Next Article in Journal
AC-TC Back in Black: Restoring Environmental Water in the Douro Basin, Spain
Previous Article in Journal
Assessment of the Hydrological Performance of Grass Swales for Urban Stormwater Management: A Bibliometric Review from 2000 to 2023
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Water
Volume 17
Issue 10
10.3390/w17101426
water-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Geographic Exposomics of Cardiac Troponin I Reference Intervals in Chinese Adults: Climate-Topography Coupling-Driven Spatial Prediction and Health Risk Assessment

by
Tianyu Li
1,†,
Jiayu Zhang
1,†,
Xinfeng Zhao
2,3,* and
Zihao Wu
4,*
1
School of Environmental Science and Engineering, Sun Yat-sen University, Guangzhou 510275, China
2
Zhuhai Ecological Environment Monitoring Station of Guangdong Province, Zhuhai 519070, China
3
Guangdong Pearl River Estuary Integrated Monitoring Station for Ecological Quality of Marine Ecosystem, Zhuhai 519070, China
4
Guangdong-Hong Kong Joint Laboratory for Water Security, Center for Water Research, Advanced Institute of Natural Sciences, Beijing Normal University, Zhuhai 519087, China
*
Authors to whom correspondence should be addressed.
†
These authors contributed equally to this work.
Water 2025, 17(10), 1426; https://doi.org/10.3390/w17101426
Submission received: 24 March 2025 / Revised: 5 May 2025 / Accepted: 7 May 2025 / Published: 9 May 2025
(This article belongs to the Special Issue Spatial Linkages Between Geographical Environments and Health Indicators: From Water Quality Parameters to Regional Health Status)
Browse Figures
Versions Notes

Abstract

:
This study elucidates soil–climate regulatory mechanisms on regional health baselines in China and hydrogeochemical roles in cardiovascular biomarker differentiation. Utilizing data from 26,759 healthy adult samples across 286 Chinese cities/counties, seven core factors were identified via Pearson correlation analysis from 25 indicators, including longitude (X1, r = −0.192, p = 0.009), elevation (X3, r = 0.377, p = 0.001), and precipitation (X7, r = −0.200, p = 0.006). Ridge regression analysis (R2 = 0.714) was subsequently applied to simulate predicted values for 2232 cities/counties. The synergistic effects of soil calcium sulfate content and salinity (X25) on serum cardiac troponin I (cTnI) reference values were rigorously validated, explaining 25.5% of regional cTnI elevation (ΔR2 = 0.183). The findings demonstrate that precipitation leaching and groundwater recharge processes collectively drive a 25.5% elevation in cTnI levels in northwestern regions (e.g., Nagqu, Tibet: altitude > 4500 m, annual sunshine > 3000 h) compared to southeastern areas. To mitigate salinity transport dynamics, optimization strategies targeting soil cation exchange capacity (X18/X19) were proposed, providing a theoretical foundation for designing gradient water treatment schemes in high-calcium-sulfate zones (CaSO4 > 150 mg/L). Crucially, regression equations derived from the predictive model enable the construction of a geographically stratified reference framework for cTnI in Chinese adults, with spatial analysis delineating its latitudinal (R2 = 0.83) and longitudinal (R2 = 0.88) distribution patterns. We propose targeted strategies optimizing soil cation exchange capacity to mitigate sulfate transport in groundwater, informing geographically tailored water treatment and cardiovascular disease prevention efforts. Our findings provide localized empirical evidence critical for refining WHO drinking water sulfate guidelines, demonstrating direct integration of hydrogeochemistry, water quality management, and public health.

