In 1969, Hurwitz1 commented that the more drugs a patient receives, the greater the risk of an untoward reaction. Multiple drugs need to be used in surgery, since no single drug supplies all pharmacological needs, and this encourages unexpected drug interaction and pharmacological response. Problems are compounded by the necessity to use the intravenous route for drug administration, thus bypassing the bodys primary immune filters and presenting high concentrations of noxious chemicals directly to sensitive cells, notably mast cells and basophils.2 The substances liberated include histamine, eicosanoids and cytokines. The clinical outcome is not infrequently an immediate exaggerated systemic inflammatory response presenting hypotension and bronchospasm with varying degree of severity and which may, or may not, be immune (antibody) mediated. All such clinical presentations are usually referred to as anaphylactoid until laboratory and in vivo analysis can establish a genuine immediate immune-mediated hypersensitivity response (type I, IgE-mediated anaphylaxis) or otherwise, and the causative agent responsible. There are other types of drug hypersensitivity response, notably IgG (type III)-mediated response to certain clinically used dextrans and a multiplicity of responses to penicillins (potentially types I, II and III). In the majority, cause and effect is somewhat transparent and the immediate hypersensitivity response, with its requirement for immediate clinical intervention, is the greatest threat to the patient. Identification of the reaction mechanism is important both for the patients surgical future and as defence for the anaesthetist in the increasing tide of litigation. The predominant cause of such a reaction is a neuromuscular blocking agent (NMBA) or, rarely, a hypnotic agent.3
While sequential analysis of plasma tryptase levels as a measure of mast cell degranulation is the most reliable marker of a drug-related event,4 tryptase release per se is not a specific marker of anaphylaxis.5 6 Anaphylaxis is a quantitative, not a qualitative, phenomenon and it becomes necessary to demonstrate not only the level of tryptase release but also the specific antibody responsible. The predominant test for the latter has been skin testing against the drugs administered, with radioimmunoassays carried out either supplementary to the skin test, at autopsy or when urgent surgery is necessary before the skin test can be carried out (usually 46 weeks after the event). Incidentally, most of the current commercially available immunoassays employ colorimetry or fluorimetry for quantification, rather than radioactivity.
For more than two decades, the skin testing procedure has been dominated by two groups, led by Fisher and Baldo in Australia and Laxenaire and Moneret-Vautrin in France. It should be pointed out that the term skin testing is widely used in the literature as synonymous with intradermal skin testing. However dermatologists also routinely use prick testing for common allergens and this test depends largely on the presence of sensitized lymphocytes. Whereas intradermal drug tests present high concentrations of active chemicals directly to skin mast cells, indirect contact occurs with the scratch test. Although the visual endpoint of either test is the classic weal-and-flare response, the latter test should largely filter out reactions caused by non-immunological mast cell activation mechanisms.7 In a comparative prospective study, Fisher and Bowey8 found no statistically significant difference between the two, but a comparative trial in reactors is lacking. Ideally, both tests should be carried out on the patient.7 8
Modern developments of NMBAs centre around two main chemical classes, the aminosteroids and the benzylisoquinolines, and these have given rise to clinically safer compounds. However, Levy and his colleagues report in this issue of the journal9 that these drugs can induce positive weal-and-flare responses independent of mast cell degranulation and that this can confuse allergy testing and interpretation between different investigators, particularly at the test concentrations usually recommended. Although a prospective study, it does receive support from reactor studies (rocuronium) carried out by Neal and colleagues.10 Threshold test concentrations have been described in the past, and are usually 1:1000 of ampoule concentration but as high as 1:10 for rocuronium. For the new drugs these concentrations may still be too high and the high incidence of positive cutaneous responses to rocuronium reported in France11 may well contain a high proportion of false positives.
The major problem with all skin testing is what you see is all that you get, a weal and flare. The problem lies in both the heterogeneity of the mast cell and its heterogeneity of activation.12 The mast cell is a key inflammatory cell; it can be activated, releasing histamine and tryptase, by multiple mechanisms, including specific surface IgE activation (immune), complement activation and direct chemical (drug) effects.13 14 With the demise of Cremophor-solubilized drugs (e.g. Althesin), the very severe life-threatening complement (C3)-mediated effects have disappeared. The antigenicity of NMBAs is complex. Their pharmacological action involves the unique bisquaternary ammonium ions present in the most potent of these compounds. The interonium distance is also critical, indicating that these compounds act by bridging receptors at the neuromuscular junction. However, monoquaternary and even some monotertiary compounds have strong neuromuscular blocking properties and it may be that, in these cases, a pair of molecules acting together provide the necessary bridge. Immune triggering of mast cells and basophils requires an analogous bridging of adjacent, surface-bound specific IgE antibody molecules, but here antigenicity appears to be much more intimately attributed to the bisquaternary ions.15 Thus, despite its well-established concentration-dependent release of plasma histamine, tubocurarine (monotertiary, monoquaternary) reactions are rarely immune-mediated, in sharp contrast to the behaviour of the bisquarternary drugs, succinylcholine and alcuronium. All the benzylisoquinolines exhibit the bisquaternary structure and are complex flexible molecules. High doses or fast administration lead to some plasma histamine release but the drugs are cardiovascularly stable. Atracurium readily produces a false positive weal-and-flare response unless care is taken with dilutions.
