Anesthesia related hypersensitivity. Anesthesie gerelateerde overgevoeligheid

Transcription

1 Faculteit Geneeskunde & Gezondheidswetenschappen Anesthesia related hypersensitivity New diagnostics and novel epidemiological data Anesthesie gerelateerde overgevoeligheid Nieuwe diagnostiek en epidemiologische data Proefschrift voorgelegd tot het behalen van graad van doctor in de medische wetenschappen aan de Universiteit Antwerpen te verdedigen door Astrid Uyttebroek Promotoren: Prof. Dr. D. G. Ebo, Dr. V. Sabato, Prof. Dr. H. Lapeere Antwerpen, 2017

2 2017 Astrid Uyttebroek Anesthesia-related allergy: New diagnostics and novel epidemiological data. Astrid Uyttebroek Faculteit Geneeskunde en Gezondheidswetenschappen, Universiteit Antwerpen Thesis Universiteit Antwerpen with references with summary in English and Dutch ISBN D/2017/12.293/17 Lay-out en fotografie: Chris Bridts met dank aan het museum Het Huis van Alijn, Gent, ( Cover Lay-out: Nieuwe Mediadienst Universiteit Antwerpen Universiteit Antwerpen Faculteit Geneeskunde en Gezondheidswetenschappen Immunologie Allergologie Reumatologie Campus Drie Eiken Universiteitsplein 1 B2610 Antwerpen Printing of this thesis was financially supported by: ALK Abello BV, Almere, The Netherlands Novartis Pharma, Vilvoorde, Belgium Shire Belgium BVBA, Brussels, Belgium ThermoFisher Diagnostics NV, Groot-Bijgaarden, Belgium 2

3 Examencommissie Promotoren Prof. Dr. D.G. Ebo Dr. V. Sabato Prof. Dr. H. Lapeere Faculteit Geneeskunde en Gezondheidswetenschappen Universiteit Antwerpen Faculteit Geneeskunde en Gezondheidswetenschappen Universiteit Antwerpen Faculteit Geneeskunde en Gezondheidswetenschappen Universiteit Gent Interne juryleden Prof. V. Van Tendeloo Prof. G. De Meyer Faculteit Geneeskunde en Gezondheidswetenschappen Universiteit Antwerpen Faculteit Farmaceutische, Biomedische en Diergeneeskundige Wetenschappen Universiteit Antwerpen Externe juryleden Prof. Dr. L. Garvey Allergy Clinic, Herlev and Gentofte Hospital Danish Anaesthesia Allergy Centre, University Hospital, Copenhagen, Denmark Prof. Dr. P.M. Mertes Nouvel Hôpital Civil, Hôpiteaux Universitaires de Srasbourg, Service d Anesthésie-Réanimation Chirurgicale, Strasbourg cedex, France 3

4 Table of contents Abbreviations... 5 Chapter 1: Introduction and aims of the thesis... 7 Part 1: Introduction... 8 Part 2: Aims of the thesis Chapter 2: Optimizing the diagnosis of IDHR to Atracurium Part 1: Flow cytometric diagnosis of atracurium-induced anaphylaxis Part 2: Immunoglobulin E antibodies to atracurium: a new diagnostic tool? Chapter 3: Optimizing the diagnosis of cefazolin hypersensitivity Part 1: Cefazolin hypersensitivity: towards optimized diagnosis Part 2: Diagnosing cefazolin hypersensitivity: lessons from dual labelling cytometry Chapter 4: Perioperative hypersensitivity reactions in Antwerp: a 16 year survey Chapter 5: Critical reflection and future perspectives Chapter 6: General conclusion Dankwoord Curriculum vitae Curriculum vitae Publications Abstracts Addendum

5 Abbreviations AB Antibiotics MDM Minor Determinant Mixture AC Amoxicillin Clavulanic Acid ME Meropenem ADR Adverse drug reaction MIV Mivacurium AMP Ampicillin MOR Morphine Apy Ampicilloyl MRGPRX2 Mas-related G-protein receptor ATR Atracurium NA Not applicable AX Amoxicillin NAV Not available Axy Amoxicilloyl NMBA Neuromuscular blocking agents AZ Aztreonam NR Non Responder BAT Basophil Activation test NRL Natural rubber latex BPO Benzy Penicilloyl NSAID Non-steroidal Anti-Inflammatory Drugs CF Ceftazidim PAPPC p-aminophenyl phosphoryl choline CI Confidence Interval PENI G Penicillin G CIS Cisatracurium PPL Benzyl penicilloyl-polylysine CL Clavulanic Acid PyG Penicilloyl G CR Ceftriaxon PyV Penicilloyl V CU Cefuroxim QAS Quaternary Ammonium Structure DHR Drug hypersensitivity reaction RAST Radio allergo sorbent test DPT Drug provocation test RIA Radio immuno assay ENDA European Network of Drug allergy ROC Rocuronium F Female SCT Subcutaneous test FcεRI High affinity receptor for IgE SEM Standard error of the mean FEIA Fluorescence Enzyme Immuno Assay SPT skin prick test H History STAT Signal Transducer and Activation of IDHR Immediate type drug hypersensitivity reactions ST Transcription Skin testing IDT Intradermal test SUX Suxamethonium IgE Immunoglobulin E TG-ROC Two graph receiver operating curve IM Imipenem Th2 Type 2 helper T cell M Male THIQ Tetrahydroisoquinoline MAPK Mitogen-Activated Protein Kinase Tot Total 5

6 6

7 Chapter 1: Introduction and aims of the thesis 7

8 Part 1: Introduction Adverse drug reactions Adverse drug reactions (ADR) are defined as any noxious, unintended, and undesired effect of a drug that occurs at correct posology used for prevention, diagnosis, or treatment (1). As displayed in figure 1, adverse drug reactions are stratified into two groups. Figure 1: adverse drug reactions, adapted from (2) The large majority of ADR, i.e. 80%, are type A reactions, which are predictable and related to the pharmacological effects of the drug (3). In contrast, type B reactions, also designated as drug hypersensitivity reactions (DHR), account for the remaining 20% of the adverse drug reactions. Drug hypersensitivity reactions are further subdivided according to the (alleged) underlying pathomechanism. When no immunological mechanism is clearly delineable, these reactions are designated as non-allergic DHR. In contrast, when the underlying mechanism involves the immune system these reactions are designated as allergic DHR or drug allergies (3,4). Drug allergies account for 15% of all DHR and are traditionally classified according to the Gell and Coombs classification into four types (see table 1). 8

9 Table 1: immunopathological reactions according Gell and Coombs (5) B cell (humoral) T cell (cellular) Immediate onset Late onset Type I: IgE/FcεRI-mediated Type IVa-d Type II: cytophylic antibodies (IgG, IgM) Type III: circulating immune complexes (IgG, IgM) This thesis focuses upon perioperative immediate drug hypersensitivity reactions (IDHR). In general, IDHR can be immune mediated, e.g. IgE/FcRεI-mediated, or they can belong to the group of nonallergic IDHR, e.g. as by non-specific immediate histamine releasers (opiates, iodinated radio contrast media) or cyclo-oxygenase 1 inhibition by non-steroidal anti-inflammatory drugs (NSAIDs), or as recently elegantly demonstrated by McNeil et al, by ligation of the Mas-related G-protein receptor MRGPRX2 by various compounds that contain a tetrahydroisoquinoline (THIQ) motif (6). Traditionally, IgE/FcεRI-mediated IDHR are based upon a two-step process. First, the asymptomatic sensitization phase, during which exposure to an allergen leads to activation of specific Th2 cells by dendritic cells. This Th2 polarization induces B cell isotype switching resulting in the production of allergen-specific IgE antibodies by plasma cells. Subsequently, these allergen-specific IgE antibodies, via their Fc portion of the heavy chain, bind to the high affinity receptor for IgE (FcεRI) present on the outer membrane of tissue resident mast cells and circulating basophils. Re-exposure to the allergen can cross-link adjacent IgE/FcεRI complexes and elicit degranulation of the mast cell/basophil with release of various mediators and cytokines that will cause the allergic symptoms. However, in the context of drug hypersensitivity, the reaction seems frequently to result from a cross-reactivity to a structural similar compound, which is often unknown (8). 9

10 Perioperative hypersensitivity reactions Background A worldwide accepted definition of perioperative anaphylaxis, for the moment, is lacking. However, using the current nomenclature, immediate perioperative hypersensitivity reactions can be defined as each reaction fulfilling the anaphylaxis criteria according to Brown (9) (table 2), that emerges during or within 2 hours after surgery. These reactions are rare but are frequently serious and potentially lifethreatening. Therefore, prompt recognition of symptoms and correct treatment is mandatory (10). However, perioperative hypersensitivity has its own scene with several peculiarities. First of all, some symptoms compatible with a hypersensitivity reaction could be due to the pharmacological effect of the anesthetic drugs (e.g. hypotension) or to mechanical manipulation (e.g. isolated bronchospasm). Also, it is not always easy to recognize an immediate hypersensitivity reaction since the symptoms are often hidden (e.g. skin symptoms under the sterile surgical drape). Furthermore, during surgery, patients are exposed to a variety of drugs, often simultaneously administered and it is known that, although of the utmost important information, the time between exposition and onset of symptoms is not a reliable marker to predict the elicitor (11). This hampers the diagnosis of perioperative hypersensitivity reactions, especially since the gold standard diagnostic test, namely the drug provocation test, is not always feasible for ethical and practical reasons (vide infra). The incidence of perioperative hypersensitivity is estimated between 1/ and 1/1.750 interventions (12). The exact rate of morbidity and mortality as a consequence of perioperative anaphylaxis is not exactly known. Reports from large registries in the USA and Japan estimate this mortality rate between 4% and 4.76% (13,14). Alternatively, this seems to be much lower in western Australia (0-1.4%) (15). For the time being, data concerning overall mortality rates due to perioperative anaphylaxis in Europe are lacking. A large UK register will publish results on this issue in the near future. Mortality due to reactions involving neuromuscular blocking agents (NMBAs) - a major cause of perioperative anaphylaxis (vide infra) - has been estimated at 9% in the United Kingdom (16), and 4.1% in France (17). Taken into account the overall case fatality rate of anaphylaxis to be estimated at under % (18); it cannot be excluded these percentages could be an overestimation. Therefore, additional surveys concerning this issue are needed. 10

11 Table 2: Brown Criteria for Anaphylaxis Grade Symptoms Examples 1 Mild (skin and subcutaneous tissues only) Generalized erythema, urticaria, periorbital edema or angioedema 2 Moderate (features suggesting respiratory, cardiovascular, or gastrointestinal involvement) 3 Severe (hypoxia, hypotension, or neurologic compromise) Dyspnea, stridor, wheeze, nausea, vomiting, dizziness (presyncope), diaphoresis, chest or throat tightness, or abdominal pain Cyanosis or SpO2 # 92% at any stage, hypotension (SBP < 90 mmhg in adults) confusion, collapse, LOC, or incontinence Adapted from: Clinical features of and severity grading of anaphylaxis, Brown SG, 2004 (9) Diagnosis of perioperative hypersensitivity reactions: techniques History taking As a first measure, a profound history taking with in depth revision of the anesthetic and surgical charts is essential to correctly identify all the drugs and related compounds that were used in the perioperative period. Therefore, a close cooperation between referring anesthesiologists, surgeons, and the allergologists is necessary to orient the diagnostic approach. In this context we have developed, a standardised questionnaire that is available for anasthesiologists having witnessed anesthesiarelated hypersensitivity reaction (see supplement). This questionnaire comprises specific inquiries concerning applied products such as anesthetics, neuromuscular blocking agents, latex, antiseptics, antibiotics, hypnotics, analgesics, dyes, plasma expanders, iodinated contrast; time-of-onset of symptoms; clinical symptoms; acute treatment; dosage of peak tryptase and continuation of surgery. Drug provocation testing According to the literature, the gold standard test to correctly diagnose an IDHR remains the controlled drug provocation test (DPT). However, because drug provocation tests entail a significant risk of severe, life-threatening complications, they can simply be (relatively) contraindicated. Furthermore, these tests are not absolutely predictive with a risk for false negative results, particularly when fulldose provocations are impossible (e.g. drug provocation tests with curarizing neuromuscular blocking agents). Therefore, other diagnostic techniques are used in the search for the culprit compound(s) (19). These techniques comprise both in vivo and in vitro allergy testing for all compounds (drugs and related products) the patient was exposed to in the perioperative period. 11

12 Skin testing Generally skin prick testing (SPT) and intradermal testing (IDT), constitute the primary step in the diagnostic approach/work-up. Several guidelines concerning the optimal technique and concentrations have been published over time (19 24). Briefly, skin prick tests are performed on the ventral part of the forearm and imply histamine 10mg/mL as a positive control to assess skin test reactivity, a saline buffer solution as a negative control to exclude cutaneous hyperreactivity, and the involved drugs. SPT are read after 15 minutes and are considered positive when the wheal reaction exceeds 3mm with a flare. When appropriate, patients with negative SPT receive additional IDT. IDT responses are considered positive when the wheal equals or exceeds 5 mm or is doubled as compared to the injection bleb. For curares a tresholdof 8 mm is required (19). Hitherto, positive (PPV) and negative predicative value (NPV) of skin testing has not always been established since this would require additional drug provocation testing. Concerning the PPV, from ethical point of view a provocation test is not justified in patients with a positive skin test (25). Therefore, the current approach is to determine a maximal non-irritating concentration by using at least 20 exposed controls. In chapter two, we used this approach to validate and optimize the skin test concentration for cefazolin (26). To assess the negative predictive value a provocation test is mandatory. However, as already outlined above, this is often contraindicated in patients with a perioperative anaphylaxis due to several practical and ethical reasons. Drug-sIgE measurement To date, quantification of drug sige antibodies generally relies upon the fluorescent enzyme immune assay (FEIA) ImmunoCAP Thermofisher, Uppsala Sweden (27). This technique is used according to the manufacturer s recommendations. Currently, only a limited number of drug-sige immunoassays is available. These comprise the drug-sige assays that are commercially available from Thermofisher namely penicilloyl G (c1), penicilloyl V (c2), ampicilloyl (c5) and amoxicilloyl (c6) determinants, cephaclor (c7), chlorhexidine (c8, antiseptic), chymopapain (c209), (bovine) gelatin (c74), human (c73), bovine (c71) and porcine (c70) insulin, morphine (c260, marker for sensitization to tertiary and quaternary substituted ammonium determinants), pholcodine (c261) and suxamethonium (c202). In addition, and for research purposes only, adrenocorticotropic hormone, atracurium, bacitracin, carboplatin, cefamandole, cefoxitin, cefotaxime, cefuroxime, cisplatinum, mepivacaine, methylprednisolone-21-succinate, nafamostat (4-guanidinobenzoic acid), oxaliplatin, penicillin minor determinants (e.g., penicillanyl), propyphenazone, protamine, rocuronium, and tetanus toxoid are offered via the Thermofisher Scientific special allergen service (28). Unfortunately, the large majority 12

13 of these available tests are not yet thoroughly been clinically validated. This is mainly due to the unavailability of sufficient numbers of accurately phenotyped patients and exposed control individuals. Mediator release tests and the basophil activation test Activation of basophils and mast cells leads to the degranulation and release of several mediators that will give rise to the clinical symptoms of immediate hypersensitivity reactions. This activation can follow an IgE/FcεRI pathway, in which cross linking of the surface bound highaffinity IgE receptor (FcεRI) by the allergen will trigger degranulation. Alternatively, as already exemplified higher, activation of these cells may also result from different pathways independent from IgE/FcεRI cross-linking, that cannot be detected by a sige antibody assay. Therefore, in an attempt to more adequately mimic the in vivo situation, various mediator release assays have been developed and described since the 1960s. These in vitro tests measure the amount of mediators (histamine, tryptase, leukotrienes and others) in the supernatant at baseline, or after stimulation with an allergen (29). Unfortunately, these assays need a two-step approach. In a first phase there is need for cell isolation and incubation. Secondly, the quantification of released mediators is performed. Due to this time-and cost consuming character, technical difficulties (e.g. quantification of low concentrations of highly unstable mediators) and the lack of a well standardized technique, these assays are to date, only seldom used as a mainstream diagnostic aid (30). Actually, in the meantime, the basophil activation test (BAT) that is described below has largely supplanted these mediator release techniques. The foundations of current flow-assisted BAT were laid a quarter of a century ago (31). The principles and requirements of BAT have been detailed extensively elsewhere (32). Traditional BAT relies on a flow cytometric analysis of activation and degranulation markers on the surface membrane. These changes can be detected and quantified on a single-cell level using specific monoclonal antibodies conjugated with different LASER-excitable fluorochromes. Basophils are commonly identified by surface markers such as CCR3 (CD193)/CD3, CD123/HLA-DR or IgE/CD203c. Only CD203c, the ectonucleotide pyrophosphatase/phosphodiesterase family member 3 enzyme (E-NPP3), is lineagespecific. After activation, the appearance and/or upregulation of surface activation and/ or degranulation markers, such as CD63 and/or CD203c, is quantified. Furthermore, histamine release can also be quantified at single cell level by flow cytometry (33). Recently, this technique has been shown to be applicable in the diagnosis of drug hypersensitivity (34). 13

14 Figure 2: principle of the basophil activation test. The allergen of interest is added to a fresh blood sample of the patient. When the allergen is able to cross-link IgE/FcεRI complexes bound on the surface of the quiescent basophils (upper mid panel), activation of the basophil occurs (upper right panel). This activation leads to an up-regulation of activation marker CD203c and can culminate in anaphylactic degranulation with the appearance of CD63 (bottom left panel). The HistaFlow technique enables to quantify intracellular histamine content and its release using the histaminase d-amine oxidase that is conjugated to a fluorochromes (33). As displayed in the bottom right panel histamine release is restricted to the cells demonstrating an appearance of CD63. Diagnosis of perioperative hypersensitivity reactions: drugs Neuromuscular blocking agents (NMBA) Neuromuscular blocking agents are quaternary ammonium compounds used in general anesthesia to obtain muscle relaxation since 1942 (35). NMBA can be divided into suxamethonium; the aminosteroids rocuronium and vecuronium and the benzylisoquinolines atracurium, cisatracurium and mivacurium. Concerning the allergenicity, it is assumed that the tertiary and quaternary substituted ammonium ions, present on all NMBA, are the major but not the sole allergenic epitope (36). Cross reactivity between curares has long been studied using IgE binding and IgE binding-inhibition studies. These studies have led to varying and often very high frequencies of cross-reactivity (37). However, to date, this finding is not endorsed by other studies and the overall clinical experience. Our 14

15 study group showed that sige towards rocuronium could be inhibited with other curares while clinically patients had a good tolerance towards these other NMBA (38). Furthermore, in our experience, many patients displaying a hypersensitivity towards rocuronium show negative investigations to the benzylisoquinolines (39). In many countries neuromuscular blocking agents represent an important cause of perioperative anaphylaxis (12). As a first step, skin testing is used to establish a diagnosis of immune mediated drug hypersensitivity reactions towards NMBA (19). Since these skin tests do not display an absolute predictive value, additional in vitro tests are frequently needed. Unfortunately, as addressed above, for most of these drugs no sige assay is readily available. For this reason, in the past decades several attempts have been undertaken to validate home-made sige assays (40 43). To date, tertiary and quaternary ammonium structures have been demonstrated to constitute the major allergenic epitopes of the NMBA (44,45). Therefore, different methods measuring sige towards these structures have been developed to document immediate hypersensitivity reactions to NMBA. Three techniques are applied, viz. choline chloride based assays (40,41,46 48), p-amionophenyl phosphoryl cholin (PAPPC) assay (40,46,47,49), and a morphine-based assays (40 42,49 53). Available ImmunoCAP FEIA assays from Thermofisher Scientific are those for suxamethonium, rocuronium (research allergen), pholcodine and morphine. As shown in table 3, sensitivity and specificity for these assays varies between 38.5%-92% and 85.7%-100%. Importantly, sige antibodies towards these quaternary or tertiary substituted ammonium structures can remain positive for several years after the acute reaction (54). Our research group established positivity thresholds for these assays using ROC analysis. A threshold of 0.13 kua/l for rocuronium, 0.11 kua/l for suxamethonium, 0.36 kua/l for morphine, and 0.43 kua/l for pholcodine was obtained, leading to a sensitivity 92, 72, 88, and 86%, respectively (52). Of note is that IgE reactivity to tertiary and quaternary substituted ammonium structures seems to be prevalent in the general population (51 53). Therefore, quantification of sige towards these structures cannot be used as a screening tool to depict immediate hypersensitivity towards NMBA (39). Whether IHR to rocuronium might be IgE independent but related to MRGPRX2 mediated activation of mast cells, as suggested by Spoerl et al (7), remains to be confirmed. In the next chapter of this thesis, the sige atracurium ImmunoCAP available from ThermoFisher for research purposes only is clinically validated by our research group (55). 15

16 Table 3: Specific IgE to NMBA and quaternary and tertiary substituted ammonium structures Compound Reference test Various NMBA H + ST RIA RIA RIA Various NMBA H + ST RIA RAST RAST Various NMBA H + ST RIA RIA Various NMBA H + ST CAP-FEIA CAP-FEIA Assay Sensitivity Specificity N Reference PAPPC: 97% MOR: 83% QAS: 86% QAS: 87.9% SUC: 66.7% Alcuronium: 40.7% MOR: 85% NMBA-specific: 52% SUX: 38.5% MOR: 67.7% Rocuronium 1 H + ST CAP-FEIA SUX: 72% 2 SUX: 60% 3 ROC: 92% 2 ROC: 68% 3 MOR: 88% PHOL: 86% PAPPC: 97% NA NA 75 (44) NA 83 (45) 98% 118 (57) SUX: % MOR: 90-95% SUX: 100% 2 SUX: ROC: 93% 2 ROC: 93% 3 MOR: 100% PHOL: 100% 866 (58) 82 (59) Various NMBA 1 H + ST CAP-FEIA QAS 4 : 87.7% QAS 4 : 90.7% 168 (61) Atracurium H+ST CAP-FEIA SUX: 28.6% ATR: 57.1% MOR: 14.2% SUX: 85.7% ATR: 100% MOR: 85.7% 78 (55) 1 applying ROC-generated drug-specific thresholds, 2 for a ROC-generated threshold of 0.11 kua/l for suxamethonium and 0.13 kua/l for rocuronium, 3 for a traditional threshold of 0.35 kua/l. 4 unclearly defined optimized morphine-based assay H: history, ST: skin tests, RIA: radio immunoassay, RAST: radio allergosorbent test, CAP-FEIA: fluorescence enzyme immunoassay available from Phadia Thermofisher, PAPPC: p-aminophenyl phosphoryl choline, MOR: morphine, QAS: quaternary ammonium structure, SUC: succinyl choline, SUX: suxamethonium, ROCU: rocuronium, QAM: quaternary ammonium morphine, N: number. NA: not available. During the past 10 years, the BAT has increasingly been used in the diagnosis of immediate hypersensitivity towards NMBA. In general, sensitivity of the technique varies between 36% and 92%, whereas specificity reaches 95% (table 4). In chapter 3 the BAT for atracurium was validated. An important advantage of the BAT, is that it not only detects the culprit drug, but also allows to study cross reactivity between different NMBAs. This way, a safe alternative for future surgeries can be identified (39). 16

