Abstract
Diagnosis of perioperative anaphylaxis (POA) and identification of causative agents remain challenging. This study aimed to describe the estimated incidence, characteristics and causative agents of POA in Malaysia. This is a retrospective review of all cases of suspected POA referred to the only anaesthetic allergy centre in Malaysia from March 2014 to December 2022. One hundred and ninety patients with suspected POA of Grade 2 and above were included. Data on clinical presentation, severity, management, serum tryptase and subsequent allergy workup (including skin and serum testing results) of these patients were extracted from the database. Dynamic tryptase was elevated in half of the cases where tryptase results were available and skin tests were positive in 96% of these cases. Skin testing was positive in 113 patients (60%) overall, and more than 70% of Grades 3 and 4 anaphylaxis cases. Neuromuscular blocking agents (NMBAs) and antibiotics were the most commonly identified causative agents (27.4% and 23% respectively). The commonest NMBAs were rocuronium and atracurium, both commonly cross-reacting with cisatracurium. The overall cross-reactivity rate among NMBAs was 58%. For antibiotics, the commonest causative agents were cefuroxime, ceftriaxone and amoxicillin/clavulanic acid. Using these data, the estimated incidence of Grades 2–4 POA over this period in Malaysia was approximately 1 in 30,000 anaesthetics. However, owing to the voluntary nature of reporting, it is possible that this is an underestimate, particularly in relation to some milder Grade 2 cases which may have gone unrecognised or unreported.
Keywords
Introduction
Perioperative anaphylaxis (POA) is an uncommon but potentially life-threatening event. Data on POA incidence are still limited in the Southeast Asia region, with a reported incidence that varies from 1:10,000 in Singapore 1 to 1:5500 in Thailand. 2 Nevertheless, many studies report neuromuscular blocking agents (NMBAs) as the first or second most commonly identified cause of POA in several countries.3 –9 Other commonly identified agents worldwide include antibiotics, chlorhexidine, dyes, latex and gelatins.3 –5 Concurring with these findings, a Singapore study in 2016 found that NMBAs were the most common cause of POA (37.5%), followed by antibiotics (25%). 1 A more recent study in a paediatric hospital in Singapore revealed antibiotics as the most commonly identified agent. 10 These studies were, however, limited by their small sample sizes. In Thailand, a study on perioperative allergic reactions found that the three most suspected causative (culprit) agents were NMBAs, antibiotics and opioids. 11 A study in Hong Kong had similar findings, with NMBAs being the commonest (25%) cause of POA followed by antibiotics (23%). 12 A recent study in Japan found that rocuronium, sugammadex and cefazolin were the top causative agents. 13 Sugammadex-induced anaphylaxis had also been reported in other countries,3,14 and there were case reports that the rocuronium–sugammadex complex was responsible for anaphylaxis.3,15,16
It is of note that some of the above-mentioned studies were not based on formal testing such as skin tests on the agents exposed during the suspected POA.2,11 The workup following suspected POA is often problematic as anaesthesia as a procedure involves multiple agents administered in succession. In addition, many of these agents may cause allergic (immune-mediated) or non-allergic (non-immune-mediated) reactions.5,17 Furthermore, the commonly available diagnostic tests are imperfect as they carry false positive or false negative results. Nonetheless, typical investigations involving skin tests with the prick and/or intradermal techniques have been used to identify the triggering agent(s). Further, the skin tests are complemented by in vitro tests, for example serum tryptase level, specific IgE (SIgE) test for the relevant agents involved, and basophil activation test (BAT) to add diagnostic accuracy, especially when drug provocation tests are contraindicated.5,17 –19
Based on the need to provide safer anaesthesia, the Anaesthetic Allergy Clinic, Hospital Kuala Lumpur was established in 2014 as the only referral centre to investigate suspected POA in Malaysia. Suspected POA cases in Malaysia have been referred to this clinic for investigation. A database of these referred cases has been kept at the clinic to date. While the occurrence of POA has been well described internationally from many countries, evidence on the incidence and characteristics of POA in Malaysia is still lacking. Therefore, we conducted a retrospective analysis from the available database to determine the estimated incidence, characteristics and causative agents of POA in Malaysia.
