Drug-Associated Ketoacidosis: A Comprehensive Disproportionality Analysis Based on the FAERS Database.

OBJECTIVE: Ketoacidosis is a critical adverse event requiring close clinical monitoring. Based on the Adverse Event Reporting System (FAERS) database of the U.S. Food and Drug Administration (FDA), this study aimed to identify the drugs with the closest association and the highest signal strength with the risk of drug-induced ketoacidosis by performing disproportionality analysis, thereby providing a reference for clinical medication safety. METHODS: Adverse event data from the first quarter of 2004 to the second quarter of 2025 were retrieved from the FAERS database. Preferred Terms (PTs) related to 'ketoacidosis' in the Medical Dictionary for Regulatory Activities (MedDRA) were used as outcome indicators. The generic names of drugs were standardized with reference to the DrugBank database. Disproportionality analysis was conducted using the Reporting Odds Ratio (ROR), Proportional Reporting Ratio (PRR) and Information Component (IC) of Bayesian Confidence Propagation Neural Network and the time trend of drug-induced ketoacidosis was analysed by combining the Time-to-Onset (TTO) analysis. RESULTS: A total of 22 375 298 adverse event reports were included in this study, among which 6977 were related to ketoacidosis. The top 5 drugs with the highest number of reports were metformin, quetiapine, empagliflozin, canagliflozin and dapagliflozin in sequence; whereas the top 5 drugs with the highest ROR values were dapagliflozin, dapagliflozin-saxagliptin, ipragliflozin, ertugliflozin-sitagliptin and empagliflozin. Among the top 50 drugs ranked by ROR, 36 (72%) had no mention of ketoacidosis risk in their package inserts. TTO analysis showed that the incidence frequency of ketoacidosis induced by quetiapine, olanzapine, aripiprazole, pembrolizumab, empagliflozin/linagliptin and nivolumab remained stable with the duration of medication; the incidence frequency induced by tirzepatide and alpelisib increased gradually over time; while that induced by empagliflozin, canagliflozin and other drugs decreased gradually with time. CONCLUSION: The high-risk drugs for inducing ketoacidosis are mainly sodium-glucose cotransporter 2 (SGLT2) inhibitors (e.g., dapagliflozin and empagliflozin), followed by psychotropic drugs (e.g., quetiapine). Attention should also be paid to the potential risk of ketoacidosis induced by some drugs that do not mention this adverse reaction in their package inserts. The findings of this study can provide data support for clinical medication monitoring and the revision of drug package inserts.