Department of Paediatrics, Division of Paediatric Pulmonology and Paediatric Allergy, and GRIAC Research Institute University Medical Centre Groningen, University of Groningen, Groningen, the Netherlands
Correction added on 17 January 2014, after first online publication: spelling of author Niggemann and initials of author Santos corrected.
To establish the effectiveness of interventions for the acute and long-term management of anaphylaxis, seven databases were searched for systematic reviews, randomized controlled trials, quasi-randomized controlled trials, controlled clinical trials, controlled before–after studies and interrupted time series and – only in relation to adrenaline – case series investigating the effectiveness of interventions in managing anaphylaxis. Fifty-five studies satisfied the inclusion criteria. We found no robust studies investigating the effectiveness of adrenaline (epinephrine), H1-antihistamines, systemic glucocorticosteroids or methylxanthines to manage anaphylaxis. There was evidence regarding the optimum route, site and dose of administration of adrenaline from trials studying people with a history of anaphylaxis. This suggested that administration of intramuscular adrenaline into the middle of vastus lateralis muscle is the optimum treatment. Furthermore, fatality register studies have suggested that a failure or delay in administration of adrenaline may increase the risk of death. The main long-term management interventions studied were anaphylaxis management plans and allergen-specific immunotherapy. Management plans may reduce the risk of further reactions, but these studies were at high risk of bias. Venom immunotherapy may reduce the incidence of systemic reactions in those with a history of venom-triggered anaphylaxis.
Anaphylaxis can be defined as a ‘severe, life-threatening generalized or systemic hypersensitivity reaction’ [1, 2]. Several working definitions of anaphylaxis have been formulated to aid clinical diagnosis and management [3-6]. The most well known of these is the consensus clinical definition proposed by Sampson et al., which involved representatives of a number of international allergy organizations, including the European Academy of Allergy and Clinical Immunology (EAACI) .
Care of patients with anaphylaxis involves consideration of both the acute, emergency treatment of reactions and long-term care, which aims to reduce the risk of further reactions and improve outcomes if, despite these measures, a further reaction ensues . It remains very difficult to predict the severity of a reaction and, in fatal episodes, death may occur within minutes of an anaphylactic reaction, these observations underscoring the importance of effective, emergency management.
EAACI is in the process of developing the EAACI Guidelines for Food Allergy and Anaphylaxis, and this systematic review is one of seven interlinked evidence syntheses that were undertaken in order to provide a state-of-the-art synopsis of the current evidence base in relation to epidemiology, prevention, diagnosis and clinical management, and impact on quality of life, which were used to inform clinical recommendations.
The aims of this systematic review were to assess the effectiveness of (i) interventions for the acute management of anaphylaxis and (ii) pharmacological and nonpharmacological approaches for the long-term management of anaphylaxis.
Details of the methodology for the identification, selection and inclusion of the studies have been previously reported [9, 10]. In summary, our inclusion criteria were systematic reviews with or without meta-analyses, randomized controlled trials (RCTs), quasi-RCTs, controlled clinical trials (CCTs), controlled before–after (CBA) designs, interrupted time series (ITS) studies and case series, with a minimum of 10 patients, for studies investigating the use of adrenaline (Fig. 1).
Standard methods for performing systematic reviews were used. A descriptive summary with data tables was produced to summarize the literature. Quality assessments of studies were undertaken using appropriate tools (see online supplement). We preferentially extracted data on risk ratios and mean differences. Because data were not suitable for meta-analysis , a narrative synthesis of the data was undertaken.
Further details can be found in the online Supporting Information.
The searches identified a total of 8929 potentially eligible studies, of which 55 satisfied our eligibility criteria and were therefore included in this review (Fig. 2). The key characteristics and main findings of all included studies are detailed in Table S1, and the quality assessment of these studies is summarized in Tables S2 (systematic reviews) and S3 (primary studies). The main findings are further discussed below.
Acute management of anaphylaxis
We identified 15 studies: four systematic reviews [12-15] (including one update), four RCTs [16-19], two case series [20, 21] and five fatality register-based reports [22-26] (including two updates) on the effectiveness of adrenaline.