Materials and methods
This is a retrospective review of the database of all cases suspected of POA referred to the Anaesthetic Allergy Clinic, Hospital Kuala Lumpur, the only referral centre in Malaysia, from March 2014 to December 2022. This study was performed in accordance with the Declaration of Helsinki. Due to the retrospective nature of the study, de-identified data were extracted from the database and subsequently used in the study. Therefore, the acquisition of informed consent was deemed not necessary by the National Institutional Review Board. The study was approved by the National Institutional Review Board, the Malaysian Research Ethics Committee (NMRR ID-23-01197-FLA), dated 16 May 2023.
For entry of relevant information to the database, the referring anaesthesiologist was requested to fill in a referral form for formal testing that included a copy of the anaesthetic record and the following details in the referral: patient demographics, underlying medical illnesses, allergy history, list of agents and timing of exposure before onset of the event, clinical presentation including cardiovascular, respiratory and mucocutaneous involvement.
The criteria for referral of cases to the Anaesthetic Allergy Clinic were any of the following during anaesthesia: unexplained cardiac arrest with or without association of cutaneous manifestation, unexplained and unexpected hypotension which requires active treatment with or without association of cutaneous manifestation, unexplained and unexpected bronchospasm with or without association of cutaneous manifestation, generalised rash or urticaria and/or angioedema.
As reporting and referral were voluntary, data on unreported cases were not available. No travel expenses were provided for patients, thus, a few patients could not attend the clinic appointment due to financial constraints. Some patients were not contactable, and some refused testing despite repeated advice. These three groups of patients were excluded from the database.
The severity of POA is graded according to the Australian and New Zealand Anaesthetic Allergy Group (ANZAAG) guidelines 20 (Table 1), which is similar to the modified Ring and Messmer scale.5,17 As we aim to describe multisystemic involvement in POA, cases of suspected Grade 1 anaphylaxis were excluded as per the guidelines of the World Allergy Organization. 21 Management of anaphylaxis, clinical course of the patient, admission to critical care, and whether surgery was abandoned or completed were also included. Mast cell tryptase results were reviewed if available. A serum tryptase level taken within 3 h of reaction of at least more than 2 µg/l + (1.2 × baseline tryptase level) was defined as an elevated dynamic tryptase level. 22
Australian and New Zealand Anaesthetic Allergy Group grading of perioperative anaphylaxis. 20
Formal testing involved skin testing offered at four to six weeks after the event or later depending on the urgency of allergy testing. A detailed clinical and allergy history was taken together with written informed consent for skin and serum testing. Skin testing was performed by the anaesthesiologist in charge of the Anaesthetic Allergy Clinic in accordance with the ANZAAG testing guidelines. 6 Skin prick and intradermal tests were performed on the patient’s forearm using non-irritating concentrations for every exposed agent available, including latex and antiseptics if any. Readings were measured after 15 min and results were interpreted as positive if the size of wheal was double that of the initial intradermal injection or more than 3 mm for skin prick tests. Results of skin testing were recorded in the patient’s case notes. If a NMBA was identified as the causative agent, alternative NMBAs would be tested for cross-reactivity. If the culprit was an antibiotic, the patient would be referred to the allergist from the Institute for Medical Research, Malaysia for differential antibiotic testing.
Serum testing consisted of acute and baseline mast cell tryptase level, total IgE and SIgEs, depending on the agents exposed perioperatively. Available SIgEs included IgE to morphine (to test for NMBAs), suxamethonium, chlorhexidine, latex and latex components, bovine gelatin and alpha-gal, amoxicilloyl, ampicilloyl, penicilloyl G and penicilloyl V. BAT was not performed routinely for the cases in this study.
Statistical analysis
All data were analysed using Statistical Package for the Social Science (SPSS) V.27.0 IBM, Armonk, NY, USA). Normally distributed continuous variables were calculated to obtain the mean while non-normally distributed variables were expressed as median. Categorical data were tabulated to describe frequencies and percentages. Associations of categorical data were performed using chi-square test or Fisher’s exact test. A P-value of <0.05 was considered to be statistically significant.
Results
A total of 315 patients were referred to the Anaesthetic Allergy Clinic, Kuala Lumpur Hospital from the period of March 2014 to December 2022 for investigation of suspected POA. Thirty-one patients did not attend the clinic due to various reasons stated above. The total number of patients who attended the clinic during the study period was 284. Thirty-nine patients were excluded from the study as the clinical scenario was not perioperative in nature. A further 54 patients with suspected POA were excluded as the clinical presentation was Grade 1. One patient had a vague clinical presentation and we were unable to trace formal documentation to support the diagnosis of anaphylaxis. The final study population consisted of 190 patients (Figure 1). Patient demographics are listed in Table 2. The median age was 37 years with a greater proportion of the cases being female (57%). The youngest patient was eight years old and the oldest was 82 years old.
