Statin therapy for primary prevention of microvascular complications in patients with heart failure and type 2 diabetes: a target trial emulation
Q W Ren, M A Mastumoto, Q Y I Qi, W C X Wang, Y Q H Yiu, Z J N Zhang, G R GuoAbstract
Background
Patients with heart failure (HF) combined with type 2 diabetes (DM) are at high risk for microvascular complications (MVCs). Although statins are routinely used for cardiovascular prevention, their primary preventive effects and pleiotropic mechanisms on diabetic MVCs in this population remain unclear.
Objective
To evaluate the effectiveness of statin therapy for the primary prevention of incident diabetic MVCs in patients with HF and T2DM.
Methods
We will conduct a target trial emulation (TTE) study using the territory-wide electronic health record (EHR) from the Clinical Data Analysis and Reporting System (CDARS). The cohort will include adults with a new diagnosis of both HF and T2DM from 2010 to 2024, without pre-existing MVCs. Exposure is defined as statin initiation within one year of the dual diagnosis versus no initiation. The primary composite outcome is time to first MVC, including sight-threatening retinopathy, diabetic kidney disease, or symptomatic neuropathy. We will use inverse probability of treatment weighting to adjust for confounders and estimate hazard ratios.
Results
We identified 85,678 patients with newly diagnosed coexisting HF and T2DM and without baseline microvascular complications from CDARS. Of these, 46,695 (54.5%) initiated statin therapy within one year and 38,983 (45.5%) did not or delayed initiation. Over a median follow-up of 4.2 years, 5,603 and 6,812 composite microvascular events occurred in the statin and non-statin groups, respectively. After inverse probability of treatment weighting, compared to the statin group, the non-statin group had a significantly higher risk of the primary outcome (adjusted hazard ratio [aHR] 1.42, 95% confidence interval [CI] 1.31–1.54; P < 0.001), corresponding to an approximately 30% risk reduction associated with early statin initiation (HR 0.70). The 5-year cumulative incidence was 13.8% in the statin group versus 19.5% in the non-statin group (absolute risk difference -5.7%, 95% CI -6.9% to -4.5%). A significant dose-response relationship was observed among statin users: compared to use <1 year, the adjusted HRs were 0.92 (95% CI 0.84–1.01) for 1–3 years, 0.81 (95% CI 0.73–0.90) for 3–5 years, and 0.68 (95% CI 0.60–0.77) for ≥5 years (P for trend <0.001). The protective association was consistent across subgroups of age, sex, and HF phenotype. Sensitivity analyses, including 1:1 propensity score matching (HR 0.72, 95% CI 0.66–0.78) and a negative control outcome analysis (HR for acute upper respiratory infection 1.02, 95% CI 0.96–1.08), supported the robustness of the primary findings.
Conclusion
We found that early statin use is associated with a reduced risk of MVCs in this high-risk population. This study will provide crucial real-world evidence on the potential pleiotropic benefit of statins, informing a more comprehensive management strategy for the growing population with cardiometabolic multimorbidity.