Background: Observational studies suggested that residual risk of cardiovascular events after LDL (low-density lipoprotein) cholesterol lowering may be linked to remnant cholesterol (RC). We conducted a large-scale Mendelian randomization study to investigate the causal role of RC to predict coronary artery disease (CAD), myocardial infarction (MI), and stroke risk. Methods: We extracted single-nucleotide polymorphisms for RC and LDL from large-scale genome-wide association databases. We estimated the genetic association with outcomes from the CARDIoGRAMplusC4D consortium (Coronary Artery Disease Genome-Wide Replication and Meta-Analysis Plus the Coronary Artery Disease Genetics), the Metastroke consortium, as well as the GLGC (Global Lipids Genetics Consortium). Genetic variants were used as instruments, thereby minimizing residual confounding and reverse causation biases of observational studies. Results: By leveraging data from a combined sample of 958 434 participants, we found evidence for a significant causal effect of RC on the risk of CAD (odds ratio [OR], 1.51 per SD unit increase in RC [95% CI, 1.42–1.60]; P =5.3×10 ^-5 ), MI (OR, 1.57 [95% CI, 1.21–2.05]; P =9.5×10 ^-4 ), and stroke (OR, 1.23 [95% CI, 1.12–1.35]; P =3.72×10 ^-6 ). There was no evidence of pleiotropy. The effect of RC on CAD and MI remained consistent after accounting for the effects of RC-associated genetic variants on LDL cholesterol: OR, 1.49 (95% CI, 1.37–1.61) for CAD and OR, 1.80 (95% CI, 1.70–19.1) for MI without a meaningful indirect effect exerted on these outcomes via the LDL cholesterol mediator. Conclusions: This large-scale Mendelian randomization study showed a robust genetic causal association between RC and cardiovascular outcomes. The effect on CAD and MI is independent of LDL cholesterol. Early screening for RC along with long-term inhibition of RC should be the focus of future therapeutic interventions.