Longitudinal assessment of coronary plaque regression related to sodium-glucose cotransporter-2 inhibitor using coronary computed tomography angiography
Tianhao Zhang,Xuelian Gao,Tianlong Chen,Hongkai Zhang,Xiaoming Zhang,Yu Xin,Dongmei Shi,Yu Du,Lei Xu,Yujie Zhou
DOI: https://doi.org/10.1186/s12933-024-02368-y
2024-07-22
Abstract:Background: Sodium-Glucose Cotransporter-2 Inhibitor (SGLT2i) is a novel oral drug for treating type 2 diabetes mellitus (T2DM) with demonstrated cardiovascular benefits. Previous studies in apolipoprotein E knockout mice have shown that SGLT2i is associated with attenuated progression of atherosclerosis. However, whether this effect extends to T2DM patients with coronary atherosclerosis in real-world settings remains unknown. Methods: In this longitudinal cohort study using coronary computed tomography angiography (CCTA), T2DM patients who underwent ≥ 2 CCTA examinations at our center between 2019 and 2022 were screened. Eligible patients had multiple study plaques, defined as non-obstructive stenosis at baseline and not intervened during serial CCTAs. Exclusion criteria included a CCTA time interval < 12 months, prior SGLT2i treatment, or initiation/discontinuation of SGLT2i during serial CCTAs. Plaque volume (PV) and percent atheroma volume (PAV) were measured for each study plaque using CCTA plaque analysis software. Patients and plaques were categorized based on SGLT2i therapy and compared using a 1:1 propensity score matching (PSM) analysis. Results: The study included 236 patients (mean age 60.5 ± 9.5 years; 69.1% male) with 435 study plaques (diameter stenosis ≥ 50%, 31.7%). Following SGLT2i treatment for a median duration of 14.6 (interquartile range: 13.0, 20.0) months, overall, non-calcified, and low-attenuation PV and PAV were significantly decreased, while calcified PV and PAV were increased (all p < 0.001). Meanwhile, reductions in overall PV, non-calcified PV, overall PAV, and non-calcified PAV were significantly greater in SGLT2i-treated compared to non-SGLT2i-treated plaques (all p < 0.001). PSM analysis showed that SGLT2i treatment was associated with higher reductions in overall PV (- 11.77 mm3 vs. 4.33 mm3, p = 0.005), non-calcified PV (- 16.96 mm3 vs. - 1.81 mm3, p = 0.017), overall PAV (- 2.83% vs. 3.36%, p < 0.001), and non-calcified PAV (- 4.60% vs. 0.70%, p = 0.003). These findings remained consistent when assessing annual changes in overall and compositional PV and PAV. Multivariate regression models demonstrated that SGLT2i therapy was associated with attenuated progression of overall or non-calcified PV or PAV, even after adjusting for cardiovascular risk factors, medications, and baseline overall or non-calcified PV or PAV, respectively (all p < 0.05). The effect of SGLT2i on attenuating non-calcified plaque progression was consistent across subgroups (all p for interaction > 0.05). Conclusions: In this longitudinal CCTA cohort of T2DM patients, SGLT2i therapy markedly regressed coronary overall PV and PAV, mainly result from a significant reduction in non-calcified plaque.