17 Table 4: BAT in immediate neuromuscular blocking agent (NMBA) hypersensitivity Stimulus Reference Test Activation marker Various NMBA H CD63 CD45 Sensitivity (%) Specificity (%) N Ref. 26 (109) Various NMBA H + ST CD (110) Various NMBA H CD63 CD203c (111) Various NMBA H + ST CD (112) Rocuronium H + ST CD (65) Various NMBA H + ST + IgE CD (113) Rocuronium H CD (62) Various NMBA H + ST CD (114) Atracurium H ± ST CD (56) NMBA: neuromuscular blocking agent, H: history, ST: skin test, N: number of patients and control individuals 1 Increasing sensitivity when only the reactions that occurred during the 3 years were taken into account, 2 taking into account the non-responders sensitivity is 76%, taking into account the non-responders sensitivity is 63%. 17

18 β lactam antibiotics sige assays for β-lactams antibiotics especially amoxicillin and benzyl penicilloyl are thoroughly studied. However, in clinical practice, they seem to be of rather limited use. Firstly, these assays generally express a low sensitivity (table 5). Moreover, this sensitivity has the tendency to decrease over time (57). Despite this low sensitivity, there are several cases of skin test negative patients with a positive sige described in IHR (58 62). In addition to this low sensitivity, specificity of these tests is not absolute, with a wide spread ranging from 52% to 100% (59,61,63 66). This could be due to nonspecific binding to the solid phase assay in patients with a high total IgE (64,66). On the other hand, it has been shown that in a quarter of the patients with a positive sige for penicillin the positivity resulted from clinically irrelevant sige antibodies to phenylethylamine (67). Therefore, as advised by the European Network of Drug Allergy (ENDA) (4), sige measurement should not be used in isolation to diagnose IDHR to β-lactam antibiotics but should be preceded/supplemented by skin testing and in some cases, the basophil activation test, who can be of important value (68). 18

19 Table 5: Specific IgE to β-lactams Compound Various β-lactams Various β-lactams Various β-lactams Various β-lactams Reference test H + ST H + ST + DPT H + ST + DPT H Assay Sensitivity Specificity N Reference FEIA FEIA FEIA CAP- CAP- CAP- CAP- FEIA Various H + ST + DPT CAPβ-lactams 1 FEIA Various β-lactams Various β-lactams H + ST H + ST RAST 2 FEIA FEIA CAP- CAP- CAP- FEIA BPO + AXO + peni G + AMP: 31.8% BPO: 32% AXO: 43% BPO+AXO: 50% BPO: 10-68% AXO: 41-53% BPO + AXO + peni G + AMP: 88.6% BPO: 98% AXO: 98% BPO+AXO: 96% BPO: 98% AXO: 95% 58 (98) 129 (99) 410 (26) 37.9% 86.7% 58 (100) 0-25% %2 45 (22) % % 2 85% 3 54%3 44% 4 80% (30) 66% 52% 293 (31) 1 home-made assay, 2 sensitivity and specificity vary according to clinical manifestations, 3 for a threshold of 0.10 kua/l, 4 for a threshold of 0.35 kua/l. H: history, ST: skin test, DPT: drug provocation test, N: number CAP-FEIA: fluorescence enzyme immunoassay available from Phadia Thermofisher. RAST: radio allergo sorbent test. Peni G: penicillin G, AMP: ampicillin, BPO: benzyl penicilloyl, AXO: amoxicillin 19

20 Beside our recent study on cefazolin 9 studies investigated BAT for β-lactam antibiotics (see table 6). The majority of studies focused on the BAT for amoxicillin. As displayed in table 6, the BAT shows a discrete superior sensitivity (approximately 50%) and specificity (approximately 95%) than sige for these drugs. In chapter 3 of this thesis, we have developed and validated the BAT for cefazolin (69). As for specific IgE, sensitivity of BAT to β-lactams is rather low and decreases over time but both tests can remain positive for years (57,69). This is important, since also skin test responsiveness to β-lactams decreases over time (70,71). However, as will emerge from our epidemiological survey, a prolonged interval between reaction and investigations does certainly not preclude testing for β-lactam antibiotics (chapter 4). Table 6: BAT in immediate β-lactam hypersensitivity Stimulus Reference test Activation marker Sensitivity (%) Specificity (%) Number of patients and controls β-lactam H CD (100) β-lactam H + DPT 1 CD (101) β-lactam H + ST + IgE + DPT CD (102) Amoxicillin H + ST CD203c CD Ref. 41 (103) β-lactam H CD (104) β-lactam H + ST + IgE CD63-CCR3 CD63-IgE (105) Amoxicillin H CD63 29 / 14 patients, no controls Amoxicillin H + ST + DPT CD63 50 / 61 patients, number of controls not mentioned Amoxicillin H + ST CD63 50 / 30 patients (108) (106) (107) Cefazolin H + ST CD63 CD203c H: history, ST: skin test, DPT: drug provocation test, IgE: immunoglobulin E patients, 17 controls (69) Natural rubber latex In many countries, natural rubber latex is still an important cause of perioperative anaphylaxis. Hitherto, diagnosis of latex allergy mainly relies upon skin testing and/or quantification of sige towards latex. However, in some cases, the diagnosis is hampered, mainly because of false-positive sige results (72,73), especially in patients suffering from grass and weed pollen allergy who are sensitized to crossreactive carbohydrate determinants and/or profilin. Therefore, in a significant number of patients additional tests such as component resolved diagnosis (74 77) and basophil activation test (73,78 81) are needed. 20

21 Chlorhexidine Chlorhexidine, a cationic bisguanide antiseptic and disinfectant, is used as the (di)acetate or (di)glucuronide salt and is a rising cause of perioperative anaphylaxis in certain countries. This disinfectant can trigger irritant dermatitis, allergic contact dermatitis (82), IDHR (including lifethreatening anaphylaxis) (83 86), or a combination of both. Diagnosis of IDHR to chlorhexidine is mainly based on skin testing and quantification of sige toward this antiseptic (85). Traditionally, the threshold for positivity of this assay was arbitrarily set at 0.35 kua/l. For this threshold, the sensitivity of sige chlorhexidine varied between % and specificity between %. However, using a ROC-generated threshold of 0.20 kua/l sensitivity increases up to 94.1% and specificity remains good (90.7%) (85,86). Importantly, the assay is influenced by a raised total IgE level above 500 kua/l and more particularly at levels higher than 2,000 kua/l (86). As for timing of sampling for sige chlorhexidine, it was shown there is an optimal timeframe between 1 and 4 months (87). However, sige to chlorhexidine might persist for years (88). As shown by Ebo et al (83), in some cases, the basophil activation test can also be of help in the diagnosis of IDHR to chlorhexidine. Opiates Although frequently used, genuine IgE-mediated allergy to opiates (morphine, codeine, pholcodine) is rare. Because of difficulties and uncertainties concerning quantification of drug-specific IgE antibodies and skin testing, diagnosis of this rare phenomenon is not always easy (89). In a recent study, it has been suggested that the two commercially available sige assays for a Papaver somniferum (poppy seed) extract and morphine can add to the diagnosis of IgE-mediated opiate allergy (90,91). However, in our own research group, using drug provocation tests we failed to confirm these data (92), mainly because of the high prevalence of sige antibodies to these compounds in an allergic population. This observation is highly relevant in perioperative reactions, since correct identification of the causative compound(s) in this setting is hampered because of simultaneous intake or administration of different agents. Incorrect opiate allergy diagnosis has important consequences. First of all, it leads to unnecessary avoidance measures but also, it puts patients at risk by overlooking other possible diagnoses such as an allergy to rocuronium or suxamethonium. Therefore, diagnostic work up should also comprise of the BAT. Recently, we observed that the opiates morphine and codeine do not trigger basophil degranulation in opiate-tolerant individuals, who tested positive in a solid phase morphine SgE antibody assay and could therefore erroneously have been diagnosed as opiate-allergic (92). Therefore, up to now, the sole in vitro method to accurately document opiate allergy is the BAT, as basophils, unlike cutaneous mast cells, are unresponsive to opiates (92,93). Moreover, negative BAT, 21

22 along with negative skin testing for different NMBA and negative provocation tests for the structurally almost similar opiates suggest these drugs probably to be safe in pholcodine hypersensitivity (93). Miscellaneous Bovine gelatin constitutes the active component in certain plasma substitutes, hemostatic sponges and can be present in various other drugs such as vaccines. Since the first descriptions of the allergenicity of gelatin (94), IgE/FcεRI-mediated IHR to this compound, including fatal anaphylaxis, have been increasingly reported. Today, 2 distinct types of IgE/FcεRI-mediated bovine gelatin allergy have been described. First, genuine gelatin allergy that results from sensitization to the protein part of the molecule. Second, gelatin allergy resulting from a sensitization to a glycan moiety of the molecule, i.e. galactose-α(1,3)-galactose (α-gal) (95 97), as first described by Chung et al (98) and Commins et al (99). To our knowledge, there are no studies that have determined the diagnostic accuracy of sige gelatin. However, it is of note that patients with life-threatening anaphylaxis to gelatin as a result of α- gal sensitization are generally overlooked by traditional gelatin-sige assay and need additional testing including quantification of α-gal sige antibodies and gelatin skin testing (95 97). Epidemiology of perioperative hypersensitivity reactions Despite standardized testing and clear guidelines concerning diagnostic procedures for perioperative anaphylaxis (19,21 24), up to now, the culprit compound(s) is/are identified in only about 60% of the cases with alleged anesthesia related anaphylaxis (table 7). This stresses the need for improvement of the diagnostic approach and the optimisation of currently available diagnostic tools. In this thesis, we focused on the optimisation of the diagnosis of IHR to atracurium (a benzylisoquinoline NMBA) on the one hand (chapter 2) and cefazolin (a first generation cephalosporin) on the other hand (chapter 3). 22

23 Table 7: Epidemiological studies performed in the past 20 years (adapted from: Mertes et al, (12)) Publica tion date Countr y Inclusion period 1996 France France Denma rk France France France France Norwa y Spain Austral ia France USA France Belgiu m Belgiu m 2011 N Diagnos is made (%) Unkno wn (%) Causal agent (n 1) (%) Causal agent (n 2) (%) Causal agent (n 3) (%) Ref NMBA NRL (19) Hypnotic (100) 0 (59.2) s (6) NMBA Latex AB (8.3) (101) 8 (61.6) (16.6) Chlorhexidi ne (22) Antibiotics (14) NRL (2) (102) 467 / / NMBA (69) Latex (12) AB (16) (46) NMBA Latex AB (47) (58.2) (16.7) (15.1) NMBA (55) Latex AB (103) (22.3) (14.7) 68 * 75 25* NMBA (60) NRL (27) Colloids (104) (7) NMBA (93) NRL (5) Colloids (2) (49) AB (44) NMBA Pyrazolo (105) (37) nes (7) NRL (7) NMBA (68) Opioids AB (8) (106) (12) NMBA (58) NRL (20) AB (13) (48) AB (50) Induction drugs (16) NMBA (11) (107) NMBA (47) NRL (20) AB (18) (108) NMBA NRL AB (88) NMBA AB (14.5) Latex (109) (61.8) (9.2) 23

24 Publica Countr Inclusion N Diagnos Unkno Causal Causal Causal Ref tion date y period is made (%) wn (%) agent (n 1) (%) agent (n 2) (%) agent (n 3) (%) 2014 UK NMBA (38) AB (8) Blue (110) 2012 dyes (6) 2015 Italy Latex AB NMBA (111) USA AB/Latex/o NMBA Tetracai (112) 2012 ndansetron (23 each) (15) ne /Fentany l (7 each) 2015 USA AB (59) Latex (18) NMBA (23) (113) AB: antibiotics, N: number of patients included in the study; NMBA: neuromuscular blocking agents; NRL; natural rubber latex *Only children included in the study. cumulative data. % calculated relative to entire patient group (opposed to % calculated relative to patients with a known culprit drug) For many years, large multicentre studies from France have identified the curarizing NMBA, natural rubber latex, antibiotics and hypnotics to be the most important causes of IgE/FcεRI-mediated anaphylaxis during surgery (108). However, as displayed in table 6, it has emerged perioperative anaphylaxis in terms of offending compounds to display distinct geographical differences (12). For instance, in Denmark, chlorhexidine seems to be an important culprit (85) whereas, for the time-being, in France, there is no mentioning of this antiseptic as a potential cause of perioperative hypersensitivity reactions. Furthermore, a shift in causes over time has been observed with the rise of some new and sometimes hidden allergens (for example patent blue and gelatine) (114). The exact reasons for these geographical differences remain elusive but could relate to a large number of variables. Among other factors these include: the ability to correctly recognize a reaction and consequently appropriately refer the patient, the severity of the reactions studied, the market share of the employed agents, the comprehensiveness of the evaluation, possible sensitizing substances in a region and the availability of in vitro testing to medications used or even differences in diagnostic techniques (e.g. concentrations used for skin testing) (12). Chapter 4 of this thesis is an epidemiological survey on perioperative hypersensitivity comprising 626 patients referred to the Department of Immunology in the University hospital Antwerp between January 2001 and November With this study we aimed at providing information concerning demographics, clinics, causes and treatment, of perioperative anaphylaxis in Flanders. 24

25 Part 2: Aims of the thesis The overarching aim of this thesis was the further optimisation of the diagnosis of perioperative hypersensitivity reactions. This has led to following aims: Chapter 2 To improve the diagnosis of immediate hypersensitivity reactions to atracurium by means of 2 novel in vitro tests: o validation of the basophil activation test (BAT) for this curare, and o validation of the modified tetrahydropapaverine ImmunoCAP assay engineered to detect serum sige antibodies to atracurium. Chapter 3 to optimise the diagnosis of immediate hypersensitivity reaction to cefazolin by means of: o validation of an adapted skin test protocol and o validation of the basophil activation test for this antibiotic Chapter 4 To report the epidemiological data of patients being investigated for alleged perioperative hypersensitivity reactions referred between 2001 and

26 References International drug monitoring. The role of the hospital. World Health Organ Tech Rep Ser. 1969;425:5 24. Johansson SGO, Bieber T, Dahl R, Friedmann PS, Lanier BQ, Lockey RF, et al. Revised nomenclature for allergy for global use: Report of the Nomenclature Review Committee of the World Allergy Organization, October J Allergy Clin Immunol May;113(5): Doña I, Barrionuevo E, Torres MJ, Fernandez TD, Mayorga C, Canto G, et al. Trends in Hypersensitivity Drug Reactions: More drugs, More Response Patterns, More Heterogeneity. J Investig Allergol Clin Immunol. 2014;24(3): Demoly P, Adkinson NF, Brockow K, Castells M, Chiriac a M, Greenberger P a, et al. International Consensus on drug allergy. Allergy. 2014;69(4): Rajan T V. The Gell-Coombs classification of hypersensitivity reactions: A re-interpretation. Trends Immunol. 2003;24(7): McNeil BD, Pundir P, Meeker S, Han L, Undem BJ, Kulka M, et al. Identification of a mastcell-specific receptor crucial for pseudo-allergic drug reactions. Nature. 2015;519(7542): Spoerl D, D incau S, Roux-Lombard P, Harr T, Czarnetzki C. Non-IgE-Dependent hypersensitivity to rocuronium reversed by sugammadex: Report of three cases and hypothesis on the underlying mechanism. Int Arch Allergy Immunol. 2016;169(4): Baldo BA, Fisher MM, Pham NH. On the origin and specificity of antibodies to neuromuscular blocking (muscle relaxant) drugs: an immunochemical perspective. Clin Exp Allergy Mar;39(3): Brown SG. Clinical features and severity grading of anaphylaxis. J Allergy Clin Immunol. 2004;114(2): Ebo DG, Fisher MM, Hagendorens MM, Bridts CH, Stevens WJ. Anaphylaxis during anaesthesia: diagnostic approach. Allergy. 2007;62(5): Krøigaard M, Garvey LH, Menné T, Husum B, Kroigaard M, Garvey LH, et al. Allergic reactions in anaesthesia: Are suspected causes confirmed on subsequent testing? Br J Anaesth Aug 12;95(4): Mertes PM, Volcheck GW, Garvey LH, Takazawa T, Platt PR, Guttormsen AB, et al. Epidemiology of perioperative anaphylaxis. Presse Med. 2016;45(9): Hepner DL, Castells MC. Anaphylaxis during the perioperative period. Anesth Analg. 2003;97: Mitsuhata H, Hasegawa J, Matsumoto S, Ogawa R. The epidemiology and clinical features of anaphylactic and anaphylactoid reactions in the perioperative period in Japan: a survey with a questionnaire of 529 hospitals approved by Japan Society of Anesthesiology; Masui. 1992;41(11): Gibbs NM, Sadleir PH, Clarke RC, Platt PR. Survival from perioperative anaphylaxis in Western Australia Br J Anaesth. 2013;111(4): Light KP, Lovell AT, Butt H, Fauvel NJ, Holdcroft A. Adverse effects of neuromuscular blocking agents based on yellow card reporting in th U.K.: Are there differences between males and females? Pharmacoepidemiol Drug Saf. 2006;15(3):

27 Reitter M, Petitpain N, Latarche C, Cottin J, Massy N, Demoly P, et al. Fatal anaphylaxis with neuromuscular blocking agents: A risk factor and management analysis. Allergy Jul;69(7): Panesar SS, Javad S, de Silva D, Nwaru BI, Hickstein L, Muraro A, et al. The epidemiology of anaphylaxis in Europe: a systematic review. Allergy Eur J Allergy Clin Immunol. 2013;68(11): Mertes PM, Jouffroy L, Working Group for the SFAR, SFA, Aberer W, Terreehorst I, et al. Reducing the Risk of Anaphylaxis During Anesthesia 2011 Updated Guidelines for Clinical Practice. J Investig Allergol Clin Immunol. 2011;21(6): Brockow K, Garvey LH, Aberer W, Atanaskovic-Markovic M, Barbaud A, Bilo MB, et al. Skin test concentrations for systemically administered drugs -- an ENDA/EAACI Drug Allergy Interest Group position paper. Allergy Jun;68(6): Ebo D, Hagendorens M, Bridts C, De Clerck L, Stevens W. Allergic reactions occuring during anaesthesia: diagnostic approach. Acta Clin Belg. 2004;59(1): Moscicki RA, Sockin SM, Corsello BF, Ostro MG, Bloch KJ. Anaphylaxis during induction of general anesthesia: subsequent evaluation and management. J Allergy Clin Immunol. 1990;86(3): Mertes PM, Laxenaire MC, Lienhart A, Aberer W, Ring J, Pichler WJ, et al. Reducing the risk of anaphylaxis during anaesthesia: guidelines for clinical practice. J Investig Allergol Clin Immunol. 2005;15(2): Société Françaises d Anesthésie et de Réanimation. Reducing the risk of anaphylaxis during anaesthesia. Abbreviated text. Ann Fr Anesth Reanim. 2002;21 Suppl 1:7s 23s. Park M a, Solensky R, Khan D a, Castells MC, Macy EM, Lang DM. Patients with positive skin test results to penicillin should not undergo penicillin or amoxicillin challenge. J Allergy Clin Immunol. American Academy of Allergy, Asthma & Immunology; 2014 Oct 28;1 2. Uyttebroek AP, Decuyper II, Bridts CH, Romano A, Hagendorens MM, Ebo DG, et al. Cefazolin Hypersensitivity: Toward Optimized Diagnosis. J Allergy Clin Immunol Pract. 2016;4(6):3 7. Hamilton RG. Clinical laboratory assessment of immediate-type hypersensitivity. J Allergy Clin Immunol. 2010;125(2):S Decuyper II, Ebo DG, Uyttebroek AP, Hagendorens MM, Faber MA, Bridts CH, et al. Quantification of specific IgE antibodies in immediate drug hypersensitivity: More shortcomings than potentials? Clin Chim Acta. 2016;460. Demoly P, Lebel B, Arnoux B. Allergen-induced mediator release tests. Allergy. 2003;58(7):553. Uyttebroek AP, Sabato V, Faber M a, Cop N, Bridts CH, Lapeere H, et al. Basophil activation tests: time for a reconsideration. Expert Rev Clin Immunol. 2014;10(10): Knol EF, Mul FPJ, Jansen H, Calafat J, Roos D, Knol, E F; Mul, F P J; Jansen, H; Calafat, J; Roos D. Monitoring human basophil activation via CD63 monoclonal antibody 435. J Allergy Clin Immunol. 1991;88(1): Bridts CH, Sabato V, Mertens C, Hagendorens MM, Clerck LS De, Ebo DG. Flow Cytometric Allergy Diagnosis: Basophil Activation Techniques. In: Gibbs BF, Falcone FH, editors. Basophils and Mast Cells: Methods and Protocols, Methods in Molecular Biology. New York, NY: Springer New York; p (Methods in Molecular Biology; vol. 1192). 27