Effectiveness and timing
Three well-conducted systematic reviews [19-21] that looked at the use of adrenaline for the treatment of anaphylaxis and adrenaline auto-injectors found no RCTs or quasi-RCTs. Weaker evidence from a case series of 27 patients receiving emergency prehospital treatment found that prompt use of adrenaline may reduce the risk of fatality . The second case series found that a fifth of children experiencing anaphylaxis needed more than one dose of adrenaline; healthcare professionals administered this second dose in 94% of cases. The five fatality register-based reports [12-16] provided important insights into the difficulties of predicting the severity of subsequent reactions, risk factors for fatality, co-existing asthma and the under-issuing, poor carriage, under-use, delayed use and incorrect use of adrenaline auto-injectors. These reports revealed that fatalities can occur, even if adrenaline is used correctly.
Site and route of delivery
Two RCTs [22, 24] found that, in both children and adults, maximal plasma concentration occurs quicker with the intramuscular than with subcutaneous route. The latter trial in adults also concluded that the optimum site of injection was the vastus lateralis muscle. A systematic review of poor quality found that the use of subcutaneous adrenaline was not contraindicated in patients older than 35 years without a history of coronary artery disease .
Dose in children
One RCT compared the effectiveness of the previous designs of the Epipen Junior and Epipen in children weighing 15–30 kgs. The authors concluded that children should be prescribed the 0.3 mg Epipen from 30 kgs .
Two systematic reviews investigating the use of glucocorticosteroids in the acute management of anaphylaxis [27, 28] failed to identify any RCTs or quasi-RCTs and so were unable to make any recommendations.
Two systematic reviews investigating the use of H1-antihistamines in the acute management of anaphylaxis [29, 30] identified no suitable RCTs or quasi-RCTs. Two RCTs that investigated the use of H1- and H2-antihistamines in acute allergic reactions (only a small proportion had anaphylaxis) [31, 32] revealed that the combination of H1- and H2-antihistamines was superior to H1-antihistamines alone in treating urticaria, but not angioedema; the second showed that pruritus was better controlled by H1-antihistamines than by H2-antihistamines and that combined treatment offered no advantage.
We found one systematic review investigating the effectiveness of methylxanthines in the acute management of anaphylaxis; this found no trials in humans .
Long-term management of anaphylaxis
Anaphylaxis management plans
We identified two systematic reviews investigating anaphylaxis management plans [34, 35]. The first found no evidence from RCTs or quasi-RCTs investigating the clinical effectiveness of anaphylaxis management plans for the prevention of recurrences or improvement in outcomes from anaphylaxis. The second systematic review had a wider focus including qualitative, epidemiological and experimental studies undertaken with or without a control group. This demonstrated that there were no universally accepted anaphylaxis management plans. It, however, identified four studies, which suggested that anaphylaxis management plans may substantially reduce the risk of future, severe reactions.
Venom immunotherapy (VIT)
We found three systematic reviews and meta-analyses [36-38] and one further systematic review without a meta-analysis  investigating the effectiveness of VIT. These four systematic reviews included a total of 23 unique studies of varying quality. Twelve of these studied patients with a history of anaphylaxis using eligible study designs and have therefore also individually been assessed (see Tables S1 and S3) [40-51].
The high-quality systematic review by Boyle et al. identified six RCTs and 1 quasi-RCT (392 participants). Five studies involved subcutaneous immunotherapy and one sublingual immunotherapy. Patients treated with VIT had less systemic allergic reactions (RR = 0.10, 95% CI 0.03–0.28) and better quality of life (mean difference = 1.21 points on a 7-point scale (95% CI 0.75–1.67)). Subcutaneous VIT-treated patients had more systemic adverse reactions (RR = 8.16, 95% CI 1.53–43.46). One systematic review of low quality showed that VIT was effective in reducing the risk of further systemic reactions .
The systematic review by Hockenhull et al.  investigated the clinical and cost-effectiveness of a specific VIT subcutaneous preparation (Pharmalgen, ALK-Abello, Reading, UK) and included nine trials (four RCTs and five quasi-RCTs), all judged to be of poor quality. They modelled cost-effectiveness showing a cost of £8–20 (€10–25) million/life year gained, assuming a base-case scenario of no improvement in quality of life. A sensitivity analysis showed that VIT was more cost-effective in those at high risk of further stings or if improvements in quality of life and anxiety associated with VIT were included in the model.
A quasi-experimental trial evaluated whether an educational intervention could improve compliance with carrying an in-date adrenaline auto-injector in high school children with food allergy . The intervention failed to demonstrate any significant difference in carriage rates in the intervention and control groups.
One systematic review of low quality investigated the management of anxiety related to children with a history of anaphylaxis . It concluded that anaphylaxis can place a substantial psychological burden on children, adolescents and carers and that dealing with this anxiety may improve outcomes.
