Abstract
Increasing and compelling evidence has been proved that heavy metal exposure is involved in the development of insulin resistance (IR). We trained an interpretable predictive machine learning (ML) model for IR in the non-diabetic populations based on levels of heavy metal exposure. A total of 4354 participants from the NHANES (2003–2020) with complete information were randomly divided into a training set and a test set. Twelve ML algorithms, including random forest (RF), XGBoost (XGB), logistic regression (LR), GaussianNB (GNB), ridge regression (RR), support vector machine (SVM), multilayer perceptron (MLP), decision tree (DT), AdaBoost (AB), Gradient Boosting Decision Tree (GBDT), Voting Classifier (VC), and K-Nearest Neighbour (KNN), were constructed for IR prediction using the training set. Among these models, the RF algorithm had the best predictive performance, showing an accuracy of 80.14%, an AUC of 0.856, and an F1 score of 0.74 in the test set. We embedded three interpretable methods, the permutation feature importance analysis, partial dependence plot (PDP), and Shapley additive explanations (SHAP) in RF model for model interpretation. Urinary Ba, urinary Mo, blood Pb, and blood Cd levels were identified as the main influencers of IR. Within a specific range, urinary Ba (0.56–3.56 µg/L) and urinary Mo (1.06–20.25 µg/L) levels exhibited the most pronounced upwards trend with the risk of IR, while blood Pb (0.05–2.81 µg/dL) and blood Cd (0.24–0.65 µg/L) levels showed a declining trend with IR. The findings on the synergistic effects demonstrated that controlling urinary Ba levels might be more crucial for the management of IR. The SHAP decision plot offered personalized care for IR based on heavy metal control. In conclusion, by utilizing interpretable ML approaches, we emphasize the predictive value of heavy metals for IR, especially Ba, Mo, Pb, and Cd.
Similar content being viewed by others
Data Availability
Data are accessible via a public repository with open access. On the NHANES website, data are provided with open access: https://www.cdc.gov/nchs/nhanes/index.htm.
References
Li M, Chi X, Wang Y, Setrerrahmane S, Xie W, Xu H (2022) Trends in insulin resistance: insights into mechanisms and therapeutic strategy. Signal Transduct Target Ther 7(1):216
Petersen MC, Shulman GI (2018) Mechanisms of insulin action and insulin resistance. Physiol Rev 98(4):2133–2223
Fliser D, Pacini G, Engelleiter R, Kautzky-Willer A, Prager R, Franek E et al (1998) Insulin resistance and hyperinsulinemia are already present in patients with incipient renal disease. Kidney Int 53(5):1343–1347
Liu X, Tan Z, Huang Y, Zhao H, Liu M, Yu P et al (2022) Relationship between the triglyceride-glucose index and risk of cardiovascular diseases and mortality in the general population: a systematic review and meta-analysis. Cardiovasc Diabetol 21(1):124
Wang F, Han L, Hu D (2017) Fasting insulin, insulin resistance and risk of hypertension in the general population: a meta-analysis. Clin Chim Acta 464:57–63
da Silva AA, do Carmo JM, Li X, Wang Z, Mouton AJ, Hall JE (2020) Role of hyperinsulinemia and insulin resistance in hypertension: metabolic syndrome revisited. Can J Cardiol 36(5):671–682
Wang J, Wu D, Guo H, Li M (2019) Hyperandrogenemia and insulin resistance: the chief culprit of polycystic ovary syndrome. Life Sci 236:116940
Whyte MB, Joy M, Hinton W, McGovern A, Hoang U, van Vlymen J et al (2022) Early and ongoing stable glycaemic control is associated with a reduction in major adverse cardiovascular events in people with type 2 diabetes: a primary care cohort study. Diabetes Obes Metab 24(7):1310–1318
Laiteerapong N, Ham SA, Gao Y, Moffet HH, Liu JY, Huang ES et al (2019) The legacy effect in type 2 diabetes: impact of early glycemic control on future complications (the diabetes & aging study). Diabetes Care 42(3):416–426
Gerstein HC, Miller ME, Byington RP, Goff DC Jr, Bigger JT, Buse JB et al (2008) Effects of intensive glucose lowering in type 2 diabetes. N Engl J Med 358(24):2545–2559
Pharmacologic approaches to glycemic treatment (2019) standards of medical care in diabetes-2019. Diabetes Care 42(Suppl 1):S90-s102
Ausk KJ, Boyko EJ, Ioannou GN (2010) Insulin resistance predicts mortality in nondiabetic individuals in the U.S. Diabetes Care 33(6):1179–85