28 Ebo DG, Bridts CH, Mertens CH, Hagendorens MM, Stevens WJ, De Clerck LS. Analyzing histamine release by flow cytometry (HistaFlow): a novel instrument to study the degranulation patterns of basophils. J Immunol Methods. 2012;375(1 2):30 8. Cop N, Uyttebroek AP, Sabato V, Bridts CH, De Clerck LS, Ebo DG. Flow cytometric analysis of drug-induced basophil histamine release. Cytom Part B Clin Cytom. 2015;90(3): Griffith H, Johnson G. The use of curare in general anesthesia. Anesth Analg. 1942;3: Baldo BA, Fisher MM. Substituted ammonium ions as allergenic determinants in drug allergy. Nature. 306(5940): Laxenaire MC, Gastin I, Moneret-Vautrin DA, Widmer S, Guéant JL. Cross-reactivity of rocuronium with other neuromuscular blocking agents. Eur J Anaesthesiol Suppl Sep;11: Ebo DG, Venemalm L, Bridts CH, Degerbeck F, Hagberg H, De Clerck LS, et al. Immunoglobulin E antibodies to rocuronium: a new diagnostic tool. Anesthesiology Aug;107(2): Leysen J, Uyttebroek A, Sabato V, Bridts C, De Clerck L, Ebo D. Predictive value of allergy tests for neuromuscular blocking agents: tackling an unmet need. Clin Exp allergy. 2014;44: Guilloux L, Ricard-Blum S, Ville G, Motin J. A new radioimmunoassay using a commercially available solid support for the detection of IgE antibodies against muscle relaxants. J Allergy Clin Immunol. 1992;90(2): Gueant JL, Mata E, Monin B, Moneret-Vautrin DA, Kamel L, Nicolas JP, et al. Evaluation of a new reactive solid phase for radioimmunoassay of serum specific IgE against muscle relaxant drugs. Allergy. 1991;46(6): Harle DG, Baldo BA, Fisher MM. Immunoassays employing substituted ammonium compounds other than neuromuscular blocking drugs to increase the detection of IgE antibodies to these drugs. Mol Immunol. 1990;27(10): Baldo BA, Fisher MM. Diagnosis of IgE-dependent anaphylaxis to neuromuscular blocking drugs, thiopentone and opioids. Ann Fr Anesth Reanim Jan;12(2): Didier A, Cador D, Bongrand P, Furstoss R, Fourneron P, Senft M, et al. Role of the quaternary ammonium ion determinants in allergy to muscle relaxants. J Allergy Clin Immunol. 1987;79(4): Baldo BA. Substituted amonium ions as allergenic determinants in drug allergy. Nature. 1983;306(5940): Laxenaire MC, Mertes PM, le Group d études des reactions anaphylactoïdes peranesthésiques. Anaphylaxis during anaesthesia. Results of a two-year survey in France. Br J Anaesth. 2001;87(4): Mertes PM, Laxenaire M-C, Alla F, le Group d études des reactions anaphylactoïdes peranesthésiques. Anaphylactic and anaphylactoid reactions occurring during anesthesia in France in Anesthesiology. 2003;99(3): Mertes PM, Alla F, Tréchot P, Auroy Y, Jougla E, le Group d études des reactions anaphylactoïdes peranesthésiques. Anaphylaxis during anesthesia in France: An 8-year national survey. J Allergy Clin Immunol. 2011;128(2): Harboe T, Guttormsen AB, Irgens A, Dybendal T, Florvaag E. Anaphylaxis during Anesthesia in Norway. Anesthesiology. 2005;102(5):

29 Fisher MM, Baldo BA. Immunoassays in the diagnosis of anaphylaxis to neuromuscular blocking drugs: the value of morphine for the detection of IgE antibodies in allergic subjects. Anaesth Intensive Care Apr;28(2): Florvaag E, Johansson SG, Oman H, Venemalm L, Degerbeck F, Dybendal T, et al. Prevalence of IgE antibodies to morphine. Relation to the high and low incidences of NMBA anaphylaxis in Norway and Sweden, respectively. Acta Anaesthesiol Scand. 2005;49(4): Ebo DG, Venemalm L, Bridts CH, Degerbeck F, Hagberg H, De Clerck LS, et al. Immunoglobulin E Antibodies to Rocuronium. Anaesthesiology. 2007;107(2): Laroche D. Evaluation of a new routine diagnostic test for immunogloulin e sensitization to neuromuscular blocking agents. Anesthesiology. 2011;114:91 7. Fisher MM, Baldo BA. Persistence of allergy to anaesthetic drugs. Anaesth Intensive Care. 1992;20(2): Uyttebroek AP, Sabato V, Bridts C, De Clerck L, Ebo D. Immunoglobulin E antibodies to atracurium: a new diagnostic tool. Clin Exp Allergy. 2014;45(2): Uyttebroek AP, Sabato V, Leysen J, Bridts CH, De Clerck LS, Ebo DG. Flowcytometric diagnosis of atracurium-induced anaphylaxis. Allergy Jun 24;69(10): Fernández TD, Torres MJ, Blanca-López N, Rodríguez-Bada JL, Gomez E, Canto G, et al. Negativization rates of IgE radioimmunoassay and basophil activation test in immediate reactions to penicillins. Allergy. 2009;64(2): Fontaine C, Mayorga C, Bousquet PJ, Arnoux B, Torres M-J, Blanca M, et al. Relevance of the determination of serum-specific IgE antibodies in the diagnosis of immediate betalactam allergy. Allergy. 2007;62(1): Silva R, Cruz L, Botelho C, Castro E, Cadinha S, Castel-Branco MG, et al. Immediate hypersensitivity to penicillins with negative skin tests--the value of specific IgE. Eur Ann Allergy Clin Immunol. 2009;41(4): Qiao H-L, Li Z, Yang J, Tian X, Gao N, Jia L-J. Hypersensitivity reactions to penicillins: studies in a group of patients with negative benzylpenicillin G skin test. J Clin Pharm Ther. 2009;34(3): Hjortlund J, Mortz CG, Skov PS, Bindslev-Jensen C. Diagnosis of penicillin allergy revisited: the value of case history, skin testing, specific IgE and prolonged challenge. Allergy. 2013;68(8): Torres MJ, Romano A, Mayorga C, Moya MC, Guzman AE, Reche M, et al. Diagnostic evaluation of a large group of patients with immediate allergy to penicillins: the role of skin testing. Allergy. 2001;56(9): Macy E, Goldberg B, Poon K-YT. Use of commercial anti penicillin IgE fluorometric enzyme immunoassays to diagnose penicillin allergy; Ann Allergy, Asthma Immunol. 2010;105(2): Aberer W, Zidarn M, Kranke B. IgE antibodies to penicillin are indicative for but not conclusive proof of penicillin allergy. Br J Dermatol. 2006;154(6): Vultaggio A, Virgili G, Gaeta F, Romano A, Maggi E, Matucci A. High Serum β-lactams Specific/Total IgE Ratio Is Associated with Immediate Reactions to β-lactams Antibiotics. PLoS One. 2015;10(4):e Vultaggio A, Matucci A, Virgili G, Rossi O, Filì L, Parronchi P, et al. Influence of total serum IgE levels on the in vitro detection of β-lactams-specific IgE antibodies. Clin Exp Allergy. 2009;39(6):

30 Johansson SGO, Adédoyin J, Van Hage M, Grönneberg R, Nopp A. False-positive penicillin immunoassay: An unnoticed common problem. J Allergy Clin Immunol. 2013;132(1): Blanca M, Romano A, Torres MJ, Férnandez J, Mayorga C, Rodriguez J, et al. Update on the evaluation of hypersensitivity reactions to betalactams. Allergy. 2009;64(2): Uyttebroek AP, Sabato V, Cop N, Decuyper II, Faber MA, Bridts CH, et al. Diagnosing cefazolin hypersensitivity: Lessons from dual-labeling flow cytometry. J Allergy Clin Immunol Pract. 2016;4(6): Blanca M, Torres MJ, Garcia JJ, Romano A, Mayorga C, de Ramon E, et al. Natural evolution of skin test sensitivity in patients allergic to beta-lactam antibiotics. J Allergy Clin Immunol. 1999;103(5 Pt 1): Romano A, Gaeta F, Valluzzi RL, Zaffiro A, Caruso C, Quaratino D. Natural evolution of skintest sensitivity in patients with IgE-mediated hypersensitivity to cephalosporins. Allergy. 2014;69(6): Ebo DG, Stevens WJ, Bridts CH, De Clerck LS. Latex-specific IgE, skin testing, and lymphocyte transformation to latex in latex allergy. J Allergy Clin Immunol. 1997;100(5): Ebo DG, Lechkar B, Schuerwegh AJ, Bridts CH, De Clerck LS, Stevens WJ. Validation of a twocolor flow cytometric assay detecting in vitro basophil activation for the diagnosis of IgEmediated natural rubber latex allergy. Allergy. 2002;57(8): Quercia O, Stefanini GF, Scardovi A, Asero R. Patients monosensitised to Hev b 8 (Hevea brasiliensis latex profilin) may safely undergo major surgery in a normal (non-latex safe) environment. Eur Ann Allergy Clin Immunol. 2009;41(4): Garnier L, Selman L, Rouzaire P, Bouvier M, Roberts O, Bérard F, et al. Molecular allergens in the diagnosis of latex allergy. Eur Ann Allergy Clin Immunol. 2012;44(2):73 9. Ebo DG, Hagendorens MM, De Knop KJ, Verweij MM, Bridts CH, De Clerck LS, et al. Component-resolved diagnosis from latex allergy by microarray. Clin Exp Allergy. 2010;40(2): Schuler S, Ferrari G, Schmid-Grendelmeier P, Harr T. Microarray-based componentresolved diagnosis of latex allergy: isolated IgE-mediated sensitization to latexprofilin Hev b8 may act as confounder. Clin Transl Allergy. 2013;3(1):11. Sanz ML, Gamboa PM, García-Avilés C, Vila L, Diéguez I, Antépara I, et al. Flow-cytometric cellular allergen stimulation test in latex allergy. Int Arch Allergy Immunol. 2003;130(1):33 9. Nettis E, Colanardi MC, Dambra PP, Capuzzimati L, Loria MP, Ferrannini A, et al. Flow cytometric basophil activation test: detection of CD63 expression as a useful aid to diagnosis of latex allergy. Ann Allergy Asthma Immunol. 2006;97(5): Sanz ML, Garcia-Aviles MC, Tabar AI, Anda M, Garcia BE, Barber D, et al. Basophil Activation Test and specific IgE measurements using a panel of recombinant natural rubber latex allergens to determine the latex allergen sensitization profile in children. Pediatr Allergy Immunol. 2006;17(2): Faber MA, Sabato V, Bridts CH, Nayak A, Beezhold DH, Ebo DG. Clinical relevance of the Hevea brasiliensis lipid transfer protein Hev b 12. J Allergy Clin Immunol. 2015;135(6): e1. Opstrup MS, Johansen JD, Zachariae C, Garvey LH. Contact allergy to chlorhexidine in a tertiary dermatology clinic in Denmark. Contact Dermatitis. 2016;74(1):

31 Ebo, D G; Bridts, C H; Stevens WJ. IgE-mediated anaphylaxis from chlorhexidine: diagnostic possibilities. Contact Dermatitis. 2006;55: Faber M, Leysen J, Bridts C, Sabato V, De Clerck LS, Ebo DG. Allergy to chlorhexidine: beware of the central venous catheter. Acta Anaesthesiol Belg. 2012;63(4): Opstrup MS, Malling HJ, Krøigaard M, Mosbech H, Skov PS, Poulsen LK, et al. Standardized testing with chlorhexidine in perioperative allergy - a large single-centre evaluation. Allergy Eur J Allergy Clin Immunol. 2014;69: Anderson J, Rose M, Green S, Fernando SL. The utility of specific IgE testing to chlorhexidine in the investigation of perioperative adverse reactions. Ann Allergy, Asthma Immunol. 2015;114(5): e1. Opstrup MS, Poulsen LK, Malling HJ, Jensen BM, Garvey LH. Dynamics of plasma levels of specific IgE in chlorhexidine allergic patients with and without accidental re-exposure. Clin Exp Allergy. 2016;46(8): Leysen J, Witte L De, Bridts CH, Ebo DG. Anaphylaxis during general anaesthesia : a 10-year survey 1 at the University Hospital of Antwerp. P Belg Roy Acad Med. 2013;2: Baldo BA, Pham NH. Histamine-releasing and allergenic properties of opioid analgesic drugs: resolving the two. Anaesth Intensive Care. 2012;40(2): Armentia A, Ruiz-Munoz P, Quesada JM, Postigo I, Herrero M, Martin-Gil FJ, et al. Clinical value of morphine, pholcodine and poppy seed IgE assays in drug-abusers and allergic people. Allergol Immunopathol (Madr). 2013;41(1): Armentia A, Pineda F, Palacios R, Martin-Gil FJ, Miguel AS, Arenal JJ, et al. Utility of opium seed extract tests in preventing hypersensitivity reactions during surgery. Allergol Immunopathol (Madr). 2014;42(1): Van Gasse AL, Hagendorens MM, Sabato V, Bridts CH, De Clerck LS, Ebo DG. IgE to Poppy Seed and Morphine Are Not Useful Tools to Diagnose Opiate Allergy. J Allergy Clin Immunol Pract. 2015;3(3): Leysen J, De Witte L, Sabato V, Faber M, Hagendorens M, Bridts C, et al. IgE-mediated allergy to pholcodine and cross-reactivity to neuromuscular blocking agents: Lessons from flow cytometry. Cytometry B Clin Cytom. 2013;84(2): Ratner B, Crawford L V. The allergenicity of gelatin. Int Arch Allergy Appl Immunol. 1955;6(4 6): Mullins RJ, James H, Platts-Mills TAE, Commins S. Relationship between red meat allergy and sensitization to gelatin and galactose-α-1,3-galactose. J Allergy Clin Immunol. 2012;129(5): e1. Ebo D, Faber M, Sabato V, Leysen J, Gadisseur A, Ch B, et al. Sensitization to the mammalian oligosaccharide galactose-a-1,3-galactose (a-gal): experience in a Flemish case series. Acta Clin Belg. 2013;68(3): Uyttebroek A, Sabato V, Bridts CH, De Clerck LS, Ebo DG. Anaphylaxis to succinylated gelatin in a patient with a meat allergy: Galactose-α(1, 3)-galactose (α-gal) as antigenic determinant. J Clin Anesth. 2014;26(7): Chung CH, Mirakhur B, Chan E, Le Q-T, Berlin J, Morse M, et al. Cetuximab-induced anaphylaxis and IgE specific for galactose-alpha-1,3-galactose. N Engl J Med. 2008;358(11): Commins SP, Platts-Mills T a E. Delayed anaphylaxis to red meat in patients with IgE specific for galactose alpha-1,3-galactose (alpha-gal). Curr Allergy Asthma Rep. 2013;13(1):

32 Laxenaire M, le Group d études des reactions anaphylactoïdes peranesthésiques. Substances responsables des chocs anaphylactiques peranesthétiques. Troisième enquête multicentrique francaise ( )*. Ann Fr Anesth Réanim. 1996;15: Laxenaire M, le Group d études des reactions anaphylactoïdes peranesthésiques. Epidémiologie des reactions anaphylactoides peranesthesiques. Quatrième enquête multicentrique (juillet 1994-déembre 1996)*. Ann Fr Anesth Réanim. 1999;18: Garvey LH, Roed-Petersen J, Menné T, Husum B. Danish Anaesthesia Allergy Centre - preliminary results. Acta Anaesthesiol Scand. 2001;45(10): Mertes P-M, Laxenaire M-C, Les membres dus GERAP. Épidémiologie des réactions anaphylactiques et anaphylactoïdes peranesthésiques en France. Septième enquête multicentrique (Janvier 2001 Décembre 2002). Ann Fr Anesth Reanim. 2004;23(12): Karila C, Brunet-Langot D, Labbez F, Jacqmarcq O, Ponvert C, Paupe J, et al. Anaphylaxis during anesthesia: results of a 12-year survey at a French pediatric center. Allergy. Munksgaard International Publishers; 2005;60(6): Lobera T, Audicana M, Pozo M, Blasco A, Fernández E, Cañada P, et al. Study of Hypersensitivity Reactions and Anaphylaxis During Anesthesia in Spain. J Investig Allergol Clin Immunol. 2008;18(185): Chong YY, Caballero MR, Lukawska J, Dugué P. Anaphylaxis during general anaesthesia: one-year survey from a Britisch allergy clinic. Singapore Med J. 2008;49(6): Gurrieri C, Weingarten TN, Martin DP, Babovic N, Narr BJ, Sprung J, et al. Allergic reactions during anesthesia at a large united states referral center. Anesth Analg. 2011;113(5): Dong SW, Mertes PM, Petitpain N, Hasdenteufel F, Malinovsky J, Members of the GERAP. Hypersensitivity reactions during anesthesia. Results from the ninth French survey ( ). Minerva Anestesiol. 2012;78: Antunes J, Kochuyt A-MM, Ceuppens JL. Perioperative allergic reactions: Experience in a Flemish referral centre. Allergol Immunopathol (Madr). 2014;42(4): Krishna MT, York M, Chin T, Gnanakumaran G, Heslegrave J, Derbridge C, et al. Multi-centre retrospective analysis of anaphylaxis during general anaesthesia in the United Kingdom: Aetiology and diagnostic performance of acute serum tryptase. Clin Exp Immunol. 2014;178(2): Mirone C, Preziosi D, Mascheri A, Micarelli G, Farioli L, Balossi LG, et al. Identification of risk factors of severe hypersensitivity reactions in general anaesthesia. Clin Mol Allergy. 2015;13(1):11. Guyer AC, Saff RR, Conroy M, Blumenthal KG, Camargo CA, Long AA, et al. Comprehensive allergy evaluation is useful in the subsequent care of patients with drug hypersensitivity reactions during anesthesia. J allergy Clin Immunol Pract. 2015;3(1): Gonzalez-Estrada A, Pien LC, Zell K, Wang X-FF, Lang DM. Antibiotics are an important identifiable cause of perioperative anaphylaxis in the United States. J Allergy Clin Immunol Pract. 2015;3(1): e1. Garvey LH. Old, New and Hidden Causes of Perioperative Hypersensitivity. Curr Pharm Des. 2016; 22(45):

33 Chapter 2: Optimizing the diagnosis of IDHR to Atracurium 33

34 34

35 Part 1: Flow cytometric diagnosis of atracuriuminduced anaphylaxis AP Uyttebroek V Sabato J Leysen CH Bridts LS De Clerck DG Ebo Published as Original article in Allergy 2014; 69(10):

36 36

37 Abstract Background: Allergy to atracurium is a rare condition with serious consequences of diagnostic error. However, correct diagnosis is not always straightforward. Objectives: To assess the utility of the basophil activation test (BAT) in atracurium sensitization and to investigate its role in identifying cross-reactivity between muscle relaxants. Methods: For validation 8 patients with perioperative anaphylaxis to atracurium and 7 individuals experiencing perioperative anaphylaxis but not exposed to neuromuscular blocking agents (NMBA) were included. Furthermore, 5 other patient groups were included in the study, all individuals exposed to different NMBA, either sensitized or not to the drug. Basophil activation with atracurium was analysed flow cytometrically. Results: TG-ROC analyses between 8 atracurium sensitized patients and 7 non-exposed controls allowed identification of 5% as the decision threshold for BAT positivity. For this cut off the BAT attained a sensitivity of 63%, specificity of 100%, positive predictive value of 100% and negative predictive value of 70%. Of the atracurium exposed individuals with a negative atracurium skin test, 2 individuals had a clear positive BAT. BAT atracurium was positive in one cisatracurium sensitized patient and negative in all cisatracurium exposed patients with a negative skin test to cisatracurium. All rocuronium and suxamethonium-sensitized patients displayed a negative BAT with atracurium. Conclusions: The BAT proves to be a useful diagnostic for atracurium induced anaphylaxis and may be complementary to skin tests. The technique enables quick and simultaneous testing of potentially cross-reactive NMBA and the identification of safe alternatives for future surgery. 37

38 Introduction Genuine IgE-mediated atracurium allergy is a rare condition with severe consequences of diagnostic error (1,2). In the absence of a validated atracurium-specific IgE assay, diagnosis of atracurium allergy and identification of safe alternatives relies almost uniquely upon skin testing and eventually measurement of sige towards substituted ammonium compounds such as the opiate morphine (3). However, although at the moment considered to be the gold standard; skin testing to benzylisoquinoline drugs does not demonstrate an absolute predictive value (3, 4). False negative skin tests (ST) might severely endanger the life of our patients, whereas false positive skin tests can lead to overdiagnosis and unnecessary avoidance strategies. Alternatively, the sensitivity of morphine sige in benzylisoquinoline sensitization has been reported to vary considerably between 25% (5) and 75% (6). Therefore, additional safe and reliable diagnostics are of tremendous interest. Upon activation with a specific allergen basophils up-regulate various degranulation and activation markers such as CD63 and CD203c that can be quantified with flow cytometry in the basophil activation test (BAT) (7, 8). With respect to drugs, it appears that the BAT can be valuable in the diagnostic management of allergy to neuromuscular blocking agents (NMBA) (9-16). However, results of basophil activation experiments with the benzylisoquinoline-derived NMBA atracurium are limited to small series and are generally poorly validated (11-16). This study aims at assessing the utility of the BAT in atracurium sensitization and to study its contribution in the identification of cross-reactivity between atracurium and other NMBA. Materials and methods Study population As displayed in figure 1, a total of 75 patients, who were referred for diagnostic evaluation after experiencing an hypersensitivity reaction during anesthesia, were included in the study. Inclusion was based on the clinical history and skin test (ST) results as described in (3). The + and - in the group names represent a positive or negative skin test to the NMBA used during operation. First, 8 patients who had presented hypotension and/or bronchospasm within 5 minutes after injection of atracurium, demonstrating a positive atracurium ST were included (ATR+). As shown in table 1, 6 patients tested positive in the prick test (SPT), 2 displayed a positive intradermal test (IDT). Second, 7 individuals who experienced anaphylaxis during surgery but did not receive a NMBA and who demonstrated entirely negative ST served as a non-exposed control group (NMBA-). In these patients diagnosis of possible other causes (e.g. latex, chlorhexidine, antibiotics and hypnotics) was established as described elsewhere (3). Third, 29 patients exposed to atracurium who suffered from perioperative anaphylaxis from an alternative cause and who demonstrated a negative ST for atracurium (ATR-) were studied. 38

39 To investigate the potential of BAT in identifying cross-reactivity between NMBA, 4 additional groups were studied; i) 3 cisatracurium-sensitized patients (CIS+), ii) 9 cisatracurium-exposed patients with negative ST to cisatracurium (CIS-), iii) 14 rocuronium-sensitized patients (ROC+), and finally iv) 5 suxamethonium-sensitized patients (SUX+). Peri-operative anaphylaxis (n=75) NMBA exposed (n=68) NMBA not exposed (n=7) NMBA- Atracurium (n=37) (diagnosis) Other NMBA (n=31) (crossreactivity) Atracurium sensitized (n=8) ATR+ Atracurium ST negative (n=29) ATR- Cisatracurium (n=12) non- benzylisoquinolone NMBA (n =19) Cisatracurium sensitized (n=3) CIS+ Cistracurium ST negative (n=9) CIS- Rocuronium sensitized (n=14) ROC+ Suxamethonium sensitized (n=5) SUX+ Figure 1: Study population. Table 1: see page 42 to 45 39

40 PATIENT SEXE AGE IgE tot sige morphine (<=0.35kU/L)* ST SUX ST VEC ST ROC ST ATR ST CIS BAT ATR 500** BAT culprit NMBA Delay*** (m) NMBA- F NA 10 NMBA- F NA 2 NMBA- F NA 149 NMBA- M NA 2 NMBA- F NA 5 NMBA- F 63 NA NA 2 NMBA- M NA 7 ATR+ F SPT1/100 SPT1/ ATR+ F SPT1/100 SPT1/ ATR+ F SPT1/100 SPT1/ ATR+ F SPT1/ ATR+ M SPT1/10 SPT1/ ATR+ F IDT1/100 - IDT1/ ATR+ F SPT1/5 IDT1/ IDT1/ ATR+ F SPT1/100 SPT1/10 NR NR 7 ATR- M SPT1/10 45 NA 34 ATR- F IDT1/100 0 NA 10 ATR- M NA 8 ATR- M NA 1 ATR- M NA 3 ATR- M NA 33 ATR- M NA 3 ATR- M NA 53 40

