Diagnosis of penicillin allergy by means of Phadebas RAST® penicilloyl G and V and skin tests

Fifty patients with suspected allergy to penicillin were tested. Skin tests were done with Na-penicillin G and penicilloyl-polylysin. Specific IgE antibody assays were done with penicilloyl G and V conjugates by means of RAST. The overall agreement between skin test and RAST results was 87%, borderline cases not included. In one case, skin tests were positive to penicillamine only, while RAST for penicilloyl G and V both proved to be positive. One case of penicillin allergy could be diagnosed in vitro post mortem only. Two cases of Hoigné syndrome showed no evidence of allergy. Patterns of skin manifestations varied but urticaria was the most commonly seen feature. Twenty patients without adverse reactions to penicillin treatment and seven patients who had not received penicillin over the last 10 years served as controls. None of them were positive in either skin tests or RAST. Two of our twenty control patients developed penicillin allergy during the study. Both showed positive RAST results.     

IgE antibodies against penicillin as determined by Phadebas RAST

Serum samples from eighty-one patients with suspected penicillin allergy were investigated with Phadebas RAST using the penicillin derivatives Benzylpenicilloyl-human serum albumin (PBO-HSA) and Phenoxymethylpenicilloyl-human serum albumin (PMPO-HSA) and the results were compared with skin test results and clinical data. Of the sixty-one patients who had anaphylactic shock and/or urticaria as a possible consequence of penicillin administration, reagins against PBO-HSA and PMPO-HSA could be detected in thirty-four cases (56%). Five per cent of these patients, with positive RAST results, showed negative skin tests; in the other 95% both RAST and skin tests were positive. All, except eight, of the RAST-negative patients had had their adverse reactions at least 2 years prior to the blood sampling and in some of these cases skin tests were also negative. RAST and provocation test results agreed in 80% of the cases where exposition was performed. It is concluded that the RAST technique is a valuable diagnostic tool for the detection of immediate type hypersensitivity to penicillin.     

Detection of penicillin allergy of the immediate type by radioimmunoassay of reagins (IgE) to penicilloyl conjugates

Reaginic antibodies (IgE) to penicilloyl were detected by an in vitro method, a radioimmunological technique (RAST), in the sera of nine patients out of eleven with a recent history of penicillin hypersensitivity of the immediate type. Provocation tests in two patients with negative RAST were negative as well as in twelve patients having had a possible penicillin hypersensitivity reaction 7 months to 14 years ago. Results from skin tests and RAST agreed. Skin tests and RAST reactions were negative for thirty-two patients with a history of penicillin reactions of the delayed type. The results of this study indicate that the RAST is a valuable alternative to the more dangerous intracutaneous test with penicilloyl-polylysine for-detection of penicillin hypersensitivity of the immediate type.     

Skin test for diagnosis of penicillin allergy--current status

The skin test for evaluating allergy to penicillin is reviewed. The reagents, penicillin, penicilloylpolylysine, benzyl penicilloate, and benzyl penilloate provide a safe and effective skin test for screening out the likelihood of severe allergic reactions to penicillin. The skin test gives much more accurate information than the patient's history and will enable many patients to receive penicillin despite a past history of allergy to the drug. Recent developments in the standardization and stabilization of the minor determinant mixture should result in making this reagent more available.     

Elective penicillin skin testing in a pediatric outpatient setting.

BACKGROUND: Adverse reactions associated with penicillin-type antibiotics are common in pediatric practice, leading to the subsequent unnecessary use of alternative antibiotics. IgE-mediated penicillin allergy represents only a fraction of these adverse reactions. OBJECTIVES: To examine (1) the trend of penicillin skin test reactivity during a recent 10-year interval, (2) the relative distribution of specific reagents related to a positive skin test result, and (3) skin test reactivity as a function of reaction history. METHODS: Penicillin testing using 3 reagents--benzylpenicilloyl polylysine, penicillin G, and sodium penicilloate (penicillin A)--was conducted in a prospective study of 359 consecutive patients referred to an outpatient pediatric allergy clinic between January 1, 1993, and May 31, 2003. We also retrospectively reviewed penicillin skin test results for 562 children previously tested between January 1, 1979, and December 31, 1992. RESULTS: Between 1993 and 2003, the prevalence of penicillin skin test sensitivity markedly declined. Of all the positive skin test results between 1979 and 2002, either penicillin G or sodium penicilloate or both identified 34%, with sodium penicilloate alone responsible for 8.5%. The rate of positive skin test reactions was not significantly different between patients with vs without a history of suggestive IgE-mediated reactions. CONCLUSIONS: A marked decline in penicillin skin test sensitivity in the pediatric age group is identified. The minor determinant reagents penicillin G and sodium penicilloate are both necessary for determining potential penicillin allergy. Relating history alone to potential penicillin sensitivity is unreliable in predicting the presence or absence of a positive skin test result.     

