PERIOPERATIVE anaphylaxis may be a life-threatening clinical condition and is typically a result of drugs or substances used for anesthesia or surgery. After anaphylaxis, allergologic assessment is essential to identify the offending agent and prevent recurrences, because no preemptive therapeutic strategies exist. This review seeks to (1) identify the clinical diagnostic pathway necessary to distinguish anaphylaxis from confounding clinical diagnoses, (2) discuss the more common allergens that cause anaphylaxis during anesthesia, (3) discuss a rational approach to the identification of the offending allergen through blood and skin testing that allows for the safe future clinical management of patients experiencing perioperative anaphylaxis, and (4) discuss new therapeutic perspectives for the management of patients whose hemodynamic collapse is unresponsive to catecholamines, the initial recommended pharmacologic intervention.

Anaphylaxis occurring during the perioperative period is a clinical syndrome involving multiple organ systems. The clinical manifestations are the consequences of the immediate as well as ongoing release of preformed mediators from mast cells and basophils. Careful consideration must therefore also be given to the monitoring of the patient after the management of anaphylaxis.1,2 

In the early 2000s, the European Academy of Allergology and Clinical Immunology proposed to define this acute nosologic entity as “a severe, life-threatening, generalized or systemic hypersensitivity reaction,”3primarily mediated by type E immunoglobulins (IgEs). More recently, this clinical entity was defined by the second National Institute of Allergy and Infectious Disease/Food Allergy and Anaphylaxis Network symposium as follows: “Anaphylaxis is a serious allergic reaction that is rapid in onset and may cause death.”4The European Academy of Allergology and Clinical Immunology committee recommended that the term anaphylactoid , introduced for non–IgE-mediated anaphylactic reactions, should no longer be used,3but this proposal has not been universally accepted.4 

Classically, IgE isotype antibodies are produced on an initial exposure to an allergen in susceptible individuals and bind to high-affinity FCϵRI receptors located in the plasma membrane of tissue mast cells and blood basophils, whereas lymphocytes, eosinophils, and platelets bind IgE antibodies via  low-affinity FCϵRII receptors. This initial phase of sensitization is clinically silent. On reexposure, the multimeric allergen cross-links two specific IgE receptors, creating a bridge between two IgEs. The two IgE receptors aggregate and induce a signal transduction cascade releasing systemically preformed biochemical mediators, including histamine, neutral proteases (tryptase, chymase), and proteoglycans (heparin) from intracellular granules in cells within tissues and blood. Some of these mediators, such as histamine, lead to increased nitric oxide production.5Newly formed proinflammatory phospholipid-derived mediators, including prostaglandin D2, leukotrienes, thromboxane A2, and platelet-activating factor, are released soon after. Thereafter, mast cells release numerous chemokines and cytokines that initiate recruitment and activation of additional inflammatory cells.

One key issue of anaphylaxis is that a very small amount of allergen is sufficient for the cells to react. The involved target organs commonly include the skin, mucous membranes, cardiovascular and respiratory systems, and the gastrointestinal tract. The corresponding clinical signs are erythema, edema, pruritus, arterial hypotension, tachycardia, and bronchial and gastrointestinal smooth muscle constriction, described by the Ring and Messmer clinical severity scale, which was adapted to describe perioperative immediate reactions (table 1).6Although this scale does not take into account the pathophysiologic mechanisms, it is appropriate for grading the clinical severity and guiding the clinical care of immediate reactions.1,7 In vivo  and in vitro  procedures can be used to differentiate between allergic (where an immune mechanism can be demonstrated, i.e. , an anaphylactic mechanism) and nonallergic (also previously called anaphylactoid , where an immune mechanism is ruled out) hypersensitivity reactions.

Table 1. Clinical Severity Scale of Immediate Hypersensitivity Reactions Adapted from  Ring and Messmer 6 

Table 1. Clinical Severity Scale of Immediate Hypersensitivity Reactions Adapted from  Ring and Messmer 6
Table 1. Clinical Severity Scale of Immediate Hypersensitivity Reactions Adapted from  Ring and Messmer 6

The overall incidence of perioperative anaphylaxis is estimated at 1 in 10,000–20,000 anesthetic procedures, whereas it is estimated at 1 in 6,500 administrations of neuromuscular blocking agents (NMBAs).8Its exact incidence remains underestimated as reactions are underreported.

Anaphylaxis is one of the rare clinical events specifically related to anesthesia leading to perioperative morbidity and mortality. Morbidity remains unknown. In France, 3% of the deaths that are partially or totally anesthesia related involve anaphylaxis,9whereas 10% of perioperative immediate hypersensitivity reactions reported to the United Kingdom Medicines Control Agency are fatal.2Nevertheless, this latter datum should be interpreted cautiously because many less severe reactions are probably not reported.2 