Overview of recruitment of patients with perioperative anaphylaxis.
Baseline characteristics of study population.
IQR: interquartile range.
The clinical features and management of POA in this study population of 190 patients are listed in Table 3. The majority of patients presented with cardiovascular manifestations (82%), in which hypotension was the most common presentation. About half of the patients presented with respiratory system involvement (54%), which was mainly bronchospasm and desaturation. There was no significant association between pre-existing cardiovascular disease and cardiovascular manifestations (odds ratio (OR) 0.78, 95% confidence interval (CI) 0.35 to 1.78, P = 0.560). However, the presence of pre-existing respiratory diseases was significantly associated with increased odds of respiratory manifestation, in which 75.6% of patients with respiratory diseases presented with a respiratory manifestation (OR: 3.69, 95% CI: 1.69 to 8.07, P = 0.003).
Clinical presentation, severity and management of perioperative anaphylaxis.
ICU: intensive care unit; PACU: post-anaesthesia care unit; HDU: high dependency unit.
More than half of the study population suffered a Grade 3 anaphylaxis (60%). Fourteen patients experienced cardiopulmonary arrest (Grade 4, 7.4%) but were successfully resuscitated. Sixty-two percent of the reactions occurred within 30 min post-induction of anaesthesia. The majority of cases received intravenous steroids as part of treatment, and adrenaline bolus and infusion were required in 68% and 44% respectively. Surgery was abandoned in one-third of the cases. More than 50% of patients required intensive care unit admission (Table 3).
Serial mast cell tryptase (acute tryptase within 3 h of reaction paired with baseline sample at 24 h) was sent in only 50% of patients (95 patients). Among these 95 patients, dynamic tryptase was elevated in 46 patients (48%). Interestingly, among the cohort of these 46 cases with raised dynamic tryptase, 44 patients (96%) eventually had a causative agent identified, whereas a causative agent was identified only in 23 patients (47%) when dynamic tryptase was negative.
Among the study population of 190 cases, skin testing was positive in 113 patients (60%), while in eight patients skin testing failed to identify any culprit (‘causative’ agent and ‘culprit’ means the same and are used interchangeably in this manuscript) despite the clinical picture being typical of anaphylaxis in terms of organ involvement and relevant timing. Of these eight patients, two patients whose dynamic tryptase was positive but skin test negative likely suffered from anaphylaxis due to blood products as both received blood transfusion intraoperatively. The remaining six patients, unfortunately, did not have tryptase sent but the clinical picture was strongly suggestive of anaphylaxis. Among these six cases, four patients had reactive skin leading to multiple positive responses at skin testing, including a patient whose history was strongly suggestive of latex allergy. The last two remaining patients had negative skin testing without acute tryptase level but clinically highly suggestive of anaphylaxis. The diagnosis of anaphylaxis was uncertain in the remaining 69 patients as skin testing was negative, acute tryptase was not elevated or not available and the clinical picture may be explained by other, alternative, diagnoses. The total number of anaesthetics delivered in Malaysia from 2014 to 2022 was 3,616,631. Based on the 121 cases highly suggestive of POA, we could estimate the incidence of Grade 2–4 POA in Malaysia to be approximately 1 in 30,000.
Identification of a causative agent by skin testing was more likely in higher grades of POA, where a causative agent was identified in 75% of Grade 3 and 71% of Grade 4 anaphylaxis, as opposed to only 30% in Grade 2 (Table 4).
Comparison between different severities of perioperative anaphylaxis and positive identification of causative agent by skin testing.
Serum for SIgE was available in 176 patients, depending on the exposed agents. Among these patients, skin testing identified a causative agent in 104 patients but only 44 had a positive SIgE correlating with the positive skin testing, yielding a sensitivity of 42%. Twenty-two patients had a raised SIgE but negative skin testing; these patients were labelled as possible sensitisation to the agent without clinical allergy. Among the patients with a positive skin test to NMBA, SIgE to morphine was only positive in three patients. The sensitivity of SIgE to morphine in patients with positive skin test to rocuronium was 21.4% vs 11.8% in patients with positive skin test to benzylisoquinolines (atracurium and cisatracurium). It should be noted that SIgE to pholcodine was not available at our centre, which may have impacted on these results. The sensitivity of SIgE to penicillins (penicilloyl G, penicilloyl V, ampicilloyl or amoxycilloyl) was higher at 62.5% in patients with positive skin test to penicillins (without taking into account the combination antibiotics such as amoxicillin–clavulanic acid and ampicillin–sulbactam), compared with 38.9% in patients with positive skin test to cephalosporins.