Tsai SF, Yang CT, Liu WJ, Lee CL (2023) Development and validation of an insulin resistance model for a population without diabetes mellitus and its clinical implication: a prospective cohort study. EClinicalMedicine 58:101934
Jiang J, Cai X, Pan Y, Du X, Zhu H, Yang X et al (2020) Relationship of obesity to adipose tissue insulin resistance. BMJ Open Diabetes Res Care 8(1):e000741
Rohm TV, Meier DT, Olefsky JM, Donath MY (2022) Inflammation in obesity, diabetes, and related disorders. Immunity 55(1):31–55
Wellen KE, Hotamisligil GS (2005) Inflammation, stress, and diabetes. J Clin Invest 115(5):1111–1119
Yang BY, Qian ZM, Li S, Chen G, Bloom MS, Elliott M et al (2018) Ambient air pollution in relation to diabetes and glucose-homoeostasis markers in China: a cross-sectional study with findings from the 33 communities Chinese health study. Lancet Planet Health 2(2):e64–e73
Zhang S, Mwiberi S, Pickford R, Breitner S, Huth C, Koenig W et al (2021) Longitudinal associations between ambient air pollution and insulin sensitivity: results from the KORA cohort study. Lancet Planet Health 5(1):e39–e49
Vella RE, Pillon NJ, Zarrouki B, Croze ML, Koppe L, Guichardant M et al (2015) Ozone exposure triggers insulin resistance through muscle c-Jun N-terminal kinase activation. Diabetes 64(3):1011–1024
Xu X, Yavar Z, Verdin M, Ying Z, Mihai G, Kampfrath T et al (2010) Effect of early particulate air pollution exposure on obesity in mice: role of p47phox. Arterioscler Thromb Vasc Biol 30(12):2518–2527
Yu Z, Han J, Li L, Zhang Q, Chen A, Chen J et al (2023) Chronic triclosan exposure induce impaired glucose tolerance by altering the gut microbiota. Food Chem Toxicol 183:114305
Choi YH, Kim JH, Lee BE, Hong YC (2014) Urinary benzene metabolite and insulin resistance in elderly adults. Sci Total Environ 482–483:260–268
Wang N, Gao X, Huo Y, Li Y, Cheng F, Zhang Z (2023) Lead exposure aggravates glucose metabolism disorders through gut microbiota dysbiosis and intestinal barrier damage in high-fat diet-fed mice. J Sci Food Agric
González-Domínguez Á, Millán-Martínez M, Domínguez-Riscart J, Lechuga-Sancho AM, González-Domínguez R (2023) Metal homeostasis and exposure in distinct phenotypic subtypes of insulin resistance among children with obesity. Nutrients 15(10):2347
Wang X, Mukherjee B, Karvonen-Gutierrez CA, Herman WH, Batterman S, Harlow SD et al (2020) Urinary metal mixtures and longitudinal changes in glucose homeostasis: the Study of Women’s Health Across the Nation (SWAN). Environ Int 145:106109
Bertinato J, Wang KC, Hayward S (2017) Serum magnesium concentrations in the Canadian population and associations with diabetes, glycemic regulation, and insulin resistance. Nutrients 9(3):296
Barra NG, Anhê FF, Cavallari JF, Singh AM, Chan DY, Schertzer JD (2021) Micronutrients impact the gut microbiota and blood glucose. J Endocrinol 250(2):R1-r21
Tan PY, Soma Roy M (2021) Dietary copper and selenium are associated with insulin resistance in overweight and obese Malaysian adults. Nutr Res 93:38–47
Hu J, Cao J, Xu Q, Lu M (2022) Dose-response relationships between urinary cobalt concentrations and obesity, insulin resistance, and metabolic-related disorders in the general population. Environ Sci Pollut Res Int 29(20):29682–29688
Li W, Jiao Y, Wang L, Wang S, Hao L, Wang Z et al (2022) Association of serum magnesium with insulin resistance and type 2 diabetes among adults in China. Nutrients 14(9):1799
Li X, Zhao Y, Zhang D, Kuang L, Huang H, Chen W et al (2023) Development of an interpretable machine learning model associated with heavy metals’ exposure to identify coronary heart disease among US adults via SHAP: findings of the US NHANES from 2003 to 2018. Chemosphere 311(Pt 1):137039
Wu X, Zhou Q, Mu L, Hu X (2022) Machine learning in the identification, prediction and exploration of environmental toxicology: challenges and perspectives. J Hazard Mater 438:129487
Li W, Huang G, Tang N, Lu P, Jiang L, Lv J et al (2023) Effects of heavy metal exposure on hypertension: a machine learning modeling approach. Chemosphere 337:139435
Ban MJ, Lee DH, Shin SW, Kim K, Kim S, Oa SW et al (2022) Identifying the acute toxicity of contaminated sediments using machine learning models. Environ Pollut 312:120086
Li T, Zhang Q, Peng Y, Guan X, Li L, Mu J et al (2023) Contributions of various driving factors to air pollution events: interpretability analysis from machine learning perspective. Environ Int 173:107861
