Patricia Takach

Evaluations of Patients with a History of Antibiotic Allergy

Beta-lactam antibiotics are among the most important causes of allergic reactions. This group includes penicillin and its semisynthetic chemical derivatives (such as ampicillin and amoxicillin) and other beta-lactam antibiotics including cephalosporins, carbapenems, and monobactams. Approximately 10% of patients report a history of penicillin allergy; however, up to 90% of these patients may be able to tolerate penicillin after a complete evaluation (Figure 32.1). A carefully obtained history of previous drug reactions to this class of antibiotics is critical to the allergy evaluation. An accurate history may be challenging in the ICU setting where patients are frequently sedated or intubated and may be receiving multiple medications. Family members or the patient’s primary care physician as well as outpatient electronic medical records may be valuable sources of the relevant history. Inquiry should be obtained for the name of the drug, time when previous exposure occurred, reason for treatment, timing of previous reaction (time elapsed from drug administration to the occurrence of adverse events), characteristics of the reaction (organ systems involved), medications in use at the time of the reaction, management required for treatment of the reaction, subsequent exposure to the drug, similar symptoms in absence of the drug (such as chronic urticaria), and whether the patient had an underlying condition that favored a reaction to a medication. For example, patients with Epstein Barr virus infections who are given ampicillin tend to develop a rash that looks like an allergic reaction but is not immunologically mediated. Information is then collected regarding all the current medications, both prescription and nonprescription, and similarly reviewed. Physical examination and laboratory studies to assess for organ system involvement are also crucial to the evaluation.

Allergic drug reactions may be classified according to the temporal relationship of onset of symptoms that include immediate (minutes to an hour), accelerated (an hour to 3 days), and delayed (beyond 3 days). If the initial reaction is immediate (i.e., occurring within 30 to 60 minutes of the drug administration), the patient is considered to be at high risk for a potentially life-threatening immediate hypersensitivity reaction such as anaphylaxis. Anaphylaxis is an immediate, immunologically mediated, systemic reaction to antigen and is characterized by smooth muscle contraction and capillary dilatation resulting from the release of mediators from mast cells and basophils.

The time interval since the previous ADR is also significant: ADRs occurring within the last year are more likely to recur on challenge with the antibiotic than are reactions that took place more than 10 years in the past. Patients are at intermediate risk for a life-threatening allergic reaction if they have only a remote (> 10 years prior) history of anaphylaxis.

Skin Testing for Beta-Lactam Antibiotics

Skin testing is the most reliable method of determining if the patient with a prior allergic history to beta-lactam antibiotics is truly at risk for an immediate hypersensitivity reaction to penicillin or other beta-lactam antibiotics. This procedure is performed using reagents containing the major antigenic determinant and penicillin G. The major determinant, benzylpenicilloyl poly-L-lysine (PPL), is available commercially as Pre-Pen. Approximately 95% of penicillin is converted in the body to the major antigenic determinant, and the remainder includes the minor antigenic determinants. Minor determinants mixture (MDM) of penicillin is not commercially available in the United States at this time. Penicillin challenges, however, of subjects with negative skin test responses to PPL and penicillin G have similar reaction rates compared with those in subjects with negative skin test responses to the full set of major and minor penicillin determinants. A less common antigen includes the R-group side chain found on beta-lactams. An example would be a selective allergy to amoxicillin but tolerance of other penicillins. Antihistamines must be avoided for 1 week prior to skin testing. Approximately 10% of patients with a history of a previous allergic reaction to penicillin have positive skin test results to the drug. Patients with a positive skin test result detect penicillin-specific IgE antibody with a wheal and flare reaction to the major determinant or penicillin G compared to the control reagents. If testing is positive, then treatment with desensitization or “induction of tolerance” is required.

Predictive Value of Skin Testing

The negative predictive value of a negative skin test to both major and minor determinants, representing the percent of patients with negative skin test results who will tolerate beta-lactam therapy without an immediate allergic reaction, is 97%, with 1% to 3% of patients with negative skin test responses (with both major and minor determinants) experiencing mild and self-limiting reactions on being challenged with the drug.

In addition, patients with negative skin tests who do react on challenge with a penicillin tend to have mild reactions. However, negative penicillin skin test results have no predictive value for the development of non-IgE-mediated reactions such as serum sickness, Stevens-Johnson syndrome, hemolytic anemia, maculopapular rashes, drug fever, or interstitial nephritis. Patients may also experience urticarial eruptions late in the course of therapy despite an initial negative skin test result.

The positive predictive value of the penicillin skin test (i.e., the number of patients with positive skin tests who will experience anaphylactic reactions on receiving beta-lactam antibiotics) is between 40% and 100%.

Indications for Skin Testing

Skin tests are indicated for any patient with a history of penicillin allergy who requires a beta-lactam drug to treat a life-threatening condition when no other alternative antibiotic is appropriate. Skin tests should be avoided in those patients at high risk for an anaphylactic reaction as determined by history. These patients should be desensitized to penicillin if the drug must be given. Contraindications to skin testing include the lack of a suitable site to perform the test (e.g., secondary to rash), the use of antihistamines, the presence of dermatographism (may cause false-positive reaction), or inadequate skin test controls.