The causative agents identified by skin testing in the above 113 patients are illustrated in Figure 2. NMBAs were the most common cause of POA (31 patients, 27.4%) followed by antibiotics (26 patients, 23%). Chlorhexidine (16%) and modified gelatin intravenous fluids (12.4%) were the other common causative agents. Latex was the sole causative agent in only six cases (5%). Two or more agents were identified in eight patients, in 50% of whom an NMBA was identified as one of the causative agents (four patients). Two patients tested positive to both antibiotics and latex whereas another two patients tested positive to both radiocontrast medium and chlorhexidine. Other agents identified include dexamethasone, tramadol, ranitidine, ondansetron and povidone-iodine.
Causative agents of perioperative anaphylaxis in Malaysia. ‘Causative agents’ is used interchangeably with ‘culprit agents’. Multiple = more than one agent identified. NMBA: neuromuscular blocking agent; TXA: tranexamic acid
Among the 31 cases of NMBA anaphylaxis, the commonest NMBA identified was rocuronium (45%), followed closely by atracurium (42%). In two patients, more than one NMBA was given prior to the onset of anaphylaxis and both agents turned out positive at skin testing (Figure 3). The rate of cross-reactivity among NMBA was 58%, or 18 patients, in whom one or more other NMBAs was positive at skin testing when a primary culprit NMBA was identified. Cisatracurium was the commonest cross-reacting NMBA (94%). Among patients with rocuronium anaphylaxis (n = 14), four patients (28.6%) showed cross-reactivity to both cisatracurium and atracurium and one patient showed cross-reactivity to cisatracurium only (7%). In patients with atracurium anaphylaxis (n = 13), 10 patients (77%) showed cross-reactivity to cisatracurium and were skin test negative to other NMBAs.
Primary neuromuscular blocking agent (NMBA) identified as the causative agent in perioperative anaphylaxis.
Beta-lactam antibiotics, including penicillins and cephalosporins, were responsible for all 26 cases of anaphylaxis caused by antibiotics. The top three culprit antibiotics were cefuroxime (seven cases), ceftriaxone (six cases) and amoxicillin–clavulanic acid (five cases).
Discussion
To our knowledge, this study in Malaysia involved the largest POA cohort reported so far in Southeast Asia. The estimated incidence of Grade 2–4 POA during the study period was approximately 1 in 30,000; however, this may not reflect the true incidence as it was based on voluntary reporting. The authors suspect that under-reporting and lack of awareness contributed to this low incidence. Moreover, in some cases, milder reactions such as Grade 2 may not have been diagnosed as anaphylaxis, and therefore not referred for formal testing. This suspicion of under-reporting has led the authors to conduct periodic awareness programmes in the form of workshops to increase the referral of suspected Grade 2 reactions and the use of acute tryptase when anaphylaxis is suspected.
In this study, the majority of patients presented with cardiovascular manifestations, of which hypotension was the main feature. Nevertheless, this was independent of the presence of pre-existing cardiovascular disease. In comparison, patients with pre-existing respiratory disease were significantly associated with respiratory system involvement, with bronchospasm being the most common presentation. In these cases, there may be other differential diagnoses, such as airway hyper-reactivity or respiratory complications, other than anaphylaxis.
In our cohort, more than half of the cases presented with Grade 3 anaphylaxis with life-threatening hypotension, but accompanying tachycardia was not frequent (only 25% of cases). Reactions more commonly occurred within a short period after induction of anaesthesia, similar to other studies.1,10,12,23 Management of POA in Malaysia followed international guidelines where adrenaline remained the mainstay of treatment along with fluid resuscitation and steroids.17,18,23 Antihistamines were still administered in a significant number of cases although their benefits had not been proven in studies.4,23 Surgery was abandoned in one-third of cases and 70% required admission to critical care or high-dependency areas. This represented a significant healthcare burden attributable to POA in Malaysia.