Phung VLH, Oka K, Hijioka Y, Ueda K, Sahani M, Wan Mahiyuddin WR (2022) Environmental variable importance for under-five mortality in Malaysia: a random forest approach. Sci Total Environ 845:157312
Lee CL, Liu WJ, Tsai SF (2022) Development and validation of an insulin resistance model for a population with chronic kidney disease using a machine learning approach. Nutrients 14(14):2832
Park S, Kim C, Wu X (2022) Development and validation of an insulin resistance predicting model using a machine-learning approach in a population-based cohort in Korea. Diagnostics (Basel) 12(1):212
Chakradar M, Aggarwal A, Cheng X, Rani A, Kumar M, Shankar A (2023) A non-invasive approach to identify insulin resistance with triglycerides and HDL-c ratio using machine learning. Neural Process Lett 55(1):93–113
Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC (1985) Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 28(7):412–419
Muniyappa R, Lee S, Chen H, Quon MJ (2008) Current approaches for assessing insulin sensitivity and resistance in vivo: advantages, limitations, and appropriate usage. Am J Physiol Endocrinol Metab 294(1):E15-26
Cheng J, Sun J, Yao K, Xu M, Cao Y (2022) A variable selection method based on mutual information and variance inflation factor. Spectrochim Acta A Mol Biomol Spectrosc 268:120652
Yu Q, Ji W, Prihodko L, Ross CW, Anchang JY, Hanan NP (2021) Study becomes insight: ecological learning from machine learning. Methods Ecol Evol 12(11):2117–2128
Zhang Z, Xu B, Xu W, Wang F, Gao J, Li Y et al (2022) Machine learning combined with the PMF model reveal the synergistic effects of sources and meteorological factors on PM(2.5) pollution. Environ Res 212(Pt B):113322
Park J, Lee WH, Kim KT, Park CY, Lee S, Heo TY (2022) Interpretation of ensemble learning to predict water quality using explainable artificial intelligence. Sci Total Environ 832:155070
Dong Z, Wang Q, Ke Y, Zhang W, Hong Q, Liu C et al (2022) Prediction of 3-year risk of diabetic kidney disease using machine learning based on electronic medical records. J Transl Med 20(1):143
Xia F, Li Q, Luo X, Wu J (2022) Machine learning model for depression based on heavy metals among aging people: a study with National Health and Nutrition Examination Survey 2017–2018. Front Public Health 10:939758
Xia F, Li Q, Luo X, Wu J (2022) Identification for heavy metals exposure on osteoarthritis among aging people and machine learning for prediction: a study based on NHANES 2011–2020. Front Public Health 10:906774
Zhao M, Wan J, Qin W, Huang X, Chen G, Zhao X (2023) A machine learning-based diagnosis modelling of type 2 diabetes mellitus with environmental metal exposure. Comput Methods Programs Biomed 235:107537
Liang C, Zhang Z, Cao Y, Wang J, Shen L, Jiang T et al (2022) Exposure to multiple toxic metals and polycystic ovary syndrome risk: endocrine disrupting effect from As. Pb and Ba Sci Total Environ 849:157780
Menke A, Guallar E, Cowie CC (2016) Metals in urine and diabetes in U.S. adults. Diabetes 65(1):164–71
Padilla MA, Elobeid M, Ruden DM, Allison DB (2010) An examination of the association of selected toxic metals with total and central obesity indices: NHANES 99–02. Int J Environ Res Public Health 7(9):3332–3347
Li XT, Yu PF, Gao Y, Guo WH, Wang J, Liu X et al (2017) Association between plasma metal levels and diabetes risk: a case-control study in China. Biomed Environ Sci 30(7):482–491
Liu B, Feng W, Wang J, Li Y, Han X, Hu H et al (2016) Association of urinary metals levels with type 2 diabetes risk in coke oven workers. Environ Pollut 210:1–8
Martins AC, Ferrer B, Tinkov AA, Caito S, Deza-Ponzio R, Skalny AV et al (2023) Association between heavy metals, metalloids and metabolic syndrome: new insights and approaches. Toxics 11(8):670
Moon SS (2013) Association of lead, mercury and cadmium with diabetes in the Korean population: the Korea National Health and Nutrition Examination Survey (KNHANES) 2009–2010. Diabet Med 30(4):e143–e148
Wang Y, Shi P, Zhao C, Shi J, Qi Z, Xu S et al (2023) Identification of the regulatory network and potential markers for type 2 diabetes mellitus related to internal exposure to metals in Chinese adults. Environ Geochem Health 45(9):6889–6902
Park YJ, Jung Y, Oh CU (2019) Relations between the blood lead level and metabolic syndrome risk factors. Public Health Nurs 36(2):118–125
