In 1928, Sir Alexander Fleming discovered that the active component of the penicillium fungus had the ability to kill bacteria in a Petri dish, and named it penicillin. In 1945, Fleming, Florey and Chain were jointly awarded the Nobel Prize in Physiology or Medicine "for the discovery of penicillin and its curative effect on various infectious diseases." Since the early 1950s, penicillin has saved millions of lives exposed to life-threatening infections. Penicillin G remains the only recommended treatment for the prevention of transmission of congenital syphilis.
The first case of anaphylaxis associated with penicillin was reported in 1945, and a World Health Organization report in 1968 stated that the death rate from anaphylaxis was 0.002%.1 There are no data to suggest that the frequency of Allergic reactions have increased over the past 60 years, and there is compelling evidence that sensitization to penicillin is lost over time.2
Anaphylaxis induced by penicillin exposure has been observed with oral, subcutaneous, and intravenous administration.3 Based on a national survey conducted in 1957, covering 827 hospitals in the United States, it was estimated that a total of 1000 deaths related to penicillin during the first 10 years of its use.1,4
Additionally, the increased use of penicillin since 1950 led to estimates that from 1965 to 1968 there were 300 deaths annually from anaphylactic shock due to penicillin use in the United States, but these data were not verifiable.1
A review of 151 deaths due to penicillin use published in the medical literature between 1951 and 1965 showed no predominance by sex;1 more than 50% of the people were between 25 and 65 years old, 44% had respiratory infections, 28 % pre-existing allergies or asthma, and 69% had prior exposure to penicillin, of which 36% had prior reactions to the medication.
The mean interval between penicillin administration and onset of symptoms was less than 15 minutes in 85% of cases, and most patients died within one hour of administration.
| Current epidemiology and geographic relevance |
Penicillin allergy is the most common allergy identified in medical records, with a prevalence ranging from 6 to 25% in various regions and populations.5, 6 Benign skin reactions such as urticaria and late maculopapular rash are the most common type. of reactions.
The incidence of new reports of penicillin allergy in 2007 in the United States was 1.4% for women and 1.1% for men in a study that extracted data from the electronic health records of 411,534 patients who had received permanent attention.7
A study in 1966 showed a 7.8% incidence of allergic reactions, with 22% of cases confirmed based on positive penicillin skin tests;8 however, single-center longitudinal studies in the United States showed that The rate of positive penicillin skin tests decreased from 15% in 1995 to 3% in 2007 and 0.8% in 2013.9,10
Penicillins have been the most common cause of fatal and nonfatal drug-induced drug reaction anaphylaxis in the United States11, 12 and the United Kingdom. The lowest rate of anaphylaxis is for oral penicillins, with one report from the United Kingdom of one case of fatal anaphylaxis from oral amoxicillin in 35 years in 100 million treatment cycles.13
Aminopenicillins are among the highest-risk medications causing benign late rashes, which commonly occur in the setting of acute Epstein-Barr virus infection.14
Aminopenicillins are considered the most common cause of acute generalized exanthematous pustulosis (AGEP).15 Penicillins have been associated with other serious skin reactions, such as drug reaction with eosinophilia and systemic symptoms (DRESS), Stevens-Johnson syndrome, and toxic epidermal necrolysis (SJS-TEN).16
| Penicillin and beta-lactams |
Unlike other beta-lactams, penicillins have a thiazolidine ring and, unlike cephalosporins and carbapenems, they have no R2 or additional side chain structures.
The side chains of first-generation penicillins and cephalosporins are less complex than the side chains of later-generation cephalosporins, and although early studies indicated more than 5% cross-reactivity between penicillins and cephalosporins, contamination of the early preparations of cephalosporins with penicillins.17,18
Currently, no more than 2% of patients with positive multiple skin test reactions to penicillin have a reaction to cephalosporins,19 with the exception of patients allergic to aminopenicillins but not to benzyl penicillin, penicillin VK, and other penicillins. .twenty
Allergy to such selective aminopenicillin has been reported infrequently in the United States,21 but appears to account for one-third of penicillin allergy cases in southern Europe, where 25 to 35% of patients who are selectively allergic to aminopenicillin have cross-reactivity with aminocephalosporins.14,19,20
In 99% of patients with a history of penicillin allergy, a skin test and carbapenem challenge are associated with an acceptable side effect profile.22
There appears to be no immunological or clinical cross-reactivity between penicillins and monobactam aztreonam; However, in patients allergic to ceftazidime, aztreonam reactions have been reported, which are due to a shared R1 side chain.22,23
| Mechanisms of allergy to penicillin |
Penicillins are small molecules that covalently bind to proteins in plasma and create hapten-carrier complexes. The beta-lactam ring binds to lysine residues on serum proteins, and when bound to a polylysine matrix, creates the major antigenic determinant, penicilloyl polylysine.24 Activation of covalent attachment to carboxyl and thiol groups leads to the creation of several minor determinants.25
The hapten-prohapten model is applied to immediate or antibody-mediated hypersensitivity to penicillin (Gell-Coombs reactions type I, II and III). In IgE-mediated reactions, dendritic cells bind and internalize penicillin-bound proteins for presentation to naïve CD4 + T cells (type 0 helper T cells).
