AIMS: Emerging evidence has linked cholesterol metabolism with platelet responsiveness. We sought to examine the dose-response relationship between low-density lipoprotein cholesterol (LDL-C) and major in-hospital bleeds in acute coronary syndrome (ACS) patients. METHODS AND RESULTS: Among 42 378 ACS patients treated with percutaneous coronary intervention (PCI) enrolled in 240 hospitals in the Improving Care for Cardiovascular Disease in China-ACS project from 2014 to 2019, a total of 615 major bleeds, 218 ischaemic events, and 337 deaths were recorded. After controlling for baseline variables, a non-linear relationship was observed for major bleeds, with the higher risk at lower LDL-C levels. No dose-response relationship was identified for ischaemic events and mortality. A threshold value of LDL-C <70 mg/dL was associated with an increased risk for major bleeds (adjusted odds ratio: 1.49; 95% confidence interval: 1.21-1.84) in multivariable-adjusted logistic regression models and in propensity score-matched cohorts. The results were consistent in multiple sensitivity analyses. Among ticagrelor-treated patients, the LDL-C threshold for increased bleeding risk was observed at <88 mg/dL, whereas for clopidogrel-treated patients, the threshold was <54 mg/dL. Across a full spectrum of LDL-C levels, the treatment effect size associated with ticagrelor vs. clopidogrel on major bleeds favoured clopidogrel at lower LDL-C levels, but no difference at higher LDL-C levels. CONCLUSIONS: In a nationwide ACS registry, a non-linear association was identified between LDL-C levels and major in-hospital bleeds following PCI, with the higher risk at lower levels. As the potential for confounding may exist, further studies are warranted. TRIAL REGISTRATION: ClinicalTrials.gov Identifier: NCT02306616.