In France, NMBAs were shown to be the most common agents involved, followed by latex and antibiotics. In a Norwegian single-center study, NMBAs were most commonly involved, with latex implicated in very few cases and with no causal agent identified in one third of the cases.10Conversely, in two Spanish centers, antibiotics followed by NMBAs were the main agents involved.11Anaphylaxis to NMBAs is not uncommon in patients without any known previous exposure to any NMBA.2,11In this particular clinical setting of anaphylaxis after a first-time NMBA administration, the source and the nature of the sensitizing agent remain unknown.12However, quaternary ammonium ions are suggested to be the allergenic determinants in NMBAs.12Commonly used chemicals, such as toothpastes, detergents, shampoos, and cough medicines, share these determinants with NMBAs. In a predisposed individual, these common chemicals might be one of the contributing factors promoting sensitization to quaternary ammonium ions and creating the risk for developing anaphylaxis to NMBAs.2Recently, it has been suggested that the sensitization to quaternary ammonium ions seems to be related to exposure to pholcodine, a cough-relieving medicine.12Moreover, an individual can have uneventful previous exposures to a specific drug (e.g. , NMBAs, antibiotics), which does not preclude the risk of anaphylaxis after a subsequent administration of this drug.2 

The etiologic diagnosis of an immediate reaction occurring during anesthesia relies on a triad including clinical, biologic, and allergologic evidence (fig. 1).

Fig. 1. The etiologic diagnosis of perioperative anaphylaxis relies on a triad including clinical, biologic, and allergologic evidence. 

Fig. 1. The etiologic diagnosis of perioperative anaphylaxis relies on a triad including clinical, biologic, and allergologic evidence. 

Close modal

First Evidence: The Clinical History Is Crucial

The initial diagnosis of anaphylaxis is presumptive, although essential, because anaphylaxis may progress within minutes to become life-threatening. The first line of evidence for the diagnosis of anaphylaxis includes the features and severity of clinical signs and the timing between the introduction of a suspected allergen and the onset of symptoms, whereas the required dosage for resuscitative medications gives insight as to the severity of the reaction.1The clinical features occurring during anesthesia may involve cardiovascular symptoms (tachycardia, bradycardia, cardiac arrhythmias, hypotension, cardiovascular collapse, cardiac arrest), bronchospasm, and cutaneous–mucous signs (erythema, urticaria, angioedema).1,2,7,8They are described according to the Ring and Messmer four-step grading scale adapted as follows (table 1).1,6,7Grade I involves cutaneous–mucous signs, whereas grade II corresponds to mild cutaneous–mucous features that may be associated with cardiovascular and/or respiratory signs. The cardinal sign of grade III is cardiovascular collapse that may be associated with cutaneous–mucous signs and/or bronchospasm, and grade IV is cardiac arrest. Grades I and II are usually not life-threatening conditions, whereas grades III and IV correspond to emergency situations necessitating prompt resuscitation.

Perioperative anaphylaxis usually occurs within minutes, even 1 min, after anesthetic induction and is primarily linked to agents administered intravenously.1The most commonly reported initial clinical features during severe reactions (grade III or IV) are pulselessness, desaturation, and difficult lung inflation due to severe bronchospasm. Respiratory signs are enhanced in patients with underlying asthma and/or chronic obstructive pulmonary disease. Asthmatic patients experiencing perioperative anaphylaxis are therefore more likely to exhibit bronchospasm.1,2Cardiovascular symptoms often include hypotension and tachycardia but may rapidly progress into severe arrhythmias and cardiovascular collapse if not recognized and treated in a timely fashion.1Because cardiovascular disturbances are the hallmark of severe anaphylaxis, cardiovascular collapse as the sole feature or cardiac arrest may be the inaugural event.1,2,7,8Acute coronary events associated with hypersensitivity reactions were recently referred to as Kounis syndrome , which is also called allergic angina  or allergic myocardial infarction .13Two variants were described as the consequences of mast cell mediator release: type I includes patients without predisposing factors for coronary artery disease, whereas type II includes patients with predisposing factors. The allergic reaction may induce coronary artery spasm with normal cardiac-specific enzymes in variant I, whereas acute infarction may be seen in both groups (types I and II). A threshold level of mast cell content has been suggested to be involved in the occurrence of such events.13 

Predictive Criteria of Anaphylaxis Severity.

Three predictive criteria of the severity of an ongoing anaphylactic reaction, which may or may not be associated, include the following: (1) The more rapidly anaphylaxis occurs after allergen exposure, the more likely the reaction is to be severe and potentially life-threatening.14(2) Cutaneous signs may be absent in rapidly progressive anaphylaxis.14Accordingly, the subcutaneous vascular bed is susceptible to vasoconstrictive influences during anaphylaxis (being the first vascular bed compromised when circulatory homeostasis is threatened). Cutaneous signs may therefore only appear after the normalization of blood pressure. Consequently, the absence of initial cutaneous vasodilatation should not preclude the diagnosis of anaphylaxis. (3) Another potentially confounding sign might be bradycardia, which can be seen in massive hypovolemia, probably as a result of the Bezold–Jarisch reflex, a cardioinhibitory reflex that has its origin in sensory receptors of the left ventricle transmitted by unmyelinated vagal C fibers.14This paradoxical bradycardia occurring during extreme hypovolemia has been reported to occur in as many as approximately 10% of patients with anaphylaxis during anesthesia.2In this case, bradycardia may be considered as a life-protecting adaptive mechanism that allows the ventricles to fill before they start contracting again despite a massive hypovolemia. Therefore, this specific event during anaphylaxis must be recognized by the anesthesiologist caring for the patient because the administration of atropine in direct response to the symptom of bradycardia might directly induce a circulatory arrest, the adequate treatment in this setting being a large volume expansion followed by epinephrine.

