Predictive value of systemic inflammation response index for atherosclerotic cardiovascular disease risk in patients with hypercholesterolemia: a machine learning study with dual-cohort validation

Chen et al. Lipids in Health and Disease (2025) 24:350 https://doi.org/10.1186/s12944-025-02765-6 Lipids in Health and Disease Open Access RESEARCH Predictive value of systemic inflammation response index for atherosclerotic cardiovascular disease risk in patients with hypercholesterolemia: a machine learning study with dual-cohort validation Yu Chen1,2, Weikang Huang2, Shihan Zhao2, Zhuoqi Ge1, Yan Liu1, Ruibing Huang1, Dongmei Li1, Qing Xu2, Xingzhen Long3, Kai Wei2, Qi Chen2, Changcheng Sheng2, Cailin Tang2* and Xue Bai2* Abstract Background Residual cardiovascular risk persists in patients with hypercholesterolemia despite lipid-lowering therapy, underscoring the importance of inflammation in ASCVD development. This study evaluated the relationship between Systemic Inflammation Response Index (SIRI) (a composite biomarker derived from neutrophil, monocyte, and lymphocyte counts) and ASCVD in patients with hypercholesterolemia. And to develop an interpretable machine learning (ML) model for predicting ASCVD risk in patients with hypercholesterolemia. Methods This study utilized data from the National Health and Nutrition Examination Survey (2001–2018), including a total of 6,645 patients with hypercholesterolemia. Additionally, a independent external cohort of 357 patients from Guizhou Provincial People’s Hospital served as the validation cohort. Then, we used ML to analyze the effect of SIRI on ASCVD in patients with hypercholesterolemia and established four risk prediction models. Area under the receiver operating characteristic curve (AUC) was used to evaluate model performance. Shapley Additive Explanations (SHAP) were applied for model interpretation, and a web-based application was developed for clinical use. Results Our findings indicated that SIRI is consistently associated with ASCVD in hypercholesterolemia patients in both cohorts. SIRI and other seven features were used to construct ML models. The XGBoost model achieved an AUC of 0.8001 in the internal validation cohort and 0.7030 in the external cohort. The model retained strong clinical relevance. SHAP analysis highlighted elevated SIRI levels as an important predictor of ASCVD risk in patients with hypercholesterolemia. The inclusion of novel inflammatory markers such as SIRI enhanced the model’s discriminative capability. *Correspondence: Cailin Tang Xue Bai Full list of author information is available at the end of the article © The Author(s) 2025. Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creati vecommons.org/licenses/by-nc-nd/4.0/. Chen et al. Lipids in Health and Disease (2025) 24:350 Page 2 of 14 Conclusions The study findings revealed that high SIRI levels were an independent risk factor for ASCVD in patients with hypercholesterolemia. In addition, this study constructed the first interpretable ML model combined with SIRI for ASCVD prediction in hypercholesterolemia patients. The model demonstrated acceptable performance and moderate generalizability. While its external specificity was limited, the model may still serve as a useful risk stratification aid to support early identification of high-risk individuals with hypercholesterolemia. Keywords Systemic inflammation response index, Hypercholesterolemia, Atherosclerotic cardiovascular disease, Machine learning, Risk prediction Introduction Atherosclerotic cardiovascular disease (ASCVD) is a major cause of death and illness worldwide [1, 2]. Hypercholesterolemia, particularly characterized by elevated low-density lipoprotein cholesterol (LDL-C), is a wellestablished factor in atherosclerosis formation and longterm cardiovascular risk [3–5]. While pharmacological interventions such as statin therapy significantly reduce LDL-C levels, a significant proportion of cardiovascular events, commonly referred to as residual cardiovascular risk, continue to occur despite optimal lipid control [6, 7]. Like lipids, inflammation is essential for the development and progression of atherosclerosis [8]. Recent studies have shown that Systemic Inflammation Response Index (SIRI) as a marker of inflammation may contribute to this residual risk [9, 10]. The SIRI integrates neutrophil, monocyte, and lymphocyte counts to reflect systemic immune activation. Compared with traditional markers, SIRI may provide a broader measure of inflammatory burden and immune dysregulation relevant to ASCVD [11]. Although prior studies have linked SIRI to cardiovascular outcomes in various populations, its predictive value in ASCVD patients with hypercholesterolemia remains insufficiently characterized. Therefore, early diagnosis and treatment are essential to reduce the risk of ASCVD in hypercholesterolemia. Recently, advances in Machine learning (ML) and big data have gained attention in medicine, especially for disease prediction and risk assessment [12–14]. The ML models can identify complex patterns that traditional models may miss, these intricate patterns hold significant potential for enhancing the accuracy and reliability of disease prediction. However, there are few studies that have evaluated the integration of SIRI with clinical and lipid parameters using ML approaches to predict ASCVD risk specifically in patients with hypercholesterolemia. Therefore, in this study, we comprehensively evaluated the impact of SIRI on ASCVD in patients with hypercholesterolemia. In addition, we used SIRI to construct ML models in combination with other clinical features to predict ASCVD in such patients. The model integrates Shapley Additive Explanations (SHAP) to provide global and personalized insights and is deployed through a webbased application to enhance clinical utility. Materials and methods Data source and study population This retrospective study utilized data from the National Health and Nutrition Examination Survey (NHANES), a program conducted by the Centers for Disease Control and Prevention to evaluate the health and nutritional status of adults and children in the United States. The NHANES dataset has been approved by the Research Ethics R (...truncated)