41 PATIENT SEXE AGE IgE tot sige morphine (<=0.35kU/L)* ST SUX ST VEC ST ROC ST ATR ST CIS BAT ATR 500** BAT culprit NMBA Delay*** (m) ATR- F IDT1/ NA 2 ATR- F NA 12 ATR- M NA 1 ATR- F NAV 0 NA 3 ATR- F NA 3 ATR- F NA 134 ATR- F NA 12 ATR- F SPT1/10 0 NA 1 ATR- F NA 8 ATR- M SPT1/10 2 NA NAV ATR- F NA 16 ATR- M NA 1 ATR- F NA 4 ATR- F SPT1/ NA 2 ATR- F NA 2 ATR- F NA NAV ATR- F NA NAV ATR- M NA 1 ATR- F NAV NA 6 ATR- F IDT1/ NA 2 ATR- M NA 4 CIS+ M SPT1/100 SPT1/ CIS+ M IDT1/ CIS+ F IDT1/ SPT1/ CIS- F NA 20 41

42 PATIENT SEXE AGE IgE tot sige morphine (<=0.35kU/L)* ST SUX ST VEC ST ROC ST ATR ST CIS BAT ATR 500** BAT culprit NMBA Delay*** (m) CIS- F NA NAV CIS- M NA NAV CIS- F NA 1 CIS- M NA 133 CIS- F 71 NAV NAV NA 29 CIS- F NAV NA 76 CIS- M NAV NA NAV CIS- M NA 8 ROC+ F SPT1/ ROC+ M IDT1/ SPT1/ ROC+ F NAV SPT1/ ROC+ M SPT1/ ROC+ F SPT1/1 SPT1/ IDT1/100 ROC+ M SPT1/1 SPT1/ ROC+ F SPT1/ ROC+ F SPT1/1 SPT1/ ROC+ F SPT1/1 SPT1/ ROC+ F SPT1/ ROC+ F SPT1/1 SPT1/1 SPT1/ ROC+ M SPT1/5 SPT1/1 SPT1/ NAV ROC+ M SPT1/5 - SPT1/ ROC+ F SPT1/ SUX+ F SPT1/ NAV SPT 1/5 SUX+ F SPT1/

43 PATIENT SEXE AGE IgE tot sige morphine (<=0.35kU/L)* ST SUX ST VEC ST ROC ST ATR ST CIS BAT ATR 500** BAT culprit NMBA Delay*** (m) SUX+ F SPT1/ SUX+ F SPT1/10 NAV SUX+ M SPT1/ Table 1: Characteristics of the 75 patients included in the study. Demographics, Skin test results, sige and percentage activated basophils for each patient are given. NA: not applicable, NAV: not available, ST: skin tests, IDT: Intradermal testing, SPT: skin prick testing, NR: non responder in BAT, M: male, F: female, - : negative test *normal value,**positive if net percentage of CD63 upregulation > 5%,***time between reaction and investigations, expressed in months NMBA-:patients with peri-operative anaphylaxis but not exposed to a NMBA (NMBA-). ATR+: patients with atracurium sensitization. ATR-: patients with peri-operative anaphylaxis exposed to atracurium presenting a negative skin test for atracurium. CIS+: patients sensitised to cisatracurium. CIS-: patients with peri-operative anaphylaxis exposed to cisatracurium but presenting a negative skin test for cisatracurium. ROC+ patients with rocuronium sensitization. SUX+ patients with suxamethonium sensitization The + and - in the group names represent a positive or negative skin test to the NMBA used during operation. 43

44 Basophil activation test The BAT for NMBA is detailed elsewhere (17). Briefly, within 3h after sampling in endotoxin-free heparinised tubes, aliquots of 100 μl whole blood were stimulated (20 min, 37 C) with buffer containing 2 ng/ml IL-3 (negative control), positive control (anti-ige 10μg/mL, Pharmingen BD Biosciences, Erembodegem, Belgium) or serial dilutions of atracurium 0.5-5,000 µg/ml. Cells were stained with 2 μl of a mixture containing anti-cd123-pe, anti-human leukocyte antigen DR-PerCP and anti-cd63-fitc conjugated antibodies (BD, Erembodegem, Belgium). Basophil activation was evaluated by flow cytometric analysis (FACSCalibur, BD, Immunocytometry Systems, San Jose, CA, USA). Results were expressed as the net percentage of CD63+ basophils, i.e. by subtracting spontaneous CD63 expression (negative control) from the value obtained with allergen (drug) stimulation. A maximum of 5% spontaneous expression of CD63 after stimulation with buffer solution (negative control) was tolerated. The threshold for positivity was calculated by TG-ROC analysis. BAT was considered positive when the optimal dilution of atracurium yielded a positive value. Furthermore, cross reactivity was checked using optimal stimulation concentrations from prior studies (10) or preliminary dose finding experiments. Total and specific IgE Total and specific IgE to substituted ammonium compounds (morphine) were quantified with an ImmunoCAP technique (ThermoFisher Scientific, Uppsala, Sweden). Results of morphine sige 0.35 kua/l were considered positive (18). 44

45 Statistics Two-graph receiver operating characteristics (TG-ROC) curve analyses was performed to calculate the optimal cut-off value [and its 95% confidence interval (CI)] corresponding to the best sensitivity and specificity (19). Results Atracurium-exposed patients and control individuals (diagnosis) One ATR+ was non-responder to positive control stimulation and atracurium. She was withdrawn from further analysis but considered false negative for calculation of sensitivity and predictive values. Figure 2A displays the dose-finding experiments between atracurium-sensitized patients (ATR+) and patients not exposed to an NMBA (NMBA-). It emerges ATR+ to show a dose-dependent basophilic appearance of CD63 in response to atracurium. In contrast, expression of CD63 in NMBA- remained merely unchanged and comparable to spontaneous expression of this surface marker. As illustrated by figure 2A dose-finding reveals 500 µg/ml to be the best discriminative stimulation concentration between ATR+ patients and NMBA- individuals, generating a diagnostic threshold value of 5% CD63 up-regulation (fig 2B). For this threshold the BAT was positive in 5/7 ATR+ responders (sensitivity 71%; 95%CI: 29-95%) and none of 7 NMBA- (estimated specificity 100%; 95%CI: %). Taking into account the non-responder; overall sensitivity was 63% (95%CI: 25-91%) and the BAT showed an absolute positive predictive value and a negative predictive value of 70% (95%CI: 35-93%). Note that all except one ATRA+ patients had their BAT experiments within 1 year after the acute event. The individual BAT results are displayed in table 1 and figure 3. From these it appears that 2 out of 29 ATR- patients test clearly positive in the BAT atracurium. The individual total IgE and sige results are summarized in table 1. It appears that 1 ATR+, 2 ATR- and 1 NMBA- patient have a positive sige morphine. 45

46 Figure 2: A: Mean percentage of CD63+ basophils for different stimulation concentrations of atracurium in atracurium sensitized (ATR+) patients (n = 7, circles) and control individuals whom experienced a peri operative anaphylaxis, not exposed to NMBA (n = 7, squares). Net percentage was obtained by subtracting the spontaneous CD63 expression from the value obtained with the drug. B: TG-ROC curve for atracurium-induced CD63 expression generated between ATR+ patients (n = 7) and NMBA- individuals (n=7). Stimulation concentration of atracurium: 500 µg/ml and the optimal threshold of 5% is indicated with the dashed line. Open symbols represent sensitivity whereas closed symbols represent specificity. 46

47 Figure 3: Individual percentages of atracurium activated (CD63+) basophils in patients with peri-operative anaphylaxis but not exposed to a NMBA (NMBA-), patients with atracurium sensitization (ATR+), patients with peri-operative anaphylaxis exposed to atracurium presenting a negative skin test for atracurium (ATR-), patients sensitized to cisatracurium (CIS+), patients with peri-operative anaphylaxis exposed to cisatracurium but presenting a negative skin test for cisatracurium (CIS-) patients with rocuronium sensitization (ROC+), patients with suxamethonium sensitization (SUX+). The dashed line indicates the TG-ROC calculated decision threshold for positivity (5%). Patients exposed to cisatracurium, rocuronium and suxamethonium (assessment of crossreactivity) Figure 3 and table 1 summarize the individual percentages of CD63 positive basophils in the patients exposed and/or sensitized to other NMBA. None of the individuals from groups CIS-, ROC+ and SUX+ demonstrated a %CD63 expression exceeding 5% for atracurium. However, BAT atracurium clearly exceeded 5% in 1 out of 3 CIS+ individuals. In these patients exposed or sensitized to other NMBA a positive sige to morphine was found in 1 CIS-, 12 ROC+ and 4 SUX+. 47

48 Discussion Studies about BAT as a diagnostic instrument for rare allergic conditions pose significant challenges and face various difficulties such as accurate identification of sufficient numbers of well-phenotyped patients, identification of optimal agent-specific concentrations and stimulation conditions, establishment and comparison of correct decision thresholds. For the time being, as addressed in the introductory paragraph, the BAT atracurium has been rarely studied, and its methodology and interpretation of results are both topics necessitating attention. From the current study, which compares the outcome of the BAT atracurium in 8 atracurium sensitized patients to 7 patients not exposed to NMBA, the optimal stimulation concentration was found to be similar to our prior observations with rocuronium (10), enabling the proposal to apply a single protocol and threshold value of 5% for both NMBA. In contrast to former publications (11-15), in our study the decision threshold was calculated by TG-ROC graph analysis and not set arbitrarily. Applying this threshold, and taking into account non-responders, our BAT technique attains a sensitivity of 63%, specificity of 100%, an absolute positive predictive value and a negative predictive value of 70%. These sensitivity data are in line with the literature showing sensitivity of the BAT NMBA to range from 36 to 79% and from 81 to 100% respectively (for review: (8)). A limitation of our calculations is the small number of patients. However, atracurium allergy is a very rare condition, compared with rocuronium or suxamethonium allergy (4). Another limitation could relate to the fact not having included control individuals with uneventful exposure to atracurium. However, such individuals are difficult to enroll for obvious reasons, mainly because they are not referred for diagnostic evaluation. Alternatively, in clinical practice physicians will need to correctly identify the causative compound(s), rather than to dichotomize between patients and asymptomatic control individuals. Clear positive BAT atracurium responses were found in 2 patients exposed to atracurium demonstrating a negative ST to this NMBA. Although in the absence of a gold standard it is impossible to draw definite conclusions, in our opinion the high percentage of drug induced basophil responses exceeding 40% suggests an atracurium sensitization overlooked by skin testing, which is currently considered as the reference test to depict NMBA sensitization. The performance of the BAT atracurium was additionally assessed in various groups of NMBA sensitized or tolerant patients. From this analysis it emerged that BAT atracurium is mirroring absence of ST responsiveness and is negative in patients sensitized to rocuronium and suxamethonium. Thereby indicating that patients who suffer from allergy to aminosteroid-derived NMBA and suxamethonium generally tolerate administration of a benzylisoquinoline (20). The observation that sige towards morphine was positive in the majority of patients sensitized to rocuronium and suxamethonium, but in only one of the 8 atracurium sensitized patients seems to 48

49 confirm the previously reported lower sensitivity of morphine-based assays for benzylisoquinoline NMBA compared to its sensitivity for the aminosteroids and suxamethonium (5, 6). In these studies sensitivity of the morphine IgE assay has been reported to vary considerably between 25% (5) and 75% (6) for atracurium, and between 97% (5)and 88.9%(6) for suxamethonium. Clearly, these findings point to the potential complementarity of BAT to skin testing for NMBA, not only in the diagnosis of atracurium sensitization but also in the study of cross-reactivity between benzylisoquinolines, i.e. atracurium and its cis-isomer cisatracurium. As a matter of fact, as already pointed out elsewhere in the Journal (9, 10), it appears BAT can be useful in patients with high clinical suspicion of NMBA allergy with negative skin tests and can contribute in the identification of safe alternatives for the future. Therefore, because of the unavailability of a gold standard, the potential serious consequences of misdiagnosis, and in line with prior findings for rocuronium (9), our current practice is to recommend BAT atracurium in all patients with a suspicion of benzylisoquinoline allergy that demonstrate negative or equivocal ST, and to dissuade any further use of atracurium in all patients with a positive BAT, irrespective the ST response to atracurium. However, one should bear in mind that the technique is of no value in non-responders to positive control and allergen stimulation. Alternatively, our data indicate that a morphine-based IgE immunoassay might be unreliable to depict sensitization to benzylisoquinoline NMBA. Conclusion In conclusion, the BAT constitutes a useful diagnostic instrument for perioperative anaphylaxis induced by atracurium. It may serve as a complementary tool that enables quick and simultaneous testing of different potential cross-reactive NMBA and consequently, it is helpful in the search for a safe alternative. In those patients displaying negative or equivocal skin tests, the BAT atracurium will be of great assistance. 49

50 References Lafforgue E, Sleth JC, Pluskwa F, Saizy C. Successful extracorporeal resuscitation of a probable perioperative anaphylactic shock due to atracurium. Ann Fr Anesth Reanim. 2005;24(5): Miraj A, Foaud A, Seth B. Cardiac arrest following an anaphylactic reaction to atracurium. BMJ Case Rep Jan;1 2. Mertes PM, Malinovsky JM, Jouffroy F, Aberer W, Terreehorst I, Brockow K, et al. Reducing the Risk of Anaphylaxis During Anesthesia: 2011 Updated Guidelines for Clinical Practice. J Investig Allergol Clin Immunol. 2011;21(6): Dong S, Mertes P, Petitpain N, Hasdenteufel F, Malinovsky J. Hypersensitivity reactions during anesthesia. Results from the ninth French survey ( ). Minerva Anestesiol. 2012;78: Fisher M, Baldo BA. Immunoassays in the diagnosis of anaphylaxis to neuromuscular blocking drugs: the value of morphine for the detection of IgE antibodies in allergic subjects. Anaesth Intensive Care. 2000;28(2): Laroche D, Chollet-Martin S, Léturgie P, Malzac L, Vergnaud M, Neukirch C, et al. Evaluation of a New Routine Diagnostic Test for Immunoglobulin E Sensitization to Neuromuscular Blocking Agents. Anesthesiology. 2011;114(91-97). Ebo DG, Bridts CH, Hagendorens MM, Aerts NE, De Clerck LS, Stevens WJ. Basophil activation test by flow cytometry: present and future applications in allergology. Cytometry B Clin Cytom. 2008;74(4): Leysen J, Sabato V, Verweij MM, De Knop KJ, De Clerck LS, Ebo DG. The basophil activation test in the diagnosis of immediate drug hypersensitivity. Expert Rev Clin Immunol. 2011;7(3): Leysen J, Bridts CH, De Clerck LS, Vercauteren M, Lambert J, Weyler JJ, et al. Allergy to rocuronium: from clinical suspicion to correct diagnosis. Allergy. 2011;66(8): Ebo DG, Bridts CH, Hagendorens MM, Mertens CH, De Clerck LS, Stevens WJ. Flow-assisted diagnostic management of anaphylaxis from rocuronium bromide. Allergy. 2006;61(8): Abuaf N, Rajoely B, Ghazouani E, Levy DA, Pecquet C, Chabane H, et al. Validation of a flow cytometric assay detecting in vitro basophil activation for the diagnosis of muscle relaxant allergy. J Allergy Clin Immunol. 1999;104(2): Monneret G, Benoit Y, Debard AL, Gutowski MC, Topenot I, Bienvenu J. Monitoring of basophil activation using CD63 and CCR3 in allergy to muscle relaxant drugs. Clin Immunol. 2002;102(2): Sudheer PS, Hall JE, Read GF, Rowbottom AW, Williams PE. Flow cytometric investigation of peri-anaesthetic anaphylaxis using CD63 and CD203c. Anaesthesia. 2005;60(3): Kvedariene V, Kamey S, Ryckwaert Y, Rongier M, Bousquet J, Demoly P, et al. Diagnosis of neuromuscular blocking agent hypersensitivity reactions using cytofluorimetric analysis of basophils. Allergy. 2006;61(3): Sainte-Laudy J, Orsel I. Interest of a new flow cytometric protocol applied to diagnosis and prevention of per anaesthetic accidents induced by neuromuscular blockers. Rev Fr Alllergol. 2008;48:

51 Johansson SGO, Oman H, Degerbeck F, Florvaag E, Nopp A. Anaphylaxis to atracurium - a non-qai-dependent reaction? Acta Anaesthesiol Scand. 2012;56(2): Ebo DG, Sainte-Laudy J, Bridts CH, Mertens CH, Hagendorens MM, Schuerwegh AJ, et al. Flow-assisted allergy diagnosis: current applications and future perspectives. Allergy. 2006;61(9): Ebo DG, Venemalm L, Bridts CH, Degerbeck F, Hagberg H, De Clerck LS, et al. Immunoglobulin E Antibodies to Rocuronium. Anaesthesiology. 2007;107(2): Greiner M, Pfeiffer D, Smith RD. Principles and practical aplication of the receiver-operating characteristic analysis for diagnostic tests. Prev Vet Med. 2000;45: Leysen J, Uyttebroek A, Sabato V, Bridts C, De Clerck L, Ebo D. Predictive value of allergy tests for neuromuscular blocking agents: tackling an unmet need. Clin Exp Allergy. 2014;44:

52 52

53 Part 2: Immunoglobulin E antibodies to atracurium: a new diagnostic tool? AP Uyttebroek V Sabato CH Bridts LS De Clerck DG Ebo Published as Letter to the editor in: Clinical & Experimental Allergy 2014; 45(2):

54 54

55 To the editor, Correct diagnosis of atracurium allergy is not straightforward because drug-specific skin tests and measurement of specific IgE (sige)-antibodies against tertiary and quaternary substituted ammonium ions do not demonstrate an absolute predictive value [1-4]. Recently a solid phase prepared from tetrahydropapaverine, modified with b-propiolactone, and conjugated to poly-l-lysine was developed to detect serum IgE-antibodies to atracurium [5]. However, before acceptance into routine diagnostic practice, this assay requires validation. In order to evaluate the diagnostic potential of this newly developed atracurium allergosorbent assay, we included 78 patients suffering from an allergic reaction during general anesthesia and referred for diagnostic work-up. This was done in a similar way as in [6]. First, 7 patients who presented with hypotension and/or bronchospasm within 5 minutes after injection of atracurium, demonstrating a positive atracurium ST were included (ATR+). Second, 7 individuals who experienced anaphylaxis during surgery but did not receive an NMBA and who demonstrated negative ST for NMBA served as a non-exposed control group (NMBA-). In these patients diagnosis of possible other causes (e.g. latex, chlorhexidine, antibiotics and hypnotics) was established according to [3]. Third, 21 patients exposed to atracurium who suffered from perioperative anaphylaxis from an alternative cause and who demonstrated a negative ST for atracurium were studied (ATR-). For further evaluation of the utility of sige atracurium, 4 additional groups were studied; 6 cisatracurium-sensitized patients (CIS+), 10 cisatracurium-exposed patients with negative ST to cisatracurium (CIS-), 21 rocuronium-sensitized patients (ROC+), and finally 6 suxamethonium-sensitized patients (SUX+). Participants gave a written informed consent as approved by the local Ethical Committee [B ]. Total IgE and sige levels for atracurium, morphine, rocuronium and suxamethonium were quantified by ImmunoCAP system (ThermoFisher Scientific, Uppsala, Sweden). Results > 0.35 kua/l were considered positive for all except rocuronium, where the threshold of 0.13 kua/l was used [7]. Furthermore, each patient received a BAT atracurium as described in [6]. Statistical analysis was performed with IBM SPSS Statistics 22. Fisher s Exact tests were used to evaluate the significant differences between different patient groups. Sensitivity of the atracurium-immunoassay, calculated from the results obtained on 7 atracuriumsensitized patients (ATR+) and 7 non-exposed control individuals (NMBA-), was 57%. Specificity was 100%. (P = 0.07, Fisher s Exact test). Positive and negative predictive values of the assay were 100% and 70%, respectively. Figure 1A-D displays the individual sige results for atracurium, morphine, rocuronium and suxamethonium. 55

56 The atracurium-immunoassay (Fig. 1A) was positive in 4/7 atracurium-sensitized patients (ATR+), in 1/6 cisatracurium-sensitized patients (CIS+) and in 1/6 suxamethonium-sensitized patients (SUX+). In contrast, no sige to atracurium was demonstrable in 7 individuals not exposed to a NMBA (NMBA-), in 21 atracurium-exposed patients with a negative atracurium skin test (ATR-) or in 10 cisatracuriumexposed patients with negative cisatracurium and atracurium skin tests (CIS-). None of the 21 rocuronium-sensitized patients (ROC+) displayed a positive sige atracurium. Of interest is that morphine-reactive sige-antibodies were barely demonstrable in patients sensitized to benzylisoquinolines (fig 1B). This in contrast to IgE reactivity to rocuronium that was also demonstrable in 4/7 atracurium-sensitized (ATR+) and 1/6 cisatracurium-sensitized (CIS+) patients (fig 1C). Most remarkable, as shown in figure 1E, positivity of sige to rocuronium in the 5 patients sensitized to these benzylisoquinolines was absolutely restricted to those patients who also tested positive in the atracurium assay. It is of note that these 5 patients demonstrated a negative skin test to rocuronium and one case was subsequently uneventfully exposed to this aminosteroid compound. Remarkably, 4 of these 5 patients sensitized to a benzylisoquinoline and showing a positive sige rocuronium demonstrated a negative sige morphine and suxamethonium result. Our data illustrate that this tetrahydropapaverine-based immunoassay yields a moderate sensitivity but an absolute specificity. These findings are in line with the observations by others who detected sige reactivity towards atracurium in 2/ 6 [1] and 2/4 patients [5]. Furthermore, our data re-emphasize that the commercially sige morphine assay advocated to detect IgE antibodies to NMBAs is unreliable as an aid in the diagnosis of allergic sensitization to atracurium and cisatracurium [1, 6]. Since the atracurium and cisatracurium molecules each contain 2 quaternary methylammonium groups, the explanation for this antibody recognition difference is not readily apparent. Figure 1E suggests that detection of IgE antibodies to atracurium indicates the existence of two different groups of benzylisoquinoline-sensitive patients, i.e., those who react in the atracurium-specific IgE assay and a group who are negative in all sige assays, i.e. atracurium, rocuronium, morphine and suxamethonium. Patients in the first group show positive rocuronium sige results and, except for one case, a negative response in the morphine immunoassay. The first, most obvious possible explanation for positive rocuronium result is the presence of similar antigenic structures on atracurium and rocuronium. Examination of both structures reveals an N-allyl group i.e., a N+-CH2-CH-CH2 sequence at position 16 of rocuronium while atracurium has a similar N+-CH2-CH2-C sequence (connecting bridge between the 2 Ns). These similar large quaternary ammonium ion structures [8] appear to be the most likely explanation for the apparent cross-reactivity between the two NMBAs. There is also another, 56

