A common question I receive is, “My patient is allergic to penicillin, what can I use to treat this infection?” My first response is generally to ask them to describe their allergic reaction because, as you are aware, taking a detailed history of prior allergic reactions is an integral part of patient evaluation. However, many patients do not recall what the allergic reaction actually was or they were told they were allergic by a parent because it happened when they were a child.
Penicillin allergy is the most commonly reported medication allergy, with 10% to 20% of hospitalized patients reporting this allergy. This self-reported history of penicillin allergy is unreliable in predicting an immediate allergic reaction.
Data suggest that up to 90% of these patients are not truly allergic as determined by allergy skin testing followed by oral penicillin challenge. Penicillin agents continue to be workhorse agents for the treatment of a variety of bacterial infections. Patients who carry the diagnosis of penicillin allergy tend to be treated with broad-spectrum antibiotics, which are often more expensive, more toxic and, possibly, less effective. Fluoroquinolones, clindamycin, vancomycin, aztreonam and carbapenems are frequently used as alternative antibacterials in this population.
Skin testing to confirm allergy
Penicillin skin testing is a safe and reliable method for determining whether a person is at risk for an immediate (immunoglobulin E-mediated) allergic reaction. However, it will not predict non–IgE-mediated reactions, such as interstitial nephritis, Stevens-Johnson syndrome, toxic epidermal necrolysis and others. Penicillin skin testing is a two-step process involving both a prick and intradermal skin test to both the major and minor determinants of penicillin — penicilloyl-polylysine (Pre-Pen, Hollister-Stier Laboratories) and diluted penicillin G, respectively.
After a negative skin test, it also is recommended that patients undergo an oral penicillin dose challenge. The negative predictive value of penicillin skin testing is more than 95%, and when reactions occur after a negative skin test result, they are usually mild. Despite this, most suspected penicillin allergic patients are not skin tested. It has been reported that less than 0.1% of approximately 25 million patients with a penicillin allergy undergo penicillin skin testing in the United States.
To date, there have been several small studies evaluating penicillin skin testing for patients in the hospital, including those in intensive care, ED and preoperative settings. Those performing the skin tests have ranged from trained allergy and immunology specialists, infectious disease specialists, nurses, as well as pharmacists. The key piece is that the person placing the skin test and interpreting the test results must be adequately trained. If the skin tests are not done correctly, the accuracy of the test results will suffer.
Ramzy and colleagues recently published their experience utilizing penicillin skin testing during a 5-month period in a large tertiary care hospital. Infectious disease fellows were responsible for administering the test, as well as interpretation of the results. During this time, they tested 146 adult and pediatric patients claiming to be allergic to penicillin, of whom 145 (99%) tested negative to the skin test. All 145 went on to tolerate the oral challenge and were successfully transitioned to a beta-lactam antibiotic without any reported reactions after 24 hours.
The most common infections of this population were pneumonia (27.4%); urinary tract infections (19.7%); intra-abdominal infections (16.4%); and skin and soft-tissue infections (13.7%). The researchers of this study estimated that the annualized cost savings for their hospital would be approximately $82,000 from altering antibiotic therapy to a beta-lactam regimen.
A pharmacy-managed penicillin skin testing program was reported by Wall and colleagues. Before starting the program, a group of pharmacists underwent training by a board-certified allergist, which included both didactic lectures as well as practice administering the scratch and intradermal tests. Once training was complete, a protocol was developed and approved by the appropriate hospital committees. Twenty-six patients were included in this study. In three of these patients, an IgE-mediated allergy was ruled out after the pharmacist conducted a thorough history, and the remaining 23 went on to complete penicillin skin testing. The skin test was negative in 22 of the 23 patients, with one patient having an indeterminate result. All 26 patients were given penicillin or other beta-lactam antibiotic without any significant adverse reactions being noted.
Del Real and colleagues performed a retrospective study of a combination of both outpatients and hospitalized patients, including ICU patients, with a history of penicillin allergy. Of 596 patients, 25.3% were outpatients, 50.3% were inpatients and 24.3% were ICU patients. The results of the skin testing were that 88.4% of patients were negative, 8.2% tested positive to penicillin and 3.4% of the tests were indeterminate. Of the 527 patients who tested negative to the penicillin skin test, 290 were treated with a beta-lactam antibiotic. Of those given a beta-lactam antibiotic, only five patients (1.7%) went on to develop adverse reactions, which were reported as flushing, rash and urticaria. The negative predictive value for an IgE-mediated event reported in this study was 99.3%.
Any patient with a history of a reaction to a penicillin antibiotic that may have been an IgE-mediated reaction is a candidate for testing, especially if the reaction occurred in the distant past. Contraindications to skin testing include patients with non–IgE-mediated reactions, such as Stevens-Johnson syndrome, toxic epidermal necrolysis or serum sickness. Patients who are known to be extremely hypersensitive to penicillin also should not be tested because testing may reactivate their disease.
The available evidence suggests that performing the test in the hospital setting is a safe and effective way to reduce the use of non-penicillin antibiotics. However, performing skin testing in the outpatient setting before the need for hospitalization may be preferable. Patients are often started on medications upon admission to the hospital that can interfere with skin test results (eg, antihistamines, H2 receptor antagonists, etc), and the logistics are often complicated because the patients may be undergoing other tests and procedures. If testing is performed before the need for hospitalization, beta-lactam therapy may be initiated at the time it’s needed vs. having to utilize alternative antibiotics while awaiting skin testing. Upon completion of the skin test, it is imperative to adequately educate the patients on the results and to correct all allergy records for the patient, particularly if they are found not to have penicillin allergy.
del Real GA. Ann Allergy Asthma Immunol. 2007;98:355-359.
Macy E. J Allergy Clin Immunol. 2014; in press.
Rimawi RH. J Hosp Med. 2013;8:341-345.
Wall GC. Am J Health-Syst Pharm. 2004;61:1271-1275.
For more information:
Jeff Brock, PharmD, MBA, is an infectious disease pharmacy specialist at Mercy Medical Center in Des Moines, Iowa. He can be reached at JBrock@mercydesmoines.org.
Disclosure: Brock reports no relevant financial disclosures.