44 - Antibiotic Class-Based Distribution and Analysis of Reported Beta-Lactam Allergies amongst Hospitalized Patients

Background: Reported β-lactam allergy (BLA) is very common, yet less than 10% of these patients exhibit true hypersensitivity.  When faced with reported BLAs, physicians often choose alternative antibiotics which can be associated with C. difficile infection, drug-resistance development, poorer outcomes, & increased costs.  Effective identification of these patients is necessary for subsequent, appropriate BLA "de-labeling."  Here, we conducted a single-center analysis of alternative antibiotic utilization amongst patients reporting BLA and compare the frequency of drug-resistant infections and C. difficile infection in allergic & non-allergic patients.

Methods: This is a retrospective review of adult patients hospitalized at The University of Texas Medical Branch from 1/1/2015 to 12/31/2019.  Pooled electronic medical records were filtered by antibiotic orders and reported allergies to penicillins or cephalosporins.  Patients with drug-resistant and/or C. difficile infection (CDI) were identified by ICD-10 codes.  Microsoft Excel & MedCalc were used for statistical calculations.

Results: Data were available for 118,326 patients and 9.3% (11,982) reported a BLA, with the highest rates seen in those receiving aztreonam (85.9%, 530/617) & clindamycin (33.7%, 3949/11718).  Amongst patients reporting BLA, high ratios-of-consumption (relative to all patients receiving antibiotics) were seen with aztreonam (7.0), clindamycin (2.7), cephalosporin/β-lactamase inhibitors (2.4), & daptomycin (2.1).  Compared to the non-BLA population, BLA patients more frequently experienced MRSA infection (3.0% vs 1.5%, OR 1.99, 95% CI 1.79-2.23, p< 0.0001), β-lactam resistance (1.2% vs 0.6%, OR 2.07, 95% CI 1.72-2.49, p< 0.0001), and CDI (1.2% vs 0.7%, OR 1.85, 95% CI 1.54-2.23, p< 0.0001).     

Conclusion: Our measured BLA rate matches approximate expectations near 10%.  Moreover, these patients experienced significantly higher frequencies of drug-resistant bacterial infections and CDI.  Targeted inpatient penicillin allergy testing stands to be particularly effective in those patients receiving disproportionately utilized alternative agents (e.g. aztreonam, clindamycin, daptomycin).  β-lactam allergy “de-labeling” in these patients is likely a valuable antimicrobial stewardship target.