In the presence of interleukin-4, naive T cells are converted into penicillin T-type T helper (Th2), which then produce interleukin-4 and interleukin-13, which induce the differentiation of B cells into plasma cells that secrete interleukin-specific IgE. penicillin, which binds to Fc epsilon receptors on the surface of basophils and mast cells.
Upon rechallenge, cross-linking of polyvalent Fc epsilon penicillin receptors bound to IgE antibodies induces mast cell degranulation and the release of soluble inflammatory mediators such as tryptase, histamine, prostaglandins, and leukotrienes, leading to the clinical manifestations of anaphylaxis.
Late reactions are often associated with models involving non-covalent binding, such as the drug-drug interaction model or altered specificity of HLA peptide presentation (altered peptide repertoire model).26,27
Common phenotypes of penicillin allergy include reactions within 1 to 6 hours after exposure (e.g., hives and anaphylaxis) and reactions occurring more than 6 hours after administration of a single dose or after multiple doses ( for example, maculopapular rashes). Late T cell-mediated reactions with systemic involvement include severe skin reactions (SJS-TEN, DRESS, and AGEP).
| Diagnosis of IgE-mediated penicillin allergy |
After years of widespread use of penicillin antibiotics, penicillin reagents were identified to detect populations at risk for allergic reactions and anaphylaxis.28,29 This led to the initial use of the major determinant (penicilloyl polylysine), which is penicilloyl coupled to lysine for stabilization as a diagnostic agent; The use of a minor determinant mixture for this purpose began in the early 1960s.25
A 1971 study of the prospective use of penicillin skin testing with penicilloyl polylysine and a minor determinant mixture in nonconsecutive hospitalized patients with a clinical indication for therapeutic use of penicillin identified 54 patients with a history of penicillin hypersensitivity but with skin testing. non-reactive who were treated with penicillin; only 1 patient had a reaction (hives and arthralgias within 24 hours of treatment).30
Based on this study and others, the positive predictive value for penicillin skin testing with these reagents was established at 50 to 75%, and the negative predictive value at more than 93%.21,30
| Penicillin skin test for IgE-mediated reactions |
Pharmacological challenges with penicillin are considered the gold standard for assessing drug tolerance. Challenges can be performed by administering increasing amounts of drug over time (for example, one-tenth of the dose followed after 30 minutes to 1 hour by the full dose) or by administering a single full dose followed by at least 1 hour of observation. .
More recent studies of penicillin skin testing have evaluated the negative predictive value after a penicillin challenge. The current negative predictive value with the use of a complete set of major and minor determinants is estimated to be approximately 98%, with a false negative reaction rate of 2 to 3% after penicillin challenge and generally mild skin reactions.21
In the United States, a complete panel of minor determinants that includes amoxicillin has never been commercially available; Penicilloyl polylysine and benzyl penicillin are the most common reagents used to evaluate penicillin allergy.
Of the reagents available globally for skin testing, penicilloyl polylysine used as a major determinant and benzyl penicillin used as a minor determinant, followed by amoxicillin challenge, have been shown to have a negative predictive value of more than 95% in populations without high risk with a history of remote reactions to penicillin.
In Europe and Australia, selective sensitization to aminopenicillins and occasionally clavulanic acid in patients with negative skin test results with penicilloyl polylysine and a minor determinant mixture has been described more frequently, with all of these reagents commercially available for testing. 31
Patients with specific side chain reactions appear to be less common in the United States. However, a minor determinant panel that included amoxicillin would provide more confidence for testing in high-risk patients, and the Food and Drug Administration is evaluating a full test kit.21
In the absence of global availability of these reagents, the use of an amoxicillin swallow challenge following a negative penicillin skin test with penicilloyl polylysine and benzyl penicillin is considered an acceptable method to examine the possibility of an IgE-mediated reaction to amoxicillin. and other penicillins, although patients with severe or recent IgE-mediated reactions are excluded from these tests.
| Direct challenge without skin testing for children |
Penicillin skin testing is safe and effective in the evaluation of children with a history of penicillin allergy.32 A retrospective cohort study involving 369 children with negative penicillin skin tests who were challenged with penicillin showed that 14 patients (3.8% ) had a mild reaction.
