Impact of a Pharmacist-Driven Penicillin Allergy Skin Testing Protocol on Antimicrobial Stewardship in a Tertiary Care Hospital

Abstract

To the Editor: Approximately 10% of the general population and up to 20% of hospitalized patients report an allergy to penicillin antibiotics.1,2 Of these patients, only about 10% experience a true IgE-mediated hypersensitivity.1,2 IgE antibodies also wane over time, up to an 80% decrease every 10 years. Therefore, a patient who experienced an IgE-mediated reaction over 10 years ago might not experience that same reaction today.1,2 Patients with a reported penicillin allergy have a higher incidence of C. diff (Clostridium difficile), MRSA (methicillin-resistant Staphylococcus aureus), VRE (vancomycin-resistant enterococci), and deviation from the standard of care.2,3 Penicillin skin testing can determine with a 97% to 99% negative predictive value if a patient does not have an IgE-mediated penicillin allergy with a negative result.2 This simple test offers the possibility of removing a label that otherwise, if untrue, would negatively impact patient care. We conducted a single-center retrospective chart review at Lutheran Hospital in Fort Wayne, Indiana, before and after the initiation of a protocol that incorporated pharmacists in determining the appropriateness of an ordered penicillin skin test. Patients were included if they were admitted to the hospital and had a reported penicillin allergy with a penicillin allergy skin test ordered during the time before the protocol (December 1, 2016 to May 31, 2017) or after (November 3, 2017 to March 3, 2018). Exclusion criteria included people under the age of 18 years. Information collected included allergy and reaction reported, outcome and adverse effects of penicillin skin test, antibiotics used before and after penicillin skin test, and number of penicillin skin tests ordered but not performed due to pharmacist intervention. We identified 19 penicillin skin tests that were ordered and were performed before the protocol was initiated and 14 penicillin skin tests ordered (of those 8 were performed) after the protocol initiation. After protocol initiation, pharmacists’ recommendations on the appropriateness of performing tests and/or antibiotic choice were accepted in 12 of 14 cases. Antibiotics were changed as a result of ordering the penicillin skin test in 7 of 19 patients before initiation of the protocol and in 10 of 14 patients after the initiation of the protocol. In the before protocol group, reasons for allergy included hives/rash, unknown or not documented, throat swelling/difficulty breathing, numbness in extremities, or nausea/vomiting/diarrhea (N/V/D). These results can be seen in Figure 1. Before the protocol was initiated (before pharmacist became involved in this care process), making no change based on a negative result was the most common response to testing. This result, along with other changes, can be seen in Figure 2. In the group of patients assessed after the protocol was implemented, the most common reason for a documented allergy was unknown or undocumented. The full results of reasons for allergy can be seen in Figure 3. To determine the impact of penicillin testing paired with pharmacist intervention, we looked at patients whose test was negative and what was done with therapy after that result if the pharmacist’s intervention was accepted, if it was not taken, or if no intervention was given. The decision to change from a non-β lactam to a β lactam was highest when a pharmacist’s recommendation was accepted. This result can be seen along with the others in Figure 4. Prior to protocol initiation, pharmacists had minimal involvement in penicillin allergy skin testing. Of all the subjects reviewed, 30% had reported a reaction to penicillin that was unknown or not documented. This provides an opportunity for pharmacists to intervene and investigate further with thorough medical chart reviews and patient interviews. Also, pharmacists’ recommendations were accepted 86% of the time which represents the already established respect between pharmacists and prescribers. After protocol initiation with pharmacist involvement in the penicillin skin testing process, 43% of tests ordered were not performed due to the inclusion/exclusion criteria not being met and 100% of subjects with a negative test had antibiotic changes. Before protocol initiation, all of the tests were performed and only 42% of patients with a negative result had antibiotics changed. Further study is warranted to evaluate the 897072 HPXXXX10.1177/0018578719897072Hospital PharmacyKurtz et al letter2019