Scholarlink

Background <p>Screening for obstructive coronary artery disease (CAD) is crucial. Conventional cardiovascular risk models have relatively poor performance in predicting obstructive coronary artery disease. CAC score (CACS) is the strongest predictor of obstructive CAD. This study developed and validated the performance of obstructive CAD prediction model (PREDICT-OCAD) and further studied that PREDICT-OCAD can save unnecessary expense by correctly reclassifying patients as low likelihood for obstructive CAD without further testing.</p> Objective <p>To develop and validate the performance of coronary artery calcium score (CACS) based nomogram for predicting obstructive coronary artery disease (CAD).</p> Methods <p>4994 participants undergoing CACS and coronary CT angiography (CCTA) from four medical centers were retrospectively included for model development (<i>n</i> = 2649), internal (<i>n</i> = 1135) and external validation (<i>n</i> = 1210). Coronary artery stenosis on CCTA and CACS were measured. Regression models were developed and compared. A nomogram was generated to identify potential predictors associated with obstructive CAD.</p> Results <p>In development sets, CACS had incremental value for prediction of obstructive CAD compared to cardiovascular risk factors-based model, with higher area under the curves (AUCs) (0.848 [95%CI: 0.830–0.866] vs. 0.879 [95%CI: 0.857-0.900]; <i>P</i> = 0.029), net reclassification improvement (NRI) index (0.577 [95%CI: 0.482–0.672]; <i>P</i> &lt; 0.001) and integrated discrimination improvement (IDI) index (0.186 [95%CI: 0.159–0.213]; <i>P</i> &lt; 0.001). A nomogram (PREDICT-OCAD) provided significantly incremental discriminate value to CACS (AUC: 0.893 [95%CI: 0.869–0.917]; <i>P</i> = 0.027), which was confirmed by continuous NRI (0.392 [95%CI: 0.247–0.538]; <i>P</i> &lt; 0.001), categorical NRI (0.069 [95%CI: 0.001–0.137]; <i>P</i> = 0.045) and IDI (0.039 [95%CI: 0.02–0.0580]; <i>P</i> &lt; 0.001), resulted in the revision of management plan as determined by cardiovascular risk factors in 18.6% (492/2649) of patients, of whom 77.2% (380/492) were correctly reclassified.</p> Conclusion <p>Our developed simple-to-use CACS based nomogram model (PREDICT-OCAD) performs better than conventional clinical model or CACS alone, which is a useful tool for risk stratification of CAD in primary health system.</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

Development and validation of a simple-to-use coronary calcium score based nomograph for predicting obstructive coronary artery disease

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年4月9日

查看全文
  • Rui Zuo,
  • Tongyuan Liu,
  • Fan Zhou,
  • Linfeng Zhai,
  • Wei Xu,
  • Jiani Zou,
  • Junhao Li,
  • Changsheng Zhou,
  • Ya Liu,
  • Chun Xiang Tang,
  • Bin Hu,
  • Liang Li,
  • Xiangwei Luo,
  • Jingjing Pan,
  • Zijian Chen,
  • Yu Zhang,
  • Long Jiang Zhang

摘要

Background

Screening for obstructive coronary artery disease (CAD) is crucial. Conventional cardiovascular risk models have relatively poor performance in predicting obstructive coronary artery disease. CAC score (CACS) is the strongest predictor of obstructive CAD. This study developed and validated the performance of obstructive CAD prediction model (PREDICT-OCAD) and further studied that PREDICT-OCAD can save unnecessary expense by correctly reclassifying patients as low likelihood for obstructive CAD without further testing.

Objective

To develop and validate the performance of coronary artery calcium score (CACS) based nomogram for predicting obstructive coronary artery disease (CAD).

Methods

4994 participants undergoing CACS and coronary CT angiography (CCTA) from four medical centers were retrospectively included for model development (n = 2649), internal (n = 1135) and external validation (n = 1210). Coronary artery stenosis on CCTA and CACS were measured. Regression models were developed and compared. A nomogram was generated to identify potential predictors associated with obstructive CAD.

Results

In development sets, CACS had incremental value for prediction of obstructive CAD compared to cardiovascular risk factors-based model, with higher area under the curves (AUCs) (0.848 [95%CI: 0.830–0.866] vs. 0.879 [95%CI: 0.857-0.900]; P = 0.029), net reclassification improvement (NRI) index (0.577 [95%CI: 0.482–0.672]; P < 0.001) and integrated discrimination improvement (IDI) index (0.186 [95%CI: 0.159–0.213]; P < 0.001). A nomogram (PREDICT-OCAD) provided significantly incremental discriminate value to CACS (AUC: 0.893 [95%CI: 0.869–0.917]; P = 0.027), which was confirmed by continuous NRI (0.392 [95%CI: 0.247–0.538]; P < 0.001), categorical NRI (0.069 [95%CI: 0.001–0.137]; P = 0.045) and IDI (0.039 [95%CI: 0.02–0.0580]; P < 0.001), resulted in the revision of management plan as determined by cardiovascular risk factors in 18.6% (492/2649) of patients, of whom 77.2% (380/492) were correctly reclassified.

Conclusion

Our developed simple-to-use CACS based nomogram model (PREDICT-OCAD) performs better than conventional clinical model or CACS alone, which is a useful tool for risk stratification of CAD in primary health system.