Given the current low prevalence of confirmed penicillin allergy, several studies have evaluated the safety and effectiveness of performing direct penicillin challenges without initial skin testing.
Most of these studies have involved children with a low rate of confirmed penicillin allergy, even when analyzed 2 months after a benign rash in reaction to amoxicillin.18
In a prospective and retrospective observational study involving 818 young children with a low-risk history of reactions to amoxicillin (children with a history of anaphylaxis were not included), Mill et al performed amoxicillin challenges with two graduated doses administered with 20 minutes of difference.
The authors reported that 2.1% of children had immediate reactions, and 3.8% had mild non-immediate reactions.33 Ibanez et al conducted a multicenter prospective study that included 732 children with a history of mild reactions to penicillins.
Using a multi-step challenge for the responsible penicillin, the authors found that 0.8% of children had immediate reactions and 4.0% had delayed reactions, with one patient requiring treatment with epinephrine.34
These and other studies suggest that a direct penicillin challenge without skin testing is probably appropriate for children with a history of benign rash but no history of anaphylaxis.
However, all studies to date that have examined direct challenges to penicillin have been performed by allergy specialists or in urgent care settings, and the safety of such challenges when performed in non-specialist clinics and populations is unknown. adults.14
Other indications for challenges include a recorded history of penicillin allergy involving symptoms not suggestive of allergy (eg, nausea or headache), family history of penicillin allergy, unknown reactions, and pruritus without rash. A direct challenge to penicillin is not recommended as a general approach until larger studies can confirm its safety and effectiveness.
| Delayed penicillin allergy test |
Skin testing procedures for delayed reactions to penicillins include patch, delayed puncture, and intradermal testing.27
Penicillin and major and minor antigenic determinants penetrate the epidermis (patch test and prick test) or dermis (intradermal test) 27 and interact covalently or noncovalently with proteins in the skin to form antigenic conjugates recognized by presenting cells. antigens expressing major histocompatibility complex class I or II.
These cells present the antigen-peptide complex to effector T cells, leading to the proliferation of CD4+ T cells, CD8+ T cells, or both, resulting in the local release of cytokines and an inflammatory response. 27
A 3-year prospective multicenter study conducted to determine the sensitivity of patch testing in identifying the perpetrator of severe skin reactions suggested that the sensitivity of delayed intradermal testing may exceed that of patch testing, particularly for rashes. maculopapular, DRESS, and AGEP.35 Patch testing has poor sensitivity for SJS-TEN (<40%), and delayed intradermal testing is not recommended due to anecdotal reports of reproducing initial reactions.
In vitro tests for late reactions are only available at research or specialized centers, and their sensitivity and specificity vary depending on the drug and the specific test.
These tests are performed by exposing the patient’s lymphocytes to the drug involved. They include the lymphocyte transformation test, which measures the proliferation of the patient’s T cells over a period of 5 to 7 days,36 and the enzyme-linked immunosorbent spot test (ELISPOT), which detects specific cytokine-producing cells. antigen after 24 hours of incubation with polymorphonuclear cells. Both tests are performed in the presence of the drugs involved.36
| Genetic risk |
The discovery of HLA associations with drug hypersensitivity syndromes has provided screening strategies to improve drug safety and has increased understanding of the immunopathogenesis of late drug reactions.26
There were no significant genetic associations for immediate allergic reactions to penicillins, and candidate gene studies have shown the strongest association with genes involved in IgE synthesis, HLA class II antigen presentation, and cytokines such as interleukins; 4, 10 and 18 however, none are currently in use for prevention or diagnosis. 37
Drug-induced liver injury related to flucloxacillin, a semisynthetic antistaphylococcal penicillin used in the United Kingdom, Europe and Australia, has been strongly associated with HLA-B*57:01 in a genome-wide association study38 and drug-induced liver injury. Drug therapy associated with amoxicillin-clavulanic acid has been associated in multiple studies with HLA-DRB1*15:01 and its haplotype, DQB1*06:02, and with HLA-A*02:01 in northern European populations.26, 39
The induced liver injury is selective for these drugs based on HLA restriction, and no cross-reactivity is observed with other beta-lactams. Given the low positive predictive value of these HLA alleles for drug-induced liver injury (<1%), testing them as a means of determining the possible presence of penicillin allergy is not currently used in routine clinical practice.
| Natural history of penicillin allergy |
The natural history of IgE-mediated penicillin allergy has been the most studied hypersensitivity reaction.