Determinations of Penicilloyl Specific IgE Antibodies for the Evaluation of Hypersensitivity against Penicillin

Results from skin tests and radioallergosorbent test (penicilloyl G and penicilloyl V) were compared in 76 patients with suspected penicillin hypersensitivity and in 22 control subjects without such clinical symptoms. The test results were also correlated to the clinical history of adverse reactions and to the levels of total serum IgE. The overall agreement between skin test and RAST results was 89% in the patient group. Two of the control subjects without symptoms were professionally employed with penicillin production; one of them had circulating IgE-antibodies and the other exhibited positive skin tests. The skin tests were more frequently positive than the RAST test. By means of skin titration some of the skin tests were demonstrated to be false positive. The discrepancy can also be explained by the fact that skin reactions also occur in other situations than IgE-mediated adverse reactions. Circulating IgE-antibodies were not found in any of 23 cases where adverse reactions appeared later than 24 h after the latest penicillin administration. It is concluded that the measurement of circulating IgE-antibodies is a valuable aid in the diagnosis of penicillin allergy.     

Anaphylaxis to amoxycillin but good tolerance for benzyl penicillin

Three patients are reported on who suffered anaphylactic reactions after amoxycillin (AX) treatment and challenge but tolerated benzylpenicillin (BP) parenterally and orally. Two of the three patients had positive skin tests and RAST to AX reagents but negative responses to benzyl penicilloyl (BPO) specific skin tests and RAST and the minor determinant mixture (MDM) skin test reagent. The third case was negative to all skin tests and RAST. RAST and RAST inhibition on the two positive sera suggest that the response is related to the acyl side chain of AX.     

Allergy to penicillin unsuccessfully treated with a haptenic inhibitor (benzyl-penicilloyl-N2-formil-lysine; BPO-Flys)

In a patient with a past history of allergy to penicillin and requiring treatment, skin tests with penicillin and penicilloyl-polylysine (PPL) were positive, and a penicilloyl RAST was strongly positive, although the Prausnitz-Kiistner test to penicillin and PPL were negative. The hapten BPO-Flys was administered together with penicillin to try to prevent reactions. An anaphylactic reaction occurred on the fifth day, and treatment was stopped. Serum total IgE values increased markedly after the reaction and PK titres to penicillin and PPL reached values of 1/256. The skin test reaction to PPL was negative the day after the clinical reaction, but became strongly positive again a few days later. The penicilloyl RAST remained strongly positive throughout. The authors consider that there was hypersensitivity to a penicillin metabolite other than the penicilloyl group, e.g. to minor determinants, and for this reason the hapten inhibitor failed.     

Radioallergosorbent test with conjugates specific for 'minor' haptenic determinants in the diagnosis of IgE-mediated penicillin allergy in man

Penicillamine-polylysine, benzylpenicillanyl-human serum albumin and ampicillin-polymers were coupled to CNBr-activated paper discs and used in addition to penicilloyl G and penicilloyl V discs in RAST investigations. Sera from sixty patients with case histories of penicillin allergy and with positive or negative skin tests to different penicillin determinants and from seventeen subjects with atopic diseases caused by allergens other than penicillin were tested. The penicilloyl-(‘major’ determinant) specific RAST (Phadebas RAST® Penicilloyl G and V) was positive in twenty out of twenty-four patients with positive skin tests to penicilloyl-polylysine, in eight out of twelve patients with sole skin reactivity to ‘minor’ determinants, and in five out of twenty-four patients with negative skin tests to all antigens used. The penicillamine-specific RAST was the only positive in vitro test in four patients with negative skin tests (two) or sole positive skin reactivity to ‘minor’ determinants (two), whereas benzylpenicillanyl and ampicillin-polymer discs added no more information than the penicilloyl structures in all patients tested. Antibodies specific to ‘minor’ determinants apparently show strong in vitro cross-reactivity with the penicilloyl determinant. It is, therefore, concluded that ‘minor’ determinant specific conjugates are of marginal importance for in vitro diagnosis of penicillin-allergic patients.     