Which Drugs or Agents?

The most commonly involved agents in perioperative anaphylaxis are NMBAs, latex, and antibiotics. Anaphylaxis usually occurs shortly after induction,1,7with NMBAs or antibiotics being primarily involved,7but anaphylaxis may occur any time with all potentially allergenic agents.1Dyes, hypnotic agents, local anesthetics, opioids, colloids, aprotinin, protamine, chlorhexidine, and contrast agents are less frequently involved.1,2,7Latex-induced anaphylaxis usually occurs up to 30–60 min after the beginning of the surgery but may occur immediately.2,15Severe and even fatal reactions to latex are now reported.16Attempts to ban latex from use in clinical products is therefore encouraged because the incidence of latex-induced anaphylaxis has increased dramatically during the past two decades, especially in high-risk groups of patients such as children or adults requiring multiple surgical procedures or healthcare workers.17Anesthesiologists themselves have a high prevalence of latex sensitization, which may have a genetic basis.18Accordingly, in a pediatric hospital including operating rooms and perioperative care areas, no allergic reaction to latex has been reported in 25,000 anesthetized children or in healthcare workers when a latex-free environment policy has been adopted.17 

Finally, the onset and type of symptoms depend on the allergen concentration,19whereas the length of symptoms (up to 36 h) may vary according to the mode of injection (e.g. , when using dyes such as isosulfan blue or patent blue for lymphatic mapping and sentinel lymph node biopsy). Moreover, the patient’s sensitivity level and route of administration are important in determining the onset and severity of symptoms. For example, intravenous and mucous membrane exposure are likely to be associated with a faster onset of more severe symptoms.19 

Second Evidence: Biologic Assessment Contributes to the Diagnosis

Biochemical tests are available, either in vivo  or in vitro .

Primary Investigation.

Histamine is a preformed inflammatory mediator contained in granules of mast cells and basophils. An early increase in plasma histamine concentration indicates activation of mast cells and/or basophils and is observed during allergic or nonallergic mechanisms. Conversely, the absence of histamine increase does not preclude an immunologic or nonimmunologic mechanism. The plasma half-life of histamine is assumed to be very short (15–20 min).7Blood samples for histamine measurement should therefore be drawn within 30 min after a grade I or II reaction. After severe reactions (grades III and IV), histamine may still be increased 2 h after the reaction, probably because of the saturation of the enzymatic metabolism of histamine allowing for the detection of increased histamine levels for this prolonged period after the reaction.7 

Tryptase is a mast cell neutral serine protease and a preformed enzyme. Two major forms have been identified2,20: α-Tryptase is secreted constitutively and is increased in mastocytosis, whereas pro–β-tryptase also secreted constitutively serves as a measure of mast cells mass. Mature β-tryptase is preferentially stored in mast cells granules and, when systemically released, reflects mast cell activation with mediator release.

Serum tryptase concentrations reach a peak between 15 min and 1 h and decline under apparent first-order kinetics with a half-life of approximately 2 h.1,2,7Therefore, blood samples may be drawn within 15 and 60 min in grade I or II reactions and within 30 min and 2 h in grade III or IV reactions.1,7An increase of total tryptase concentrations (i.e. , the sum of α- and β-tryptase) is highly suggestive of mast cell activation as seen in anaphylaxis, but its absence does not preclude the diagnosis (grades I and II). Conversely, an increase of tryptase may be seen during nonallergic mechanisms, although less pronounced when compared with immunologic mechanisms.2,20To compare the concentrations with baseline levels, a new sample of tryptase can be taken either at least more than 24 h after the reaction or when the patient is referred for investigation.1,7,20Finally, tryptase concentrations may remain increased in cases of late-onset, biphasic anaphylaxis or when associated with an underlying mastocytosis2,19,20(fig. 2).

Fig. 2. The clinical pathway allows for the identification of the culprit agent, documenting the pathophysiologic mechanism (allergic  vs. nonallergic) of the perioperative immediate reaction and allows for advice regarding subsequent anesthetics. NMBA = neuromuscular blocking agent. 

Fig. 2. The clinical pathway allows for the identification of the culprit agent, documenting the pathophysiologic mechanism (allergic  vs. nonallergic) of the perioperative immediate reaction and allows for advice regarding subsequent anesthetics. NMBA = neuromuscular blocking agent. 

Close modal

In clinical situations of an immediate nonallergic reaction (e.g. , histamine release), histamine may be increased, whereas tryptase usually remains normal. In contrast, because histamine and tryptase concentrations correlate with the severity of the allergic reaction,7combined histamine and tryptase measurements have been recommended for the diagnosis of immediate reactions,7whereas others recommend only tryptase.1,2,20This constitutes the second component of the triad.

Secondary Investigation.