Interventions have been studied to prevent anaphylaxis. A major limiting factor with these studies is the fact that the groups studied were not known to be at high risk of anaphylaxis.
Adrenaline admixture with snakebite antivenom
A systematic review and meta-analysis demonstrated that adrenaline premedication reduces adverse reactions when administering snakebite antivenom  (RR = 0.32, 95% CI 0.18–0.58). A subsequent factorial designed RCT investigated the use of adrenaline, antihistamines and glucocorticosteroids given alone or in combination . This also found adrenaline, but not other interventions, to be effective in reducing the risk of anaphylaxis.
Pharmacological interventions for the prevention of anaphylaxis to iodinated contrast media
One systematic review of nine trials involving 10 011 unselected patients failed to demonstrate that premedication with glucocorticosteroids, H1-antihistamines or a combination of H1- and H2-antihistamines prevented anaphylaxis triggered by iodinated contrast media .
This comprehensive review of the international literature has found little robust evidence for the acute or long-term management of anaphylaxis. The only trial evidence uncovered for the emergency management of anaphylaxis was in relation to adrenaline, but these trials have been undertaken in patients who were at the time not experiencing anaphylaxis [22-25]. Taken together with the methodologically lower-quality evidence from case series and fatality registers, there is some evidence to support the use of adrenaline for the emergency management of anaphylaxis. The evidence points to the importance of injecting this by the intramuscular route into the anterolateral aspect of the thigh. In relation to longer-term management considerations, anaphylaxis management plans may be effective in reducing the risk of recurrence. VIT reduces the severity of reactions to subsequent stings and improves quality of life, and concerns around cost-effectiveness remain.
Acute management of anaphylaxis
Case fatality register studies have demonstrated that deaths can occur within minutes of the onset of a reaction and have highlighted how difficult it is to predict the severity of reactions . It is therefore consistently advocated by guidelines that all reactions are promptly managed by initiating a range of pharmacological and nonpharmacological interventions. This review failed to identify any high-quality studies investigating the role of nonpharmacological interventions such as the role of cardiopulmonary resuscitation and posture. In contrast, we found some evidence investigating the role of adrenaline – the main drug treatment advocated in guidelines. This evidence was derived from case series, fatality registers and a limited number of trials in people not experiencing anaphylactic reactions. Overall, this body of evidence was weak, but pointed to the importance of the early administration of adrenaline, in an appropriate dose, administered by the intramuscular route into the anterolateral aspect of the thigh. Given the considerable ethical and logistical challenges in undertaking trials of parenteral adrenaline in patients experiencing anaphylaxis, it is unlikely that stronger evidence will be forthcoming.
We found no evidence from primary studies for other potential treatments, such as fluid replacement, oxygen, glucocorticosteroids, antihistamines, methylxanthines and bronchodilators, and it is therefore not possible to offer any recommendations for the use of these treatments.
Long-term management of anaphylaxis
The long-term management of anaphylaxis centres on the need to identify triggers and co-factors, providing advice on how to minimize further reactions, and equipping individuals with the skills and equipment needed to manage further reactions . Consideration also needs to be given to ameliorate any psychological consequences of a diagnosis of anaphylaxis. Researchers have therefore thus far focused attention on the role of anaphylaxis management plans, immunomodulatory interventions and a variety of educational and psychological interventions.
The formal experimental evidence in support of anaphylaxis management plans is limited. Studies that have, however, used before–after designs and that therefore did not satisfy our inclusion criteria have found that these may result in substantial reductions in the risks of further reactions [58, 59]. Given the high risk of confounding with such study designs, this evidence must be interpreted with caution.
The single educational study included failed to show a positive effect on carriage of in-date adrenaline auto-injectors by high school children with previous anaphylaxis. A study with a before–after design found that the input of a multidisciplinary allergy clinic was effective in improving parents' knowledge of food allergy and in reducing subsequent reactions . This evidence is encouraging, but due to the high inherent risk of bias associated with such a design, these findings need to be treated with caution until more evidence from studies employing more robust study designs is forthcoming. The systematic review on psychological interventions for children with a history of anaphylaxis and their parents/carers was difficult to interpret because of its poor quality and reporting. Whilst clearly demonstrating that a number of studies have found that a diagnosis of anaphylaxis can be associated with anxiety and impaired quality of life, there was very little in the way of primary evidence demonstrating that intervening could improve outcomes in such individuals/families.