The aminosteroids vary in their ammonium ion distribution, do not release plasma histamine and rarely show a response to intradermal tests. Pancuronium, widely considered the safest NMBA in this respect, has less than critical interonium spacing (for blockade and possibly for immune bridging) while vecuronium and rocuronium are monotertiary, monoquaternary drugs. Thus, one would expect that fewer immune reactions will be observed with aminosteroids than with the benzylisoquinolines. Despite this, the overall clinical safety profiles of the two classes are virtually identical.16 Just why some NMBAs are so antigenic is obscure; succinylcholine is generally considered the most antigenic,3 despite its very close structural relationship to acetylcholine. Further, some patients appear to carry cross-reacting antibodies to succinylcholine without previous exposure to that agent, and these antibodies can persist for many years. It has been suggested that some patients become cross-sensitized to NMBAs as a result of the widespread exposure to the quaternary ammonium structure in nature (e.g. in foodstuffs and cosmetics) or perhaps as an autoimmune response to acetylcholine. However, most reactors have evidence of previous uneventful exposure to that drug and hence some sort of conventional immune sensitization.
The mechanisms of non-immunological mast cell activation are obscure. They may involve degranulation to varying degrees, with both histamine and tryptase release or the release of other mediators of the inflammatory response, the eicosanoids, without significant degranulation. While morphine is considered to be the classic opiate, its potential for releasing significant histamine appears unique and is not shared by the fentanyl compounds. The tertiary ammonium group of morphine provides a potent blocker in vitro of IgE antibodies to NMBAs. Local anaesthetic agents offer a halfway stage structurally to NMBAs, since they are bistertiary ammonium compounds. However, hypotension is most commonly the result of the local anaesthetic blockade of sympathetic fibres and not histamine release, although severe histaminoid responses have been encountered in Biers block procedures. The latter may relate to lipid membrane solubility or effects on sodium ion channels. Intradermal testing against these compounds has, at best, been equivocal.
The mechanism of non-immunological histamine and other mediator release appears to involve membrane-associated enzymes such as phospholipase A2 and phospholipase C, involved in arachidonic acid synthesis and metabolism.6 Antigenic responses to Diprivan may involve the effect of its detergent carrier on mast cell membranes as much as the active ingredient, diisopropylphenol. The initial solution of this drug in Cremophor by its then manufacturers, ICI, gave rise to a very high incidence of life-threatening complement-mediated histamine release and the drug had to be reformulated. Like the aminosteroid neuromuscular blocking drugs, Diprivan has a tendency to target the lungs as the principal shock organ. Indications of positive skin test results to Diprivan seem too high to account for genuine allergy to this simple phenol-based molecule, particularly when compared with the low incidence of classic sensitization to the barbiturate hypnotics.
Finally, one must address the problem of the site chosen for skin testing. Skin testing is usually carried out on the anterior face of the forearm or on the patients back. The latter is favoured in France17 since this skin is less exposed to environmental modifications and has similar characteristics in all individuals. Moneret-Vautrin and Laxenaire17 also comment that the thin skin of the forearm is more likely to release histamine non-specifically. Drug test doses are selected on the basis of the concentration least likely to initiate a non-specific weal-and-flare response and these have largely been based on the observation of sequential concentration increases in reactors. The elegant studies reported by Levy and colleagues9 must be tempered with the realization of the potential for volunteer sensitization, fortunately absent in this study. Mast cells isolated from different anatomical sites show marked heterogeneity when challenged with different neuromuscular blocking drugs in vitro. Thus, in studies reported by Marone and Stellato,18 succinylcholine did not induce histamine release from any type of mast cell. Vecuronium induced minimal (5%) release from skin and lung mast cells but not at all from heart, in contrast to atracurium, which showed concentration-dependent release up to 46%. Commenting on the variation in the histamine-releasing activity of atracurium between individuals, now extended to other anaesthetic drugs, Stellato and colleagues introduced another variable: mast cell releasability factor.19 Levy and his colleagues have extended these observations with skin biopsy and light and electron microscopy.9 Here, the non-immune-induced weals revealed moderate degranulation when generated by cis-atracurium but not by rocuronium.
