Scholarlink

Background <p>Stroke is a leading public health burden in China, highlighting the need for accessible biomarkers to support risk stratification. The C-reactive protein/high-density lipoprotein cholesterol ratio (CHR), integrating inflammatory and lipid pathways, has emerged as a novel composite biomarker, yet its association with stroke across different clinical subgroups, particularly in the Chinese population with varying glycemic status, remains inadequately explored.</p> Methods <p>The analysis was conducted using the longitudinal observation dataset (five waves) of the China Health and Retirement Longitudinal Study (CHARLS), collected from 2011 to 2020. It included 10,484 community-dwelling adults aged ≥ 45 years. Incident stroke risk was modeled with Cox regression. Receiver operating characteristic (ROC) assessed predictive performance, and Restricted Cubic Splines (RCS) examined non-linearity.</p> Results <p>The cumulative incidence of stroke over the 9-year observation period was 9.19%, accounting for a total of 963 cases. In fully adjusted analyses, CHR per IQR increase was weakly associated with higher stroke risk in prediabetes (HR 1.06, 95% CI 1.00–1.13) and diabetes, but not in normal glucose regulation (NGR). Compared with Q1, Q4 showed higher risk in NGR (HR 1.94, 95%CI 1.36–2.77) and in prediabetes (HR 1.40, 95%CI 1.02–1.93); in NGR, a stronger association was observed in those aged &lt; 60 years (Q4 vs. Q1: HR 2.77, 95% CI 1.61–4.76). RCS suggested non-linear associations, with a pattern consistent with a J-shaped curve in NGR and weaker nonlinearity in prediabetes. Sensitivity analyses yielded broadly consistent patterns, particularly for the elevated risk in the highest CHR quartile in NGR. Discrimination was low for CHR in predicting stroke based on ROC analysis (AUC 0.572; 95% CI 0.533–0.598).</p> Conclusions <p>Higher CHR was associated with an increased risk of stroke in Chinese adults, with heterogeneity across glycemic status, showing stronger and nonlinear (J-shaped) associations in NGR than in diabetes. CHR demonstrated limited discriminative ability for stroke, suggesting its utility may lie in risk stratification rather than standalone prediction.</p> Graphical Abstract <p></p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Association of C-reactive protein/high-density lipoprotein cholesterol ratio with stroke in Chinese adults across different glycemic status

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年2月11日

查看全文
  • Wei Huang,
  • Jun Gu,
  • Xueli Sun,
  • Zhimei Shen,
  • Qin Yin,
  • Xu Wang

摘要

Background

Stroke is a leading public health burden in China, highlighting the need for accessible biomarkers to support risk stratification. The C-reactive protein/high-density lipoprotein cholesterol ratio (CHR), integrating inflammatory and lipid pathways, has emerged as a novel composite biomarker, yet its association with stroke across different clinical subgroups, particularly in the Chinese population with varying glycemic status, remains inadequately explored.

Methods

The analysis was conducted using the longitudinal observation dataset (five waves) of the China Health and Retirement Longitudinal Study (CHARLS), collected from 2011 to 2020. It included 10,484 community-dwelling adults aged ≥ 45 years. Incident stroke risk was modeled with Cox regression. Receiver operating characteristic (ROC) assessed predictive performance, and Restricted Cubic Splines (RCS) examined non-linearity.

Results

The cumulative incidence of stroke over the 9-year observation period was 9.19%, accounting for a total of 963 cases. In fully adjusted analyses, CHR per IQR increase was weakly associated with higher stroke risk in prediabetes (HR 1.06, 95% CI 1.00–1.13) and diabetes, but not in normal glucose regulation (NGR). Compared with Q1, Q4 showed higher risk in NGR (HR 1.94, 95%CI 1.36–2.77) and in prediabetes (HR 1.40, 95%CI 1.02–1.93); in NGR, a stronger association was observed in those aged < 60 years (Q4 vs. Q1: HR 2.77, 95% CI 1.61–4.76). RCS suggested non-linear associations, with a pattern consistent with a J-shaped curve in NGR and weaker nonlinearity in prediabetes. Sensitivity analyses yielded broadly consistent patterns, particularly for the elevated risk in the highest CHR quartile in NGR. Discrimination was low for CHR in predicting stroke based on ROC analysis (AUC 0.572; 95% CI 0.533–0.598).

Conclusions

Higher CHR was associated with an increased risk of stroke in Chinese adults, with heterogeneity across glycemic status, showing stronger and nonlinear (J-shaped) associations in NGR than in diabetes. CHR demonstrated limited discriminative ability for stroke, suggesting its utility may lie in risk stratification rather than standalone prediction.

Graphical Abstract