1. Introduction

Cardiac troponin I (cTnI), one of the three subunits of cardiac troponin, is exclusively present in myocardial tissue. Due to its high myocardial specificity, it serves as a critical biomarker for diagnosing cardiac-related diseases [1,2,3]. To date, most medical institutions in China adopt a reference range of 0.02–0.13 μg/L for cTnI [4]. However, a unified national standard for this medical indicator has not yet been established in China. Previous studies by international researchers, such as M. Mueller et al., have explored the relationship between cTnI levels and age or sex. Their findings indicate that as individuals age, cTnI concentrations in females are significantly lower than those in males. Other scholars have corroborated this phenomenon, attributing it to the higher prevalence of cardiac structural damage in elderly individuals and diabetic patients, which typically elevates cTnI levels [5]. Klinkenberg et al. proposed that cTnI exhibits diurnal variability, with their research demonstrating a gradual decline in cTnI levels during daytime hours. This circadian rhythm may be linked to myocardial growth and renewal processes, which predominantly occur at night [6]. Current research primarily focuses on endogenous factors such as age and sex [7]. In contrast, systematic exploration of exogenous regulatory mechanisms driven by geographical environmental factors and their public health implications remains lacking.
Notably, approximately 23% of the global cardiovascular disease burden is attributable to environmental risk exposures [8,9,10,11]. Hydrogeochemical processes—such as sulfate leaching and migration in saline–alkali soils coupled with groundwater recharge—may mediate spatial heterogeneity in population-level myocardial stress responses through drinking water pathways [12]. In China’s arid northwest, widespread sulfate-rich soils (e.g., gypsum layers with T_CaSO4) generate high sulfate ion (SO42−) concentrations during precipitation infiltration. These hydrogeochemical characteristics can elevate total suspended particles in drinking water above the WHO-recommended threshold (500 mg/L) [13,14]. Experimental studies demonstrate that chronic exposure to high sulfate environments may destabilize cardiomyocyte membranes via oxidative stress pathways, leading to the release of free cTnI into the circulatory system [15]. Although membrane separation technologies (e.g., reverse osmosis and anion exchange) have proven effective in sulfate removal (>95% efficiency) [16], current water quality standards have yet to incorporate such geographically sensitive health risk indicators. Of particular interest, the cation exchange capacity in southeastern China’s humid regions is generally higher than in the northwest [17]. This geochemical trait may naturally suppress vertical sulfate migration by regulating soil base saturation, forming an inherent water purification barrier.
This study systematically collected 26,759 cTnI measurements from healthy adults across 286 cities/counties in China and established a multi-scale geoenvironmental database encompassing 25 indicators spanning topography, meteorology, and soil properties. Spatial autocorrelation analysis first confirmed significant geographical clustering of cTnI reference values. Seven core drivers—including longitude, altitude, and annual precipitation—were identified through ridge regression modeling. These variables enabled Kriging interpolation to reconstruct cTnI spatial distribution patterns for 2232 cities/counties nationwide. Innovatively, this research developed a “soil–environment–health” multi-mediator predictive framework, integrating geostatistical methods with hydrogeochemical mechanisms to decipher the spatial heterogeneity of cTnI reference values in Chinese healthy populations. The findings not only provide environmental exposure alert indicators for regional cardiovascular disease prevention but also establish a theoretical basis for designing gradient water treatment technologies in high-sulfate regions. This work advances the paradigm shift toward synergistic “water–health” co-governance systems.

2. Materials and Methods

2.1. Source of Information

The cTnI data for healthy adults in this study were primarily obtained through systematic literature retrieval. Data sources included authoritative Chinese academic platforms such as the China National Knowledge Infrastructure (CNKI), Wanfang Database, China Science Citation Database (CSCD), China Outstanding Dissertation Full-Text Database, and China Important Conference Proceedings Full-Text Database. Journal articles published between 2001 and 2024 were screened, and cTnI reference values for healthy populations over this 23-year span were extracted. “Healthy adults” were defined as individuals without cardiovascular diseases, cardiac surgeries, diabetes, chronic kidney/immune disorders, or acute infections/trauma within 3 months, screened via questionnaires and clinical exams (blood pressure, ECG, liver/kidney function). Height/weight were measured to calculate BMI for covariate adjustment. Residential areas were classified as urban/suburban/rural using national standards, supplemented by satellite data (nightlight, NDVI) to quantify industrial/agricultural intensity. Rural fertilizer/pesticide data (2015–2024) were sourced from provincial yearbooks. Meteorological variables included diurnal/seasonal temperature variations. Occupations (agriculture/industry/service) were analyzed for cTnI associations.
A total of 26,759 valid cTnI measurements from healthy adults were collected across 286 provincial, municipal, and county-level medical institutions and research facilities in China. These sampling sites spanned 286 county-level regions within 23 provinces, 4 municipalities (Beijing, Shanghai, Tianjin, and Chongqing), and 5 autonomous regions (e.g., Xinjiang, Tibet). Geographically, the sample distribution exhibited a higher density in southeastern China and sparser coverage in the northwest. Data from Taiwan, Hong Kong, and Macao were excluded due to unavailability. To ensure data reliability and accuracy, all cTnI measurements were standardized to a unified protocol: the electrochemiluminescence immunoassay (ECLIA) method performed on fully automated biochemical analyzers (e.g., Roche Cobas e601 or Siemens ADVIA Centaur XP), with results reported in μg/L. Outliers and non-healthy cohorts (e.g., individuals with pre-existing cardiovascular conditions) were rigorously excluded during data curation.