57 seemingly less likely, structural explanation. At rocuronium position 17 there is an acetyl group (-O- CO-CH3) so, considering positions 16 and 17 of rocuronium, the sequence N+-CH-CH-O-CO-C is seen and via the connecting bridge in atracurium, the sequence N+-CH2-CH2-CO-O-C occurs. Note, however, that three of the above sequences are internal, not terminal, and hence not easily accessible structures [9]. Although these possible structural explanations are speculative, they do have the virtue of being amenable to experimental examination, in particular by quantitative side-by-side inhibition studies with carefully selected analogues designed to determine the precise structures recognised by the IgE antibodies [9]. In any case, since our results show that skin testing and BAT with rocuronium were negative in all benzylisoquinoline-sensitized patients demonstrating a positive sige test for rocuronium and negative in the test for morphine, it remains to be established whether the observed in vitro cross-reactivity between atracurium and rocuronium is clinically relevant. Moreover, one patient even tolerated administration of rocuronium during subsequent anesthesia. In addition, the facts that rocuronium antibodies are also found in patients not sensitized to atracurium (ATR-) and in patients who did not receive an NMBA (NMBA-) (fig 1C), needs to be taken into account. These patients do not react in the atracurium assay but show reactivity in the morphine assay suggesting that, in these individuals, the rocuronium assay recognizes the substituted ammonium groups. In the second group of atracurium patients, that is those who were negative in all sige assays namely atracurium, rocuronium, morphine and suxamethonium, the BAT for atracurium was positive for 2/3 patients (fig 1E), making the possibility of false positive skin tests unlikely. Whether the findings presented here indicate a non-ige mediated hypersensitivity to atracurium or an IgE mediated mechanism that cannot be demonstrated with the currently available diagnostic tools. We can conclude that the results of IgE antibody profiles in this study suggest the possibility of different atracurium sensitization patterns. The newly developed atracurium sige assay proved specific but demonstrated only moderate sensitivity. Additional follow-up studies in larger cohorts are needed 57

58 before the assay can be confidently applied or rejected as a routine diagnostic test for atracurium allergic sensitivity. Figure 1 A-D: Individual results of sige to atracurium, morphine, rocuronium and suxamethonium in patients with perioperative anaphylaxis but not exposed to a NMBA (NMBA-), patients with atracurium sensitization (ATR+), patients with peri-operative anaphylaxis exposed to atracurium presenting a negative skin test for atracurium (ATR-), patients sensitized to cisatracurium (CIS+), patients with peri-operative anaphylaxis exposed to cisatracurium but presenting a negative skin test for cisatracurium (CIS-) patients with rocuronium sensitization (ROC+), patients with suxamethonium sensitization (SUX+). The dashed line indicates the threshold for positivity which was 0.35 kua/l for atracurium, morphine and suxamethonium. For rocuronium the threshold was set at 0.13kU/L based on (7). E. Graphical representation after classifying (dendogram) of all individual sige. sige towards suxamethonium (sigesuxa), morphine (sigemorph), rocuronium (sigerocu) and atracurium (sigeatra) are shown.the asterix denotes patients with a positive BAT atracurium. 58

59 References Fisher M, Baldo BA. Immunoassays in the diagnosis of anaphylaxis to neuromuscular blocking drugs: the value of morphine for the detection of IgE antibodies in allergic subjects. Anaesth Intensive Care. 2000;28(2): Laroche D, Chollet-Martin S, Léturgie P, Malzac L, Vergnaud M, Neukirch C, et al. Evaluation of a New Routine Diagnostic Test for Immunoglobulin E Sensitization to Neuromuscular Blocking Agents. Anesthesiology. 2011;114(91-97). Mertes PM, Malinovsky JM, Jouffroy F, Aberer W, Terreehorst I, Brockow K, et al. Reducing the Risk of Anaphylaxis During Anesthesia: 2011 Updated Guidelines for Clinical Practice. J Investig Allergol Clin Immunol. 2011;21(6): Dong S, Mertes P, Petitpain N, Hasdenteufel F, Malinovsky J. Hypersensitivity reactions during anesthesia. Results from the ninth French survey ( ). Minerva Anestesiol. 2012;78: Johansson SGO, Oman H, Degerbeck F, Florvaag E, Nopp A. Anaphylaxis to atracurium - a non-qai-dependent reaction? Acta Anaesthesiol Scand. 2012;56(2): Uyttebroek AP, Sabato V, Leysen J, Bridts CH, De Clerck LS, Ebo DG. Flowcytometric diagnosis of atracurium-induced anaphylaxis. Allergy Jun 24;69(10): Ebo DG, Venemalm L, Bridts CH, Degerbeck F, Hagberg H, De Clerck LS, et al. Immunoglobulin E Antibodies to Rocuronium. Anaesthesiology. 2007;107(2): Pham NH, Baldo BA, Puy RM. Studies on the mechanism of multiple drug allergies. Structural basis of drug recognition. J Immunoassay Immunochem. 2001;22: Baldo BA, Pham NH. Drug Allergy. Clinical Aspects, Diagnosis, Mechanisms, Structure- Activity Relationships. New York: Springer;

60 60

61 Chapter 3: Optimizing the diagnosis of cefazolin hypersensitivity 61

62 62

63 Part 1: Cefazolin hypersensitivity: towards optimized diagnosis Astrid P. Uyttebroek Ine I.Decuyper Chris H. Bridts Antonino. Romano Margo M. Hagendorens Didier G. Ebo Vito Sabato Published as Original article in JACI in practice 2016;4(6):

64 Highlight box What is already known about this topic? Up to now the recommended maximum non-irritating concentration for skin testing with cefazolin is 2mg /ml. It seems that cefazolin hypersensitivity is not a class hypersensitivity. What does this article add to our knowledge? Increasing cefazolin concentration for skin tests up to 20 mg/ml increases the sensitivity of the test without affecting its specificity. How does this study impact current management guidelines? We recommend to use 20mg/mL as maximum non-irritating concentration in the diagnostic work up of immediate cefazolin hypersensitivity. This study confirms that cefazolin hypersensitivity is a selective allergy with good tolerance to other β-lactam antibiotics. 64

65 Abstract Background: Correct diagnosis of cefazolin hypersensitivity is not straightforward, mainly because of the absence of in vitro tests and uncertainties concerning the optimal cefazolin concentration for skin testing. Cross-reactivity studies suggest cefazolin hypersensitivity to be a selective hypersensitivity. Objective: The first objective was to confirm that the application of a higher than 2 mg/ml test concentration could increase skin test sensitivity. A second part aimed at investigating the crossreactivity between cefazolin and other β-lactam antibiotics. Methods: 66 patients referred to our clinic after experiencing perioperative anaphylaxis, and exposed to cefazolin, underwent skin testing with cefazolin up to 20 mg/ml. Patients exhibiting a positive skin test with cefazolin had a panel of skin tests with other β-lactams and, if indicated, graded drug challenges to study cross-reactivity. Results: Increasing skin test concentration from the recommended 2 mg/ml to 20 mg/ml identified an additional 7/19 (27%) patients, who would otherwise have displayed negative skin testing. The concentration was proven non-irritating in 30 cefazolin exposed control individuals in which an alternative culprit for perioperative anaphylaxis was identified. Graded challenge testing, following negative skin testing, displayed that all patients tolerated alternative β-lactam antibiotics (i.e., amoxicillin, cephalosporins, monobactams, carbapenems). Of them, 11 individuals also tolerated an alternative cephalosporin, suggesting cefazolin hypersensitivity (generally) is a selective allergy. Conclusions: Increasing cefazolin concentration for skin tests up to 20 mg/ml benefits the sensitivity of diagnosis. Furthermore, our data confirm that cefazolin hypersensitivity seems to be a selective allergy with good tolerance to other β-lactam antibiotics. 65

66 Introduction Cefazolin, a first generation cephalosporin, is widely used for preoperative antibiotic prophylaxis [1, 2]. Some studies regarding perioperative anaphylaxis indicate that β-lactam antibiotics are a relevant cause of IgE-mediated hypersensitivity reactions, cefazolin being responsible for the majority of these reactions [3-7]. Although cefazolin hypersensitivity constitutes a potential life-threatening condition with serious consequences, correct diagnosis of cefazolin hypersensitivity is not straightforward for various reasons. First of all, drug provocation tests with this parenteral cephalosporin are hazardous and time consuming. Secondly, no reliable cefazolin-specific IgE antibody assay is available. Therefore, clinical suspicion of cefazolin hypersensitivity is generally confirmed with skin tests [8]. However, for the time being, skin testing with cephalosporins are not entirely standardized and optimal skin test concentrations remain to be established [8]. Actually, sensitivity, specificity and predictive values of this diagnostic method remain unknown, mainly as the maximal non-irritating skin test concentrations need to be established. To date, the European Network on Drug Allergy (ENDA) recommends a maximal non-irritating skin test concentration of 2 mg/ml for all the cephalosporins [9]. On the other hand, as reported in the same guidelines [9], there is evidence that, for several cephalosporins, skin test concentrations up to 20 mg/ml are probably also not irritant. As a matter of fact, for cefazolin a concentration up to 33 mg/ml has been described as being non-irritant in some studies [10-12]. The fact that cephalosporin hypersensitivity is not a class hypersensitivity has recently been reported in a series of patients by Romano et al. [12]. Regarding cefazolin, studies conducted up to now showed that IgE-mediated hypersensitivity towards cefazolin appears to be selective in the great majority of allergic subjects [12-17]. As a first objective of this study, we sought to confirm whether a test concentration higher than 2 mg/ml could increase the sensitivity of the skin tests, and add to the diagnosis in patients who would otherwise yield negative skin test responses. The second part of our study aimed at investigating the cross-reactivity between cefazolin and other β-lactam antibiotics. 66

67 Methods Study population During the period from January 2013 to December 2015 we evaluated 157 patients who were referred to our outpatients clinic for diagnostic evaluation after experiencing a perioperative hypersensitivity reaction grade 1-3 according to the Brown criteria [18, 19]. Of them, 66 received cefazolin as a perioperative antibiotic prophylaxis. All these patients underwent the standardized protocol for all potential offenders of perioperative anaphylaxis [20]. Furthermore, all patients underwent skin testing with cefazolin as described below. Patients exhibiting a positive skin test with cefazolin had a panel of skin tests with other β-lactams and, if indicated, graded challenges with some of them. Skin testing Skin tests were performed on different days [21]. Firstly, minor determinant mixture ((MDM) consisting of sodium benzylpenicillin, benzylpenicilloic acid, and sodium benzylpenicilloate), penicilloyl-polylysine (PPL), and benzylpenicillin (BP); subsequently amoxicillin (clavulanic-acid) and cefazolin were tested. Finally, patients displaying a positive skin test towards cefazolin underwent skin tests with alternative β-lactams. We applied maximal skin test concentrations recommended by ENDA, except for cefazolin and cefuroxime [9]. For skin testing with penicillin reagents, the maximum nonirritating concentration was 5x10-5 mm/l for PPL, 2x10-2 mm/l for MDM, UI/mL in normal saline for BP, and 20 mg/ml in normal saline for amoxicillin, ampicillin, and amoxicillin (20 mg/ml) + clavulanic acid (4 mg/ml). For cefazolin and cefuroxime, maximum non irritating concentrations were obtained by performing skin testing with 100 mg/ml and 20 mg/ml in normal saline in 10 healthy controls individuals [10, 12, 22]. As skin testing for cefazolin and cefuroxime using 100 mg/ml yielded irritative positive results in up to 7/10 healthy control individuals, for both drugs, a maximal concentration of 20 mg/ml was subsequently used to test cefazolin exposed patients. Skin testing with the monobactam aztreonam was performed at a concentration of 2 mg/ml, with imipenemcilastatin at a concentration of 0.5 mg/ml for each component, and with meropenem at a concentration of 1 mg/ml in normal saline. All reagents were diluted no more than 2 hours before testing. Results were considered positive when wheal/flare equaled or exceeded 3/3 mm. IDTs were considered positive when the wheal/flare equaled or exceeded 5 mm. Total and specific IgE measurement In all patients total IgE and sige levels for the commercially available β-lactam determinants i.e. penicilloyl G, penicilloyl V, amoxicilloyl, ampicilloyl and cefaclor were quantified by FEIA ImmunoCAP 67

68 system (ThermoFisher Scientific, Uppsala, Sweden). Results equalling or exceeding 0.35 kua/l were considered positive. Graded challenges When indicated, a graded challenge up to a cumulative dose (CD) equalling or exceeding the therapeutic dose, was performed in order to study cross-reactivity and to identify a safe alternative β- lactam antibiotic for the future. Oral challenges were performed with amoxicillin (CD 901 mg), amoxicillin-clavulanic acid (CD 901 mg) and cefuroxime axetil (CD 661 mg). Intramuscular challenges were performed with aztreonam (CD 1 g) and ceftriaxone (CD 1 g). Results Skin testing Figure 1 displays the results of the skin testing. In total, 66 cefazolin exposed patients who experienced a perioperative anaphylaxis underwent the standardized diagnostic protocol for all potential offenders. In 30 patients a cause other than cefazolin was found (e.g. curares, latex, chlorhexidine). These 30 patients received the maximal non irritating cefazolin skin testing up to 20 mg/ml. In all these individuals, skin testing with cefazolin was negative. These patients were considered as exposed control individuals. In contrast, 19 of the remaining 36 individuals exposed to cefazolin, in whom no other cause for perioperative anaphylaxis was found, displayed positive skin test responses to this cephalosporin. 12 patients had a positive intradermal test at a concentration up to 2 mg/ml, as recommended by current guidelines, and an additional 7 (27%) had a positive intradermal test at a ten-fold higher concentration of 20 mg/ml. In the remaining 17 patients, no causative agents responsible for the perioperative anaphylaxis could be identified (Fig. 1). All the patients with an IgE-mediated hypersensitivity towards cefazolin displayed negative skin tests for a panel of β-lactam antibiotics as displayed in table 1. 68

69 Figure 1: Flowchart displaying skin test results in patients and control individuals. Total and specific IgE measurement Total and specific IgE results are displayed in table 1. Only one patient displayed a weak positive specific IgE toward penicilloyl V (0.36 kua/l). Graded challenges Challenges with alternative β-lactams were performed in 16/19 patients (Table 1). Five patients were challenged with only amoxicillin or amoxicillin-clavulanic acid; 2 with only cefuroxime axetil, whereas in 9 patients, controlled administrations of amoxicillin or amoxicillin-clavulanic-acid, as well as one or more cephalosporins and/or aztreonam were performed. All challenges were negative. In 3/30 skin test negative patients a challenge test with cefazolin was negative. 69

70 Drug challenge IgE measurement (ku/l) C* Skin testing with alternative betalactam(s) with alternative betalactam(s) PG PPL MDM AX AC AP CU CR CF AZ IM ME TOT PyG PyV APy AXy CL AX AC CU CR CF A Z 2 _ _ 7 20 _ 13 _ 2 _ U _ 0.2 _ 6 20 _ 122 _ _ U _ 14 _ Table 1. Allergologic test results of the 21 individuals with cefazolin allergy. C*concentration at which the IDT with cefazolin was positive, TOT: total IgE, PyG: penicilloyl G, PyV: penicilloyl V, APy: Ampicilloyl, CL: Cefaclor, PF: penicillin G, PPL: penicilloylpolylysine, MDM: minor determinant, AX: amoxicillin, AC: amoxicillin-clavulanic acid, AP: ampicillin, CU: cefuroxime, CR: ceftriaxone, CF: ceftazidime, AZ: aztreonam, IM: imipinem, ME: meropenem. 20 _ U _ U _ 20 _ _ 2 _ 61 _ _

71 Discussion To our knowledge, this is the largest monocentric study about immediate perioperative hypersensitivity to cefazolin. It demonstrates that increasing cefazolin concentration for skin testing up to 20 mg/ml probably improves the sensitivity of the test without affecting its specificity. This observation is in line with previous studies [10, 12]. As a result, 7 additional patients could be identified as possibly cefazolin allergic who would otherwise have not been diagnosed because of negative skin testing. Moreover, in our series these additional patients represent almost one-third of the cefazolin allergic population. Hitherto, current guidelines [9, 23, 24] have recommended a maximal nonirritating concentration for all cephalosporins of 2 mg/ml. Therefore, we believe that the observation that a concentration of 20 mg/ml that allows to identify an additional 30% of patients is relevant, especially as this concentration was established by using a titrated skin test procedure. However, to really calculate the negative predictive value of the maximal non irritating skin test concentration for cefazolin one should perform challenge tests in patients displaying a negative skin test to 20 mg/ml. The main limitation of this study is that we did not challenge the majority of patients displaying a negative 20mg/mL intradermal test. One could argue that increasing cefazolin concentration could entail a risk for false positive skin test results. Although the determination of the precise test accuracy (sensitivity/specificity) would require provocation tests in all patients including those with a positive skin test, such an approach cannot be justified and has been dissuaded for obvious reasons [25]. Therefore, as an alternative approach, it is common practice to identify non-irritating skin test concentrations, ideally by enrolling at least 20 control individuals [9]. We have followed these recommendation and have performed titrated skin testing up to 100 mg/ml that was found irritative, whereas 20 mg/ml did not trigger a skin test response in 30 exposed control individuals. Previous cross-reactivity studies [12-17] demonstrated that cefazolin hypersensitivity is mainly a selective hypersensitivity, i.e., it does not involve cross-reactivity with other cephalosporins and/or penicillins. In effect, in these studies, the great majority, namely 20/22, of patients suffering from an IgE-mediated hypersensitivity to cefazolin displayed a pattern of selective response to it [12-17]. Our study of 19 cefazolin skin test-positive patients, confirms this data as no patient reacted to an alternative β-lactam antibiotic and none of the patients except one displayed positive results in the ImmunoCAP. Of them, 11 individuals tolerated alternative cephalosporins; specifically, 9 patients were able to tolerate cefuroxime axetil. In Belgium this is the only (second generation) cephalosporin available in oral formulation. The patient displaying positive specific IgE towards penicilloyl V with a specific/total IgE ratio of [25] tolerated an oral challenge with amoxicillin clavulanic acid pointing to a false positive IgE result [26]. 71

72 Regarding evaluation of cross-reactivity, a limitation of this study is that, since some data was collected in retrospect, not all the patients received the same battery of skin testing and drug provocations with alternative β-lactams. These data confirm that in the majority of cases cefazolin hypersensitivity seems to be an isolated allergy with tolerance for alternative β-lactam antibiotics. This probably relates to the fact that cross-reactivity of cephalosporins is mainly determined by the R1 side-chain structure [12]. The R1 side-chain of cefazolin consists of a heterocycle N-methylthiodiazole structure [13] which is different to other cephalosporin R1 side chains [8, 12, 27]. In conclusion, our study demonstrates that diagnosis of cefazolin hypersensitivity benefits from a drugspecific intradermal test concentration up to 20 mg/ml. Furthermore, study of cross-reactivity reveals that, according to literature data, cefazolin hypersensitivity in the great majority of cases is a selective allergy with tolerance to other β-lactam antibiotics. 72

73 References Unger NR, Stein BJ. Effectiveness of pre-operative cefazolin in obese patients. Surg Infect (Larchmt). 2014;15(4): Bratzler DW, Dellinger EP, Olsen KM, Perl TM, Auwaerter PG, Bolon MK, et al. Clinical Practice Guidelines for Antimicrobial Prophylaxis in Surgery. Surg Infect (Larchmt). 2013;14(1): Dong S, Mertes P, Petitpain N, Hasdenteufel F, Malinovsky J, GERAP. Hypersensitivity reactions during anesthesia. Results from the ninth French survey ( ). Minerva Anestesiol. 2012;78: Leysen J, De Witte L, Bridts CH, Ebo DG. Anaphylaxis during general anaesthesia: a 10-year survey 1 at the University Hospital of Antwerp. P Belg Roy Acad Med. 2013;2: Gonzalez-Estrada A, Pien LC, Zell K, Wang XF, Lang DM. Antibiotics are an important identifiable cause of perioperative anaphylaxis in the United States. J Allergy Clin Immunol Pract. 2015;3(1):101-5 e1. Antunes J, Kochuyt AM, Ceuppens JL. Perioperative allergic reactions: experience in a Flemish referral centre. Allergol Immunopathol (Madr). 2014;42(4): Gurrieri C, Weingarten TN, Martin DP, Babovic N, Narr BJ, Sprung J, et al. Allergic reactions during anesthesia at a large United States referral center. Anesth Analg. 2011;113(5): Kim M-H, Lee J-M. Diagnosis and Management of Immediate Hypersensitivity Reactions to Cephalosporins. Allergy Asthma Immunol Res. 2014;6(6): Brockow K, Garvey LH, Aberer W, Atanaskovic-Markovic M, Barbaud A, Bilo MB, et al. Skin test concentrations for systemically administered drugs -- an ENDA/EAACI Drug Allergy Interest Group position paper. Allergy. 2013;68(6): Testi S, Severino M, Iorno M, Capretti S, Ermini G, Macchia D, et al. Nonirritating concentration for skin testing with cephalosporins. J Investig Allergol Clin Immunol. 2010;20(2): Empedrad R, Darter AL, Earl HS, Gruchalla RS. Nonirritating intradermal skin test concentrations for commonly prescribed antibiotics. J Allergy Clin Immunol. 2003;112(3): Romano A, Gaeta F, Valluzzi RL, Maggioletti M, Zaffiro A, Caruso C, et al. IgE-mediated hypersensitivity to cephalosporins: Cross-reactivity and tolerability of alternative cephalosporins. J Allergy Clin Immunol. 2015;136(3): e3. Pipet A, Veyrac G, Wessel F, Jolliet P, Magnan A, Demoly P, et al. A statement on cefazolin immediate hypersensitivity: data from a large database, and focus on the cross-reactivities. Clin Exp Allergy. 2011;41(11): Igea J, Fraj J, Davila I, Cuevas M, Cuesta J, Hinojosa M. Allergy to cefazolin: study of in vivo cross reactivity with other betalactams. Ann Allergy. 1992;68(8): Weber EA. Cefazolin specific side chain hypersensitivity. J Allergy Clin Immunol. 1996;98(4): Somech R, Weber EA, Lavi S. Evaluation of immediate allergic reactions to cephalosporins in non-penicillin-allergic patients. Int Arch Allergy Immunol. 2009;150(3): Warrington RJ, McPhillips S. Independent anaphylaxis to cefazolin without allergy to other beta-lactam antibiotics J Allergy Clin Immunol. 1996;98(460-2). 73

