appropriateness of carbapenem prescription of each patient was evaluated using institutional guidelines based on previously published recommendations. Results: Our analysis revealed that among 212 patients with recorded penicillin allergy, of the 247 carbapenem treatment episodes, 79 (32%) were unjustified. Abdominal and lower respiratory tract infections were most frequently associated with inappropriate carbapenem use (OR 2.64, 95% CI 1.22-5.71, P = .014 and OR 2.26, 95% CI 1.08-4.73, P = .031). The recorded allergy type was not documented or unclear in 153 patients (72%) and penicillin allergy was only confirmed in 2 patients (0.9%). Inconsistencies in allergic symptom documentation and allergy types were found between the institution's two software programs. Conclusion: While a multimodal approach to identify and accurately label penicillin allergies remains essential to reduce inappropriate carbapenem use, our findings highlight the need for comprehensive and easily accessible guidelines for carbapenem utilization and structured history-based allergy assessment as an initial screening tool, embedded in a tailored digital allergy record template.
RESUMEN
PURPOSE: Cefepime is recommended for treating infections caused by AmpC beta-lactamase-producing Enterobacterales (AmpC-PE), though supporting evidence is limited. Therefore, this study compared outcomes associated with cefepime versus carbapenem therapy for bloodstream infections (BSIs) caused by AmpC-PE after phenotypic exclusion of ESBL-co-producing isolates. METHODS: This retrospective cohort study compared definite cefepime versus carbapenem treatment for AmpC-PE BSI in hospitalized patients of the University Hospital Basel, Switzerland, between 01/2015 and 07/2020. Primary outcomes included in-hospital death, renal impairment and neurologic adverse events; secondary outcomes included length of hospital stay and recurrent infection. RESULTS: Two hundred and seventy episodes of AmpC-PE BSI were included, 162, 77 and 31 were treated with a carbapenem, cefepime and other antibiotics, respectively. Patients treated with carbapenems were more likely to be transferred to the ICU on admission and more frequently had central venous catheter as a source of infection. In uni- and multivariable analyses, primary and secondary outcomes did not differ between the two treatment groups, except for more frequent occurrence of neurological adverse events among patients treated with carbapenems and shorter length of hospital stay among survivors treated with cefepime. CONCLUSION: After excluding isolates with phenotypic ESBL-co-production, cefepime was not associated with adverse outcomes compared to carbapenems when used to treat BSIs caused by AmpC-PE. Our study provides evidence to support the use of cefepime as a safe treatment strategy for AmpC-PE BSI, particularly in clinically stable patients without initial renal impairment or increased susceptibility to neurological adverse events.