WANG Qi-ping, MIAO Hui-qing, ZHONG Guo-rui, ZHANG Jin-bo, ZHU Yi-xi, LIU Peng-cheng, HUANG Ling
Objective: This study aims to conduct safety surveillance of infusion pump adverse events based on the U.S. Food and Drug Administration (FDA) Manufacturer and User Facility Device Experience (MAUDE) database and identify potential safety signals associated with infusion pumps. Methods: Adverse event reports related to infusion pumps recorded in the FDA MAUDE database between 2019 and 2023 were analyzed. Five signal detection methods were applied, including the Reporting Odds Ratio (ROR), Proportional Reporting Ratio (PRR), Bayesian Confidence Propagation Neural Network (BCPNN), Yule's Q, and the Poisson Probability Method (PPM). Adverse event terms were standardized using the Medical Dictionary for Regulatory Activities (MedDRA) and the terminology of the International Medical Device Regulators Forum (IMDRF). Results: A total of 2,235,079 device problem reports and 2,003,247 patient problem reports were included. Multiple risk signals related to infusion pump device failures were identified, particularly those involving material integrity and infusion or flow issues. Additionally, infusion pump device failures were found to be associated with patient harm signals, such as hyperglycemia, hypoglycemia, and diabetic ketoacidosis. Conclusion: This study identified the characteristics of patients using infusion pumps and detected potential safety signals associated with infusion pump use, including signals not described in the product labeling. However, due to the inherent limitations of the MAUDE database, the detected signals cannot establish causal relationships, and further studies are required to validate these findings.