Other in vitro  tools have been proposed in clinical practice but are not available everywhere. No NMBA-specific IgE assay is currently available, except for suxamethonium, and its sensitivity remains low (30–60%).20Some authors suggested the use of a quaternary ammonium (choline analog) or morphine-based solid-phase IgEs to evaluate sensitization to the quaternary ammonium of NMBAs.12,20,In vitro  specific IgE assays are also available for a few other anesthetics (thiopental, propofol), antibiotics (amoxicillin, cefaclor, penicillin G and V), or latex.2,7,20Nevertheless, specific IgE assays seem to be less sensitive than skin tests.12Finally, serum IgEs provide a possible explanation of the mechanism but do not prove that the drug or agent is responsible for the reaction.2 

Several other tests are proposed for the indirect detection of specific IgEs to anesthetic drugs. The leukocyte histamine release test is reliable when cross-reactivity among NMBAs is investigated for NMBA reintroduction.1,7The quantification of in vitro –activated basophils (CD63 and CD203c) after challenge with the culprit agent using flow cytometric analysis might be a useful tool for the diagnosis of NMBA-induced anaphylaxis1,2,20but requires further investigation.

Third Evidence: Skin Tests

Skin tests remain the gold standard for the detection of IgE-mediated reactions by exposing the mast cells of the skin to the suspected allergen in patients having experienced anaphylaxis.8The results of skin testing constitute the third component of the triad.

Why and When to Perform Skin Tests?

Because premedication with H1and/or H2receptor antagonists and/or corticosteroid has not been proven to prevent anaphylaxis,1,7,14skin tests constitute the tool to (1) identify the culprit agent (and therefore to definitely avoid it), (2) prove the pathophysiologic mechanism of the reaction (allergic vs.  nonallergic), and (3) suggest a safe alternative drug for future exposures (fig. 2). A 4- to 6-week delay after the reaction is required to avoid a false-negative test result because of mast cell depletion.1,7 

How to Perform Skin Tests?

Skin testing should always be performed according to the clinical history to provide a pertinent diagnosis. Therefore, all drugs injected just before the reaction, as well as latex, must be skin tested, and immediate hypersensitivity reactions necessitate immediate reading of skin tests (i.e. , after 15–20 min). Before test agents, the cutaneous reactivity should be assessed with a negative (saline) and a positive control (codeine and/or histamine). Investigation of drugs is performed by prick tests (PTs), followed by intradermal tests (IDTs) using freshly prepared commercial solutions, diluted or undiluted, without exceeding the maximal concentrations to avoid false-positive results (table 2).1,2,7,20IDTs are more sensitive but less specific than PTs.21However, IDTs are more likely to trigger systemic allergic reactions and thus should be only performed after PTs.21Therefore, if the PT result is negative, the IDT is performed by injecting 0.02–0.05 ml of the corresponding drug. If the IDT is negative, 10-fold increasing concentrations are used with incremental 15- to 20-min intervals between each IDT until the test result is positive or the highest nonirritant concentration is achieved.1,2,7,20The criteria of PT/IDT positivity and the different concentrations of anesthetic drugs that are normally nonreactive have been strictly defined in France (table 2).7These rules have been recommended in Scandinavia1and adapted by others20(for extensive review, see Dewachter and Mouton-Faivre8).

Table 2. Concentrations of Anesthetic Agents Normally Nonreactive during Skin Tests 

Table 2. Concentrations of Anesthetic Agents Normally Nonreactive during Skin Tests 
Table 2. Concentrations of Anesthetic Agents Normally Nonreactive during Skin Tests 

Which Drugs and Substances Are Involved?

Neuromuscular Blocking Agents.

Neuromuscular blocking agents are frequently involved during perioperative anaphylaxis at a range of 50–70% according to different reports in Europe,2,7,10,11whereas limited data are available in the United States. There has not been an epidemiologic study of the causative agents of perioperative anaphylaxis in the United States. Instead, data rely on reports to the US Food and Drug Administration, and because of a lack of standardized skin testing, overall incidences in the United States are unknown. All NMBAs may elicit anaphylaxis.8,20The sensitivity of skin tests for NMBAs in patients having experienced anaphylaxis after an NMBA injection is greater than 95%, and their reproducibility is excellent.7,8The maximal concentration recommended should not be exceeded (table 2). Cross-reactivity between NMBAs is common and approximates 60–70%.1,12,19,20Consequently, to be complete, investigation for cross-reactivity with the other commercialized NMBAs should be performed to identify safe alternative regimens (i.e. , negative skin-tested NMBAs) during the diagnostic approach of NMBA-induced anaphylaxis.1,2,7In France and as promoted by others, cross-reactivity is performed by PTs followed by IDTs, because PTs are less sensitive than IDTs.1,7Conversely, others recommend PTs only.2Indeed, an arbitrary contraindication to all NMBAs cannot be accepted as the rule in case of a documented anaphylaxis to a NMBA, because this places an unfavorable burden for the anesthetic options for the patient in the future.

Do Skin Tests Overestimate the Incidence of Anaphylaxis to Rocuronium?

