ESC Congress 2026 Abstract · Oct 26, 2026

The newest pillar is the most fragile: SGLT2 inhibitors show the lowest 12-month ongoing use among guideline-directed CAD medications in a 15-country European survey

By Tjon L, Gallego Llorente M, Rekker LY et al.

Background: The 2024 ESC chronic coronary syndromes guidelines assign SGLT2 inhibitors (SGLT2i) a Class I, Level A recommendation for patients with type 2 diabetes, yet SGLT2i ongoing use in coronary artery disease (CAD) has not been benchmarked against other guideline-directed medical therapies (GDMT) within the same multinational cohort. Purpose: To quantify SGLT2i ongoing-use trajectories in CAD and rank SGLT2i against other GDMT at standardised timepoints across European centres. Methods: Cross-sectional web-based survey of cardiologists across 15 European countries (CAD module: n=30). Respondents estimated the proportion of eligible patients remaining on each of eight drug classes at 3, 6, and 12 months post-index event ("estimated ongoing use"), capturing both initiation and persistence. Six GDMT classes were compared: aspirin, P2Y12 inhibitors, statins, ACEi/ARB, beta-blockers, and SGLT2i. Within-class change from 3 to 12 months was tested with paired comparisons; between-class differences at 12 months with non-parametric tests. From a separate heart failure module (n=32), SGLT2i initiation intent was extracted to distinguish prescribing gaps from persistence gaps. Results: At 12 months, SGLT2i had the lowest estimated ongoing use among all six GDMT classes (55.2%; 95% CI 46.1–64.3), followed by beta-blockers (55.9%), ACEi/ARB (67.4%), P2Y12 inhibitors (69.2%), statins (75.3%), and aspirin (85.5%). SGLT2i was already among the lowest at 3 months (65.5%, comparable to beta-blockers at 65.8%) and declined by 7.5 percentage points over the following nine months. This decline parallels the attrition observed for established therapies including aspirin (−9.7 pp), suggesting that the low 12-month level is driven by a lower starting point rather than faster attrition. At 12 months, SGLT2i was 30.3 percentage points below aspirin (p<0.0001) and 20.1 points below statins. 40% of centres reported <50% ongoing use at 12 months. Country-level means ranged from 32% to 74%, reflecting substantial cross-national variability. In the heart failure module, 91% of respondents reported initiating SGLT2i always or usually when indicated, suggesting that persistence gaps, not prescribing intent, drive the shortfall. Conclusion: Among 6 guideline-directed CAD therapies surveyed across 15 European countries, SGLT2 inhibitors show the lowest estimated 12-month ongoing use despite Class I ESC recommendations. The rate of adherence attrition is comparable to established therapies; so the problem is not the slope, it is the starting line. A 30-point gap versus aspirin and a 12-month floor of 55.2% suggest that prescribing inertia and suboptimal initiation at discharge, rather than differential patient dropout, are the primary drivers of value loss. These findings support prioritising discharge-level interventions (systematic initiation, titration protocols, and structured follow-up plans) over strategies focused solely on long-term persistence.Figure 1:Ongoing use at 3 and 12 months

doi.org/10.1093/eurheartj/ehag785.3333Open accessPopulations