Background: Approximately 10% of patients report a penicillin allergy; however, fewer than 1% of the population have an IgE-mediated allergy. Penicillin allergy evaluation involves obtaining a detailed history of the reported reaction and may include diagnostic testing such as skin testing and/or an oral penicillin challenge. When appropriate, this process can result in the removal of the allergy label from a patient's medical record, a process known as allergy de-labeling. This practice has been well studied and is considered safe among the general population. Given its safety and effectiveness, the American College of Obstetricians and Gynecologists (ACOG) encourage penicillin allergy evaluation for any patient with a documented penicillin allergy; however, despite these recommendations, this patient population is less likely to undergo evaluation for allergy de-labeling.
Antimicrobial use in pregnancy is common, particularly for managing Group B Streptococcus (GBS), cesarean section prophylaxis, and infectious complications such as chorioamnionitis. ACOG guidelines recommend beta-lactam antimicrobials, including ampicillin and cefazolin, as first-line agents for prophylaxis and treatment during the peripartum period. However, patients labeled as penicillin-allergic are often prescribed second-line, broad-spectrum antimicrobials instead. Wellstar MCG Health (WMCGH) implemented a new penicillin allergy screening assessment process for obstetrics and gynecology patients in September 2025. The purpose of this study is to expand upon existing literature by evaluating the outcomes of this new assessment.
Methods:This is a single center, retrospective, chart review study evaluating the outcomes of a newly implemented penicillin allergy screening assessment in obstetrics and gynecology patients. All patients 18 years or older who were seen at pre-specified WMCGH women’s clinic locations with a documented penicillin allergy were eligible for enrollment. Patients were excluded if they did not have established care with WMCGH prior to admission and if they were admitted for a non-related OB/GYN encounter. The primary outcome was the difference in the incidence of guideline recommended first-line antimicrobials received for GBS, chorioamnionitis, and surgical prophylaxis. Secondary outcomes included hospital length of stay, incidence of de-labeled penicillin allergy, rate of postpartum and surgical site infections, total days of antimicrobial use, 30-day hospital re-admission, and antimicrobial cost savings.
Results: A total of 140 patients were included with 97 patients in the pre-implementation group and 43 patients in the post-implementation group. For the primary outcome, 20% of patients in the pre-implementation group received first-line antimicrobials compared to 43% in the post implementation group (p = 0.106). Implementation of penicillin allergy screening was associated with a 29% relative risk reduction in patients receiving non-first line antimicrobial therapy. There was no difference in hospital length of stay between the two groups, and 30-day hospital re-admission was 10% vs. 5% in the pre- and post-implementation groups respectively. The rate of postpartum and surgical site infections was similar between pre- and post-implementation groups (4% vs 5%, p = 0.891) in addition to the total days of antimicrobial use (0.3 vs 0.2 days, p = 0.460). A total of 3 patients (7%) were de-labeled based on penicillin allergy assessment. The pre-implementation group demonstrated higher overall drug cost utilization, largely driven by more costly antimicrobial agents ordered. This difference may become more pronounced with a larger sample size, suggesting potential for meaningful cost savings at scale.
Conclusion: Patients in the post-implementation group received guideline-recommended first-line antimicrobials more often than those in the pre-implementation group; however, this study was underpowered to detect statistical significance. Patients in the post-implementation group that were appropriately identified and screened were able to be successfully de-labeled. Most de-labeling occurred through an allergy/immunology referral consult and oral amoxicillin challenge. Further studies with larger sample sizes are warranted to better evaluate the impact of this intervention.
[email protected]
