Data Mining and Analysis for Iodinated Contrast Media Adverse Event Signals Based on the Food and Drug Administration Adverse Event Reporting System Database.

Background: The purpose of this study was to employ the US Food and Drug Administration (FDA) Adverse Event Reporting System (FAERS) database to mine and analyze adverse events related to iodinated contrast media (ICM), explore the characteristics of adverse events (AEs) including their occurrence and correlation strength between AEs and drugs, and to provide valuable insights for clinical use.

Methods: The FAERS database was queried, data from Q1 of 2004 to Q2 of 2023 were extracted, and AE reports targeting 5 ICMs as the primary suspects were collected. Data mining and analysis were carried out on relevant reports using the reporting odds ratio (ROR), proportional reporting ratio (PRR), Bayesian confidence propagation neural network (BCPNN), and empirical Bayes geometric mean (EBGM), while the standardized medical dictionary for regulatory activities (MedDRA) queries (SMQ) was used for systematic classification.

Results: A total of 11,155,106 AE reports were retrieved from FAERS, with 2,412 for ioversol, 2,001 for iohexol, 987 for iodixanol, 1,154 for iopamidol, and 3,835 for iopromide. ICM-induced AE occurrence targeted 21 system organ classes (SOCs). A total of 329 significant disproportionality Preferred terms (PTs) conforming to the 4 algorithms were simultaneously retained. The results revealed that the medium and strong adverse drug reaction (ADR) signals of the 5 ICMs largely focused on "respiratory, thoracic and mediastinal disorders," "general disorders and administration site conditions," "immune system disorders," and "skin and subcutaneous tissue disorders." Ioversol (logROR = 1.21, P = 0.034) and iopromide (logROR = 1.32, P = 0.004) were both correlated with a higher incidence of a significant ADR signal, namely throat irritation, particularly in females. In addition, ioversol and iopromide also suggested that toxic nephropathy (logROR = -2.47, P < 0.001) and hyperhidrosis (logROR = -1.22, P = 0.001) were significant ADR signals, especially in males, respectively.

Conclusions: While the AE distribution of the 5 ICMs was consistent, there were variations in specific ADR signal characteristics, warranting further consideration and exploration.