In 1981, a retrospective study from the United States showed that the prevalence of positive penicillin skin testing was lower among patients evaluated 10 years or longer after a documented reaction than among patients who were evaluated 7 to 12 months after a reaction (prevalence, 22% vs. 73%).40
A prospective longitudinal study from Spain followed 31 patients with positive penicillin skin tests and showed that at 1 year, 81% of patients had positive tests, and at 5 years, 12 of 18 patients (67%) continued skin testing. positive, indicating a loss of penicillin-specific IgE over time.41
A similar decrease in the rate of cephalosporin positive skin tests over time has been shown, although patients with positive skin tests for penicillin and cephalosporins take longer to lose their sensitivity than patients sensitized to cephalosporins alone.42
Some children with a history of serum sickness-like reactions to amoxicillin have been shown to have no such reactions to amoxicillin when challenged, suggesting that the reaction is not long-lasting; Future penicillin skin testing, an ingestion challenge, or both should be considered in this population.43 The natural history of severe skin reactions to penicillins remains unknown.
| Clinical implications of a penicillin allergy marker |
Patients with a penicillin allergy receive more vancomycin, fluoroquinolones, and clindamycin than patients without an allergy.6 Penicillin is the drug of choice for syphilis44 and other infections, and a penicillin allergy “label” has associated implications, which have not always been fully appreciated.
Among patients with methicillin-susceptible Staphylococcus aureus sepsis, the risk of death within 30 days is lower with beta-lactam treatment than with vancomycin 45 and a higher rate of clinical failure has been observed with non-beta-lactam antibiotics for bacillus sepsis. gram-negative.46
Decision analysis models project that patients with methicillin-susceptible S. aureus bacteremia will have inferior outcomes if they are treated with vancomycin rather than having their penicillin allergy evaluated.47
Case-control studies in the United States and the United Kingdom involving more than 50,000 patients labeled as allergic to penicillin showed increased rates of infection with methicillin-resistant S. aureus (MRSA), vancomycin-resistant enterococcus, and Clostridioides difficile (formerly Clostridium difficile), 6.48
Prolonged hospitalizations and higher readmission rates have also been reported among patients with a penicillin “label” allergy.6,49 Surgical site infections are reported to be 50% higher among patients with a penicillin allergy label. penicillin than among those without such a label.50 A penicillin allergy label is also costly.
Several studies from North America and Europe have documented higher costs of inpatient and outpatient care for patients with penicillin allergy, 51,52 and penicillin allergy testing and delabeling are estimated to lead to cost savings, and the largest study shows a reduction in total healthcare expenditures per patient per year.53-55
| Evaluation of Penicillin Allergy in Antibiotic Stewardship Programs |
In contemporary clinical practice, more than 90% of patients labeled as allergic to penicillin can safely receive the drug. This observation, together with the estimate that, on average, 8 to 15% of unselected patients are labeled as allergic to penicillin, 56 indicates that many patients labeled as allergic could safely receive it.
The high burden of penicillin allergy labeling and growing evidence of adverse personal and public health consequences provide the rationale for a formalized hospital-based process to prioritize penicillin allergy evaluation as part of a health care program. antibiotic administration.
Given that the majority of adults with a penicillin allergy label acquired it in childhood and given that more than 90% of patients labeled penicillin allergic can “remove” the label, 57 there is an opportunity to integrate strategies formalized, risk-based testing in antibiotic stewardship programs targeting populations with the greatest need for antibiotics and at highest risk of developing antibiotic resistance and other conditions, such as C. difficile infection.58
Retrospective and observational data suggest that direct oral challenge procedures may be safe in lower-risk populations, including patients with a remote or unknown history of allergy or a mild skin reaction.33,59,60
Given the large number of patients labeled globally, an evidence base is needed to guide the safest and most effective approach to labeling patients with penicillin allergy in the context of these formalized programs.
| Best clinical practices to remove the penicillin allergy “label” |
Various methods have been used to eliminate penicillin allergy labels in inpatient and outpatient populations.
