Does thePenicillin Allergy Label Affect Outcomes of Complicated Odontogenic Infections?

Penicillins are a potent antibiotic in managing odontogenic infections, but 10% of the population is labelled as allergic to these drugs. This has limited their use and resulted in increased utilization of health care resources as well as complications associated with alternative antibiotics. The purpose of the study was to measure the association between patients labeled as penicillin allergic and treatment outcomes in a sample of patients treated for complicated odontogenic infections. Additionally, we sought to investigate antibiotic resistance patterns in these patients.

A retrospective cohort study was performed at the Michigan Medicine health care system to include patients who were treated for complicated odontogenic infections by oral and maxillofacial surgery between 2016 and 2020. Complicated odontogenic infection was defined as any odontogenic infection requiring admission and surgical management in the operating room. The primary predictor variable was the penicillin allergy label, which was determined by chart review and not confirmed with formal testing. Outcomes were measures of disease severity. The primary outcome variable was hospital length of stay. Secondary outcome variables were ICU admission (yes/no), repeat computed tomography scan(s), repeat surgery (yes/no), and re-admission (yes/no). Co-variates included were age, sex (male/female), tobacco use status, diabetes, immunocompromised state, number of spaces involved, white blood cell count upon admission and insurance status. For our secondary aim, the primary predictor variable was again penicillin allergy and outcome variable was antibiotic resistance as determined by wound culture results following surgical intervention. Negative binomial regression and logistic regression analyses were performed. P < .05 was considered significant.

A total of 150 patients met the inclusion criteria and of those 17.3% reported as penicillin allergic. Patients labelled as penicillin allergic did not differ significantly from patients without penicillin allergy label in terms of treatment outcomes. Age, diabetes, and immunosuppression were associated with an increased length of stay. Patients labelled as penicillin allergic were at significantly higher risk for antibiotic resistance (relative risk = 2.34; 95% confidence interval, 1.66 to 3.32; P < .001), specifically clindamycin resistance (relative risk = 3.17; 95% confidence interval, 1.93 to 5.18; P < .001).

Penicillin allergy was significantly associated with clindamycin resistance. There were similar outcomes amongst patients with and without a penicillin allergy label despite antibiotic differences. Delabeling efforts for patients with a reported penicillin allergy must be considered and local nomograms for antibiotic selection should be used by providers when seeking alternative antibiotics.