PURPOSE OF THE STUDY.
To retrospectively assess the diagnostic accuracy of the National Institute of Allergy and Infectious Disease/Food Allergy and Anaphylaxis Network Criteria (NIAID/FAAN) criteria for the diagnosis of anaphylaxis in the emergency department (ED).
ED patients (20% children overall) who were diagnosed with an allergic reaction or anaphylaxis by the emergency care provider. A subset of patients with related diagnoses was also included.
This was a retrospective cohort study of patients presenting to Saint Mary’s Hospital ED in Minnesota, through April 2008 and October 2008. Electronic medical records were reviewed and data were collected on inciting allergen, timing of symptoms onset, presenting signs and symptoms, and allergic history. Individual records were reviewed by 2 experienced board-certified allergists who were blind to the results regarding NIAID/FAAN criteria. The final consensus diagnosis by the 2 allergists was considered the reference standard for the diagnosis of anaphylaxis.
Two hundred fourteen patients participated in the study with a median age of 36 years (∼25% were <18 years old). About 40% of patients met NIAID/FAAN criteria for anaphylaxis. Emergency physicians diagnosed anaphylaxis in about 26%, of whom 27% were not considered to have anaphylaxis by the allergists. Compared to diagnosis by allergists, the sensitivity of the criteria was 96.7% (95% confidence interval [CI], 88.8%–99.1%), the specificity was 82.4% (95% CI, 75.5%–87.6%), positive predictive value was 68.6% (95% CI, 58.2%–77.4%), and negative predictive value was 98.4% (95% CI, 94.5%–99.6%). Other diagnoses made in those not meeting criteria were indeterminate reaction, NSAID drug reaction, other medication reaction, allergic reaction, asthma exacerbation, anxiety, carcinoid syndrome, post-viral syndrome, oral allergy syndrome.
This study demonstrated that by using expert diagnosis as the reference standard, the NIAID/FAAN criteria for making a diagnosis of anaphylaxis in the ED is highly sensitive but less specific.
Underdiagnosis of anaphylaxis in the ED is becoming less of a problem with more education and better-defined criteria. However, this study demonstrated that emergency physicians continued to miss over a third of cases that would have been considered anaphylaxis by an allergist. The NIAID/FAAN criteria are highly sensitive and have a high negative predictive value that might makes it useful in preventing underdiagnosis, although a substantial rate of false-positive results continue. The limitation here was that the standard reference was expert opinion, and interrater agreement among allergists was not ideal (κ = .77). Further prospective studies in other populations are needed to validate this study.
- Copyright © 2012 by the American Academy of Pediatrics