Predicting cardiovascular disease and all-cause mortality using the lymphocyte-to-monocyte ratio: Insights from explainable machine learning models
Background: Cardiovascular disease (CVD) is a leading cause of death globally, with its incidence and mortality rates continuing to rise. While commonly used biomarkers such as low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), and blood glucose are widely app...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-03-01
|
| Series: | International Journal of Cardiology. Cardiovascular Risk and Prevention |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2772487525000108 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1823861082203095040 |
|---|---|
| author | Jichao Wu Die Huang Jiefang Li Jingxing Yi Yu Lei Jun Yin |
| author_facet | Jichao Wu Die Huang Jiefang Li Jingxing Yi Yu Lei Jun Yin |
| author_sort | Jichao Wu |
| collection | DOAJ |
| description | Background: Cardiovascular disease (CVD) is a leading cause of death globally, with its incidence and mortality rates continuing to rise. While commonly used biomarkers such as low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), and blood glucose are widely applied, they have certain limitations. This study investigates the lymphocyte-to-monocyte ratio (LMR), a simple immune biomarker associated with inflammation, to assess whether it can serve as a new marker for predicting chronic inflammation in cardiovascular disease, and compares it to traditional biomarkers. Methods: We conducted a cross-sectional analysis of data from the National Health and Nutrition Examination Survey (NHANES) from 2007 to 2018, utilizing a cohort of 1518 participants with a median follow-up period of 150 months. During this time, 522 participants died, including 166 from cardiovascular disease. We employed various statistical methods, including weighted Cox proportional hazards models, restricted cubic spline models, and time-varying receiver operating characteristic curves, to examine the association between LMR and mortality risk. Results: The analysis revealed an L-shaped relationship between LMR and the incidence of cardiovascular disease. Lower LMR levels were negatively correlated with all-cause and cardiovascular mortality. The XGBoost model yielded the best performance metrics (AUC and F1 scores), and SHAP value analysis indicated that LMR significantly contributes to CVD outcomes. Non-linear analyses confirmed a stable negative correlation between LMR and all-cause mortality. Conclusion: The study concludes that LMR is a simple and practical indicator for predicting cardiovascular disease and its mortality. Low levels of LMR significantly increase the risk of both cardiovascular disease and all-cause mortality in patients. |
| format | Article |
| id | doaj-art-0f28f0dddf0b4c45a49fdb4cbdaaf0ea |
| institution | Kabale University |
| issn | 2772-4875 |
| language | English |
| publishDate | 2025-03-01 |
| publisher | Elsevier |
| record_format | Article |
| series | International Journal of Cardiology. Cardiovascular Risk and Prevention |
| spelling | doaj-art-0f28f0dddf0b4c45a49fdb4cbdaaf0ea2025-02-10T04:35:30ZengElsevierInternational Journal of Cardiology. Cardiovascular Risk and Prevention2772-48752025-03-0124200372Predicting cardiovascular disease and all-cause mortality using the lymphocyte-to-monocyte ratio: Insights from explainable machine learning modelsJichao Wu0Die Huang1Jiefang Li2Jingxing Yi3Yu Lei4Jun Yin5Department of Laboratory, Second Affiliated Hospital of Shantou University Medical College, 69 Dongxia North Road, Shantou, 515041, ChinaDepartment of Laboratory, Second Affiliated Hospital of Shantou University Medical College, 69 Dongxia North Road, Shantou, 515041, ChinaDepartment of Laboratory, Second Affiliated Hospital of Shantou University Medical College, 69 Dongxia North Road, Shantou, 515041, ChinaDepartment of Laboratory, Second Affiliated Hospital of Shantou University Medical College, 69 Dongxia North Road, Shantou, 515041, ChinaDepartment of Hematology, Second Affiliated Hospital of Shantou University Medical College, 69 Dongxia North Road, Shantou, 515041, ChinaDepartment of Laboratory, Second Affiliated Hospital of Shantou University Medical College, 69 Dongxia North Road, Shantou, 515041, China; Corresponding author.Background: Cardiovascular