74 Mertes PM, Malinovsky JM, Jouffroy F, Aberer W, Terreehorst I, Brockow K, et al. Reducing the Risk of Anaphylaxis During Anesthesia: 2011 Updated Guidelines for Clinical Practice. J Investig Allergol Clin Immunol. 2011;21(6): Brown SG. Clinical features and severity grading of anaphylaxis. J Allergy Clin Immunol. 2004;114(2): Ebo DG, Fisher MM, Hagendorens MM, Bridts CH, Stevens WJ. Anaphylaxis during anaesthesia: diagnostic approach. Allergy. 2007;62(5): Romano A, Gueant-Rodriguez RM, Viola M, Amoghly F, Gaeta F, Nicolas JP, et al. Diagnosing immediate reactions to cephalosporins. Clin Exp Allergy. 2005;35(9): Romano A, Gaeta F, Valluzi RL, Caruso C, C. A, Viola M, et al. Diagnosing nonimmediate reactions to cephalosporin. J Allergy Clin Immunol. 2012;129(4): Mirakian R, Leech SC, Krishna MT, Richter AG, Huber PA, Farooque S, et al. Management of allergy to penicillins and other beta-lactams. Clin Exp Allergy. 2015;45: Dickson S, Salazar K. Diagnosis and management of immediate hypersensitivity reactions to cephalosporins. Clin Rev Allergy Immunol. 2013;45: Vultaggio A, Virgili G, Gaeta F, Romano A, Maggi E, Matucci A. High serum beta-lactams specific/total IgE ratio is associated with immediate reactions to beta-lactams antibiotics. PLoS One. 2015;10(4):e Johansson SG, Adedoyin J, Van Hage M, Grönneberg R, Nopp A. False-positive penicillin immunoassay: An unnoticed common problem. J Allergy Clin Immunol. 2013;132(1): Baldo BA, Pham NH. Drug Allergy. Clinical Aspects, Diagnosis, Mechanisms, Structure- Activity Relationships. New York: Springer;

75 Part 2: Diagnosing cefazolin hypersensitivity: lessons from dual labelling cytometry A.P. Uyttebroek V. Sabato N. Cop I.I. Decuyper M.A. Faber C.H. Bridts C. Mertens M.M. Hagendorens L.S. De Clerck D. G. Ebo Published as Clinical Communication in JACI in practice 2016; 4(7):

76 Short summary Basophil activation test is the sole in vitro diagnostic for immediate drug hypersensitivity reactions to cefazolin. Dual labelling experiments show that, unlike CD203c, appearance of CD63 is not a sensitive marker for diagnostic use. To the editor, In the absence of a cefazolin-specific IgE assay and provocation tests with this intravenous cephalosporin being hazardous, diagnosis of immediate cefazolin hypersensitivity predominantly relies upon skin tests. However, validation of cefazolin skin tests has focussed on assessing the irritating potential in healthy (exposed) control individuals and data on sensitivity of cefazolin skin testing remain scarce. Therefore, the availability of another diagnostic test could be of interest to further evaluate the reliability of cefazolin skin tests and to add to the diagnosis. Upon encounter of allergen crosslinking surface bound IgE antibodies, basophils upregulate activation and degranulation markers and release bioactive mediators such as histamine. Both the immunophenotypic alterations and the release of histamine are quantifiable on a single cell level by multicolour flow cytometry in the basophil activation test (BAT/HistaFlow) [1]. BAT/Histaflow is being increasingly introduced in the diagnostic approach of immediate drug hypersensitivity reactions (IDHR) [2], and have mainly been applied to document IDHR to β-lactam antibiotics, particularly aminopenicillins and to a lesser extent quinolones. For the time being BAT has not been thoroughly validated in immediate cephalosporin hypersensitivity. In this study we sought to evaluate the BAT in cefazolin hypersensitivity, as allergy to this first generation cephalosporin can constitute an important cause of perioperative anaphylaxis with serious consequences of diagnostic error [3]. Patients and controls were selected as described in [4]. Eighteen patients suffering from perioperative anaphylaxis after intravenous injection of cefazolin demonstrating a positive skin test for the drug, and all other possible causes excluded were selected. Seventeen individuals exposed to cefazolin during anesthesia with a negative skin test and another identifiable cause served as exposed control individuals. Investigations were performed between 0.3 and 38 months (median 3 months) after the reaction. Patients provided informed consent in accordance with the Declaration of Helsinki. Skin testing with cefazolin (Cefazoline 1g, Sandoz, Vilvoorde, Belgium) was performed according to the ENDA recommendations [5], but a maximum concentration of 20 mg/ml was applied [4]. The BAT was performed as described elsewhere [1]. Briefly, within 3h after sampling in endotoxin-free heparinised tubes, aliquots of 100 μl whole blood were incubated (20 min, 37 C) with dilution buffer (negative control), positive control (anti-ige 10μg/mL, BD Biosciences, Erembodegem, Belgium) or serial dilutions of cefazolin ranging from 11 µmol/l to 1100 µmol/l-µmol/l. Cells were stained with 20 76

77 µl of monoclonal anti-human IgE (clone GE-1, Sigma Aldrich GmBH, Steinheim, Germany) labeled with Alexa Fluor 488 (Molecular Probes, Invitrogen, Paisley, UK) and 10 µl of monoclonal anti-human CD63- PE (clone H5C6, BD Bioscience), 10 µl CD203c-APC (clone NP4D6, BD Bioscience). Flow cytometric characterization of basophils relied upon a combination of side scatter (SCC), anti-ige and CD203c positivity. Results were expressed as %CD63 and %CD203c upregulated basophils. In all patients diagnosis of cefazolin hypersensitivity was established with IDR. Two patients were nonresponsive to positive control stimulation in the BAT and were withdrawn from further analysis. In the 16 responders, spontaneous and anti-ige induced appearance of CD63 and upregulation of CD203c was comparable to results in exposed control individuals (data not shown). As displayed in figure 1A and 1B, in exposed control individuals cefazolin-induced activation showed no upregulation for CD203c and CD63 on basophils. In contrast, patients demonstrated a dose-dependent up-regulation of both surface markers. For both read-outs, two-graph receiver operating characteristic (TG-ROC) analysis revealed stimulation with 1100 µmol/l to be most discriminative between patients and exposed control individuals. For this stimulation concentration TG-ROC analysis generated a diagnostic threshold value of 5 % for CD63 and CD203c net upregulation (figure 1B and 1D). However, for this threshold the CD63-BAT was positive in only 6/16 patients (sensitivity 38%, 95% CI 15%-65%) and 1/17 control individuals (specificity 94%, 95% CI %), whereas the CD203c read-out was positive in 12/16 patients (sensitivity 75%, 95% CI 47-98%) and 1/17 control individuals (specificity 94%, 95% CI %). Figure 1E and 1F displays the individual percentages of CD63 and CD203c upregulation and number of positive BAT in responsive patients and exposed control individuals. The potentials and limitations of BAT in the diagnosis of IDHR to β-lactam antibiotics has mainly been investigated in penicillins, particularly amoxicillin [2]. To our knowledge this is the first study that assesses the applicability of BAT with dual labelling of CD63 and CD203c in immediate cefazolin hypersensitivity. Our data confirm that the diagnostic outcome of the BAT significantly varies according to the applied read-out with upregulation of the lineage specific ectoenzyme nucleotide pyrophosphate phosphodiesterase-3 (CD203c) to be more sensitive than the appearance of lysosomeassociated membrane glycoprotein-3 (CD63) [6]. Actually, the IL-3 free CD63-BAT attained a sensitivity of 38% and specificity of 94%, whereas the CD203c read-out yielded a sensitivity of 75% and specificity of 94%. Taking into account the 2 non-responders in BAT; overall sensitivity of the CD203c read-out was 67% (95%CI: 41-87%). The exact reason for this dissociation between CD63 and CD203c is unclear but could mirror differential signalling and regulation of degranulation pathways [7]. One explanation for this differential signalling could relate to the fact that the activation signal of cefazolin to be too weak to trigger CD63 upregulation, as it has been hypothesized CD203c expression 77

78 to have a higher sensitivity to activation (i.e. takes fewer cross-links) [8]. Alternatively, this dissociation might suggest an alternative activatory mechanism independent from IgE/FcεRI cross-linking. The latter could explain why most patients suffered from an IDHR to cefazolin without prior exposure. Conclusion In summary, in the absence of a reliable sige assay, BAT is the sole in vitro diagnostic for IDHR to cefazolin. However, from our dual labelling experiments it appears that, appearance of CD63 is too insensitive in the diagnosis of cefazolin hypersensitivity. On the contrary, CD203c seems to be a better marker for diagnostic use. Whether the dissociation between CD63 and CD203c suggests a non-ige mediated activation of the cells remains to be investigated. 78

79 Figure 1 A-F: Dose finding experiments for CD203c+ (A) and CD63c ++ (C) between patients (open circles) and control individuals (closed circles). Net percentage was obtained by subtracting the spontaneous CD203c and CD63 expression from the value obtained with the drug. Results are expressed as mean and SEM. It discloses 1100 µmol/l to be the optimal stimulation concentration for both read outs. For those optimal stimulation concentration a two graph receiver operating curve reveals the optimal cut off to be 5% for both read outs (B and D). Individual percentages of CD203c (E) and CD63 (F) upregulation in cefazolin activated basophils in patients and controls. The dashed line indicates the TG-ROC calculated decision threshold for positivity (5%). 79

80 References Bridts CH, Sabato V, Mertens C, Hagendorens MM, De Clerck LS, Ebo DG. Flow cytometric allergy diagnosis: basophil activation techniques. Methods in molecular biology (Clifton, NJ). 2014;1192: Mangodt EA, Van Gasse AL, Decuyper I, Uyttebroek A, Faber MA, Sabato V, et al. In vitro Diagnosis of Immediate Drug Hypersensitivity: Should We Go with the Flow? Int Arch Allergy Immunol. 2015;168(1):3-12. Gonzalez-Estrada A, Pien LC, Zell K, Wang XF, Lang DM. Antibiotics are an important identifiable cause of perioperative anaphylaxis in the United States. J Allergy Clin Immunol Pract. 2015;3(1):101-5 e1. Uyttebroek AP, Decuyper II, Bridts CH, Romano A, Hagendorens MM, Ebo DG, et al. Cefazolin Hypersensitivity: Toward Optimized Diagnosis. Allergy Clin Immunol Pract. Brockow K, Garvey LH, Aberer W, Atanaskovic-Markovic M, Barbaud A, Bilo MB, et al. Skin test concentrations for systemically administered drugs -- an ENDA/EAACI Drug Allergy Interest Group position paper. Allergy. 2013;68(6): Sturm EM, Kranzelbinder B, Heinemann A, Groselj-Strele A, Aberer W, Sturm GJ. CD203cbased basophil activation test in allergy diagnosis: characteristics and differences to CD63 upregulation. Cytometry Part B, Clinical cytometry. 2010;78(5): MacGlashan D, Jr. Marked differences in the signaling requirements for expression of CD203c and CD11b versus CD63 expression and histamine release in human basophils. International archives of allergy and immunology. 2012;159(3): Knol EF. Requirements for effective IgE cross-linking on mast cells and basophils. Molecular nutrition & food research. 2006;50(7):

81 Chapter 4: Perioperative hypersensitivity reactions in Antwerp: a 16 year survey. 81

82 82

83 Introduction Perioperative hypersensitivity reactions are rare events, with an estimated prevalence varying between 1/ /1.750 anesthetics procedures (1). For example, an early study in New Zealand reported a prevalence of moderate to severe anaphylaxis in general anesthesia of 1/1750 (2). In a Japanese survey the reported prevalence was 1/ (3), while in a recent French study the estimated prevalence of 100.6/ surgeries (4). These considerable differences can be attributable to several reasons. Firstly, there is still no universally accepted definition of perioperative hypersensitivity and different classification systems are applied to identify and stratify the patients. Secondly, the perioperative hypersensitivity scene displays specific challenges with regard to fast and correct recognition of symptoms. Actually, differentiating between a genuine perioperative hypersensitivity reaction and masqueraders is not always straightforward. For example, hypotension might merely represent a pharmacological side effect of some anesthetics; bronchospasm could be a mechanical consequence of intubation or insufficient sedation. Furthermore, erythema and urticaria can easily be overlooked as they can be hidden by the surgical drape that covers the patient (1). Lastly, different diagnostic algorithms are often followed, mainly as a result of the accessibility and availability of some diagnostics. Along with the differences in incidence and prevalence, several surveys have revealed that the causes of perioperative hypersensitivity also display significant geographical differences (1,5 16). For instance, in Europe, French studies disclosed curarizing NMBA, latex, antibiotics and hypnotics being the predominant causes of perioperative hypersensitivity. In contrast, in Demark, chlorhexidine seems to be the primary culprit (5). In Spain, antibiotics, NMBA (15) and latex appear to be the most prevalent offenders (8). In the UK the top 3 culprits were antibiotics, NMBA and morphine (9). Finally, in Belgium, Antunez et al (6) reported NMBA, antibiotics, latex and chlorhexidine to be the most significant causes of perioperative hypersensitivity reactions. Findings that were largely confirmed by a prior analysis on a part of our cohort (17). This survey primarily aims at extending our preliminary analysis with an additional 5 years and to study various parameters such as discontinuation of the intervention, use of epinephrine, changes over time, optimal time-point for conformational testing, added value of basophil activation tests and significance of peak tryptase. As a matter of facts, unlike the descriptions of demographics and causative compounds, data on these parameters remain scarce. 83

84 Methods Patients Patients with a clinical suspicion of a perioperative hypersensitivity reaction were referred to the outpatients clinic of the Department Immunology Allergology Rheumatology of the Antwerp University Hospital between June 2001 and December All patients underwent a history taking by trained physicians who also scrutinized the available anesthetic charts and surgical notes. Data on demographics, prior allergies, underlying (predisposing) comorbidity and conditions, previous anesthesia, clinical presentation, time elapsed between administration of drugs and onset-of-reaction, drugs applied during anesthesia (not restricted to anesthetics), peak tryptase levels, and management of the acute reaction were collected. Perioperative hypersensitivity was defined and stratified according to the Brown s classification (18), (table 1), provided the reaction occurred during or within 2 hours after surgery. All later events or reactions not full-filing the Brown s criteria and without evidence for mast cell activation (vide infra). To study trends in causative agents over time the observation period was arbitrarily divided into two time-periods, viz and Table 1. Brown s criteria Grade Symptoms Examples 1 Mild (skin and subcutaneous tissues only) 2 Moderate (features suggesting respiratory, cardiovascular, or gastrointestinal involvement) 3 Severe (hypoxia, hypotension, or neurologic compromise) Clinical features of and severity grading of anaphylaxis (18) Generalized erythema, urticaria, periorbital edema or angioedema Dyspnea, stridor, wheeze, nausea, vomiting, dizziness (presyncope), diaphoresis, chest or throat tightness, or abdominal pain Cyanosis or SpO2 # 92% at any stage, hypotension (SBP < 90 mmhg in adults) confusion, collapse, LOC, or incontinence Tryptase Baseline and peak tryptase levels were measured using the ImmunoCAP tryptase technique (Phadia, Thermo Fisher Scientific, Uppsala, Sweden). Mast cell activation was defined when the peak tryptase level equaled or exceeded (1.2*basal tryptase +2) (19). 84

85 Total and sige measurement Total and specific IgE antibodies (sige) were quantified by FEIA ImmunoCAP, Thermo Fisher Scientific, Uppsala, Sweden. Specific IgE to latex and chlorhexidine were quantified systematically. sige to morphine, suxamethonium, pholcodine, rocuronium and atracurium; penicilloyl G, penicilloyl V, ampicilloyl, amoxicilloyl, gelatin, alfa-gal, ethylene oxide (FEIA ImmunoCAP, Thermo Fisher Scientific, Uppsala, Sweden) when indicated. ImmunoCAP for rocuronium and atracurium were obtained by experimental prototypes developed for research use. sige rocuronium became available in 2006, sige atracurium in 2012 (20,21). Decision thresholds were set at 0.35 kua/l, except for the rocuronium and suxamethonium for which a drug-specific cut-off of 0.11 kua/l and 0.13 kua/l respectively was used (21). Skin testing Skin prick tests (SPT) and intradermal tests (IDT) were performed according to the recommendations of the Société Francaise d Anesthésie et de reanimation (SFAR) (22 24), and the ENDA/EAACI drug allergy interest group (25). Briefly, SPT were performed on the ventral part of the forearm and implied histamine 10 mg/ml (HAL Allergy Benelux BV) as a positive control to assess skin test responsiveness, a saline buffer solution as a negative control to exclude cutaneous hyper-responsiveness. All administered drugs, potential cross-reactive compounds and alleged safe alternatives were tested according to (26). SPT were read after 15 minutes and considered positive when the wheal of the wheal and flare reaction exceeded 3 mm. If needed, patients with negative SPT had additional IDT. IDT responses were considered positive when the wheal of the wheal and flare reaction equaled or exceeded 5 mm ( 8 mm for curares) or when the wheal had doubled as compared to the injection bleb. Maximal concentrations used for SPT and IDT were used according to (25). During the study period, the recommendation for maximal concentration tests for IDT with rocuronium was changed from a 1/100 dilution (0.01 mg/ml) to 1/200 (0.05 mg/ml) dilution (27). For cefazolin, since 2013, the maximal skin test concentration was 20 mg/ml, as this concentration was shown to be non-irritating and more sensitive than the recommended concentration of 2 mg/ml (28). Basophil activation test The technical principles of flow cytometric analysis of in vitro activated basophils are described elsewhere (29). The test includes stimulation with a negative control (stimulation buffer), a positive control (anti-ige) and stimulation with relevant drugs or related compounds. Results are expressed as net percentage of CD63 positive basophils, that is, by subtracting spontaneous CD63 expression (negative control) from the value obtained with allergen (drug) stimulation. 85

86 Drug provocation testing (DPT) When indicated, graded drug challenges up to a cumulative dose equaling or exceeding the therapeutic dose, were performed with the drug of interest, except for NMBA because of obvious reasons. Results Patients In total, 626 patients with an alleged perioperative hypersensitivity reaction were referred for diagnostic work-up. Median age of the patients was 46.7 years (range 1.5 to 88). Based on the clinical description and timing, 534 (85.3%) patients were diagnosed as having suffered from perioperative hypersensitivity reactions and 74 (11.8%) as having presented an unrelated event. The majority of the latter group suffered from mild isolated bronchospasm or discrete hypotension without evidence of mast cell activation. Finally, 18 (2.8%) patients were excluded from further analysis, as information about the clinical reaction was incomplete or unclear. Hypersensitivity reactions Severity In the 534 patients in who a perioperative hypersensitivity reaction was suspected, 344 (64 % (95CI 60-80)) had presented with a severe reaction (grade 3), 80 (15 % (95CI 12-18)) with a moderate reaction and 110 (21 % (95CI 17-24)) with a mild reaction, respectively. Noteworthy, 57 patients (11 % (95CI 8-14)), mainly severe grade 3 reactors required cardiac massage. In 341 anesthetics records information was available about the administration of epinephrine. In total epinephrine was administered in 196 patients, viz. 180/238 patients with a severe reaction, 14/45 patients with a moderate reaction and in 2/58 patients with a mild reaction. Information about the outcome of surgery was available in 406/534 cases. In 111 (27 % (95CI 23-32)) of them, the surgical interventions were aborted. The majority of these patients (102/111) displayed a severe reaction with pronounced hypotension and or bronchospasm and had their intervention postponed. 86

87 Culprits At the end of the allergological workup, at least one culprit was identifiable in 351/534 (66 %) of the reactions. In the remaining 183 (34 %), no offender was identifiable. As displayed in figure 2 NMBA were the predominant cause of perioperative hypersensitivity reactions. Over the entire study period, the second most important cause was latex, followed by antibiotics (mainly cefazolin), and chlorhexidine respectively. Importantly, in 24 patients a double sensitization was demonstrable. The most frequent observed combinations were NMBA with latex or NMBA with chlorhexidine. A female predominance was demonstrable for NMBAs, latex and antibiotics. In contrast, chlorhexidine hypersensitivity was more prevalent in men (76%). Also, median age in patients with a hypersensitivity to this antiseptic was significantly higher (66 years) than for other patients (52 years) (Pearson Chi square, P < 0.01). Figure 2: Causal agents in patients with a perioperative hypersensitivity reaction. NMBA was the most important culprit found in 164/534 (31%) patients, followed by latex in 76/534 (14.2%), antibiotics (ABC) in 51/534 (10%) and chlorhexidine in 34/534 (6.36%) patients. In 50/535 (9%) a miscellaneous compound was identified as the offender. In 182/534 (34%) no culprit was identified. 87

88 Table 2A displays the most prevalent offenders in relationship to the clinical severity of the acute reaction. Although intravenously injected drugs such as NMBA or antibiotics generally trigger more severe reactions, the severity of the reaction is not absolutely predictive for the cause. Actually, all culprits can give rise to severe, moderate or mild reactions. Table 2B shows the relationship between the culprit and the time elapsed between induction and reaction divided into 1) before induction, 2) less than 30 minutes after induction or 3) more than 30 minutes after reaction. From this figure it emerges that also that time of onset of reaction cannot be used to predict the culprit drug. Although intravenously administered drugs predominantly triggered reactions within half an hour, we have observed that, although rare, these reactions could also appear later in surgery. Alternatively, compounds such as latex or chlorhexidine, who usually trigger reactions after a longer contact period, can also give rise to rapid, early reactions. Table 2A: Severity of reaction (Brown's criteria) Causal agents in perioperative anaphylaxis in relation to the severity of the reaction Mild Moderate Severe Culprit N % N % N % unknown Antibiotics Chlorhexidine Latex Miscellaneous NMBA Table 2B: Timing of reaction: Causal agents in relation to time elapsed between induction and reaction. Culprit before induction <30m after induction >30 min after induction N % N % N % Unknown NMBA Antibiotics Chlorhexidine Latex Miscellanneous

