Objective: The differential diagnosis of hypersensitivity reactions associated with anticonvulsants requires accuracy because of the many implications for patient management. We tested an integrated Bayesian and biochemical diagnostic approach.
Methods: The patients were analyzed clinically by two tests. One test, the Bayesian Adverse Reaction Diagnostic Instrument (BARDI), calculates the posterior probability of a drug being the cause based on epidemiologic and case data. The other, the lymphocyte toxicity assay, is an in vitro rechallenge that determines the percentage of cell death attributable to a drug's toxic metabolites. The setting for the study was an adverse drug reaction clinic at Sunnybrook Health Science Centre and the Hospital for Sick Children, Toronto, Ontario, Canada. Fifty-one patients who had hypersensitivity reactions after receiving aromatic anticonvulsants were tested. Four of these patients had more than one reaction reported, with different anticonvulsants generating 56 distinct events.
Results: Compared to the lymphocyte toxicity assay, BARDI had 94% sensitivity, 93% accuracy, and 50% specificity. When lymphocyte toxicity assay data were incorporated into BARDI, agreement rose from 93% to 100%. BARDI also identified which drug was a more likely cause for 11 patients receiving multiple anticonvulsants.
Conclusion: These findings show that BARDI and the lymphocyte toxicity assay have high concordance and, when used in an integrated approach, these tests can improve the diagnostic accuracy and enhance the management of patients with hypersensitivity reactions.