Skin test results should only be interpreted in terms of the tryptase-defined response and the patients clinical history. Despite the comparative simplicity of the testing protocols,8 17 considerable experience is necessary to exclude false positives and tests ideally should be carried out at centres with experience of drug testing, preferably by dermatologists or allergologists, and not as a one off investigation by the anaesthetist. A current unsatisfactory trend is to test simultaneously a panel of all the usual neuromuscular blocking drugs with a view to elucidating future potential cross-reactivity. This panel is inevitably tested at the same dilution and not infrequently neat, at ampoule strength. Laboratory reports are often accompanied by instruction for the patient not to receive any of the positive reactors, seriously compromising availability for future anaesthesia. Anecdotally, some anaesthetists whose patients have been screened for a whole battery of neuromuscular blocking drugs have received positive reports on a drug which has been used, without incident, to stabilize the shocked patient. This is either a non-immune response or genuine cross-reactivity but with antibody avidity insufficient to precipitate clinical anaphylaxis (cf. antibodies to house-dust mite and insect venom). Perhaps it is now time to re-examine such observations and their potential medicolegal significance. Prick testing may provide more specificity with the newer drugs while Fisher and Baldo have developed an intriguing radioimmunoassay using morphine bound in solid phase.20 They claim this is a more sensitive and efficient test for the detection of IgE antibodies to NMBAs than individual drug-specific radioimmunoassay. Either way, the anaesthetist requires clear, nationally agreed, guidelines on the investigation of hypersensitivity-type reactions, the incidence of which, despite considerable advances in patient monitoring and new drugs over the last decade, has remained virtually unchanged.
John Watkins
Sheffield
References
1 Hurwitz N. Predisposing factors in adverse reactions to drugs. Br Med J 1969; 1: 5369[ISI][Medline]
2 Genovese A, Stellato C, Marsella CV, Adt M, Marone G. Role of mast cells, basophils and their mediators in adverse reactions to general anaesthetics and radiocontrast media. Int Arch Allergy Appl Immunol 1996; 110: 1322
3 Naguib M, Magboul MA. Adverse effects of neuromuscular blockers and their antagonists. Drug Safety 1998; 18: 99116[ISI][Medline]
4 Watkins J. Tryptase release and clinical severity of anaesthetic reactions. Agents Actions 1992; Special Conference Issue: C2035
5 Fisher MM, Baldo BA. Mast cell tryptase in anaesthetic anaphylactoid reactions. Br J Anaesth 1998; 80: 269[ISI][Medline]
6 Veien M, Szlam F, Holden JT et al. Mechanisms of nonimmunological histamine and tryptase release from human cutaneous mast cells. Anesthesiology 2000; 92: 107481[ISI][Medline]
7 Wood M, Watkins J, Wild G et al. Skin testing in the investigation of reactions to intravenous anaesthetic drugs. A prospective trial of atracurium and tubocurarine. Ann Fr Anesth Reanim 1985; 4: 1769[Medline]
8 Fisher MM, Bowey CJ. Intradermal compared with prick testing in the diagnosis of anaesthetic allergy. Br J Anaesth 1997; 79: 5963
9 Levy JH, Gottge M, Szlam F et al. Weal and flare responses to rocuronium and atracurium in humans. Br J Anaesth 2000; 85: 8449
10 Neal SMcD, Manthri PR, Gadiyar V, Wildsmith JAW. Histaminoid reactions associated with rocuronium. Br J Anaesth 2000; 84: 10811[Abstract]
11 Laxenaire MC. Epidemiologie des reactions anaphylactoides peranesthesiques. Quatrieme enquete multicentrique. Ann Fr Anesth Reanim 1999; 18: 796809[ISI][Medline]
12 Benyon CR, Robinson C, Church MK. Differential release of histamine and eicosanoids from human skin mast cells by IgE dependent and non-immunological stimuli. Br J Pharmacol 1989; 97: 898904[Abstract]
13 Stellato C, Paulis AD, Cirillo R et al. Heterogenicity of human mast cells and basophils in response to muscle relaxants. Anesthesiology 1991; 74: 107886[ISI][Medline]
14 McKinnon RP, Wildsmith JAW. Histaminoid reactions in anaesthesia. Br J Anaesth 1995; 74: 21728
15 Didier A, Cador D, Bongrand P et al. Role of the quaternary ammonium ion determinants in allergy to muscle relaxants. J Allergy Clin Immunol 1987; 79: 57884[ISI][Medline]
16 Watkins J. Adverse reactions to muscle relaxants. The role of the laboratory in clinical investigations. In: Harper NJN, Pollard BJ eds. Muscle Relaxants in Anaesthesia. London: Edward Arnold, 1995; 23458
17 Moneret-Vautrin DA, Laxenaire MC. Skin tests in diagnosis of allergy to muscle relaxants and other anaesthetic drugs. Monogr Allergy 1992; 30: 14555[ISI][Medline]
18 Marone G, Stellato C. Activation of human mast cells and basophils by general anaesthetic drugs. Monogr Allergy 1992; 30: 5473[ISI][Medline]
19 Stellato C, Cirillo R, Paulis AD et al. Human basophil/mast cell releasability. IX: Heterogeneity of the effects of opioids on mediator release. Anesthesiology 1992; 77: 93240[ISI][Medline]
20 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 Intens Care 2000; 28: 16770[ISI][Medline]