A raised dynamic tryptase for diagnosing POA has been quoted to have a sensitivity and specificity of 78% and 91% respectively. 24 Unfortunately, serial mast cell tryptase was sent in only half of the cases analysed. Among the cohort of cases in which serial tryptase was sent after the event, almost all patients with a raised dynamic tryptase eventually had a causative agent identified. These findings concurred with a study in Hong Kong. 12 It is also imperative to note that a negative dynamic tryptase does not exclude anaphylaxis and further investigation such as skin testing in clinically suspected cases is still warranted. 24 This was reflected in this study, where a causative agent was identified by skin testing in 47% of patients with negative dynamic tryptase. This could be explained by the time-lag between sample collection and the acute reaction due to the inherent nature of tryptase levels during anaphylactic reaction. 19 In addition, as we followed verbatim the ANZAAG guideline for skin testing, it was possible that the Malaysian population might have responded differently to the maximum non-irritating concentrations of the guideline.
As has been noted by a consensus statement, POA of Grades 1 and 2 reactions were said to be more likely non-IgE mediated (e.g. non-specific activation of mast cells and basophils) whereas Grades 3 and 4 reactions were more likely to be IgE mediated. 23 It is interesting to note that a causative agent was identified by positive skin test in more than 70% of Grades 3 and 4 anaphylaxis in this study. This is probably because Grades 3 and 4 reactions were IgE mediated as mentioned.
This study found that NMBAs and antibiotics were the two most common causative agents of perioperative anaphylaxis in Malaysia, similar to other countries around the region and globally,1 –4,10 –12 although geographical variations between countries and regions may be present. Contrary to other studies in Southeast Asia, our findings came from more robust data with a larger cohort. The most common NMBA implicated was rocuronium, followed closely by atracurium. Interestingly, the findings of this study differed from those found in NAP6, 4 where suxamethonium was the most commonly implicated NMBA. There were no positive cases related to suxamethonium in the current study, possibly because of the lack of its routine use in Malaysia. The true incidence of anaphylaxis according to individual triggering NMBA could not be determined owing to the unavailability of accurate exposure data. In contrast to previous studies,3,4,7 cisatracurium was the most likely NMBA to cross-react in the present study. Nevertheless, this finding might be attributed to false positive results as we were unable to confirm this with graded intravenous challenges with cisatracurium for logistical reasons. Moreover, we could not adjust incidences based on the relative use of various NMBAs in Malaysia during the study period. Sugammadex was also not identified in any of the cases, most likely because it was still not used widely as a reversal agent for neuromuscular blockade in Malaysia because of its high cost.
Apart from NMBAs, we found that beta-lactam antibiotics were also one of the common causative agents in our cohort. The indiscriminate labelling of beta-lactam allergy has been shown to result in poorer clinical outcomes, increased incidence of serious antibiotic-resistant infections, prolonged hospitalisations, and increased healthcare utilisation owing to the use of antibiotics from alternative classes. 25 It is therefore of utmost importance to perform differential antibiotic testing in these cases to ascertain cross-reactivity among beta-lactam antibiotics, which would impact future perioperative management. It has been suggested that antibiotics should be administered prior to induction of anaesthesia to simplify treatment and subsequent diagnostic workup. 4 Further in our analysis, chlorhexidine was identified as the third most frequent causative agent, which, incidentally, is also a major culprit of POA worldwide.3,4 As chlorhexidine use is ubiquitous in the hospital environment, it is important to practise vigilance and caution during the perioperative period as it may be a hidden allergen.
With regard to in vitro tests, SIgEs have been said to have variable and generally low sensitivity and specificity except for latex and chlorhexidine.5,18,24 It is also only available for a limited number of agents. The available SIgEs to test for NMBA in our centre were IgE to suxamethonium and IgE to morphine. SlgE to morphine and suxamethonium may be unreliable when used in isolation in the investigation of POA as they may give false positive results, owing to possible sensitisation to tertiary and quaternary substituted ammonium which is prevalent in the general population. 19 Studies have shown that a morphine-based immunoassay is useful to detect suxamethonium- and rocuronium-reactive antibodies but not atracurium-reactive antibodies.18,19,24 However, in this study, the sensitivity of SIgE to morphine is low for both patient groups with skin test positive to rocuronium (21.4%) and to benzylisoquinolines (11.8%). SIgE for rocuronium, atracurium and pholcodine to aid diagnosis of NMBA-induced anaphylaxis was not available during the study period, thus may have impacted on the sensitivity results of the SIgE in this study. Furthermore, it is also possible that some of the NMBA anaphylaxis may be due to IgE-independent reactions such as via activation of Mas-related G protein-coupled receptor member X2 receptor.5,26
Similarly, SIgE to beta-lactam antibiotics had poor sensitivity in our study, concurring with previous studies.19,24 However, SIgE to penicillins had better sensitivity in patients with a positive skin test for penicillins (62.5%) when compared with patients with a positive skin test for cephalosporins (38.9%). Nevertheless, this higher SIgE sensitivity of 62.5% did not take into account other agents combined with the antibiotics such as amoxicillin–clavulanic acid and ampicillin–sulbactam, or the ratio of SIgE:total IgE. Thus, similar to NMBAs, SIgE to beta-lactam antibiotics should not be used in isolation to exclude or confirm the diagnosis of anaphylaxis to these antibiotics in POA. Considering that cefuroxime and ceftriaxone were the most common antibiotics identified in this study, SIgE to penicillins added minimal value to the diagnosis of anaphylaxis to cephalosporins.