Rotter I, Kosik-Bogacka D, Dołęgowska B, Safranow K, Lubkowska A, Laszczyńska M (2015) Relationship between the concentrations of heavy metals and bioelements in aging men with metabolic syndrome. Int J Environ Res Public Health 12(4):3944–3961
Scinicariello F, Buser MC, Mevissen M, Portier CJ (2013) Blood lead level association with lower body weight in NHANES 1999–2006. Toxicol Appl Pharmacol 273(3):516–523
Wang N, Sheng Z, Zhou S, Jiang F, Zhang Z (2022) Chronic lead exposure exacerbates hepatic glucolipid metabolism disorder and gut microbiota dysbiosis in high-fat-diet mice. Food Chem Toxicol 170:113451
Flores CR, Puga MP, Wrobel K, Garay Sevilla ME, Wrobel K (2011) Trace elements status in diabetes mellitus type 2: possible role of the interaction between molybdenum and copper in the progress of typical complications. Diabetes Res Clin Pract 91(3):333–41
Wang Q, Wei S (2018) Cadmium affects blood pressure and negatively interacts with obesity: findings from NHANES 1999–2014. Sci Total Environ 643:270–276
Nie X, Wang N, Chen Y, Chen C, Han B, Zhu C et al (2016) Blood cadmium in Chinese adults and its relationships with diabetes and obesity. Environ Sci Pollut Res Int 23(18):18714–18723
Salcedo-Bellido I, Gómez-Peña C, Pérez-Carrascosa FM, Vrhovnik P, Mustieles V, Echeverría R et al (2021) Adipose tissue cadmium concentrations as a potential risk factor for insulin resistance and future type 2 diabetes mellitus in GraMo adult cohort. Sci Total Environ 780:146359
Nguyen J, Patel A, Gensburg A, Bokhari R, Lamar P, Edwards J (2022) Diabetogenic and obesogenic effects of cadmium in Db/Db mice and rats at a clinically relevant level of exposure. Toxics 10(3):107
Mendel RR (2009) Cell biology of molybdenum. BioFactors 35(5):429–434
Yang J, Lu Y, Bai Y, Cheng Z (2023) Sex-specific and dose-response relationships of urinary cobalt and molybdenum levels with glucose levels and insulin resistance in U.S. adults. J Environ Sci (China) 124:42–49
Vasto S, Di Gaudio F, Raso M, Sabatino L, Caldarella R, De Pasquale C et al (2022) Impact on glucose homeostasis: is food biofortified with molybdenum a workable solution? A two-arm study. Nutrients 14(7):1351
González-Domínguez Á, Millán-Martínez M, Domínguez-Riscart J, Mateos RM, Lechuga-Sancho AM, González-Domínguez R (2022) Altered metal homeostasis associates with inflammation, oxidative stress, impaired glucose metabolism, and dyslipidemia in the crosstalk between childhood obesity and insulin resistance. Antioxidants (Basel) 11(12):2439
Zheng S, Nie Z, Lv Z, Wang T, Wei W, Fang D et al (2022) Associations between plasma metal mixture exposure and risk of hypertension: a cross-sectional study among adults in Shenzhen. China Front Public Health 10:1039514
Guney M, Zagury GJ (2012) Heavy metals in toys and low-cost jewelry: critical review of U.S. and Canadian legislations and recommendations for testing. Environ Sci Technol 46(8):4265–74
Buser MC, Scinicariello F (2017) Cadmium, lead, and depressive symptoms: analysis of National Health and Nutrition Examination Survey 2011–2012. J Clin Psychiatry 78(5):e515–e521
Author information
Authors and Affiliations
Contributions
All authors contributed to the study conception and design. Material preparation, data collection, and analysis were performed by Jun Liu and Xingyu Li. The first draft of the manuscript was written by Peng Zhu and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
Corresponding author
Ethics declarations
Ethical Approval
The NHANES program was approved by the ethics review committee of the National Center for Health Statistics in December 1998.
Consent to Participate
The participant data were obtained from the publicly available NHANES database, so no additional consent was obtained.
Competing Interests
The authors declare no competing interests.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Below is the link to the electronic supplementary material.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Liu, J., Li, X. & Zhu, P. Effects of Various Heavy Metal Exposures on Insulin Resistance in Non-diabetic Populations: Interpretability Analysis from Machine Learning Modeling Perspective. Biol Trace Elem Res (2024). https://doi.org/10.1007/s12011-024-04126-3
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s12011-024-04126-3