Implementation and outcomes of beta-lactam allergy management protocol at a comprehensive cancer center

Beta-lactam allergy (BLA) is associated with increased broad-spectrum antibiotic (Br-ABX) use and worse clinical outcomes. We evaluated our hospital-wide BLA protocol (BLA-P) that used following categories: intolerance, low-risk, and high-risk.

Hospitalized adult patients with listed BLA during 10/2021-12/2022 were eligible. Exclusions were critically ill, surgical, hospice or comfort care, or non-verbal patients. Assessment was counted each time a pharmacist evaluated BLA. Interventions were no further action (high-risk allergy, patient refusal, unstable clinical status), updated allergy label, or delabeled. Delabeling was done either based on antibiotic history (direct-delabeling), or via test-dose challenge for low-risk patients. Br-ABX usage was compared in the unique delabeled patients: the empiric antibiotic use 90 days post-delabeling versus pre-delabeling using McNemar test (SPSS).

A total of 700 assessments in 631 patients were identified. 441 assessments in 377 patients (median 63 years-old, 41% male, 50% hematological cancer) met inclusion criteria. The assessments revealed 9% intolerance, 55% low-risk, 23% high-risk and 13% unknown reaction. Interventions resulted in no further action 7%, updated label 72%, and delabeling 21%. 65% of the delabeling was via direct-delabeling and 35% test-dose challenge. Among patients who received a test-dose challenge, 36/36(97%) had no documented allergic reactions, and 1/26(3%) developed a mild rash. The use of aztreonam (pre-delabeling 28% vs. post-delabeling 1.2%, p < 0.001) and meropenem (13% vs. 2.4%, p = 0.022) significantly decreased while cefepime (24% vs. 50%, p = 0.001) and piperacillin-tazobactam (3.7% vs. 22%, p < 0.001) increased after delabeling.

BLA-P led to 21% delabeling, which resulted in increased preferred Br-ABX and decrease in aztreonam/meropenem use among delabeled patients.