From the results of the present data, it can be deduced that SIgEs only complemented skin testing or guided diagnosis when skin testing was contraindicated or could not be performed to identify the causative agents. It should be noted also that SIgE was routinely sent for all exposed agents and not specifically paired with the suspected drug in the study population, which contributed to the diagnostic inaccuracies of SIgEs.
In this study, a culprit was identified in 60% of the cases, corresponding with most studies which showed that the proportion of reactions in which a culprit is identified varies in the range of 40% to 70%. 27 As mentioned, there were a few patients with a typical clinical picture of anaphylaxis, but investigations failed to identify the culprit. It is suggested that in these patients, we need to consider provocation tests, re-evaluate for overlooked (hidden) culprits and consider re-testing in two to three months. 27 The diagnosis of anaphylaxis is uncertain in the remaining skin test-negative patients where the clinical picture might be due to surgical or anaesthetic causes, non-specific histamine release of certain drugs, or other non-immune-mediated mechanisms. This again highlighted the challenges and complexities in the diagnosis of perioperative anaphylaxis despite a systematic approach.
In common with other studies, this study has several limitations, the first of which is the inherent limitations of a retrospective study. In addition, although the current study showed skin testing was more sensitive compared with SIgE to identify the causative agent, we may wrongly diagnose patients with skin test negativity due to false negative results or incorrectly label patients with reactive skin due to false positivity. Furthermore, we did not perform provocation tests to NMBA to confirm or disprove alternative NMBA from the skin testing results because of the risks involved owing to their pharmacologic effects. Future efforts to provide the setting for provocation tests towards anaesthetic drugs may assist, particularly in determining the cross-reactivity between the NMBAs. Also, BAT has been suggested to have a better sensitivity, of 50% to 90%. 24 A recently proposed diagnostic algorithm for the investigation of perioperative hypersensitivity reactions suggested the inclusion of BAT especially if the skin test is negative. 24 This has been confirmed by a recent study from Japan that found the combination of tryptase levels, skin testing, and BAT results and clinical score improved the certainty of anaphylaxis diagnosis. 13 Nevertheless, BAT was not performed routinely but rather remained a research tool in Malaysia, mainly due to the technical laboratory limitations that come with the test. Moreover, the logistical difficulties posed an additional challenge as BAT involved the analysis of living cells. Future improvements in these respects may include the routine use of BAT. Finally, the unavailability of data on pholcodine exposure and IgE tests to pholcodine might make our analysis incomplete. However, we would have to wait and see from the recent ban of pholcodine-containing drugs in Malaysia from April 2023 whether it may alter the causative agents involved for POA in future. In conclusion, we could summarise that the diagnosis of POA relies on a relevant clinical history and a combination of tests involving skin testing supplemented with in vitro tests such as serum tryptase, specific IgEs, BAT or histamine release test. 5
Conclusion
This study of POA in Malaysia found that skin tests were positive in almost all cases that had raised dynamic tryptase, and also in more than 70% of higher grades of POA—thus identifying the causative agent(s). NMBAs and antibiotics were the two most common causative agents of POA. Rocuronium and atracurium were the main agents responsible for NMBA anaphylaxis. The cross-reactivity rate among NMBAs was found to be 58%. Further, it could be estimated that the incidence of Grade 2–4 POA in Malaysia was approximately 1 in 30,000, although this may be an underestimate due to the voluntary nature of reporting, and the possibility that some milder Grade 2 cases of POA were underdiagnosed.
Footnotes
Author Contribution(s)
Acknowledgements
We would like to thank the Director General of Health, Ministry of Health, Malaysia, for his permission to publish this study. We would also like to thank all staff at the Anaesthetic Allergy Clinic, Hospital Kuala Lumpur, for their support in this study.
Data availability statement
The data that support the findings of this study are available upon request from the corresponding author.
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