Skin test results to penicillin may remain negative up to 6 weeks after an allergic reaction to beta-lactam antibiotics. Skin testing should, therefore, be repeated after this interval if it is initially negative and the test is clinically indicated. A history of penicillin-induced exfoliative dermatitis (i.e., Stevens-Johnson syndrome or toxic epidermal necrolysis [see Chapter 43]) is an absolute contraindication to skin testing as well as to the administration of beta-lactam antibiotics. The approach to a patient with a history of a beta-lactam allergy is summarized in Figure 32.1.

Although most investigators feel that penicillin skin tests are predictive of reactions to other semi-synthetic penicillins, controversy does exist. It is recommended that patients who have a history of a recent and severe reaction to a semisynthetic penicillin (i.e., they are at high risk for an anaphylactic reaction) continue to avoid the specific antibiotic that caused the reaction, even if penicillin skin test results are negative.

The risk of having an adverse reaction to the skin test reagents is less than 1%, but systemic and fatal reactions occur. The risk of becoming sensitized to beta-lactam antibiotics as a result of penicillin skin testing is also less than 1%. Resensitization after oral treatment with penicillin is rare, and repeat skin testing is not required in patients who have tolerated 1 or more oral courses of penicillin. Resensitization after high-dose parental penicillin may be more likely. Retesting may be considered in individuals with recent or particularly severe previous reactions.

Evaluation of Allergies to Cephalosporins and Other Non–Beta-Lactam Antibiotics

Patients with a positive penicillin skin test response to penicillin have a slightly increased risk (~2%) of reacting to cephalosporins, and some of these reactions may be anaphylactic reactions. A group of studies evaluated patients with positive penicillin skin tests (PPL, penicillin G, or MDM) who were challenged with cephalosporins. The overall reaction rate was 3.4%; if the analysis was limited to studies published after 1980 when cephalosporins were no longer contaminated with penicillin, the reaction rate was reduced to 2%.

Therefore, depending on the reaction history, these patients should be administered cephalosporin through graded challenge or desensitization (induction of tolerance procedure). If a patient has a history of a cephalosporin allergy but a negative penicillin skin test result, one should consider a penicillin, not a cephalosporin (Figure 32.E1).

Allergic cross-reactivity between shared R-group side chains should also be considered. Amoxicillin and certain cephalosporins share identical R-group side chains (see Tables 32.E1 and 32.E2). Research has shown that 12% to 38% of patients proved to be selectively allergic to amoxicillin reacted to cefadroxil yet were able to receive penicillin. Therefore, patients with an amoxicillin or an ampicillin allergy should avoid cephalosporins that share identical R-group side chains or undergo desensitization of these cephalosporins.

No clinically significant cross-reactivity among the monobactam, aztreonam, and penicillin has been demonstrated. Thus, patients with positive skin test to penicillin can be given this drug safely. However, aztreonam has been known to cross-react with ceftazidime, so patients allergic to either antibiotic should avoid both. The current Joint Task Force on Drug Allergy: Updated Practice Parameters 2010 report limited data indicating a lack of significant allergic cross-reactivity between penicillin and carbapenems. These parameters noted that penicillin skin test–negative patients may receive carbapenems, whereas penicillin skin test–positive patients and patients with a history of penicillin allergy who do not undergo skin testing should receive carbapenems via graded challenge.

Reliable skin tests to other classes of antibiotics, such as vancomycin or sulfonamide-type drugs, are not commercially available. However, skin testing with nonirritating concentrations of the drug may provide useful information (see Table 32.E3). This study collected data on healthy subjects without drug allergies to determine baseline nonirritating intradermal skin test concentrations for 15 commonly prescribed parenteral antibiotics. Serial 10-fold dilutions were prepared and tested in these individuals to determine the lowest dilution that did not elicit an irritant response. This concentration was then termed the nonirritating concentration (NIC). Under these circumstances, a positive skin test using the NIC of a drug suggests that the patient has drug-specific IgE antibodies and, for that reason, is at risk for a reaction and should receive an alternate, non-cross-reacting antibiotic or undergo induction of tolerance (also called desensitization). On the other hand, a negative NIC skin test result does not rule out the presence of drug-specific IgE antibodies because it is possible that a drug metabolite not present in the test reagent may be the relevant allergen.

Vancomycin may cause non-IgE-mediated histamine release and elicit immediate cutaneous erythema, flushing, and pruritus (i.e., “red man syndrome”). Slowing the rate of infusion and premedicating with histamine1 receptor antihistamines can prevent red man syndrome. Vancomycin rarely causes IgE-mediated reactions, but anaphylaxis associated with vancomycin has been reported.

A detailed history is always critical for the appropriate evaluation of allergic reactions to medications. The specific drug in question should be avoided unless life-threatening infections demand its use and an induction of tolerance procedure (also called desensitization) would then be necessary. In situations where there is a definite need for a particular medication and no alternative medication is available, induction of tolerance should be considered. This procedure involves the administration of incremental doses of the drug, which temporarily allow tolerance to the drug. As in skin testing, induction of drug tolerance should almost never be performed if the reaction history includes a severe non-IgE-mediated reaction such as Stevens-Johnson syndrome (SJS); toxic epidermal necrolysis (TEN); drug rash, eosinophilia, and systemic symptoms (DRESS); hepatitis; or hemolytic anemia.