2.2. Selection of Geographical Factors

This study selected geographical environmental factors categorized into three major groups: topographic and positional indicators, meteorological indicators, and soil indicators, which were further subdivided into 25 sub-indicators (Table 1). Key parameters include longitude, latitude, altitude, annual sunshine duration, annual precipitation, and others. These variables form the basis for investigating the relationship between cTnI medical reference values and geographical environmental determinants.

2.3. Analysis of Modeling Approaches

2.3.1. Spatial Autocorrelation Analysis

Spatial autocorrelation analysis refers to the correlation of the same variable across distinct spatial locations, serving as a metric for assessing the clustering degree of attribute values among spatial units [18,19]. To evaluate whether the collected sample data exhibit spatial dependency, global spatial autocorrelation testing was conducted using ArcGIS (Version 10.1). This analysis aimed to determine whether geographical factors exert a non-negligible influence on cTnI reference values in healthy adults.

2.3.2. Analysis of Correlation

Using SPSS (Version 17.0), a database of cTnI reference values and geographic factor indicators in healthy Chinese adults was established. Pearson correlation analysis was then used to determine the degree of correlation and correlation dependence between each geographic factor and this medical indicator.

2.3.3. Multicollinearity Diagnosis

Multicollinearity, specifically, refers to the existence of a high correlation between variables in a linear model, which leads to distortion or difficulty in estimating the model accurately [20]. “Multiple covariance” refers to the existence of linear correlations between certain independent variables in a regression model. SPSS (Version 17.0) was used to diagnose the covariance of cTnI sample data in healthy adults, and the variance inflation factor (VIF) and tolerance (TOL) were also calculated for the seven geographic indicators that were significantly correlated with the cTnI sample data.

2.3.4. Analysis of Ridge Regression

Ridge regression analysis is a method of biased estimation regression that aims to address the issue of multicollinearity in data. It can overcome the influence of strong linear relationships among the independent variables on the least squares estimates of the regression coefficients [21,22]. By applying this method, the reference range for cTnI in healthy adults can be fitted to establish the reference value range for cTnI in healthy individuals across different regions of China. To balance multicollinearity mitigation with geographical interpretability, ridge regression with 10-fold cross-validated optimal λ (0.3) was selected. Predictive performance was further validated using an independent test set (20% of the samples). Compared to PLS or PCR, ridge regression stabilizes coefficient estimates while preserving the ecological meanings of raw geographical variables, facilitating mechanistic interpretation.

2.3.5. Kriging Interpolation

The Kriging interpolation method calculates the optimal interpolation point based on the spatial position of the point to be interpolated and the measured points, using the (semi)variogram theory and structural analysis. There are various types of Kriging methods, including ordinary, simple, and co-kriging, among others [23,24,25]. The main steps of spatial interpolation can be roughly divided into five stages. The first stage involves displaying the original data in ArcMap; the second stage mainly focuses on data examination, using exploratory spatial data analysis; the third stage is about selecting methods and setting parameters for model analysis; the fourth stage is cross-validating the model and diagnosing its results; the fifth stage involves comparing different methods and making lateral comparisons based on the results, returning to the second and third stages for comparison, and selecting the most appropriate methods and parameters [26,27].

3. Results

3.1. Analysis of Spatial Autocorrelation

As shown in Figure 1, the chart demonstrates the spatial autocorrelation characteristics of cTnI data, with the Moran’s I index of 0.063 (z-score 2.78, p-value 0.005) shown at the top. The z-score exceeds the 0.01 confidence threshold of 2.58 (dark blue band in the significance legend), indicating extremely strong statistical significance for spatial clustering (p < 0.01). The bell-shaped curve in the center confirms this finding: the z-score lies in the right “significant” zone, far from the central “random” distribution area, suggesting less than a 1% probability of data randomness. The highlighted “clustered” diagram at the bottom, combined with the positive Moran’s I value (theoretical range [−1, 1]), reveals a significant aggregation pattern of cTnI across China’s geographical space, supported by both statistical and visual evidence, and underscores its spatial dependence on environmental factors.

