Navigating the Penicillin Pseudo-Allergy Pandemic


Nine out of ten patients who claim a penicillin allergy don’t actually have one. Here’s how to handle this all-too-common situation.

A whopping 10% of patients in the United States report a penicillin allergy, earning penicillin the dubious distinction of most common drug allergy. However, the good news is that only 1 in 10 of those patients have true, IgE-mediated reactions, which means that only 1% of the population has a true penicillin allergy. [1]

More than 90% of patients with reported penicillin allergy do not have IgE antibodies on skin testing, either because their reaction was improperly labeled as allergic or because they had prior allergy, which resolved over time. [2] Fearing anaphylaxis, emergency providers are understandably hesitant to prescribe beta-lactam antibiotics in patients with reported penicillin allergy and instead may opt for 2nd and 3rd line agents, which are often broader than needed, more expensive, and may contribute to antimicrobial resistance. [3]


Health outcomes may also be worse. A study of hospitalized patients with penicillin allergy showed compared to non penicillin-allergic controls, the penicillin-allergic group had longer hospital stays and developed 23% more cases of C. difficile colitis, 14% more cases of MRSA, and 30% more VRE infections. [4] Yikes!  Here we will review the data on penicillin allergy, cephalosporin cross-reactivity, and provide a framework for managing patients with penicillin listed on their allergy band.


In order to understand what reactions are or are not allergic reactions, we have to discuss hypersensitivity reactions. Type I hypersensitivity, also known as IgE-mediated reactions, are the ones we care the most about as emergency providers. They occur rapidly, usually within minutes to an hour or two of exposure to the offending drug, through mast cell degranulation and release of histamine and other inflammatory mediators. [5] Clinical manifestations include flushing, pruritus, hives, angioedema, bronchospasm, laryngeal edema, and hypotension.


Anaphylaxis, the most severe form of Ig-E mediated allergy, is defined as involvement of two or more organ systems following exposure to the antigen. Anaphylaxis is life-threatening and is treated with immediate intramuscular epinephrine (0.3 mg IM every 3-5 minutes for adults and children over 25 kg, and 0.15 mg IM in children 10-25 kg, 0.01 mg/kg IM in children under 10 kg). [6]  All other types of hypersensitivity reactions do not involve IgE antibodies and are delayed, occurring within 48-72 hours of exposure. [7] Type II hypersensitivity reactions occur when complement-fixing antibodies bind complement and lead to celllysis and tissue injury, as seen in autoimmune hemolytic anemia and anti epidermal antibodies causing skin blistering in pemphigus vulgaris.

Type III hypersensitivity is characterized by immune complex deposition leading to tissue injury, as seen in serum sickness and drug fever. Finally, type IV hypersensitivity reactions involve a local immune and inflammatory reaction as a result of cellular immunity and proliferation of inflammatory cells at the site, seen in contact dermatitis from poison ivy. Type IV reactions are also likely responsible for the morbilliform rashes seen with amoxicillin and sulfonamides. [3]


Penicillin is a member of the class of antibiotics known as beta-lactams, the most commonly prescribed drug class in the United States. Beta-lactams include the –cillins like penicillin and amoxicillin, cephalosporins (cephalexin, ceftriaxone), carbapenems (imipenem, meropenem), and monobactams (aztreonam). The core structure is a beta-lactam ring, and various side chains contribute to different spectrum of activity and duration of action. [2]


Patients may react to allergenic epitopes of the core beta-lactam ring and adjacent ring structures, or they may react to the R side chains of aminopenicillins (i.e. amoxicillin and ampicillin), which are not present in penicillin. These R side chains are also shared by certain cephalosporins.

While penicillin is the most commonly reported drug allergy, true IgE-mediated reactions to penicillin and other beta-lactams are quite rare. [2,5] Patients often report an allergy history that is vague or unknown: “Mom told me I was allergic.” [8] They may also have experienced adverse effects like gastrointestinal upset, diarrhea and headache, which were falsely labeled as allergy. Up to 10% of patients may develop a nonpruritic, nonurticarial rash after exposure to aminopenicillins, such as amoxicillin, which is often incorrectly labeled as allergy. [5]

Nearly all patients with Epstein-Barr virus (infectious mononucleosis) develop a morbilliform rash when exposed to amoxicillin. [9] While these rashes may be undesirable, they are not true allergic reactions, and, importantly, are not associated with more severe reactions like anaphylaxis on subsequent exposure to the drug. [5] When patients experience a nonpruritic, nonurticarial rash after exposure to amoxicillin, it should be labeled as an adverse event, not an allergic reaction, and does not preclude the patient from receiving the drug in the future.