Allergological Examination of Cystic Fibrosis Patients with Skin Reactions During Carbenicillin Treatment

Sixteen of 84 cystic fibrosis patients developed skin manifestations during intensive carbenicillin therapy. The possibility of allergic drug reactions was examined. None of the patients showed any significant levels of antibodies of IgE, IgG or IgM classes against carbenicillin and different penicilloyl determinants as recorded with RAST, sandwich-radioimmunoassay and double antibody assay. The leukocytes of the patients did not release histamine on in vitro provocation with carbenicillin. Furthermore, none of the patients responded to prick test or to peroral penicillin provocation. The negative findings indicated that the recorded reactions were most probably not due to antibody-mediated allergy of type I or type III.     

Specific reactions and cross reactions of anti-penicilloyl antibodies

In human sera the specific reactions and cross reactions of antibodies directed against the penicilloyl groups were determined using different penicilloyl derivatives as antigen. The penicilloyl group providing the highest litre in one Ig class was said to define the specific reaction for this Ig class. This criterium was justified by the fact that antibodies evoked by long-term treatment with benzylpenicillin showed specific reaction to the benzylpenicilloyl group. Moreover, specific reactions of antibodies of different Ig classes in one serum were the same. Both the side chain and the thiazolidine ring of the penicilloyl group can serve as antigenic determinants.     

Determination of IgE antibodies to the benzyl penicilloyl determinant. A comparison between poly-L-lysine and human serum albumin as carriers.

The influence of two different carriers, poly-L-lysine (PLL) and human serum albumin (HSA) in the binding of specific IgE antibodies to the benzyl penicilloyl hapten (BPO) was determined in a solid-phase assay. Serum samples from patients with a history of immediate reaction to penicillin and which had shown the presence of IgE antibodies to BPO were used. Benzyl penicilloyl derivatized cellulose discs were prepared using PLL of different molecular weight and HSA as carriers. These were treated with different molar ratios of benzyl penicillin. These reagents were compared for uptake of BPO-specific IgE using a pool of sera in a radioallergosorbent test (RAST) type assay. Two PLL systems and two HSA systems were finally compared using 26 individual sera. RAST values were compared by Kruskal-Wallis and Wilcoxon tests. The relationships between the four different assays were evaluated by determining Pearson correlation coefficients and the concordance by determining intraclass correlation coefficients (ICC). Analysis of means by the Wilcoxon test revealed significant differences (P less than 0.01) only when the different carrier assays were compared. The correlation coefficients between all the assays were significant (P less than 0.0001), but the ICC was low when the different carrier assays were compared. These results indicate that the nature of the carriers studied (PLL and HSA) influences the capacity for binding IgE antibodies in the RAST procedure. The differences observed indicate that one conjugate cannot be substituted for the other in the determination of IgE antibodies to BPO and that BPO-PLL is preferable.     

Evaluation of adverse cutaneous reactions to aminopenicillins with emphasis on those manifested by maculopapular rashes

We assessed 195 subjects with histories of adverse reactions to aminopenicillins, using 1) skin tests with penicilloyl polylysine (PPL), minor determinant mixture (MDM), benzylpenicillin (PG), amoxicillin, and ampicillin (read after 20 min and 48 h); 2) patch tests with PG, amoxicillin, and ampicillin; and 3) RAST for penicilloyls G and V. Oral challenges with ampicillin, amoxicillin, and penicillin V were administered to 34/60 patients reporting maculopapular reactions. Immediate hypersensitivity (IH), in most cases for both penicillin and aminopenicillins, was diagnosed (based on skin tests, RAST, or both) in 35 subjects who had suffered anaphylactic shock, or urticaria, angioedema, or both urticaria and angioedema. Thirty-three of the 60 subjects reporting maculopapular reactions presented delayed intradermal and patch-test positivity, indicating delayed hypersensitivity (DH), for ampicillin and amoxicillin, and three were also positive for PG. Diagnoses were confirmed with oral challenges in 18/33. The remaining 27/60 were negative in all allergologic tests, with oral-challenge confirmation in 16. Our findings highlight the importance of the amino group in DH to aminopenicillins. Moreover, the mean time interval between the last reaction and our tests was significantly (P < 0.01) longer in DH subjects (54.96 months) than in those with IH (18.62 months), suggesting that the time of testing is less important in cases of DH.     

Stinging insect allergy: Detection and clinical significance of venom IgE antibodies

Venom-specific IgE antibodies in 109 sera from patients who had had immediate systemic allergic reactions following insect stings were measured by the radioallergosorbent (RAST) procedure. The majority of sera contained IgE antibodies to either bee, yellow jacket, or hornet venoms. Some sera had positive RAST reactions with 2 or 3 venoms, but others contained single venom-specific IgE antibodies. Of 24 patients who had large local reactions, the sera of 12 contained venom IgE antibodies. The RAST procedure provides an accurate means of documenting IgE-mediated allergic sensitivity to stinging insects.     

