Key Points:

• The risk that patients with a history of penicillin allergy will experience a hypersensitivity reaction if they are rechallenged with penicillin is low; however, that finding does not justify the fact that almost half of such patients are given a second penicillin prescription. • Patients with a history of penicillin allergy have a markedly increased risk of cross-reactivity to carbapenems.

PHILADELPHIA—Two recent studies provide good news and bad news about penicillin allergy. It may be less common than had been thought—even among patients with a history of hypersensitivity to that drug class.[1] However, a surprisingly high number of patients with such a history are given a second penicillin prescription. Furthermore, patients who have had an allergic reaction to penicillin have an increased risk of cross-reactivity to carbapenems.[2]

SAFETY OF PENICILLIN REPRESCRIPTION

In a retrospective study, nearly half of the patients who experienced an allergic or “allergic-like” event after being given a penicillin later received another prescription for a drug in that class (usually, amoxicillin). In these patients, the risk of a second event was about 11 times higher than it was in the patients who had not reacted to the penicillin initially, noted lead author Andrea J. Apter, MD, MSc, an Associate Professor of Pulmonary, Allergy, and Critical Care Medicine at the University of Pennsylvania in Philadelphia.

The study also found that the overall risk of a hypersensitivity reaction to penicillin was small: Among the more than three million patients in this study, fewer than 1% experienced an allergic or allergic-like event following initial administration of a drug in that class. That rate is considerably lower than the 10% to 15% prevalence previously reported.

The reason for this difference may lie in the populations studied. Earlier prevalence estimates were based predominantly on small case series of inpatients. Yet most people who are given penicillins are outpatients. To better quantify the rate of penicillin allergy in the general population, Dr. Apter’s group used a large database of primary care medical records to identify 3,375,162 patients who had been given a penicillin prescription at least once during a 14-year period. Of these, 6,212 (0.18%) experienced an allergic or allergic-like event after the initial prescription. (Allergic reactions included anaphylaxis, urticaria, angioedema, erythema multiforme, and laryngeal spasm. Bronchospasm, asthma exacerbations, and eczema were considered allergic-like events.)

Almost half—3,014 (48.5%)—of the patients who had had an initial hypersensitivity reaction were given a second penicillin prescription at least 60 days after the first one. Fifty-seven (1.9%) of these patients experienced a second reaction to the antibiotic.

A second penicillin prescription was also given to more than two million of the patients who had not suffered a hypersensitivity reaction initially. Allergic or allergic-like events developed in 3,452 (0.17%) of these patients.

Few of the hypersensitivity reactions (initial or secondary) were serious; none resulted in death. Urticaria accounted for about 75% of the events.

Life-threatening reactions, such as anaphylaxis and toxic epidermal necrolysis (TEN), were rare. Anaphylaxis accounted for only 16 (0.5%) of the events that occurred after the first penicillin prescription and 32 (0.9%) of those that followed the second prescription. Thus, the prevalence of anaphylaxis among all patients given a penicillin initially was 0.0005%; among those receiving a second prescription, it was 0.002%. TEN was even less common, affecting only six (0.2%) of the patients who had experienced hypersensitivity reactions initially and 11 (0.3%) of those who reacted to the second prescription. Furthermore, of the 22 patients who had suffered anaphylaxis or TEN initially and were then given penicillin again, only one patient experienced a second severe reaction.

However, these results must be interpreted with caution in light of the retrospective nature of the study. Patients at risk for severe reactions may have been underrepresented, and not all of the reactions patients experienced may have been listed in the database. Furthermore, the results apply only to outpatients given oral penicillins; the risks associated with intravenous administration to inpatients may be much higher.

Until these issues are clarified, Dr. Apter and colleagues urge restraint; their results should not lull physicians into thinking that administration of a penicillin to patients who report a history of allergy to that drug class is safe. Instead, physicians should be more proactive in identifying such patients. “As always, the history is extremely important,” Dr. Apter concluded.

CARBAPENEM CROSS-REACTIVITY

A separate study, also retrospective, sought to quantify the risk of cross-reactivity to carbapenems among patients with documented or self-reported penicillin allergy. It found that the risk is more than five times higher in patients with a history of penicillin allergy than in those without such a history.

“That is important to know,” emphasized Randolph E. Regal, PharmD, one of the investigators, “because the risk of carbapenem cross-reactivity in penicillin-allergic patients has not been addressed in the literature to any great degree.” Dr. Regal is a Clinical Pharmacist in Adult Internal Medicine and Infectious Diseases at the University of Michigan Health System in Ann Arbor.

This study included 211 inpatients; 100 of them (group 1) had a history of penicillin allergies, and the other 111 (group 2) did not. Slightly more than half of the patients in both groups had been treated with meropenem; the remainder had received imipenem.

The incidence of hypersensitivity reactions was 11% in group 1 and 2.7% in group 2. However, the mean time between when the antibiotic was given and when the reaction developed was 6.1 days in group 1 and 3.7 days in group 2. After adjusting their analysis for the patients’ age, sex, and prednisone use, the investigators found that a history of penicillin allergy increased the risk of cross-reactivity by a factor of 5.2.

The specific carbapenem administered did not appear to influence the risk of such a reaction, but lead author William A. Prescott, Jr, PharmD, a resident in pediatrics and pharmacotherapy at the University of Michigan Health System, acknowledged that “the study was not statistically powered to detect any differences in the odds of cross-reactivity with imipenem and meropenem.”

Based on their findings, the investigators advise physicians to be cautious when administering a carbapenem to patients with a history of penicillin allergy, particularly if the allergy has been documented by a health care professional. Cephalosporins should be used cautiously as well.

Penicillins, carbapenems, and cephalosporins are alike in that they have a characteristic bicyclic core structure, which is believed to play a large role in β-lactam hypersensitivity. “There is a perception that [patients] are safer on cephalosporins than they are on carbapenems,” Dr. Regal explained. “However, the odds of an allergic reaction to a cephalosporin in penicillin-allergic patients are about the same as the odds that we found for carbapenem cross-allergenicity in these patients.”

Thus, the most prudent course, Dr. Prescott suggested, may be to consider a different type of antibiotic, such as a fluoroquinolone, for patients with a history of penicillin sensitivity.

—Timothy Begany

References
1. Apter AJ, Kinman JL, Bilker WB, et al. Represcription of penicillin after allergic-like events. J Allergy Clin Immunol. 2004;113:764-770.
2. Prescott WA, DePestel DD, Ellis JJ, Regal RE. Incidence of carbapenem-associated allergic-type reactions among patients with versus patients without a reported penicillin allergy. Clin Infect Dis. 2004;38:1102-1107.