Of particular interest, reports have suggested an increased frequency of anaphylaxis with rocuronium in France and in Norway, leading the Norwegian Medicine Agency to publish an alert reserving its use for urgent intubations.§

Propenyl-ammonium groups may account for this increased allergenicity.12This trend was not observed in Australia or in the United States. Therefore, some authors called into question the accuracy of skin tests in the diagnosis of allergy to NMBAs, arguing that false-positive skin test results due to inappropriate dilutions of rocuronium used by either PTs or IDTs might explain the high incidence reported.22,23Dilution thresholds were debated in controls. Dhonneur considered that undiluted and diluted (1/100, 1/10) solutions of rocuronium may induce false-positive PT results,23whereas Levy recommended at least 100-fold dilution before skin testing (IDTs).22Others showed that PT results were always negative with rocuronium,24whereas nonreactive rocuronium dilutions for IDTs were considered to be either at 1/10,00024or 1/100.25However, when, as in these controls, the pretest probability is very low (absence of previous clinical history), predictive values of skin tests with NMBAs remain unknown, and skin test results are not predictive of outcome.2,7,20Biopsies of rocuronium wheals performed in these controls confirmed the absence of mast cell degranulation.22,24Consequently, the “positive” skin responses may be attributed to a direct effect of NMBAs on cutaneous vasculature.22,24In contrast to control patients, skin tests with NMBAs performed in anaphylactic patients are highly reliable and detect drug-induced IgE cross-linking with corresponding release of inflammatory mediators.12,20Therefore, skin tests performed in control versus  anaphylactic patients cannot and should not be compared.26Nevertheless, the apparent increased incidence of anaphylaxis to rocuronium might be due to (1) a reflection of usage and market share, (2) biased reporting of adverse effects of new drugs, (3) statistical issues, or (4) genotypic differences.19This question requires further epidemiologic data in order to be understood.2,19 

Other Drugs and Substances Involved.


In Europe, investigation of latex is performed by PTs using commercial extracts.1,2,7Their sensitivity is excellent (75–90%).2No commercially available skin test reagent exists in the United States, where diagnosis relies on in vitro  tests.15,19Latex gloves extracts are often used, but their amount of latex proteins is not standardized.14 


Anaphylaxis triggered by antibiotics involves primarily penicillins and cephalosporins (70%) which share the β-lactam ring.20It may occur at first exposure.1,21The European Network Drug Allergy interest group on drug hypersensitivity proposed the highest skin testing concentrations as follows for amoxicillin (20–25 mg/ml), ampicillin (20–25 mg/ml), and most cephalosporins (1–2 mg/ml). The specificity of skin testing with β-lactams is between 97% and 99%, whereas the sensitivity is around 50%.8Therefore, oral provocation tests in patients with suggestive clinical history and negative skin test results with β-lactams is recommended.1,20,21Cross-reactivity between penicillins and cephalosporins seems to be low (10%) and is attributed to the common β-lactam ring.19However, the allergenicity of the molecule is also determined by the side chains attached to the β-lactam ring. Therefore, first-generation cephalosporins and cefamandole share a similar structural side chain with penicillin and amoxicillin.2A recent meta-analysis suggested that patients allergic to penicillin or amoxicillin have a higher incidence of allergic reactions to first-generation cephalosporins and cefamandole but not to later-generation cephalosporins.2Anaphylaxis with vancomycin remains rare; IDTs with vancomycin should be performed below a concentration of 10 μg/ml.20Anaphylaxis with vancomycin should be distinguished from red man syndrome, a clinical entity resulting from nonspecific histamine release and observed when the drug is rapidly injected.14,20 


Anaphylaxis to thiopental or propofol is rarely reported, whereas anaphylaxis to etomidate and ketamine remains extremely rare.1,2,19Hypnotics may be skin tested according to the concentration limits provided in table 2.1,7,20 


Anaphylaxis to opioids is very rare.1,2,19,20Morphine induces histamine release; the maximal concentration recommended for skin testing should not be exceeded (table 2). PTs, followed by IDTs, if negative, are performed with phenylpiperidines (alfentanil, fentanyl, remifentanil, sufentanil) (table 2).1,7,20Cross-reactivity remains uncommon among phenylpiperidines.19 

Local Anesthetics.

Anaphylaxis to local anesthetics is very uncommon and has decreased in frequency because of the decreasing use of the ester group of local anesthetics. Most allergic reactions are due to the common metabolic product of the ester local anesthetic, para-amino benzoic acid.1,2,7,20This common allergenic determinant implies cross-reactivity among all local anesthetics belonging to the ester group. Allergic reactions to amide local anesthetics remain anecdotal. Ingredients included in local anesthetic solutions such as antioxidants or preservatives including metabisulfite or parabens (also metabolized to para-amino benzoic acid) may also elicit allergic or adverse reactions.19,20Finally, cross-reactivity among esters is common, whereas it is rarely seen in the amide group and is absent between esters and amides. Local anesthetics (without preservatives or epinephrine) may be skin tested according to the concentration limits provided in table 2.1,7,20 


Anaphylaxis with colloids is rare but is more frequent with gelatins (0.35%) than with hydroxyethylstarch (0.06%).19Skin tests may be performed with PTs (pure solution) followed by IDTs in case of negativity.1,7,20 


The risk of anaphylaxis with aprotinin is approximately 2.8% in reexposed patients.20Aprotinin was previously used to reduce blood loss and the need for blood transfusion during surgery, but it was recently withdrawn from the market. However, some fibrin glue products still contain aprotinin. PTs (pure solution) followed by IDTs (up to 1/10 dilution) in case of negativity are recommended.7,20 


The incidence of anaphylaxis to isosulfan or patent blue is less than 2%,19whereas methylene blue is very rarely involved. Dyes may be skin tested by PTs followed by IDTs (up to 1/100 for methylene blue because it is a histamine-releaser and up to 1/10 dilution for isosulfan/patent blue).19 

Other Drugs.