Even hives, which are a classic symptom of IgE-mediated reactions, can occur from other causes but be falsely attributed to the drug. For example, hives are a common manifestation of viral illnesses in children, and patients may be inappropriately labeled as allergic, when in fact their hives were due an underlying viral illness which was inappropriately treated with antibiotics. [8]

When assessing a penicillin allergy, ask patients about the symptoms they had. How soon after taking the medication did symptoms develop? Was the patient taking other medications at the same time, confounding the picture? Has the patient been able to take other beta-lactam antibiotics?

Symptoms that raise concern for IgE-mediated reaction include reactions that occurred immediately or within minutes to a few hours, including hives, angioedema affecting the face, extremities, oropharynx, or genitalia, and respiratory symptoms like wheezing or shortness of breath. A patient who developed facial swelling minutes after a penicillin shot is concerning for true IgE-mediated allergy; the child with a maculopapular rash 7 days after starting amoxicillin is not.

The time since the reported allergic reaction is also important, since penicillin-specific IgE antibodies wane over time. Approximately 50% of patients with penicillin allergy lose sensitivity after 5 years of the initial reaction, and 80% lose sensitivity by 10 years.


Early studies of cephalosporin allergy report high cross-reactivity rates with penicillin, as high as 41%. However, these studies were performed with crude cephalosporin preparations that were often contaminated with penicillins, and results overestimate the true incidence. [5,9] More recent studies suggest cross-reactivity rates between penicillin and cephalosporins are much lower, from 0 – 6%. [5]

While it was initially thought that the shared beta-lactam ring of penicillins and cephalosporins was responsible for their cross-reactivity, it was discovered that the degradation products of the two compounds are very different. Instead, experts believe that cross reactivity between the two compounds is more likely due to similarities in the Rside chains. [5,10] If a patient with confirmed penicillin allergy is given a cephalosporin that has a distinct side chain, the patient is extremely unlikely to have a reaction to the cephalosporin. The American Academy of Pediatrics states that the likelihood of a patient with confirmed penicillin allergy reacting to a cephalosporin with a different side chain is similar to the general rate of allergic reactions in a non-penicillin allergic patients. [10] This means that if a patient has a reported allergy to amoxicillin, he or she can still safely receive ceftriaxone, since the two compounds have distinct side chains. However, it is still wise to avoid cephalosporins in patients with a reported severe allergy to beta lactams, such as anaphylaxis.


The cross-reactivity of penicillins and carbapenems is thought to be very low. Well-designed prospective studies of patients with confirmed penicillin allergy report cross-reactivity with carbapenems to be less than 1%. [2,5] In the case of monobactams, cross-reactivity between penicillins and aztreonam is essential negligible, though there is a possibility of interaction between ceftazidime and aztreonam due to similarity in side changes. [2]


Patients with reported penicillin allergy can be referred for skin testing to confirm penicillin hypersensitivity. Penicillin skin testing involve a scratch test followed by injection of small wheals containing penicillin G and benzylpenicilloyl, the main determinant of penicillin allergy, as well as histamine and saline for positive and negative controls, respectively. [11] The test takes 15-20 minutes, and a positive test is induration at the site of the injection.

A negative test is usually followed by an oral amoxicillin challenge. [12] While penicillin skin testing is not available in most emergency departments, some hospitals utilize them to improve antibiotic stewardship, which increases antimicrobial options available for use if penicillin allergy is ruled out. A study of orthopedic patients receiving perioperative antibiotics at Mayo Clinic showed that penicillin skin testing reduced vancomycin use from 30% to 11%. [13] There may also be a potential for costs savings given shorter hospital length of stay and fewer complications in patients not allergic to-penicillin.


Let’s take a clinical example. You diagnose a 6-year-old female with pyelonephritis, and plan to administer ceftriaxone. The patient has a reported allergy to amoxicillin, and upon further questioning of parents, you learn that the patient developed a maculopapular rash 3 days after starting amoxicillin. The timing and type of rash suggests this was NOT an IgE-mediated reaction, and she likely does not have a true amoxicillin allergy. Furthermore, you astutely review Chart 1, and deduce that amoxicillin and ceftriaxone have distinct side chains, so that even if the patient did have an amoxicillin allergy, she would be very unlikely to react to ceftriaxone.


