SGLT2 Inhibitor Adverse Reaction Reporting Relative to Prescribing Volume

Background: /Objectives: Sodium-glucose cotransporter 2 (SGLT2) inhibitors share a renal mechanism but differ in SGLT1 selectivity, off-target activity and licensed indications. We compared exposure-adjusted reporting of suspected adverse reactions for canagliflozin, dapagliflozin, empagliflozin and ertugliflozin and examined whether differences between them are consistent with molecule-specific pharmacology. Methods: UK Yellow Card reaction entries for 2021–2025 were related to English NHS general practice and estimated hospital prescribed items for the same 60 months. Rates were calculated with exact Poisson intervals, reporting rate ratios (RRRs) versus dapagliflozin with Poisson regression and Holm adjustment, and annual trends with quasi-Poisson regression; within-class disproportionality was also calculated without a prescribing denominator. Results: The analysis included 8789 reaction entries and 52.19 million items. Total reporting rates were 125.47, 188.83, 143.53 and 213.64 per million items, respectively. Ketoacidosis and ketosis reporting was higher for canagliflozin (RRR 2.45, 95% CI 2.02–2.95) and empagliflozin (1.93, 1.69–2.19), as was euglycaemic ketoacidosis reporting, and peripheral tissue injury reporting was higher for canagliflozin (2.80, 1.63–4.64). Amputation reporting was higher for canagliflozin (3.69, 1.14–10.61) but not after adjustment, and 2.68 for lower-limb terms. Within the class, the entry-level ketoacidosis reporting odds ratio for canagliflozin was 3.10 (2.53–3.80). Ketoacidosis reporting declined by 32.0% per year for dapagliflozin and by 24.4% for empagliflozin. Conclusions: Higher ketoacidosis and peripheral tissue injury reporting for canagliflozin is compatible with its lower SGLT2 selectivity, but equally with population or reporting differences, which these data cannot separate. Empagliflozin does not fit a selectivity explanation. These are reporting rates, not incidence estimates.