Protamine, antiseptics (chlorhexidine, povidone iodine), and iodinated contrast agents may also induce anaphylaxis. Skin testing may be performed to prove the diagnosis.1,19 

Finally, to ensure an accurate diagnosis and to provide subsequent recommendations for future anesthetic management, the interpretation of the allergologic assessment should always be linked to the clinical event. A suggestive clinical history (most severe reaction) associated with an increased tryptase concentration (absence of tryptase increase does not preclude the diagnosis)1,2,7linked to skin tests positivity to the suspected agent confirm the diagnosis of anaphylaxis to this agent, which should be avoided.

Conversely, a suggestive clinical history (in general not a severe reaction) associated with an increase (or not) of histamine without a tryptase increase along with skin test negativity suggests a nonallergic reaction (histamine release with potent histamine-releasing drugs such as atracurium, mivacurium, and vancomycin).

Furthermore, a suggestive clinical history (essentially severe disturbances of cardiovascular homeostasis) associated with a tryptase increase and a persistent increased baseline tryptase along with skin test negativity suggests mastocytosis8(fig. 2).

Most importantly, patients should be given a detailed report with unequivocal answers to provide to caregivers before future anesthetic exposures.

Ideally, all patients experiencing an episode of perioperative anaphylaxis would have undergone an allergologic assessment before further anesthetics. The reality is very different. In many countries, the allergologic assessment is not routinely performed. Identification of at-risk patients is therefore required before any procedure requiring anesthetics which must be conducted in a manner to avoid a suspected drug or agent. Patients who have had previous uninvestigated severe immediate reactions during anesthesia are at increased risk of a recurrence during subsequent anesthetics.1,2,7Regional anesthesia is preferred whenever possible.1,7Moreover, a latex-free environment should be used to prevent a perioperative clinical reaction in patients having experienced clinical manifestations of allergy when exposed to latex, in patients who had many surgical or urologic cannulation procedures (because of the high incidence of sensitization to latex), and in patients with clinical manifestations of allergy to tropical fruits (e.g. , avocado, banana, kiwi, pineapple, papaya) because of the high rate of cross-reactivity with latex (also called latex–fruit syndrome ).7In contrast, patients who are atopic (except for latex) or those who are allergic to a drug (except for antibiotics which may be injected as perioperative antibiotic prophylaxis) that is not likely to be used during the perioperative period are not considered to be at risk for perioperative anaphylaxis.7 

The following measures should be applied whenever possible: (1) withdraw the suspected culprit drug; (2) immediately discontinue anesthetic drugs when the anaphylactic event occurs during induction; (3) maintain the airway with 100% oxygen; (4) provide early administration of epinephrine in case of grade III or IV reactions; (5) call for help, especially for grade III and IV reactions; (6) place the patient supine in the Trendelenburg position; and (7) abbreviate the surgical procedure if possible when it occurs during surgery.

Restoration of the Cardiovascular Homeostasis Is First Priority

Administration of intravenous epinephrine and an expansion of intravascular volume are the key points of perioperative management of anaphylaxis.1,2,7,27There is no absolute contraindication during anaphylaxis for the use of epinephrine when the clinical situation dictates its use,14and early administration should be the rule.1,2,7,27Therefore, poor outcomes during anaphylaxis, including deaths, are associated with either late or absent administration of epinephrine or inadequate or excessive dosing, emphasizing the need for careful titration.1 

Epinephrine should never be injected during grade I reactions, whereas titrated boluses (10–20 μg) of epinephrine may be sometimes necessary during grade II (table 1). Conversely, titrated intravenous bolus administrations of epinephrine (100–200 μg) are required in grade III reactions, renewed every 1 or 2 min if necessary according to the hemodynamic response and followed by a continuous infusion (1–4 μg/min) to prevent the need for repeated bolus injections.1,2,7,27Grade IV reactions (cardiac arrest) require cardiopulmonary resuscitation and high doses of epinephrine.7,27A commonly used sequence is to administer 1–3 mg intravenously (over 3 min) and 3–5 mg intravenously (over 3 min) followed by a continuous infusion (4–10 μg/min).27 

Fluid Therapy

Changes in vascular permeability during anaphylaxis might permit, within 10 min, a 50% transfer of the intravascular fluid into the interstitial space.14Fluid therapy initiated with either crystalloids or colloids is therefore essential to compensate for large fluid shifts and should be initiated early.1,2,7,27 


Bronchospasm is treated with inhaled β2agonist (salbutamol or albuterol).1,2,7,27In cases of persistent bronchospasm, intravenous injection of a β2agonist (salbutamol, 100–200 μg) is recommended, and a continuous infusion (5–25 μg/min) should be considered.2,7,27In contrast, when cardiovascular collapse and bronchospasm occur together, epinephrine remains the first-line therapy to correct first the cardiovascular homeostasis disturbances, whereas its β2effect is effective in relieving bronchoconstriction.7,27Intravenous corticosteroids early in the course of therapy are recommended because of their antiinflammatory effects. Their beneficial effects are delayed at least 4–6 h.1,7,27 

Additional Therapy

Corticosteroids and/or H1receptor antagonists are often recommended in the management of anaphylaxis,1,2,7,27but their effects have never been evaluated in placebo-controlled trials.14In retrospective studies analyzing recurrent episodes of anaphylaxis, corticosteroids and antihistamines did not prevent biphasic anaphylaxis, indicating that patients should be followed up carefully after apparent remission of anaphylaxis. However, corticoids are useful for angioedema.1 

Anaphylactic Shock Refractory to Catecholamines: Alternative Therapy?

