Clinical Deep Data Accumulation System (CLIDAS) reveals lipid paradox in guideline-defined high risk Japanese patients after PCI

Abstract Background Japanese clinical guidelines recommend a stratification of the risks and LDL-cholesterol (LDL-C) treatment goals for patients with coronary artery disease after percutaneous coronary intervention (PCI), i.e. <100 mg/dL for normal risk patients and <70 mg/dL for high risk patients; however, less is known about the association between baseline LDL-C values and long-term prognosis. Purpose To investigate the association between LDL-C goals and baseline LDL-C levels in relation to major adverse cardiovascular events (MACE) among high-risk patients after PCI, using a real-world database. Methods We developed the Clinical Deep Data Accumulation System (CLIDAS) that acquires clinical data directly from hospital information system, and implemented the system in 6 university hospitals and the national cardiovascular center in Japan. The CLIDAS database accumulates data regarding patient background, laboratory data, prescriptions, electrocardiogram, echocardiogram, PCI report, and long-term prognosis. We retrospectively analyzed 8540 consecutive patients who underwent PCI during April 2014 and March 2020 in participating hospitals, and classified them into the normal risk group (n=3712, 43%) and the high risk group [n=4828, 57%, with any of acute coronary syndrome (ACS), familial hypercholesterolemia (FH), or diabetes with additional risk factor(s)], for which LDL-C goals are <100 mg/dL and <70 mg/dL, respectively, according to the Japanese Atherosclerosis Society guidelines or the diagnosis and prevention of atherosclerotic cardiovascular diseases. The primary outcome was the time to first occurrence of MACE, a composite of cardiovascular death, stroke, myocardial infarction, and coronary revascularization in associations with baseline LDL-C levels and patient background. Results Proportion of male (77% vs. 77%) and age (71±11 vs. 70±11) were similar between 2 groups. The prevalence of ACS at the index PCI (0% vs. 62%), FH (0% vs. 2%), hypertension (61% vs. 86%), diabetes (11% vs. 67%), dyslipidemia (73% vs. 84%), hemodialysis (4% vs. 9%), peripheral artery disease (5% vs. 9%), smoking (16% vs. 30%), and prescription of statins (79% vs. 86%) were significantly higher in the high risk group. Among patients in the high risk group, but not in the normal risk group, baseline LDL-C <70 mg/dL was paradoxically associated with higher risk of MACE (P<0.0001 by Log-rank test) (Figure). The Cox proportional hazard model confirmed that the high risk group (risk ratio 1.54, 95% CI [1.31–1.81]), baseline LDL-C <70mg/dL (risk ratio 1.44, 95% CI [1.18–1.75]), baseline age (risk ratio 1.36, 95% CI [1.28–1.45] per 10 year), and prescription of statins (risk ratio 0.80, 95% CI [0.66–0.96]) were significantly associated with the risk of MACE in this population. Conclusion The CLIDAS real-world database revealed that baseline low LDL-C paradoxically associated with an increased risk of MACE among guideline-defined high risk patients after PCI. Funding Acknowledgement Type of funding sources: Private grant(s) and/or Sponsorship. Main funding source(s): Jichi Medical University, Tochigi, Japan, and Kowa