disease (CVD) is a leading cause of death globally, with its incidence and mortality rates continuing to rise. While commonly used biomarkers such as low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), and blood glucose are widely applied, they have certain limitations. This study investigates the lymphocyte-to-monocyte ratio (LMR), a simple immune biomarker associated with inflammation, to assess whether it can serve as a new marker for predicting chronic inflammation in cardiovascular disease, and compares it to traditional biomarkers. Methods: We conducted a cross-sectional analysis of data from the National Health and Nutrition Examination Survey (NHANES) from 2007 to 2018, utilizing a cohort of 1518 participants with a median follow-up period of 150 months. During this time, 522 participants died, including 166 from cardiovascular disease. We employed various statistical methods, including weighted Cox proportional hazards models, restricted cubic spline models, and time-varying receiver operating characteristic curves, to examine the association between LMR and mortality risk. Results: The analysis revealed an L-shaped relationship between LMR and the incidence of cardiovascular disease. Lower LMR levels were negatively correlated with all-cause and cardiovascular mortality. The XGBoost model yielded the best performance metrics (AUC and F1 scores), and SHAP value analysis indicated that LMR significantly contributes to CVD outcomes. Non-linear analyses confirmed a stable negative correlation between LMR and all-cause mortality. Conclusion: The study concludes that LMR is a simple and practical indicator for predicting cardiovascular disease and its mortality. Low levels of LMR significantly increase the risk of both cardiovascular disease and all-cause mortality in patients.http://www.sciencedirect.com/science/article/pii/S2772487525000108All-cause mortalityCardiovascular diseaseLymphocyte-to-monocyte ratio |
| spellingShingle | Jichao Wu Die Huang Jiefang Li Jingxing Yi Yu Lei Jun Yin Predicting cardiovascular disease and all-cause mortality using the lymphocyte-to-monocyte ratio: Insights from explainable machine learning models International Journal of Cardiology. Cardiovascular Risk and Prevention All-cause mortality Cardiovascular disease Lymphocyte-to-monocyte ratio |
| title | Predicting cardiovascular disease and all-cause mortality using the lymphocyte-to-monocyte ratio: Insights from explainable machine learning models |
| title_full | Predicting cardiovascular disease and all-cause mortality using the lymphocyte-to-monocyte ratio: Insights from explainable machine learning models |
| title_fullStr | Predicting cardiovascular disease and all-cause mortality using the lymphocyte-to-monocyte ratio: Insights from explainable machine learning models |
| title_full_unstemmed | Predicting cardiovascular disease and all-cause mortality using the lymphocyte-to-monocyte ratio: Insights from explainable machine learning models |
| title_short | Predicting cardiovascular disease and all-cause mortality using the lymphocyte-to-monocyte ratio: Insights from explainable machine learning models |
| title_sort | predicting cardiovascular disease and all cause mortality using the lymphocyte to monocyte ratio insights from explainable machine learning models |
| topic | All-cause mortality Cardiovascular disease Lymphocyte-to-monocyte ratio |
| url | http://www.sciencedirect.com/science/article/pii/S2772487525000108 |
| work_keys_str_mv | AT jichaowu predictingcardiovasculardiseaseandallcausemortalityusingthelymphocytetomonocyteratioinsightsfromexplainablemachinelearningmodels AT diehuang predictingcardiovasculardiseaseandallcausemortalityusingthelymphocytetomonocyteratioinsightsfromexplainablemachinelearningmodels AT jiefangli predictingcardiovasculardiseaseandallcausemortalityusingthelymphocytetomonocyteratioinsightsfromexplainablemachinelearningmodels AT jingxingyi predictingcardiovasculardiseaseandallcausemortalityusingthelymphocytetomonocyteratioinsightsfromexplainablemachinelearningmodels AT yulei predictingcardiovasculardiseaseandallcausemortalityusingthelymphocytetomonocyteratioinsightsfromexplainablemachinelearningmodels AT junyin predictingcardiovasculardiseaseandallcausemortalityusingthelymphocytetomonocyteratioinsightsfromexplainablemachinelearningmodels |