Epinephrine sometimes fails to restore the profound disturbances of cardiovascular homeostasis. This singular clinical entity is called anaphylactic shock refractory to catecholamines , although it remains undefined in the literature. Norepinephrine, metaraminol, or glucagon for patients receiving β-blocker therapy is recommended in this clinical setting.1,2,7Because desensitization of adrenergic receptors might be one of the contributing factors of catecholamine failure occurring during anaphylaxis, arginine vasopressin (AVP) may be an alternative through its vasoconstrictive effects mediated by nonadrenergic vascular AVP V1receptors. Another factor contributing to catecholamine failure may involve nitric oxide (which seems to play a pivotal role during anaphylaxis) because increased nitric oxide synthesis contributes to the hypotension and resistance to vasopressors during vasodilatory shock. Thus, AVP directly decreases intracellular concentrations of the nitric oxide second messenger, guanosine 3′,5′-cyclic monophosphate.28Although experimental work provides support for the possible use of AVP during anaphylaxis,28AVP may be detrimental when injected alone in the early course of anaphylaxis (5 min)28or when higher doses are used. However, several case reports suggest that AVP might be considered as a potential rescue therapy during anaphylaxis. Therefore, some patients experiencing anaphylaxis refractory to epinephrine, norepinephrine, and/or phenylephrine were successfully treated with AVP injected at least 10–20 min after shock onset.29AVP could therefore play a pivotal role in cases of catecholamine failure occurring during anaphylaxis. Nevertheless, it is important that both successful and unsuccessful uses of AVP during resuscitation attempts during anaphylaxis be reported, such that a fair assessment of its potential usefulness can be established.

Moreover, methylene blue is known to interfere with nitric oxide–mediated vascular smooth muscle relaxation and was recently successfully used in catecholamine- and vasopressin-resistant anaphylaxis.30Finally, preclinical studies are needed to better understand the mechanism and potential clinical value of these rescue therapies.

Perioperative anaphylaxis is a severe and rapid clinical condition that can be lethal even in previously healthy patients. The diagnosis of perioperative anaphylaxis might be missed because the clinical presentation of this infrequent pathology can be of very rapid onset with variable clinical signs. Optimal management of perioperative anaphylaxis should therefore be encouraged through teaching applications such as an anesthesia simulator to compensate for the low frequency with which the average practitioner would encounter anaphylaxis in routine clinical care. Ideally, the combination of clinical, biochemical, and skin test evidence will identify the culprit agent and will allow the informed practitioner to avoid these agents in future clinical procedures. Moreover, the failure of catecholamines to restore cardiovascular homeostasis during anaphylaxis should receive more attention, especially regarding the establishment of precise recommendations about the potential indications of AVP in this clinical setting.

