Signal mining of valproic acid-induced adverse drug events based on FAERS / 中国药房

ABSTRACT

OBJECTIVE To provide reference for clinically safe and rational drug use through mining and analyzing adverse drug event （AE） signals induced by valproic acid （VPA）. METHODS Reporting Odds Ratio （ROR） and Bayesian Confidence Propagation Neural Network （BCPNN） methods of Measures of Disproportionality were performed to mine and analyze the data of VPA-related AE reports in the US FDA Adverse Event Reporting System （FAERS） database from the first quarter of 2013 to the fourth quarter of 2022. RESULTS A total of 1 253 （ROR） and 1 109 （BCPNN） valid signals of preferred terms （PT） were obtained after data processing by the two analysis methods， involving 27 system organs （SOC）， mainly focusing on nervous system disorders， psychiatric disorders， general disorders and administration site conditions. Signals that did not appear in the instruction were associated with 2 SOCs： ear and labyrinth disorders， infections and infestations. CONCLUSIONS As a first-line broad-spectrum anti-epileptic drug， attention should also be paid to eye toxicity and infection risk in the clinical application in addition to paying attention to common adverse events in the instruction.