10% of hospitalized patients have penicillin allergy listed in their records, fewer than 1% of patients have true allergies.

Use of more-expensive and broader-spectrum antibiotics is associated with longer and more-expensive hospital stays and more side effects, nosocomial infections, and resistant organisms.

Blumenthal KG et al. Reaction risk to direct penicillin challenges: A systematic review and meta-analysis. JAMA Intern Med 2024 Sep 16; [e-pub]. (https://doi.org/10.1001/jamainternmed.2024.4606)

 researchers examined the safety of direct penicillin challenges (without preceding skin tests) for delabeling patients without true allergies. Among more than 9000 patients in these studies, 438 experienced reactions (3.5%), with only 5 reactions classified as severe: 3 episodes of anaphylaxis, 1 delayed rash with fever, and 1 kidney injury. No fatalities were reported.

NNH of 1800

The PENFAST score is a good tool to help decide which patients can undergo direct oral challenge safely (NEJM JW Gen Med Aug 1 2023 and JAMA Intern Med 2023; 183:883). In general, if a patient has a history of severe immediate reaction (angioedema or anaphylaxis), a recent urticarial reaction (within 5 years), or any severe delayed reaction (e.g., Stevens–Johnson syndrome, serum sickness, drug reaction with eosinophilia, drug-induced cytopenia, organ injury), I would refer to an allergist for evaluation.

Bottom line

We have far more patients who should have their penicillin allergy delabeled than we have allergists to perform these challenges. Primary care clinicians and hospitalists can do this easily by giving one dose of amoxicillin (500 mg) and watching the patient for 1 to 2 hours; intramuscular epinephrine and oral antihistamines must be available, but are seldom needed.