These include support from allergy-trained clinical pharmacists to perform preventative testing in patients with a history of penicillin allergy who are at high risk for antibiotic use, 61 the use of clinical decision support tools, and specific algorithms for testing. of penicillin,54 and the use of skin tests through telemedicine consultation (since there are few allergy specialists).62,63
A systematic review of penicillin testing in hospitalized patients, including studies in intensive care units, confirmed the safety and effectiveness of this approach in removing the label of penicillin allergy, with 95% of patients having skin testing negative.56
More recently, algorithms or pathways have been developed to guide non-allergists in the use of antibiotics in patients labeled with penicillin allergy, with risk assessment based on clinical history, timing and phenotype of reaction and the coexistence of associated conditions.64
Delabeling is accomplished with the use of oral or intravenous test doses and challenges for low-risk patients and with the use of skin testing for high-risk patients. A study of a decision support pathway developed as part of a guideline for antibiotic prescribing at several Partners HealthCare teaching hospitals in Boston showed that more test doses of cephalosporins were given to patients labeled as allergic to penicillin after implementation of the guideline, with a resulting reduction in the use of vancomycin, aztreonam, and fluoroquinolones.65
A follow-up study by the same researchers compared the use of penicillin skin testing and a computerized guide with usual care and showed that both approaches led to higher rates of use of third- and fourth-generation cephalosporins, but only the skin test had a higher rate of penicillin use at discharge.66
Such decision support pathways improve antimicrobial stewardship67 but do not lead to systematic removal of the penicillin allergy label. Evaluation of penicillin allergy in outpatient clinics and the use of alerts in electronic health records have facilitated the evaluation of penicillin allergy preoperatively.60,68,69
The use of an amoxicillin challenge without a penicillin skin test was associated with low morbidity in a group of Marine patients with a history of selective penicillin allergy70 and in children with a history of low-risk symptoms of penicillin allergy. .71 However, larger studies are needed to evaluate safety in these and other populations.
| Penicillin desensitization |
Patients with IgE-dependent penicillin allergy, including anaphylaxis, who require penicillin as first-line therapy are candidates for rapid desensitization. The first penicillin desensitization, attributed to O’Donovan during World War II, was performed by adding increasing amounts of oral penicillin to milk until the target dose was reached without side effects in a soldier who had had an anaphylactic reaction to intramuscular penicillin. .29
Since then, numerous patients have been successfully desensitized with intramuscular, intravenous, and oral protocols.73 The mechanisms of rapid desensitization have been studied in cellular and animal models,74 leading to the development of clinical protocols.75
In 2009, Legere et al successfully performed desensitization on 15 patients who had cystic fibrosis and a forced expiratory volume in 1 second of less than 1 liter, including 1 patient who underwent desensitization during lung transplantation, using a standard protocol. , in which the successive concentration was increased by a factor of 10 and the doses were doubled every 15 minutes until the target dose was reached in 6 hours.76
This and similar protocols have been used for intravenous and oral penicillin desensitization with 100% success, allowing administration of the target dose and maintenance of first-line therapy. Desensitization has temporary effects that last for at least two drug dosing intervals, after which desensitization needs to be repeated.
Long-acting benzathine penicillin is associated with an acceptable adverse event profile 1 to 3 weeks after penicillin desensitization.73 Empirical desensitization in the absence of positive skin tests does not answer the question of whether a patient is truly allergic. to penicillin, and follow-up formal penicillin allergy testing is recommended after completing the penicillin treatment course.
| Conclusions |
The incidence of IgE-mediated and non-IgE-mediated reactions has not increased worldwide in the last 50 years, and a penicillin allergy label has serious consequences for individual and public health.
Although a large number of patients are labeled as allergic to penicillin, more than 95% of them can safely receive penicillin when properly evaluated.
Penicillin allergy is lost over time, and the use of sensitive and specific tools to identify patients with true reactions should be a health priority implemented through algorithms and labeling programs.
Over time, delabeling patients who no longer have penicillin allergy would be expected to control the use of alternative, more expensive antibiotics and reduce the associated morbidity and mortality and the increase in organisms resistant to penicillin and beta-lactams.
Protecting patients who are truly allergic to penicillin through accurate diagnosis, appropriate labeling, and, if necessary, desensitization should be the next steps to improve the safety and quality of care in personalized medicine.
Allergists should play a central role in facilitating outpatient and inpatient testing programs aimed at correctly identifying patients with penicillin allergy. By taking appropriate history taking and risk stratification to identify patients without IgE-mediated allergy as well as low-risk patients, all healthcare providers can play a central role in alleviating the enormous burden of public and individual health related to penicillin allergy label.
