Diagnosing, managing and preventing anaphylaxis: Systematic review

Background This systematic review used the GRADE approach to compile evidence to inform the European Academy of Allergy and Clinical Immunology's (EAACI) anaphylaxis guideline. Methods We searched five bibliographic databases from 1946 to 20 April 2020 for studies about the diagnosis, management and prevention of anaphylaxis. We included 50 studies with 18 449 participants: 29 randomized controlled trials, seven controlled clinical trials, seven consecutive case series and seven case-control studies. Findings were summarized narratively because studies were too heterogeneous to conduct meta-analysis. Results It is unclear whether the NIAID/FAAN criteria or Brighton case definition are valid for immediately diagnosing anaphylaxis due to the very low certainty of evidence. There was also insufficient evidence about the impact of most anaphylaxis management and prevention strategies. Adrenaline is regularly used for first-line emergency management of anaphylaxis but little robust research has assessed its effectiveness. Newer models of adrenaline autoinjectors may slightly increase the proportion of people correctly using the devices and reduce time to administration. Face-to-face training for laypeople may slightly improve anaphylaxis knowledge and competence in using autoinjectors. We searched for but found little or no comparative effectiveness evidence about strategies such as fluid replacement, oxygen, glucocorticosteroids, methylxanthines, bronchodilators, management plans, food labels, drug labels and similar. Conclusions Anaphylaxis is a potentially life-threatening condition but, due to practical and ethical challenges, there is a paucity of robust evidence about how to diagnose and manage it.

Automated summary

This systematic review found limited and uncertain evidence on the diagnosis, management, and prevention strategies of anaphylaxis. Adrenaline is commonly used for emergency management, but robust research on its effectiveness is lacking.