Drugs Highly Associated with Infusion Reactions Reported using Two Different Data-mining Methodologies

Objective: Infusion reactions can be serious life threatening adverse events and have been associated with many drugs and biologic agents. Our objective was to report drugs associated with infusion reactions using two different data-mining methodologies. Methods: The Food and Drug Administration Adverse Event Reporting System (FAERS) was data-mined for drugs highly associated with infusion reactions. Drugs were included if there were >10 reported adverse events and if the Empirical Bayesian Geometric Mean (EBGM) score ≥ 2. Molecular Health’s MASE (Molecular Analysis of Side Effects) reports Proportional Reporting Ratios (PRR) for drugs highly associated with infusion reactions and was cross-referenced to improve detection sensitivity. Results: Using FAERS, the highest EBGM scores by class were reported as: pegloticase and α-1- antitrypsyn (enzymes), iron dextran and ferric gluconate (electrolytes and nutrients), infliximab and gemtuzumab (immunomodulators), and paclitaxel and oxaliplatin (antimetabolites). Using MASE, the highest PRR scores were reported as: idursulfase and galsulfase (enzymes), iron dextran and phytonadione (electolytes and nutrients), gemtuzumab and infliximab (immunomodulators), mercaptopurine and azathioprine (antimetabolites). Amphotericin and vancomycin had the highest scores for the antimicrobial class for both FAERS and MASE. Conclusions: Using the two statistical methods EBGM and PRR, both specificity and sensitivity were preserved. However, neither system detected several drugs with established relationships to infusion reactions, including protamine and nitroglycerine. Reactions caused by these drugs were possibly underreported because the effects have been well established or due to evolution of administration with slower administration. We hope this analysis encourages further research into overlapping mechanisms for infusion reactions