In summary, penicillin allergy is commonly reported but largely overestimated. Questioning your patient about the type of reaction that occurred can help elucidate whether the reaction was an IgE-mediated reaction. Referral for penicillin skin testing is a good option that can lead to better antibiotic stewardship, decreased costs, and better outcomes for your patient. Many cephalosporins can be safely given to patients with penicillin allergy provided they do not share a common side chain.


  1. Joint Task Force on Practice Parameters representing the American Academy of Allergy, Asthma and Immunology; American College of Allergy, Asthma and Immunology; Joint Council of Allergy, Asthma and Immunology. Drug allergy: an updated practice parameter. Ann Allergy, Asthma, Immunol. 2010;105(4):259-273.
  2. Blumenthal K, Solensky R. Allergy evaluation for immediate penicillin allergy: Skin test-hased diagnostic strategies and cross-reactivity with other beta-lactam antibiotics.
  3. Drug Allergy: An Updated Practice Parameter. Ann Allergy, Asthma Immunol. 2010;105(4):259-273.e78. doi:10.1016/j.anai.2010.08.002.
  4. Macy E, Contreras R. Health care use and serious infection prevalence associated with penicillin “allergy” in hospitalized patients: A cohort study. J Allergy Clin Immunol. 2014;133(3):790-796. doi:10.1016/j.jaci.2013.09.021.
  5. Terico AT, Gallagher JC. Beta-Lactam Hypersensitivity and Cross-Reactivity. J Pharm Pract. 2014;27(6):530-544. doi:10.1177/0897190014546109.
  6. Lexicomp: Epinephrine: Drug information.
  7. Lagacé-Wiens P, Rubinstein E. Adverse reactions to β-lactam antimicrobials. Expert Opin Drug Saf. 2012;11(3):381-399. doi:10.1517/14740338.2012.643866.
  8. Vyles D, Chiu A, Simpson P, Nimmer M, Adams J, Brousseau DC. Parent-Reported Penicillin Allergy Symptoms in the Pediatric Emergency Department. Acad Pediatr. 2017;17(3):251-255. doi:10.1016/j.acap.2016.11.004.
  9. Har D, Solensky R. Penicillin and Beta-Lactam Hypersensitivity. Immunol Allergy Clin North Am. 2017;37(4):643-662. doi:10.1016/j.iac.2017.07.001.
  10. Pichichero ME. A Review of Evidence Supporting the American Academy of Pediatrics Recommendation for Prescribing Cephalosporin Antibiotics for Penicillin-Allergic Patients. Pediatrics. 2005;115(4):1048-1057. doi:10.1542/peds.2004-1276.
  11. PRE-PEN ® [package insert]Plainville, CT: ALK-ABELLO, Inc; 2013.
  12. Macy E, Ngor EW. Safely Diagnosing Clinically Significant Penicillin Allergy Using Only Penicilloyl-Poly-Lysine, Penicillin, and Oral Amoxicillin. J Allergy Clin Immunol Pract. 2013;1(3):258-263. doi:10.1016/j.jaip.2013.02.002.
  13. LI J. Reduction of vancomycin use in orthopedic patients with a history of antibiotic allergy. J Allergy Clin Immunol. 2000;105(1):S339. doi:10.1016/S0091-6749(00)91425-8.


Karen Serrano, MD is an assistant professor in the department of emergency medicine at the University of North Carolina.

Dr. Shenvi is an assistant professor in the department of emergency medicine at the University of North Carolina. She authors RX Pad each month in EPM.


  1. John D. Dunn MD JD on

    this was a very helpful article–thanks a lot.

    now would you please educate our pharmacists to think the way you do?

  2. Chip Pettigrew MD FACEP on

    Your article was great! However, I was very interested to read some of your cited references. They did not accompany the article in EP Monthly, but there was a note at the bottom of your article stating, “References online at .”
    Actually, there are NO references to be found online for your article at .
    I tried to call EP Monthly at the phone number listed in the magazine (410) 626-1053 to ask where to get these references. I received an automated message that this is no longer a working number.
    I am left wondering what criteria the Sydney Award committee used for declaring that EP Monthly is “the best long-form journalism of the year.” Seems pretty sloppy to me that they would publish articles with no access to cited references and that they would have no good way to confirm information in real time.

Leave A Reply