This study is part of an ongoing projet to create, derive and validate a clinical decision rule with regards to the stratification of the biphasic reaction risk following an anaphylaxis event that has mandated an emergency room visit.
Phase 0: Systematic literature review (completed)
Phase 1: Statistical derivation of two clinical decision rules (retrospective observational study) - (completed)
Phase 2: Validation of clinical decision rules (prospective observational study - theoretical) - (Prematurely terminated)
Objective: The objective of phase 2 was to test the applicability and performance of our clinical decision rules in a prospective observational manner.
UPDATE (March 2023) This unfortunately failed, the rule showed underperformance and clinical utility wasn't attained / maintained in the derivation phase. Furthermore, we tried combining data for the derivation study (phase 1) and the validation study (phase 2) and found that since the phenomenon of biphasic reaction was so heterogeneous, we would need to recruit an unrealistic number of anaphylaxis cases. This endeavor for which we have nor the time, nor the ressources and nor the organizational capabilities would also statistically lead to a clinical decision rule still subject to failure in its further external validations because of the very heterogeneous nature of allergens and the subjects immune reactions to them (projected frailty index).
Methods: Based on the three clinical decision rules (2 derived from phase 1and one based on the literature review), a prospective observational study was carried out in the emergency rooms of Chicoutimi, Jonquière and Alma. Data collection for this theoretical validation phase has been underway since December 2019 and stopped in march 2023, having received the approval of the various local research committees (ethics, science and convenience).
A questionnaire was developed from the three rules derived in the previous phases. All the elements present in these rules have been grouped by categories (history, allergen, symptoms and treatment). In this way, the criteria for each rule were not placed one after the other so that no rule could be recognized or inferred. This questionnaire was presented during department meetings. It is placed on the charts of patients with suspected anaphylaxis at triage and also is available at strategic locations in all three emergency rooms (for example, in the reanimation room).
Doctors were therefore invited to complete the questionnaire anonymously following their initial management of an anaphylactic reaction. Thereafter, they treat the patient as they would have done according to their usual practice. No rules was applied to the patients and the answers provided had no impact on their future care. The questionnaires were then deposited in a secured location determined for this purpose, in each of the emergencies.
The collection of files began in December 2019 and is now terminated. The inclusion and exclusion criteria were the same as in phase 1. However, a patient who did not yet meet the criteria for anaphylaxis, but who, according to the clinician's assessment, would inevitably progress to an anaphylactic reaction could also be included if treated in this way. This was then clearly indicated on the questionnaire.
At the end of the preliminary recruitment period, all completed questionnaire files will be reviewed to determine if a biphasic reaction has occurred. The three clinical decision rules will then be applied to the population to test their performance, validity and safety. The planned interim analysis which led to preliminary termination of the study included 191 cases of anaphylaxis and 11 biphasic reactions (only 6 of which were deemed clinically significant as per protocol criteria).
Some number of anaphylaxis cases were included in the study in between the preliminary analysis (march 2022) and its the results and decision to terminate the study (march 2023). This remaining will of course be analyzed and we propose to publish our data sets in hopes it can be used by other researchers in the field.
Lastly, the investigators were proposing to run a phase 3 trial, if the phase 2 trial had identified valid and clinically usable clinical decision rule(s). That phase 3 trial would have aimed to prospectively validate in real clinical setting and in a different patient population subset the selected rule(s). This phase 3 will unfortunately not be taking place as the trial was prematurely terminated during planned interim analysis in its phase 2 due to clinical uselessness of all 3 rules being tested. Of note, investigators tried to combine data sets from both phases 1 and 2 and still could not derive a rule for which statistical solidity could be expected in the face of such an heterogeneous process as the biphasic allergic reaction is.
