May 17, 2019
5 min read

Is your patient ‘allergic’ to penicillin? Perhaps not

You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact

As readers of Infectious Diseases in Children are well aware, it is not uncommon to hear “penicillin” when parents are asked, “Does your child have any drug allergies?” However, many published studies of adults and children have demonstrated that the vast majority of patients who answer with “penicillin” — greater than 95% — are not “allergic” with respect to a demonstrable immunoglobulin E-mediated (eg, anaphylaxis) or T-lymphocyte reaction (eg, Stevens-Johnson syndrome). Most of these reported allergies are not likely to be clinically significant with repeated penicillin or beta-lactam antibiotic administration, or the reported adverse reactions were initially unrelated to penicillin administration (eg, a viral exanthem).

Epidemiology of penicillin allergy

Edward A. Bell

Although about 10% of patients in the United States report a penicillin allergy, most are shown to be not truly allergic when penicillin skin testing is performed (ie, skin testing is negative). It is also important to consider that even in individuals who have positive penicillin skin tests, the immune response decreases over time, by approximately 10% per year. Pathophysiologic theories describing the medicinal chemistry underlying penicillin allergy have changed over recent years. Although all beta-lactam antibiotics (eg, amoxicillin, cephalosporins) share a beta-lactam core ring structure, it is the chemical side chains (R-group side chains) of beta-lactam antibiotics that are responsible for inducing immediate, IgE-mediated, clinically significant allergic reactions. Thus, in many circumstances, other beta-lactam antibiotics can safely be given to children with a reported penicillin allergy. An excellent review of this subject was published in Pediatrics in 2005 by Michael E. Pichichero, MD, from the University of Rochester Medical Center.

Clinical history and presentation

Certainly, prescribers will wonder about determining the authenticity or accuracy of a reported reaction, and they will ponder if penicillin was truly the cause of the reaction and if it will occur again with repeated administration. In a review published in JAMA earlier this year, Shenoy and colleagues stratified patients’ risk for serious adverse reactions to subsequent drug administration based on their allergy history.

A low-risk history of previous reactions includes:

  • pruritus without rash;
  • gastrointestinal reactions (eg, nausea, diarrhea); and
  • family history.
  • A moderate-risk history includes:
  • urticaria or pruritic rashes; and
  • reactions with features of IgE-mediated symptoms but not anaphylaxis (eg, cutaneous itching, flushing, urticaria and angioedema plus respiratory, cardiovascular or GI symptoms).

A high-risk history includes:

  • anaphylactic symptoms;
  • positive penicillin skin testing;
  • recurrent reactions; and
  • reactions to multiple beta-lactam antibiotics.

The timing of the reaction related to drug administration is additionally important to consider because clinically significant reactions occurring within 1 hour of drug administration are more likely to be IgE mediated. However, serious reactions occurring hours to days later (eg, onset of Stevens-Johnson syndrome) may also be drug related and point to a true allergy. A 2015 review published in the Cleveland Clinic Journal of Medicine recommends asking the following specific questions for a clinical history:

  • Do you remember the details of the reaction?
  • How many years ago did the reaction occur?
  • How was the reaction managed?
  • What was the outcome?
  • Why was your child given penicillin?
  • Has your child tolerated other forms of penicillin since the reaction?

Clinical pediatric studies

Penicillin skin testing was performed in a study that assessed children aged 4 to 18 years (n = 100) who presented to a pediatric ED with symptoms suggestive of a low-risk, IgE-mediated penicillin allergy (rash, itching, diarrhea, vomiting, rhinitis, nausea, cough or family history of allergy). One hundred percent of the children reacted negatively and had their labeled penicillin allergy removed from the medical record.

Deciphering a low-risk history from a moderate-risk history, especially with regard to cutaneous reactions (eg, urticaria or a mild maculopapular rash), can be difficult in the pediatric population, given the frequent infectious illnesses infants and young children experience and the commonality of associated rashes and skin manifestations. Published studies have documented that most individuals identified with a penicillin “allergy” have low-risk histories.

In their review, Shenoy and colleagues stated that amoxicillin can be prescribed to patients with nonallergic symptoms or a family history. They recommended a direct oral amoxicillin challenge under medical observation for other individuals with a low-risk history, which includes pruritus without rash.

Mill and colleagues sought to determine the diagnostic properties of a graded provocation challenge with amoxicillin among children (median age, 1.7 years; n = 818) with a suspected nonsevere allergy history to amoxicillin. The setting was an allergy clinic in Montreal, Canada, in 2016. Among the 818 children evaluated, 770 (94%) tolerated the provocation challenge. Some children (2%) developed immediate reactions (< 1 hour) but all were deemed mild, and 3.8% developed nonimmediate reactions. All children with immediate and nonimmediate reactions tolerated cefixime administration. Children who developed an immediate reaction were skin tested with penicillin at a follow-up visit, and only one child reacted positively. Data from this study are useful, although perhaps limited by the setting. The rate of cross-reactivity between penicillin and aminopenicillins (ampicillin and amoxicillin) is less than 1.3% in the U.S. because penicillin and amoxicillin/ampicillin have different chemical R-group side chains.



Caregiver- and patient-reported penicillin allergies are common, yet many published data indicate that most reported reactions are not true IgE-mediated reactions (with concerns for anaphylaxis upon repeated use), and more likely result from infectious (such as common viral exanthems in children) or other nondrug-related causes. Although the gold standard for determining the propensity for clinically significant penicillin allergies is penicillin skin testing, pediatric providers (ie, providers not specializing in allergy/immunology) can verify a penicillin “allergy” with a detailed clinical history. Although it is not practical to refer every child with a penicillin “allergy” to an allergist/immunologist, many children can benefit from a referral. Children with an unclear history likely would benefit from an allergist/immunologist referral, especially younger children, children with a history of repeated infections and a need for antibiotic prescribing and children with multiple reported drug allergies.

Removing a child’s penicillin allergy provides numerous documented benefits. False penicillin allergies have been demonstrated to result in increased use of broad-spectrum antibiotics (which may not be more effective than narrow-spectrum antibiotics for acute respiratory tract infections), increased hospitalization time, increased risk for additional infections (such as vancomycin-resistant enterococcus, MRSA and Clostridioides difficile), increased medical visits and costs and increased antibiotic-related adverse drug effects. False penicillin allergies may follow children into adulthood and for years beyond.

Disclosure: Bell reports no relevant financial disclosures.