Allergic reactions to ampicillin. Studies on the specificity and selectivity in subjects with immediate reactions

Background and Objective Ainpicillin (AMP) is a drug that has been prescribed extensively. Reactions that have been reported include exanthema. desquamative contact eczema, urticaria and anaphylaxis. Experimental evidence indicates that the side chain of AMP is a structure that may induce a selective immune response either at the humoral or lymphocyte T-cell level. With regard to IgE reactions, the selectivity and specificity of the response needs to be studied in humans. Objective To study tbe specificity of tbe IgE response in a group of subjects who had an immediate allergic reaction after the administration of AMP. Methods Subjects developing an immediate response (anapbylaxis or urticaria) after the administration of AMP or an aminopenicillin derivative witb the same side chain as AMP were studied. Skin tests were made to determinants generated from benzyl penicillin (BP): benzyl penicilloyl (BPO) and minor determinant mixture (MDM), as well as amoxicillin (AX) and AMP. Specific IgE antibodies were determined to benzyl penicilloyl polylisine (BPO-PLL), amoxicilloyl-polyllsine (AX-PLL) and ampicilloyl-polylisine (AMPPLL). The specificity of the IgE antibody response was studied by RAST and RAST inbibition. Subjects were classified in three categories: group A: those who were skin test and/or RAST positive to determinants derived from benzylpenicllin, group B: those who were negative to determinants derived from benzylpenicillin but were skin lest and/or RAST positive to determinants derived from AX and AMP and group C: those who were exclusively positive to determinants derived from AMP. Results A total of 48 subjects was included in the study. In group A there were 35 cases, in group B 10 cases, and in group C tbree cases. RAST inhibition studies showed that in some instances tbe side chain of AMP could induce specific responses with a variable degree of crossreactivity between BP and AX. Conclusions Atbough AMP can induce an immediate IgE response in subjects allergic to betalactams and tbe structure of the side chain may contribute to the specificity of the response, our results indicate tbat in most instances crossreactivity with the other penicillins exists and that in the groups studied selective reactions to just AMP derived determinants were uncommon.     

Frequency of systematic reactions to penicillin skin tests.

BACKGROUND: Penicillin skin testing is generally considered to be safe when performed sequentially with puncture and intradermal testing although fatalities have been reported. OBJECTIVE: We analyzed the rate of systemic reactions to penicillin skin tests for a period of seven and two-thirds years. METHOD: This retrospective study used a computerized database at the Mayo Clinic. Altogether 1710 patients were skin-tested to penicillin from January 1992 to September 1999. All patients tested had a history of penicillin allergy. Patients were tested with benzylpenicilloyl polylysine (Pre-Pen) (6.0 X 10(-5) M), freshly prepared penicillin G (10,000 units/ml), and penicilloate (0.01 M). Prick tests were done first and if negative then intradermal tests. Systemic reactions were evaluated and treated by physicians. RESULTS: Eighty-six patients had positive penicillin skin tests of which two had systemic reactions. Our systemic reaction rate for all patients tested was 0.12%; and 2.3% for the penicillin skin test-positive group, with no fatalities. CONCLUSION: The incidence of systemic reaction to penicillin skin tests is low. Skin prick tests should always be done first. If there is a history of a previous serious reaction, the skin tests-if done-should be diluted to start with. Those doing penicillin skin tests should be prepared to treat a systemic reaction.     

* Immediate allergic reactions to amoxicillin

A large group of patients with suspected allergic reactions to (β-lactam antibiotics was evaluated. A detailed clinical history, together with skin tests, RAST (radioallergosorbent test), and controlled challenge tests, was used to establish whether patients allergic to β-lactam antibiotics had selective immediate allergic responses to amoxicillin (AX) or were cross-reacting with other penicillin derivatives. Skin tests were performed with benzylpenicilloyl-poly-L-lysine (BPO-PLL), benzylpenicilloate, benzylpenicillin (PG), ampicillin (AMP), and AX. RAST for BPO-PLL and AX-PLL was done. When both skin test and RAST for BPO were negative, single-blind, placebo-controlled challenge tests were done to ensure tolerance of PG or sensitivity to AX. A total of 177 patients were diagnosed as allergic to β-lactam antibiotics. We selected the 54 (30.5%) cases of immediate AX allergy with good tolerance of PG. Anaphylaxis was seen in 37 patients (69%), the other 17 (31%) having urticaria and/or angioedema. All the patients were skin test negative to BPO; 49 of 51 (96%) were also negative to MDM, and 44 of 46 (96%) to PG. Skin tests with AX were positive in 34 (63%) patients. RAST was positive for AX in 22 patients (41%) and to BPO in just 5 (9%). None of the sera with negative RAST for AX were positive to BPO. Challenge tests with AX were performed in 23 subjects (43%) to establish the diagnosis of immediate allergic reaction to AX, and in 15 cases (28%) both skin test and RAST for AX were negative. PG was well tolerated by all 54 patients. We describe the largest group of AX-allergic patients who have tolerated PG reported so far. Diagnosis of these patients can be achieved only if specific AX-related reagents are employed. Further studies are necessary to determine the exact extent of this problem and to improve the efficacy of diagnostic methods.     