89 Trends in culprit Table 3 shows a comparison of the predominant culprits between and From this figure it emerges hypersensitivity to latex to have decreased (Pearson, P=0.0052), whereas the significance of antibiotics (mainly cefazolin) has increased (Pearson, P=0.0098). As a matter of fact, antibiotics have supplanted latex as the second most important cause of perioperative hypersensitivity. Finally, it is shown that the group of miscellaneous causes has also gained importance (P=0.0453), whereas the sensitization rates for NMBA and chlorhexidine have remained unaltered over time. Table 3: comparison of the culprits found during vs NS = not significant Culprit N Column % N Column % P value (Pearson) Unknown % % NS Others % % <0.05 Antibiotics % % <0.01 Chlorhexidine % % NS Latex % % <0.01 NMBA % % NS Mast cell activation In total 139 out of the 180 (77%) patients in whom paired peak and baseline tryptase values were available demonstrated a significant mast cell activation. When compared to severity grade of the reactions, mast cell activation was predominantly seen in severe reactions (125/147; 85%) versus 8/20 (40%) in moderate and 6/13 (46 %) in mild reactions (Pearson P<0.01). Importantly, in 111 (80%) patients with a clear mast cell activation, a culprit was demonstrable. Two patients were finally diagnosed with a clonal mast cell disorder. Both displayed a baseline tryptase 11.4 µg/l. 89

90 Timing of allergological investigation Table 4 displays the data of hypersensitivity to antibiotics in 204 antibiotic exposed patients. It reveals that testing for AB should not be postponed. Actually, our data suggest that early testing - i.e. within 6 weeks after the acute event - identifies significantly more patients than when testing is performed between 6 weeks and 3 years (35% vs 19%) (Pearson, P < 0.05). For NMBA, latex and chlorhexidine, no significances were demonstrated between early and later testing (data not shown). The number of patients tested after 3 years was too small for robust statistical analyses. Table 4. Diagnosis made in AB-exposed patients in relation to timing of testing N % N % <6 weeks weeks - 3 years > 3 years AB Other diagnosis NMBA In total, 164/534 (31 %) of patients were diagnosed with an NMBA hypersensitivity. The majority of patients presented with a hypersensitivity to rocuronium (123/164; 75%), followed by atracurium (14/164; 8 %) and cisatracurium (12/164; 7 %). Suxamethonium hypersensitivity was found in 10/164 (6 %) of the patients. Vecuronium and mivacurium hypersensitivity was present in 1/164; 0.6%) and 4/164 (2.4%) patients, respectively. Rocuronium hypersensitivity In 110 out of the 123 patients with a documented rocuronium hypersensitivity results of triple testing were available. and revealed that diagnosis was unequivocally endorsed by at least 2 tests in 85/110(77%) patients. In 19 patients, diagnosis relied solely upon skin testing as quantification of sige and BAT yielded negative results. In 9 out the 11 rocuronium patients with negative skin test responses BAT contributed to correct diagnosis. In 5/123 patients, diagnosis relied solely upon the BAT while skin testing and/or sige testing remained negative. In 114 rocuronium hypersensitive patients both titers of sige rocuronium and morphine were available. 69 patients were positive in both tests; 22 patients were negative in both assays. Eleven patients displayed a positive sige rocuronium but a negative sige morphine. Vice versa, 12 patients were positive for sige morphine but not for sige rocuronium. 90

91 Atracurium hypersensitivity In total, 14 atracurium allergic patients were identified. Triple testing for atracurium was available in 12. Figure 3 displays the results of triple testing. Specific IgE morphine was known in 13/14 patients and positive in only 1 case. Alternatively, sige rocuronium was available in 8/14 patients. In three of them it was positive. Figure 3: Diagnostic Venn diagram of atracurium allergic patients (n=14) 91

92 Unrelated events Unrelated events mainly included pharmacological/toxic reactions (13/74; 17.57%) or were thought to be a consequence of the mechanic manipulation during intubation, or as a result of insufficient anesthesia (33/74; 44.59%). In the remaining 28 patients other mechanisms were suspected (e.g. aspiration, underlying comorbidities or vagal reactions). 92

93 Discussion In the current study we analysed the data from 626 patients referred to our outpatients clinic for alleged perioperative hypersensitivity reactions between January 2001 and December Based upon a careful revision of the anesthetic and surgical notes perioperative hypersensitivity was likely to have occurred in 534 cases (85%). In the remainder patients reactions appeared to be unrelated or too poorly described for correct classification. In the individuals with a clinical suspicion of a perioperative hypersensitivity reaction a culprit was demonstrable in two-thirds of the patients. This greatly parallels findings from other groups who could identify an offender up to 72 % of the cases (4,6,7,11 16,30 34). Overall, the predominant causes in our region were curarizing neuromuscular blocking agents, Hevea latex, β-lactam antibiotics and chlorhexidine. However, as recently shown by others for latex (1,5,35,36) and NMBA (5,12,13,37) it appears that the contribution of these offenders might vary over time. Actually, in our survey β-lactam antibiotics have supplanted latex in the second half of study period. Anyway, our findings confirm that perioperative hypersensitivity reactions display distinct geographical characteristics that might vary over time. Variation that might result from primary prevention measures (38) or, as shown in Norway for NMBA, the withdrawal of a potential sensitizer (39 42). Obviously, part of our findings might also be related to optimized diagnostics. Furthermore, our data confirm that perioperative hypersensitivity reactions might also display gender-related variations with hypersensitivity to NMBA, latex, β-lactams, but not chlorhexidine to be more prevalent in (middle-aged) women than (older) men. It has previously been shown chlorhexidine hypersensitivity to be more present in men undergoing urological or cardiothoracic surgery (43). In our study we also stratified the patients according to severity of the reaction. 64% of our patients had suffered from grade 3 reactions with profound hypotension and severe bronchospasm. Overall one-third of the cases had their intervention aborted and were scheduled for future surgery. 64% of these patients, mainly type 3 reactors, were treated with epinephrine. These observations are in line with the findings of others (12,44). Perioperative hypersensitivity can be life-threatening with serious consequences from diagnostic error. For correct diagnosis a thorough history, viz. revision of anesthetic and surgical notes is of the essence to orient the confirmatory testing. As with other surveys of this nature it appeared that although these notes can be particularly useful, there is no parameter that is absolutely predictive for the cause. Although reactions to intravenously injected drugs such as NMBA, anesthetics, opioids or antibiotics tend to be more severe and rapid in onset, severity nor time elapsed between administration and start of the reaction were absolutely predictive. Thereby confirming that one 93

94 cannot predict the culprit based on onset nor clinical manifestation of the reaction (45). Actually, significant hypersensitivity reactions with resuscitation and discontinuation of the surgical procedure was not restricted to intravenous drugs but also demonstrable for latex and chlorhexidine. Another relevant observation of our study relates to serum tryptase. Although this parameter does not add to the identification of the culprit it is clear that anesthesiologists should be encouraged to quantify both peak and baseline tryptase. Actually, quantification of tryptase might not only shed light upon the underlying pathomechanism of a perioperative hypersensitivity reaction but also unveil the presence of a (clonal) mast cell disorder (46,47). As addressed in (23,48), there is an ongoing debate on the optimal timing of testing. There appears to be consensus on not to postpone testing unnecessarily. However, it remains unclear whether early testing, within 4-6 weeks might negatively affect the outcome of the tests due to a refractoriness of cutaneous mast cells/basophils or temporarily negativation of specific IgE antibodies. Our data do not endorse such an assumption, as a matter a fact, it seems diagnosis of β-lactam antibiotic hypersensitivity benefits from early testing. For other culprits, early testing seems not to affect the outcome. Alternatively our study discloses that diagnosis of perioperative hypersensitivity might benefit from the introduction of novel in vitro diagnostics such as quantification of sige antibodies and basophil activation experiments. For example, for rocuronium, 11 patients negative for skin testing could correctly be diagnosed thanks to the BAT and/or sige towards rocuronium. Furthermore, in a vast majority of rocuronium patients, positive skin testing was confirmed by at least one of the two in vitro tests, making it unlikely that rocuronium hypersensitivity in our survey is overestimated due to inappropriate irritating skin test dilutions (49,50). The sensitization profiles of our rocuronium allergic patients concerning IgE for rocuronium and morphine stress the need for drug specific tests. Actually, sige morphine, a biomarker for sensitization towards tertiary and quaternary ammonium ions, structures thought to be the major allergenic epitope of the NMBA (51,52), seems far from absolutely depicting NMBA hypersensitivity. Actually sige to morphine was negative in about one-third of the patients hypersensitive to rocuronium and all but one patient with atracurium hypersensitivity. Alternatively, it should be kept in mind that quantification of sige to morphine and Papaver somniferum is of no use to diagnose opiate allergy (53). As a matter of fact, as a result of uncertainties associated with skin tests for these potent non-specific histamine releasers, the only reliable test to document genuine opiate allergy appears to be the basophil activation test (54). In essence, this study confirms that perioperative anaphylaxis displays distinct geographical differences concerning culprit drugs, and that the causal agents can vary over time. Furthermore we have shown that onset of the reaction and clinical phenotype are not absolutely predictive for the 94

95 cause and that confirmatory testing should not be restricted to severe reactors who were resuscitated and had their intervention aborted. Although skin testing still merits the status of primary diagnostic investigation to document perioperative hypersensitivity novel in vitro tests can add to the diagnostic management, particularly when skin tests yield negative or equivocal responses. Moreover, basophil activation tests might not only help to identify the culprit agent(s) but also in the assessment of crossreactivity and tailoring the safe alternative(s) for future anesthesia. 95

96 References Mertes PM, Volcheck GW, Garvey LH, Takazawa T, Platt PR, Guttormsen AB, et al. Epidemiology of perioperative anaphylaxis. Presse Med. 2016;45(9): Galletly DC, Treuren BC. Anaphylactoid reactions during anaesthesia. Seven years exeperience of intradermal testing. Anaesthesia. 1985;40: Takazawa T, Mitsuhata H, Mertes PM. Sugammadex and rocuronium-induced anaphylaxis. J Anesth. Springer Japan; 2016;30(2): Mertes PM, Alla F, Tréchot P, Auroy Y, Jougla E, le Group d études des reactions anaphylactoïdes peranesthésiques. Anaphylaxis during anesthesia in France: An 8-year national survey. J Allergy Clin Immunol. 2011;128(2): Garvey LH, Roed-Petersen J, Menné T, Husum B. Danish Anaesthesia Allergy Centre - preliminary results. Acta Anaesthesiol Scand. 2001;45(10): Antunes J, Kochuyt A-MM, Ceuppens JL. Perioperative allergic reactions: Experience in a Flemish referral centre. Allergol Immunopathol (Madr). 2014;42(4): Gonzalez-Estrada A, Pien LC, Zell K, Wang X-FF, Lang DM. Antibiotics are an important identifiable cause of perioperative anaphylaxis in the United States. J Allergy Clin Immunol Pract. 2015;3(1): e1. Berroa F, Lafuente A, Javaloyes G, Cabrera-Freitag P, De La Borbolla JM, Moncada R, et al. The incidence of perioperative hypersensitivity reactions: A single-center, prospective, cohort study. Anesth Analg. 2015;121(1): Meng J, Rotiroti G, Burdett E, Lukawska JJ. Anaphylaxis during general anaesthesia: experience from a drug allergy centre in the UK. Acta Anaesthesiol Scand. 2017;61(3): Mertes PM, Tajima K, Regnier-Kimmoun MA, Lambert M, Iohom G, Gueant-Rodriguez RM, et al. Perioperative anaphylaxis. Med Clin North Am. Elsevier Inc; 2010;94(4): Dong SW, Mertes PM, Petitpain N, Hasdenteufel F, Malinovsky J, Members of the GERAP. Hypersensitivity reactions during anesthesia. Results from the ninth French survey ( ). Minerva Anestesiol. 2012;78: Gurrieri C, Weingarten TN, Martin DP, Babovic N, Narr BJ, Sprung J, et al. Allergic reactions during anesthesia at a large united states referral center. Anesth Analg. 2011;113(5): Guyer AC, Saff RR, Conroy M, Blumenthal KG, Camargo CA, Long AA, et al. Comprehensive allergy evaluation is useful in the subsequent care of patients with drug hypersensitivity reactions during anesthesia. J allergy Clin Immunol Pract. 2015;3(1): Harboe T, Guttormsen AB, Irgens A, Dybendal T, Florvaag E. Anaphylaxis during Anesthesia in Norway. Anesthesiology. 2005;102(5): Lobera T, Audicana M, Pozo M, Blasco A, Fernández E, Cañada P, et al. Study of Hypersensitivity Reactions and Anaphylaxis During Anesthesia in Spain. J Investig Allergol Clin Immunol. 2008;18(185): Krishna MT, York M, Chin T, Gnanakumaran G, Heslegrave J, Derbridge C, et al. Multi-centre retrospective analysis of anaphylaxis during general anesthesia in the United Kingdom: Aetiology and diagnostic performance of acute serum tryptase. Clin Exp Immunol. 2014;178(2):

97 Leysen J, Witte L De, Bridts CH, Ebo DG. Anaphylaxis during general anaesthesia : a 10-year survey 1 at the University Hospital of Antwerp. P Belg Roy Acad Med. 2013;2: Brown SG. Clinical features and severity grading of anaphylaxis. J Allergy Clin Immunol. 2004;114(2): Sprung J, Weingarten TN, Schwartz LB. Presence or Absence of Elevated Acute Total Serum Tryptase by Itself Is Not a Definitive Marker for an Allergic Reaction. Anesthesiology. 2015;122(3): Johansson SGO. Anaphylaxis to atracurium - a non-qai-dependent reaction? Acta Anaesthesiol Scand. 2012;56(2): Ebo DG, Venemalm L, Bridts CH, Degerbeck F, Hagberg H, De Clerck LS, et al. Immunoglobulin E Antibodies to Rocuronium. Anaesthesiology. 2007;107(2): Société Françaises d Anesthésie et de Réanimation. Reducing the risk of anaphylaxis during anaesthesia. Abbreviated text. Ann Fr Anesth Reanim. 2002;21 Suppl 1:7s 23s. Mertes PM, Laxenaire MC, Lienhart A, Aberer W, Ring J, Pichler WJ, et al. Reducing the risk of anaphylaxis during anaesthesia: guidelines for clinical practice. J Investig Allergol Clin Immunol. 2005;15(2): Société française d anesthésie et réanimation (Sfar), Société française d allergologie (SFA). Reducing the risk of anaphylaxis during anaesthesia. Short text. Ann Fr Anesth Reanim. 2011;30(3): Brockow K, Garvey LH, Aberer W, Atanaskovic-Markovic M, Barbaud A, Bilo MB, et al. Skin test concentrations for systemically administered drugs -- an ENDA/EAACI Drug Allergy Interest Group position paper. Allergy Jun;68(6): Mertes PM, Jouffroy L, Working Group for the SFAR, SFA, Aberer W, Terreehorst I, et al. Reducing the Risk of Anaphylaxis During Anesthesia 2011 Updated Guidelines for Clinical Practice. J Investig Allergol Clin Immunol. 2011;21(6): Mertes PM, Moneret-Vautrin DA, Leynadier F, Laxenaire M-C. Skin reactions to intradermal neuromuscular blocking agent injections: a randomized multicenter trial in healthy volunteers. Anesthesiology Aug;107(2): Uyttebroek AP, Decuyper II, Bridts CH, Romano A, Hagendorens MM, Ebo DG, et al. Cefazolin Hypersensitivity: Toward Optimized Diagnosis. J Allergy Clin Immunol Pract. 2016;4(6):3 7. Bridts CH, Sabato V, Mertens C, Hagendorens MM, Clerck LS De, Ebo DG. Flow Cytometric Allergy Diagnosis: Basophil Activation Techniques. In: Gibbs BF, Falcone FH, editors. Basophils and Mast Cells: Methods and Protocols, Methods in Molecular Biology. New York, NY: Springer New York; p (Methods in Molecular Biology; vol. 1192). Mirone C, Preziosi D, Mascheri A, Micarelli G, Farioli L, Balossi LG, et al. Identification of risk factors of severe hypersensitivity reactions in general anaesthesia. Clin Mol Allergy. 2015;13(1):11. Light KP, Lovell AT, Butt H, Fauvel NJ, Holdcroft A. Adverse effects of neuromuscular blocking agents based on yellow card reporting in th U.K.: Are there differences between males and females? Pharmacoepidemiol Drug Saf. 2006;15(3): Mertes P-M, Laxenarie M-C. Anaphylactic and anaphylactoid reactions occurring during anaesthesia in France. Seventh epidemiologic survey (January 2001 December 2002). Ann Fr Anesth Rkanim. 2004;23:

98 Mertes PM, Laxenaire M-C, Alla F, le Group d études des reactions anaphylactoïdes peranesthésiques. Anaphylactic and anaphylactoid reactions occurring during anesthesia in France in Anesthesiology. 2003;99(3): Escolano F, Valero A, Huguet J, Baxarias P, de Molina M, Castro A, et al. Prospective epidemiologic study of perioperative anaphylactoid reactions occurring in Catalonia (1996-7). Rev Esp Anestesiol Reanim. 2002;49(6): Low AE, McEwan JC, Karanam S, North J, Kong KL. Anaesthesia-associated hypersensitivity reactions: Seven years data from a British bi-specialty clinic. Anaesthesia. 2016;71(1): Trautmann A, Seidl C, Stoevesandt J, Seitz CS. General anaesthesia-induced anaphylaxis: Impact of allergy testing on subsequent anaesthesia. Clin Exp Allergy. 2016;46(1): Garvey LH. Old, New and Hidden Causes of Perioperative Hypersensitivity. Curr Pharm Des. 2016; 2016;22(45): Blaabjerg MSB, Andersen KE, Bindslev-Jensen C, Mortz CG. Decrease in the rate of sensitization and clinical allergy to natural rubber latex. Contact Dermatitis. 2015;73(1):21 8. de Pater GH, Florvaag E, Johansson SGO, Irgens Å, Petersen MNH, Guttormsen AB. Six years without pholcodine; Norwegians are significantly less IgE-sensitized and clinically more tolerant to neuromuscular blocking agents. Allergy. 2017;72(5): Florvaag E, Johansson SGO, Irgens Å, De Pater GH. IgE-sensitization to the cough suppressant pholcodine and the effects of its withdrawal from the Norwegian market. Allergy Eur J Allergy Clin Immunol. 2011;66(7): Florvaag E, Johansson SGO, Öman H, Harboe T, Nopp A. Pholcodine stimulates a dramatic increase of IgE in IgE-sensitized individuals. A pilot study. Allergy Eur J Allergy Clin Immunol. 2006;61(1): Harboe T, Johansson SGO, Florvaag E, Öman H. Pholcodine exposure raises serum IgE in patients with previous anaphylaxis to neuromuscular blocking agents. Allergy Eur J Allergy Clin Immunol. 2007;62(12): Egner W, Helbert M, Sargur R, Swallow K, N H, Garcez T, et al. Chlorhexidine Allergy in 4 Specialist Allergy Centres in the UK, : Clinical Features and Diagnostic Tests. Clin Exp Immunol. 2013; Garvey LH, Belhage B, Krøigaard M, Husum B, Malling H-J, Mosbech H. Treatment with Epinephrine (Adrenaline) in suspected anaphylaxis during anesthesia in Denmark. Anesthesiology. 2011;115(1): Krøigaard M, Garvey LH, Menné T, Husum B, Kroigaard M, Garvey LH, et al. Allergic reactions in anaesthesia: Are suspected causes confirmed on subsequent testing? Br J Anaesth Aug 12;95(4): Bonadonna P, Pagani M, Aberer W, Bilò MB, Brockow K, Oude Elberink H, et al. Drug hypersensitivity in clonal mast cell disorders: ENDA/EAACI position paper. Allergy. 2015;70: Akin C. Mast cell activation syndromes presenting as anaphylaxis. Immunol Allergy Clin North Am. Elsevier Inc; 2015;35(2): Demoly P, Adkinson NF, Brockow K, Castells M, Chiriac a M, Greenberger P a, et al. International Consensus on drug allergy. Allergy. 2014;69(4):

99 Levy JH. Weal and flare responses to intradermal rocuronium and cisatracurium in humans. Br J Anaesth. 2000;85(6): Dhonneur G, Combes X, Chassard D, Merle JC. Skin sensitivity to rocuronium and vecuronium: a randomized controlled prick-testing study in healthy volunteers. Anesth Analg. 2004;98(4): , table of contents. Baldo BA, Fisher MM, Pham NH. On the origin and specificity of antibodies to neuromuscular blocking (muscle relaxant) drugs: an immunochemical perspective. Clin Exp Allergy Mar;39(3): Fisher MM, Baldo BA. Immunoassays in the diagnosis of anaphylaxis to neuromuscular blocking drugs: the value of morphine for the detection of IgE antibodies in allergic subjects. Anaesth Intensive Care Apr;28(2): Van Gasse AL, Hagendorens MM, Sabato V, Bridts CH, De Clerck LS, Ebo DG. IgE to Poppy Seed and Morphine Are Not Useful Tools to Diagnose Opiate Allergy. J Allergy Clin Immunol Pract. 2015;3(3): Leysen J, De Witte L, Sabato V, Faber M, Hagendorens M, Bridts C, et al. IgE-mediated allergy to pholcodine and cross-reactivity to neuromuscular blocking agents: Lessons from flow cytometry. Cytometry B Clin Cytom. 2013;84(2):

100 100

101 Chapter 5: Critical reflection and future perspectives 101

102 102

103 Allergological work up of immediate drug hypersensitivity aims at identifying the culprit drug and offering a safe alternative for future use. Generally this is based upon skin testing, specific IgE determination and basophil activation testing when available, and challenge testing with the culprit and/or alternative medication. In this thesis we present new data on the diagnosis of the curarizing neuromuscular blocking agent atracurium, the 1 st generation cephalosporin cefazolin and we updated the epidemiology of perioperative anaphylaxis. As with all scientific studies of this nature, we have encountered some difficulties that might have affected the outcomes. Nevertheless, we believe that our analyses are robust and not flawed by major methodological errors or limitations. One limitation of our studies is the fact that we did not perform challenge testing with the probable culprit drug, since long considered to be the gold standard test for drug hypersensitivity diagnosis. Especially in the context of perioperative anaphylaxis, where patients are exposed to several drugs administered simultaneously, the absence of drug challenges could be seen as a drawback. However, in most of our patients we deemed challenges to be unnecessary or difficult to justify for various reasons. For a majority of our patients, provocation testing would have been a dangerous endeavor as they had suffered from severe life-threatening reactions to intravenous drugs. This is also in line with current guidelines that consider a life-threatening reaction as a contra-indication for a challenge test. Furthermore, for neuromuscular blocking agents and antiseptics (full-dose) challenges are virtually impossible for obvious practical reasons. Alternatively, for several agents, we were able to perform multiple testing (i.e. the combination of sige and/or skin testing and/or basophil activation testing) allowing to ascertain correct diagnosis without absolute need for provocation testing. With our studies we have pursued this strategy for atracurium. By validating the basophil activation test and specific IgE for atracurium we are now able to more accurately diagnose this hypersensitivity when combining them with the currently accepted golden standard i.e. skin testing. For the future we plan to expand our data concerning new surgeries in our patients. This will allow us to assess the negative predictive value of our diagnostic tools. Another important limitation of validation studies of new diagnostic tools, is the lack of an appropriate control group, viz. individuals being exposed to the drug/compound of interest without a reaction. Actually, the absence of such a healthy control group mainly hampers evaluation of the specificity and negative predictive value. However, healthy individuals are difficult to enroll, because they are not referred for diagnostic evaluation. In our assessments this issue could (to some extent) be circumvented by the standardized evaluation of our patients that included systematic testing for all compounds exposed to. For example, a patient allergic to atracurium could be used as a control for cefazolin. In any way, our approach better reflects the clinical practice, in which physicians will need 103

