Background: Patients with penicillin allergies often receive alternative preoperative antibiotics despite antimicrobial stewardship guidelines supporting cefazolin in nonanaphylactic cases. These alternative regimens may be less effective, and recommendations to administer cefazolin under such circumstances have not been widely adopted. This study evaluated differences in infectious complications, preoperative antibiotic use, and timing of prophylaxis in breast reconstruction patients with penicillin allergies. Methods: This was a retrospective review of patients who underwent tissue expander breast reconstruction between 2015 and 2022. Demographic data and comorbidities were collected, and infectious complication rates were compared between penicillin-allergic and non–penicillin-allergic patients. Secondary outcomes included timing to onset of infection and infection as a function of preoperative antibiotic administration timing. Results: Among 427 patients, 96 (22.4%) reported a penicillin allergy. Baseline characteristics were comparable between groups. Infection occurred in 37.5% (36 of 96) penicillin-allergic patients versus 23.8% (79 of 331) of non–penicillin-allergic patients ( P = 0.002). Among those who received clindamycin prophylaxis, patients developing infection had a significantly shorter antibiotic administration-to-incision interval compared with those who did not ( P = 0.03). The median time from surgery to infection onset was 72 (interquartile range 24–126) days for penicillin-allergic patients and 96 (interquartile range 30–196) days for non–penicillin-allergic patients ( P = 0.15). Conclusions: In this population, penicillin-allergic patients experienced significantly higher infection rates than those who received cefazolin prophylaxis. The increased infection risk may be partially explained by protocol-driven alterations leading to a shorter interval between antibiotic administration and incision in patients receiving clindamycin, though multiple factors related to penicillin allergy and the community antibiogram could also contribute.
Quiroga et al. (Sun,) studied this question.