Desensitization of anaphylactic hypersensitivity specific for the penicilloate minor determinant of penicillin and carbenicillin

A 68-yr-old man with a history of a morbilliform rash caused by intravenous penicillin required carbenicillin (CB) therapy for refractory Serratia marcescens septicemia. Skin testing showed a positive immediate skin test to the penicilloate minor determinant in the presence of negative tests to benzylpenicilloylpolylysine (BPL) and penicillin G (PG), as well as cross-reactivity between the penicilloate derivatives of PG and CB. True densensitization was accomplished by gradual administration of CB intravenously and was accompanied by a diffuse flush reaction. There was specific loss of wheal-and-flare reactivity as well as of specific serum reaginic antibody activity during the procedure, and there was no evidence of activation of serum complement. This case illustrates the usefulness of skin tests in the prediction and management of penicillin allergy and presents data pertaining to immunologic mechanisms involved in true desensitization.     

Skin testing to detect penicillin allergy

Skin testing for penicillin allergy with penicillin G (Pen G), penicilloic acid (PA), and penicilloyl poly-L-lysine (PPL) was performed on 740 subjects, and the results were assessed from epidemiologic and immunologic perspectives. Approximately 95% of these patients had histories of apparent allergic reactions to beta-lactam antibiotics, and 63% were skin-test positive. The prevalence of positive skin tests was related to the time that had elapsed between clinical reactions and skin testing. Ninety-three percent were skin-test positive 7 to 12 mo after reactions, and 22% were positive 10 yr or more after reactions. Patients under 30 yr of age had a prevalence of positive skin tests 1.7-fold higher than older patients. Testing with PPL, PA, and Pen G detected 76.3%, 55.3%, and 57.1% of the positive patients, respectively. Omission of PPL, PA, or Pen G would have led to a failure to detect 25.6%, 7.2%, and 6.2% of the positive patients, respectively. Subjects with skin tests positive to penicillin often reacted to skin tests with other beta-lactam antibiotics; 73% (41 of 56) reacted to ampicillin and 51% (38 of 74) reacted to cephalothin. No serious allergic reactions were provoked by testing. None of the 83 skin test--negative patients treated with beta-lactam antibiotics immediately after testing experienced acute allergic reactions. Two patients developed mild urticaria beginning 3 and 5 days into therapy. One skin test--negative patient experienced urticaria 3 hr after receiving oral penicillin 6 mo after skin testing. This patient's skin-test status immediately before therapy was unknown. These results support the position that testing with PPL, PA, and Pen G is a rapid, safe, and effective method for identifying patients at risk, or not at risk, for allergic reactions to penicillin.     

Intrapartum anaphylaxis to penicillin in a woman with rheumatoid arthritis who had no prior penicillin allergy.

BACKGROUND: Little is known about the development of drug allergy during pregnancy or in patients with altered immune status. OBJECTIVE: To report a case of new-onset penicillin allergy during pregnancy in a woman with rheumatoid arthritis. METHODS: A 39-year-old woman with rheumatoid arthritis developed intrapartum anaphylaxis that led to fetal demise. She had previously received penicillin-based antibiotics without any allergic reactions. Because of group B streptococcus colonization, an intravenous infusion of penicillin G was started during labor. Within minutes, she developed severe anaphylaxis. RESULTS: A fluorescent enzyme immunoassay revealed a moderate level of specific IgE to penicilloyl G and penicilloyl V (3.15 kU/L and 2.77 kU/L, respectively). Given the patient's history, these positive results were considered confirmatory of penicillin allergy. This case raises a number of salient points. First, patients can develop severe allergy to penicillin despite having safely received penicillins in the past. Possible factors that influenced the development of severe penicillin sensitivity in this patient are discussed. Second, unexpected intrapartum anaphylaxis can occur, which can be life threatening to the mother or fetus. Third, safe and reliable methods for diagnosis of drug allergy must be available. CONCLUSIONS: This case illustrates that during the current unavailability of skin testing reagents in the United States, a positive result on in vitro testing can be helpful in confirming penicillin allergy in cases in which drug challenge is deemed unsafe.