104 to correctly identify the causative compound(s), rather than to distinguish patients from asymptomatic control individuals. In literature it has been mentioned skin testing with cefalosporines might not be irritative in concentrations higher than the one currently recommended. However, this was never studied appropriately. That is why we chose to use a titration technique in healthy controls to identify the maximal non-irritating skin test concentration. In a second study, we assessed the basophil activation test for cefazolin. Actually, the clinical relevance of a positive basophil activation test in a patient with a negative skin test remains unclear. Furthermore, the dissociation between CD63 and CD203c upregulation needs to be studied. Determining the positive predictive value of a positive skin test with a non-irritant concentration is not justified from ethical point of view. However, as also stated by Aberer (1), an important step forward would be the determination of the negative predicative value of skin test and or BAT with cefazolin as this would avert overdiagnosis of cefazolin hypersensitivity and incorrect use of alternative antibiotics. In a recent study, Kheir et al demonstrated that the majority of patients given vancomycin as alternative antibiotic are incorrectly dosed leading to insufficient prophylaxis (2). This emphasis the need for correct diagnosis of hypersensitivity. Therefore we recently decided to perform provocation tests in patients displaying negative skin tests for this cephalosporin, irrespective the result of the basophil activation test. From these data we will learn how combining these techniques can lead to a more robust diagnosis. The fact that our epidemiological survey spans over 16 years also introduced some inevitable limitations. Actually, over time there was introduction of a standardized questionnaire, there were some new skin test recommendations, changes of availability of sige tests. Furthermore, awareness of some valuable parameters (for example serum tryptase) rose over the years. This has led to different diagnostic protocols in patients over time and of course missing data. For the future we plan to continue this survey and hope to gather more information concerning the less frequent causes such as gelatin, dyes and others and to study the role of clonal mast cell disorders in perioperative hypersensitivity reactions. This way, we hope to further optimize diagnostic tests for these miscellaneous culprits. 1. Aberer W. Diagnosing Cefazolin Hypersensitivity: Important Steps Forward, but We Haven t Reached the Goal Yet. J Allergy Clin Immunol Pract. American Academy of Allergy, Asthma & Immunology; 2016;4(6): Kheir MM, Tan TL, Azboy I, Tan DD, Parvizi J. Vancomycin Prophylaxis for Total Joint Arthroplasty: Incorrectly Dosed and Has a Higher Rate of Periprosthetic Infection Than Cefazolin. Clin Orthop Relat Res. Springer US; 2017; 104

105 105

106 106

107 Chapter 6: General conclusion 107

108 108

109 The perioperative hypersensitivity scene has its own peculiarities that might affect correct diagnosis and entail substantial risk for future reactions. As emerges from the literature NMBA constitute a significant cause of perioperative hypersensitivity reactions in certain countries. For long, diagnosis of NMBA hypersensitivity was principally based upon skin testing. Actually, drug-specific in vitro tests have only become available recently and need further standardization and harmonization of protocols before they can enter mainstream application. In this thesis, we focus on atracurium and cefazolin. In the first part of this dissertation we demonstrate that both basophil activation tests and quantification of drug-specific IgE can be applied complementary to skin tests to diagnose atracurium hypersensitivity. Moreover, it appears that basophil activation experiments helps to study crossreactivity between different NMBA. In contrast, we found sige to morphine, used as a biomarker for sensitization to tertiary and quaternary ammonium structures to be of almost no value to document hypersensitivity to atracurium. Next, in the second part of the thesis, we demonstrate that the diagnosis of cefazolin hypersensitivity benefits from an optimized skin testing protocol, indicating that there is an absolute need for drugspecific protocols and that we should abandon sometimes arbitrarily chosen and generalized skin test concentrations. With respect to the cefazolin basophil activation test we observed reliability of the assay probably to be dependent on the applied read-out. Anyhow, in the absence of a cefazolin-specific IgE test, for the time-being, basophil activation remains the sole in vitro test that might help to establish the diagnosis of cefazolin hypersensitivity. Importantly, our series, which is the largest series published on IgE mediate allergy to cefazolin, confirms the previous reports about the selectivity of this allergy. Actually, patients hypersensitive to cefazolin tolerate alternative beta lactam antibiotics (including other cephalosporins). Finally, in the last part of the thesis we took the opportunity to scrutinize our database on alleged perioperative hypersensitivity reactions. In ourdatabse NMBA, latex, antibiotics and chlorhexidine appear to be the predominant causes. However, we also observed a shift in these causes over time. This could be properly related to the optimization of diagnostic protocols for some agents and also to the reduced exposition for some other possible culprits (i.e. latex). Literature about the need for resuscitation and the use of epinephrine in the treatment of perioperative hypersensitivity and numbers of aborted procedures is limited. From our study it appears that concerning the reactions suggestive for perioperative hypersensitivity as far as data are available up to 10% of the patients were resuscitated and that in more than one third of the patients 109

110 epinephrine was used. In over a quarter of the patients the intervention was aborted and postponed. We also showed that mast cell activation positively correlates with the severity of the reaction. In conclusion we can state that perioperative hypersensitivity although rare has important consequences, not only for the patient, but also for the health care community and health economics. Therefore, further studies concerning this issue are needed, to further investigate this phenomenon, unravel the underlying mechanisms and to optimise the diagnosis. 110

111 111 Algemene samenvatting

112 112

113 Perioperatieve overgevoeligheidsreacties worden gekenmerkt door een aantal bijzonderheden die een correcte diagnose kunnen beinvloeden. Uit de literatuur blijkt dat in bepaalde landen, curariserende spierontspanner (neuromuscular blocking agents, NMBA) en antibiotica belangrijke oorzaken zijn van perioperatieve overgevoeligheidsreacties. Gedurende lange tijd beruste de diagnose van een NMBA en antibiotica overgevoeligheid bijna uitsluitend op huidtesten. Medicatie specifieke in vitro testen zijn namelijk nog maar sinds kort beschikbaar en moeten nog verder gestandaardiseerd worden. Ook moeten technische protocollen nog geharmoniseerd worden alvorens deze testen algemeen kunnen worden geaccepteerd en gebruikt. Deze thesis focust zich specifiek op het optimaliseren van de diagnose van atracurium en cefazoline. In het eerste deel van dit proefschrift tonen we aan dat zowel de basofielenactivatietest als de bepaling van medicatie specifiek IgE complementair zijn aan huidtesten, om een overgevoeligheid ten opzichte van atracurium vast te stellen. Bovendien tonen we aan dat de basofielenactivatietest kan helpen in het bestuderen van de kruisreactiviteit tussen de verschillende NMBA onderling. Anderzijds stellen we vast dat sige morfine - die gebruikt wordt als merker voor een sensitisatie aan tertiair en quaternair gesubstitueerde ammonium structuren - van weinig tot geen waarde is in de diagnose van atracurium overgevoeligheid. In een volgend deel van de thesis tonen we aan dat de diagnose van een overgevoeligheid ten opzichte van cefazoline verbetert door een geoptimaliseerd huidtest protocol te gebruiken. Dit toont de noodzaak aan van geneesmiddelen-specifieke protocollen in plaats van de gangbare doch vaak arbitrair gekozen en veralgemeende huidtestconcentraties. Wat de basofielenactivatie test voor cefazoline betreft vinden we dat de betrouwbaarheid van deze test vermoedelijk afhankelijk is van de gebruikte activatiemerker. Hoe dan ook, gezien het ontbreken van een cefazoline specifieke IgE test, blijft de basofielenactivatietest voorlopig de enige in vitro test die een hulp kan bieden in het stellen van de diagnose van een overgevoeligheid ten opzichte van cefazoline. Onze reeks - de grootste serie gepubliceerd omtrent IgE gemedieerde allergie voor cefazoline - bevestigd de voorgaande rapporten omtrent de selectiviteit van de allergie. Patiënten allergisch aan cefazoline zijn namelijk in staat om andere beta-lactam antibiotica (waaronder ook cefalosporines) te tolereren. Ten slotte hebben we in het laatste deel van deze thesis de mogelijkheid benut om onze databank betreffende perioperatieve overgevoeligheidsreacties te bestuderen. In onze regio blijken NMBA, latex, antibiotica en chlorhexidine de belangrijkste oorzaken. Echter, zoals ook blijkt uit de literatuur, zagen we een verschuiving in deze oorzaken over het verloop van de jaren. Dit kon gelinkt worden aan 113

114 enerzijds de optimalisatie van diagnostische protocollen voor sommige oorzaken, en de afgenomen blootstelling voor andere oorzaken (bv. latex). Literatuur omtrent de nood aan reanimatie en het gebruik van adrenaline in de behandeling van perioperatieve anafylaxie is niet uitgebreid, evenmin de literatuur omtrent het aantal stopgezette operaties. Uit onze studie blijkt met betrekking tot de reacties suggestief voor een perioperatieve overgevoeligheidsreactie, dat voor zover de informatie beschikbaar is tot ruimschoots 1 op 10 patiënten gereanimeerd werd, en dat bij meer dan een derde van de patiënten adrenaline werd toegediend. Bij ruim een kwart van de patiënten werd de ingreep gestaakt en uitgesteld. Dit toont niet alleen de ernst van perioperatieve overgevoeligheidsreacties aan maar benadrukt ook de nood van een snelle en correcte diagnose te kunnen stellen om zo toekomstige levensbedreigende reacties te kunnen vermijden. Voorts tonen we ook aan dat mast cel activatie een positieve correlatie vertoont met de ernst van de reacties. Om te eindigen kunnen we stellen dat perioperatieve overgevoeligheidsreacties, alhoewel een zeldzaam fenomeen, een belangrijke impact hebben op zowel de patiënt, als op de gezondheidszorg. Dit maakt dat nieuwe studies binnen dit domein absoluut noodzakelijk zijn, zowel om de onderliggende mechanismen beter te begrijpen, nieuwe boosdoeners te identificeren, als de diagnostische mogelijkheden verder te optimaliseren. 114

115 115 Dankwoord

116 116

117 Een proefschrift is niet volledig zonder een dankwoord. Met het gevaar iemand te vergeten wil ik graag enkele mensen in de verf zetten. Allereerst gaat mijn dank uit naar mijn promotoren Professor Didier Ebo, Dr. Vito Sabato en Professor Hilde Lapeere. Didier, een proefschrift afleggen zonder zo n gedreven en door onderzoek gebeten promotor als u lijkt me niet haalbaar. Steeds bereikbaar voor vragen of problemen, of gewoon voor een korte babbel. U staat er steeds, van s morgensvroeg, in de week of in het weekend. Vito, jou dank ik vooral voor de kritische klinische opleiding die ik van je kreeg. De vele (soms verhitte) overlegmomenten in de kliniek zal ik niet snel vergeten. Jouw kennis van allergie, immunologie en zeldzame aandoeningen is indrukwekkend en heeft mijn klinische blik zeker en vast verruimd. Je encyclopedische kennis van de literatuur blijft verbazen. De samenwerking met jou zal van mij zeker en vast een betere arts hebben gemaakt. Ook je steun, inzet en motivatie omtrent het onderzoek verdient veel dank. Ontelbare keren stond ik in je bureau om toch nog even de aanpak, diagnosecodes en methodologie voor de studies te overlopen. Professor Lapeere, bedankt voor het vertrouwen dat u in me stelde toen u aan mij dacht als kandidaat voor dit doctoraat. Ik kijk er naar uit om mijn klinische opleiding in Gent aan te vatten. Veel dank gaat uit naar mijn mededoctoraatstudenten, die ervoor gezorgd hebben dat de doctoraatsperiode ook een leuke en gezellige tijd was. Nathalie, samen begonnen we aan een doctoraat, allebei niet goed wetend wat ons te wachten stond en samen een weg zoekend in het labo. Naarmate ik ook meer kliniek deed zagen we elkaar wat minder, maar elke week was er wel tijd om elkaar op de hoogte te brengen van de laatste nieuwtjes! Margriet, jou wil ik bedanken om me wat wegwijs te maken in de wereld van het onderzoek, maar ook in een nieuwe stad. Het was fijn om een mede dermatologie-doctoraatstudent te hebben. Je eeuwig optimisme en je talent om in alles het goede te zien maken je bijzonder. Ine en Athina, bedankt om zulke fijne collega s te zijn. Met jullie erbij viel er altijd wel iets te beleven! De (helaas) korte middagpauzes waren een leuk intermezzo tussen het werk door. Niet alleen op het werk, maar ook ernaast hebben we heel wat afgelachen en gediscussieerd. Ine, je bijzondere kijk op geneeskunde en onderzoek waren een stimulans voor mij om het beter te doen. Onze gezamenlijke yogalessen waren iets om naar uit te kijken, jammer dat ik nu wat te ver woon hiervoor. Athina, je enthousiaste en rechttoe rechtaan aanpak waren verfrissend! Bedankt om me op de hoogte te houden van alle nieuwe tv programma s en roddelfeitjes die ik moest weten om mee te zijn! 117

118 Dan ook aan Christel en Lisa, onze laboranten, voor het uitvoeren van de experimenten in dit proefschrift en hun technische expertise. Professor De Clerck, bedankt voor de ideeën en kritische analyses op de vrijdagnamiddagvergaderingen. Chris, bedankt voor je hulp bij alle statistiek en bij de lay-out van dit boekje. Kathleen, bedankt voor al het administratieve werk dat je voor deze thesis in orde bracht. Margo, jou wil ik bedanken voor je vlotheid en positivisme. Altijd paraat om een conflict te temperen, en om alles eens goed uit te praten! Veel dank gaat uit naar de secretaressen en verpleging van de afdeling Immunologie / Allergologie / Reumatologie. Annemie, Ingrid en Nicole, bedankt voor het eindeloze geduld dat jullie toonden als ik weer eens mijn agenda wou veranderen, of wanneer er weer een brief dringend getypt moest worden. Kim en Tim, bedankt voor het feilloos runnen van de dagopname en de antwoorden op al mijn praktische vragen. Dank aan de verpleging, die ondanks een steeds drukker wordende poli, steeds kwaliteitsvol werk blijft leveren. Furthermore, I want to express my gratitude to Prof. Lene Garvey and Prof. Paul Michel Mertes for their willingness to review this work. Dank ook aan de anesthesisten uit andere ziekenhuizen voor het doorverwijzen van patiënten, en natuurlijk ook grote dank aan al deze patiënten voor hun bijdrage aan het wetenschappelijk onderzoek. Bijzondere dank ook voor mijn ouders. Mama en papa, bedankt voor alles, jullie niet aflatende steun bij alles wat ik doe, zelfs als dat een verhuis naar Antwerpen betekent. Bedankt voor alle kansen die jullie me hebben gegeven. Zonder jullie stond ik niet waar ik nu sta. Niet te vergeten natuurlijk zijn mijn Gentse vriendinnen: Celine, Muriel, Griet, Liselotte, Charlotte, Leen, Sarah, Anneloor, Sofie, bedankt om er te zijn, het luisterend oor en voor al de leuke momenten! En tot slot, Pieter, bedankt voor je steun, aanmoediging en vooral je relativeringsvermogen. Bedankt om wie je bent en gewoon om er te zijn. 118

119 119 Curriculum vitae

120 120

121 Curriculum vitae Astrid Uyttebroek werd geboren te Kortrijk op 12 september Na het behalen van haar ASO diploma Wetenschappen-Wiskunde(8u), aan het Sint Lievenscollege te Gent in 2006; studeerde zij Geneeskunde aan de Universiteit Gent. Het diploma van Arts werd in 2013 met grote onderscheiding gehaald. Vervolgens vatte zij als doctoraatsbursaal van de Faculteit Geneeskunde, haar doctoraatswerk aan op het laboratorium Immunologie-Allergologie-Reumatologie aan de Universtiteit Antwerpen. In het kader van haar opleiding tot arts-specialist in de dermatologie combineert zij dit tot op heden met klinische taken op de dienst Dermatologie van het Universitair Ziekenhuis Gent. Zij behaalde voor haar onderzoeksresultaten de volgende onderscheidingen: Price of the Belgian Society of Clinical Allergy and Immunology 2015 Price of the Belgian Society of Clinical Allergy and Immunology 2016 Na het verdedigen van haar proefschrift zal zij haar voltijdse klinische opleiding in de dermatologie en venerologie aanvatten. 121

122 Publications Leysen J, Uyttebroek A, Sabato V, Bridts C, De Clerck L, Ebo D. Predictive value of allergy tests for neuromuscular blocking agents: tackling an unmet need. Clin Exp allergy. 2014;44: (IF: 5.587)/(Cited 18/01/2017: 11) Uyttebroek A, Leysen J, Bridts C, Ebo D. Letter to the authors concerning the accepted manuscript: exploring the link between pholcodine and neuromuscular anaphylaxis by Brush et al. Br J Clin Pharmacol Oct;78(4): (IF: 5.259)/(Cited 18/01/2017: 1) Uyttebroek AP, Sabato V, Bridts CH, Ebo DG. In vitro diagnosis of immediate IgE-mediated drug hypersensitivity: warnings and (unmet) needs. Immunol Allergy Clin North Am. Elsevier Inc; 2014 Aug;34(3):681 9, x. (IF: 1.97)/(Cited 18/01/2017: 5) Uyttebroek AP, Sabato V, Leysen J, Bridts CH, De Clerck LS, Ebo DG. Flowcytometric diagnosis of atracurium-induced anaphylaxis. Allergy Jun 24;69(10): (IF: 6.335)/(Cited 1801/2017: 11) Uyttebroek AP, Sabato V, Faber M a, Cop N, Bridts CH, Lapeere H, et al. Basophil activation tests: time for a reconsideration. Expert Rev Clin Immunol. 2014;10(10): (IF: 2.596)/(Cited 18/01/2017: 9) Uyttebroek A, Sabato V, Bridts CH, De Clerck LS, Ebo DG. Anaphylaxis to succinylated gelatin in a patient with a meat allergy: Galactose-α(1, 3)-galactose (α-gal) as antigenic determinant. J Clin Anesth. 2014;26(7): (IF 0.68)/(Cited 18/01/2017: 6) Uyttebroek AP, Sabato V, Bridts CH, De Clerck LS, Ebo DG. Moxifloxacin hypersensitivity: Uselessness of skin testing. J Allergy Clin Immunol Pract. 2015;3(3): (IF: 5.429)/Cited 18/01/2017: 4) Uyttebroek AP, Sabato V, Bridts C, De Clerck L, Ebo D. Immunoglobulin E antibodies to atracurium: a new diagnostic tool. Clin Exp Allergy. 2014;45(2): (IF: 5.587)/(Cited 18/01/2017: 4) Uyttebroek A, Van Gasse A, Sabato V, Bridts C, Ebo D. Basophil activation tests: A diagnostic breakthrough in opiate allergy. Allergy, Asthma Immunol Res. 2015;7(4). (IF: 2.309)/(Cited 18/01/2017: 1) Mangodt EA, Van Gasse AL, Decuyper I, Uyttebroek A, Faber MA, Sabato V, et al. In vitro Diagnosis 122

123 of Immediate Drug Hypersensitivity: Should We Go with the Flow? Int Arch Allergy Immunol. 2015;168(1):3 12. (IF: /(Cited 18/01/2017: 6) Faber MA, Van Gasse AL, Decuyper II, Uyttebroek A, Sabato V, Hagendorens MM, et al. IgEreactivity profiles to nonspecific lipid transfer proteins in a northwestern European country. Journal of Allergy and Clinical Immunology (IF: )/Cited 18/01/2017: 1) Decuyper II, Ebo DG, Uyttebroek AP, Hagendorens MM, Faber MA, Bridts CH, et al. Quantification of specific IgE antibodies in immediate drug hypersensitivity: More shortcomings than potentials? Clin Chim Acta. 2016;460 (IF: 2.799)/(Cited 18/01/2017: 1) Mangodt EA, Van Gasse AL, Bastiaensen A, Decuyper II, Uyttebroek A, Faber M, et al. Flow-assisted basophil activation tests in immediate drug hypersensitivity: Two decades of antwerp experience. Acta Clin Belgica Int J Clin Lab Med. 2016;71(1). (IF: 0.683)/(Cited 18/01/2017: 0) Cop N, Uyttebroek AP, Sabato V, Bridts CH, De Clerck LS, Ebo DG. Flow cytometric analysis of drug- Induced basophil histamine release. Cytom Part B Clin Cytom. 2015;90(3): (IF: 2.822)/(Cited 18/01/2017: 0) Uyttebroek AP, Decuyper II, Bridts CH, Romano A, Hagendorens MM, Ebo DG, et al. Cefazolin Hypersensitivity: Toward Optimized Diagnosis. J Allergy Clin Immunol Pract. 2016;4(6):3 7. (IF: 5.429)/(Cited 18/01/2017: 2) Uyttebroek AP, Sabato V, Cop N, Decuyper II, Faber MA, Bridts CH, et al. Diagnosing cefazolin hypersensitivity: Lessons from dual-labeling flow cytometry. J Allergy Clin Immunol Pract. 2016;4(6): (IF: 5.429)/(Cited 18/01/2017: 1) 123

124 Abstracts Uyttebroek AP, Sabato V, Leysen J, Bridts CH, De Clerck LS, Ebo DG. Flowcytometric diagnosis of atracurium-induced anaphylaxis. Allergy Jun 24;69(10): (IF: 6.335)/(Cited 1801/2017: 11) Functional exploration of cross-reactivity between neuromuscular-blocking drugs by basophil inhibition :a proof of concept. Poster presentation EAACI congress Barcelona 2015 Uyttebroek AP, Decuyper II, Bridts CH, Romano A, Hagendorens MM, Ebo DG, et al. Cefazolin Hypersensitivity: Toward Optimized Diagnosis. J Allergy Clin Immunol Pract. 2016;4(6):3 7. (IF: 5.429)/(Cited 18/01/2017: 2) Uyttebroek AP, Sabato V, Cop N, Decuyper II, Faber MA, Bridts CH, et al. Diagnosing cefazolin hypersensitivity: Lessons from dual-labeling flow cytometry. J Allergy Clin Immunol Pract. 2016;4(6): (IF: 5.429)/(Cited 18/01/2017: 1) 124

125 125 Addendum

126 126

127 127

128 128

129 129

130 130

131 131

132 132

133 133

134 134

135 135

136 136