Kroigaard M, Garvey LH, Gillberg L, Johansson SG, Mosbech H, Florvaag E, Harboe T, Eriksson LI, Dahlgren G, Seeman-Lodding H, Takala R, Wattwil M, Hirlekar G, Dahlen B, Guttormsen AB: Scandinavian Clinical Practice Guidelines on the diagnosis, management and follow-up of anaphylaxis during anaesthesia. Acta Anaesthesiol Scand 2007; 51:655–70
Harper NJ, Dixon T, Dugue P, Edgar DM, Fay A, Gooi HC, Herriot R, Hopkins P, Hunter JM, Mirakian R, Pumphrey RS, Seneviratne SL, Walls AF, Williams P, Wildsmith JA, Wood P, Nasser AS, Powell RK, Mirakhur R, Soar J: Suspected anaphylactic reactions associated with anaesthesia. Anaesthesia 2009; 64:199–211
Johansson SG, Hourihane JO, Bousquet J, Bruijnzeel-Koomen C, Dreborg S, Haahtela T, Kowalski ML, Mygind N, Ring J, van Cauwenberge P, van Hage-Hamsten M, Wuthrich B: A revised nomenclature for allergy: An EAACI position statement from the EAACI Nomenclature Task Force. Allergy 2001; 56:813–24
Sampson HA, Munoz-Furlong A, Campbell RL, Adkinson NF Jr, Bock SA, Branum A, Brown SG, Camargo CA Jr, Cydulka R, Galli SJ, Gidudu J, Gruchalla RS, Harlor AD Jr, Hepner DL, Lewis LM, Lieberman PL, Metcalfe DD, O’Connor R, Muraro A, Rudman A, Schmitt C, Scherrer D, Simons FE, Thomas S, Wood JP, Decker WW: Second symposium on the definition and management of anaphylaxis: Summary Report-Second National Institute of Allergy and Infectious Disease/Food Allergy and Anaphylaxis Network symposium. J Allergy Clin Immunol 2006; 117:391–7
Evora PRB, Simon MR: Role of nitric oxide production in anaphylaxis and its relevance for the treatment of anaphylactic hypotension with methylene blue. Ann Allergy Asthma Immunol 2007; 99:306–13
Ring J, Messmer K: Incidence and severity of anaphylactoid reactions to colloid volume substitutes. Lancet 1977; 1:466–9
French Society of Anesthesiology and Intensive Care Medicine: Reducing the risk of anaphylaxis during anaesthesia: Abbreviated text. Ann Fr Anesth Reanim 2002; 21(suppl 1):7–23
French Society of Anesthesiology and Intensive Care Medicine
Dewachter P, Mouton-Faivre C: What investigation after an anaphylactic reaction during anaesthesia? Curr Opin Anaesthesiol 2008; 21:363–8
Lienhart A, Auroy Y, Pequignot F, Benhamou D, Warszawski J, Bovet M, Jougla E: Survey of anesthesia-related mortality in France. Anesthesiology 2006; 105:1087–97
Harboe T, Guttormsen AB, Irgens A, Dybendal T, Florvaag E: Anaphylaxis during anesthesia in Norway: A 6-year single-center follow-up study. Anesthesiology 2005; 102:897–903
Lobera T, Audicana MT, Pozo MD, Blasco A, Fernández E, Cañada P, Gastaminza G, Martinez-Albelda I, González-Mahave I, Muñoz D: Study of hypersensitivity reactions and anaphylaxis during anesthesia in Spain. J Investig Allergol Clin Immunol 2008; 18:350–6
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 2009; 39:325–44
Kounis NG: Kounis syndrome (allergic angina and allergic myocardial infarction): A natural paradigm? Int J Cardiol 2006; 110:7–14
Practice Parameters of the Joint Task Force on Practice Parameters for Allergy and Immunology: The diagnosis and management of anaphylaxis—An updated practice parameter. J Allergy Clin Immunol 2005; 115:S483–523
Hepner DL, Castells MC: Latex allergy: An update. Anesth Analg 2003; 96:1219–29
Turillazzi E, Greco P, Neri M, Pomara C, Riezzo I, Fineschi V: Anaphylactic latex reaction during anaesthesia: The silent culprit in a fatal case. Forensic Sci Int 2008; 179:e5–8
De Queiroz M, Combet S, Berard J, Pouyau A, Genest H, Mouriquand P, Chassard D: Latex allergy in children: Modalities and prevention. Paediatr Anaesth 2009; 19:313–9
Brown RH, Hamilton RG, Mintz M, Jedlicka AE, Scott AL, Kleeberger SR: Genetic predisposition to latex allergy: Role of interleukin 13 and interleukin 18. Anesthesiology 2005; 102:496–502
Hepner DL, Castells MC: Anaphylaxis during the perioperative period. Anesth Analg 2003; 97:1381–95
Ebo DG, Fisher MM, Hagendorens MM, Bridts CH, Stevens WJ: Anaphylaxis during anaesthesia: Diagnostic approach. Allergy 2007; 62:471–7
Mirakian R, Ewan PW, Durham SR, Youlten LJ, Dugue P, Friedmann PS, English JS, Huber PA, Nasser SM: BSACI guidelines for the management of drug allergy. Clin Exp Allergy 2009; 39:43–61
Levy J, Gottge M, Szlam F, Zaffer R, McCall C: Weal and flare responses to intradermal rocuronium and cisatracurium in humans. Br J Anaesth 2000; 85:844–9
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; 98:986–9
Berg CM, Heier T, Wilhelmsen V, Florvaag E: Rocuronium and cisatracurium-positive skin tests in non-allergic volunteers: Determination of drug concentration thresholds using a dilution titration technique. Acta Anaesthesiol Scand 2003; 47:576–82
Mertes PM, Monneret-Vautrin DA, Leynadier F, Laxenaire MC: Skin reactions to intradermal neuromuscular blocking agent injections: A randomized multicenter trial in healthy volunteers. Anesthesiology 2007; 107:245–52
Dewachter P, Mouton-Faivre C: Skin testing performed in individuals cannot be compared with responses from anaphylactic patients. Anesth Analg 2008; 107:1441
American Heart Association: Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care, part 10.6: Anaphylaxis. Circulation 2005; 112:IV143–5
American Heart Association
Dewachter P, Raeth-Fries I, Jouan-Hureaux V, Menu P, Vigneron C, Longrois D, Mertes PM: A comparison of epinephrine only, arginine vasopressin only, and epinephrine followed by arginine vasopressin on the survival rate in a rat model of anaphylactic shock. Anesthesiology 2007; 106:977–83
Schummer C, Wirsing M, Schummer W: The pivotal role of vasopressin in refractory anaphylactic shock. Anesth Analg 2008; 107:620–4
Del Duca D, Sheth SS, Clarke AE, Lachapelle KJ, Ergina PL: Use of methylene blue for catecholamine-refractory vasoplegia from protamine and aprotinin. Ann Thorac Surg 2008; 87:640–2