Immunomodulatory approaches are of considerable interest as these have the potential to modify the disease course and may possibly prove curative. We found a modest body of evidence in relation to VIT in relation to the management of those with stinging insect anaphylaxis, although much of this was judged to be at high risk of bias. This body of evidence did, however, consistently demonstrate that VIT can significantly reduce the severity of subsequent systemic reactions to insect stings, but given the infrequency of further stings and the low number of fatalities, it was not possible to assess whether VIT reduced the risk of fatal reactions to insect stings. There were no formal cost-effectiveness studies identified, the only potentially relevant evidence emerging from modelling studies in relation to a specific product: this found that VIT is most likely to be cost-effective in populations at high risk of further exposure (e.g. bee keepers, their family members and those who live near bee farms) or if likely benefits to quality of life are accounted for. VIT is a treatment which has been shown to reduce subsequent reactions, and although the treatment may give rise to adverse effects, it is a treatment patients prefer over the long-term carriage of an adrenaline auto-injector. In state-funded health services, however, the cost implications of such an intervention may prevent widespread availability limiting its use to high-risk patients only where its cost-effectiveness profile is likely to be much more favourable.
We found no eligible studies investigating the role of desensitization therapy in the management of those with anaphylaxis to drugs or latex. Studies investigating the effectiveness of oral and sublingual immunotherapy have mainly been undertaken in those without a history of anaphylaxis to foods; these studies are therefore reviewed in the companion systematic review on the management of food allergy.
Prophylactic approaches can also potentially play a role in those with a history of anaphylaxis. The evidence we uncovered did not, however, directly focus on this population. Rather, the approaches studied have been used in the general population and have found that prophylactic use of adrenaline can substantially reduce the risk of anaphylaxis associated with anti-snake venom, that adding antihistamines or glucocorticosteroids conferred no additional advantage and that antihistamines and glucocorticosteroids were of questionable value in preventing anaphylaxis associated with iodinated contrast media-based diagnostic investigations in unselected populations.
Implications for future research
This review has underscored the dearth of high-quality research to guide everyday clinical decision-making. There is then a pressing need to develop the evidence base for both the acute and long-term management of this potentially life-threatening disorder. In relation to acute management, it is widely accepted that it would be unethical to undertake studies investigating the effectiveness of adrenaline when compared with placebo, but trials could potentially be undertaken investigating the optimum dose, site, route and timing of administration of adrenaline. Other important questions that need to be addressed include establishing the role of H1- and H2-antihistamines and glucocorticosteroids, and these could also potentially be investigated using formal experimental designs. Similarly, there are a range of interventions delivered with the aim of improving longer-term outcomes – for example, provision of adrenaline auto-injectors, anaphylaxis management plans, immunotherapy – and these can all potentially also be studied using formal trial designs. Ideally, these studies should investigate the impact of interventions on the outcomes that have been carefully described in recent Cochrane and other systematic reviews [12, 14, 28, 29, 35]. A fuller discussion of the research agenda will be made available in the forthcoming EAACI Anaphylaxis Guidelines.
There is at present little in the way of robust evidence to guide decisions on the acute or long-term management of anaphylaxis. Key recommendations from this review have been summarized in Box 1. Given the risk of death and the considerable morbidity associated with anaphylaxis, these evidence gaps need to be filled wherever possible . These gaps include the need for educational interventions for patients, carers and healthcare professionals, lack of data on the pharmacodynamics of adrenaline and the ideal dosage in children and to some extent adults, and a lack of effective study designs on the benefits of educational plans .
Box 1. Key recommendations
There is some evidence that prompt administration of adrenaline may be life-saving; it should therefore be used as the drug of first choice in the acute management of anaphylaxis.
Adrenaline should be administered by the intramuscular route into the anterolateral aspect of the mid-thigh.
Anaphylaxis management plans may reduce the severity of subsequent reactions.
VIT may reduce the severity of subsequent reactions and improve quality of life, but cost-effectiveness considerations should be considered, particularly in those at low risk of further stings.
Adrenaline used prophylactically can reduce severe adverse effects associated with anti-snake venom administration.
We would like to acknowledge the support of the EAACI and the EAACI Food Allergy and Anaphylaxis Guidelines Group in supporting the undertaking of this systematic review. We would also like to thank the EAACI Executive Committee for their helpful comments and suggestions.
AS, AM and GR conceived this review. It was undertaken by SD with the support of SSP, GR and AS. SD, AS and GR led the drafting of the manuscript, and all authors critically commented on drafts of the manuscript.
Ethical approval was not required.
Conflicts of interest
The authors declare that they have no conflicts of interest.