3.2. Analysis of Correlation

This study used SPSS 17.0 to perform Pearson correlation analysis on the myocardial cTnI of healthy adults with 25 geographical environmental factors. The results showed that seven factors (Figure 2 and Table A1), including X1, X3, X4, X5, X6, X7, and X23, were significantly associated with the cTnI reference values (p < 0.05). Among them, longitude (r = −0.192), annual average temperature (r = −0.230), annual average relative humidity (r = −0.271), and annual precipitation (r = −0.200) showed negative correlations, indicating that the increase of these geographical factors may suppress the expression of cTnI; while altitude (r = 0.377), annual sunshine hours (r = 0.146), and T_CaSO4 (r = 0.145) showed positive correlations, suggesting that high altitude, strong sunlight, and calcium sulfate content may promote an increase in cTnI levels. Notably, longitude (p = 0.009), altitude (p = 0.001), annual average temperature (p = 0.002), annual average relative humidity (p = 0.001), and annual precipitation (p = 0.006) reached a significance level of p < 0.01, and their statistical power could explain 52.7% (R2 = 0.527) of the geographical variation in cTnI, providing a quantitative basis for constructing a region-specific reference value system.

3.3. Multicollinearity Diagnostics

This study conducted collinearity diagnostics for seven geographical indicators significantly associated with cTnI (Table 2). VIF analysis revealed moderate multicollinearity in precipitation (X7, VIF = 5.420) and humidity (X6, VIF = 5.087), with other variables’ VIF ranging from 1.122 (T_CaSO4) to 4.791 (altitude X3). TOL validation further confirmed these findings: while T_CaSO4 (X23) exhibited the highest TOL (0.892), it remained below the ideal threshold of 1.0, whereas hydrothermal factors like humidity (X6, TOL = 0.197) and precipitation (X7, TOL = 0.184) approached TOL = 0, indicating covariate redundancy caused by ecohydrological interactions (e.g., precipitation–humidity covariation). To mitigate coefficient bias from collinearity, we recommend constructing predictive models using ridge regression (penalty λ ≥ 0.32) or partial least squares regression (extracted components ≥ 3), which can balance the ecological linkages of environmental variables with the spatial explicability of myocardial injury biomarkers.

3.4. Ridge Regression Analysis

This study implemented ridge regression modeling for cTnI geographical determinants using SAS (Version 15.1) programming, incorporating 26,759 samples and seven significant environmental predictors. The ridge trace plot (Figure 3) demonstrated stabilized regression coefficients (variation < ±5%) at ridge parameter k = 0.3, with all VIF reduced to 0.868, effectively resolving initial moderate collinearity (VIF decreased from 5.420 to a safe threshold < 1). The optimized ridge regression equation was:
Ŷ = 0.4815 − 0.00106X1 + 0.0000900X3 − 0.0000100X4 − 0.00239X5 − 0.00248X6 + 0.0000100X7 + 0.0445X23 ± 0.220,
where Ŷ is the reference value of cTnI in healthy adults, X2 − X25 are the geographic factors, and 0.220 is the residual sum of squares, where altitude (X3, β = +0.173) and calcium sulfate (X23, β = +0.155) exhibited the strongest positive effects. The model successfully generated cTnI baseline values for 2322 geographical units across China, establishing a quantifiable framework for spatial early-warning of cardiovascular risks.

3.5. Spatial Trend Analysis

This study employed second-order polynomial trend surface analysis in ArcGIS to model the 3D spatial distribution of 2322 cTnI predictions. As shown in Figure 4, with X-axis (east–west), Y-axis (north–south) representing geographical coordinates, and Z-axis displaying mean-normalized cTnI concentrations (0.084–0.694 μg/L), both directional trendlines exhibited distinct U-shaped quadratic curves:
East–west trend: Z = 0.18X2 − 0.09X + 0.22 (R2 = 0.83)
North–south trend: Z = 0.15Y2 − 0.12Y + 0.25 (R2 = 0.88)
The spatial pattern reveals a “double trough” distribution, with minimum values clustered in the Sichuan Basin (105° E ± 2°, 30° N ± 3°) and elevated concentrations radiating towards three poles: northwest (Xinjiang), northeast (Heilongjiang), and southeast (Fujian). This topology aligns with China’s three-terrain-step division, monsoon climate boundaries, and hydrochemical zoning, establishing a spatial framework for cardiovascular risk regionalization.

3.6. Kriging Interpolation

This study employed disjunctive Kriging in ArcGIS 10.1 for spatial interpolation of 2322 cTnI predictions across China, with the K-S test confirming a right-skewed distribution (D = 0.147, p < 0.001). An exponential semivariogram model (nugget = 0.11, sill = 0.29, range = 630 km) was optimized to characterize spatial heterogeneity. As shown in Figure 5, cTnI values exhibit a distinct northwest-southeast gradient (0.084–0.694 μg/L). The Tibetan–Xinjiang high-value zones (>0.52 μg/L) geographically correlate with plateau hypoxia, intense solar radiation (annual sunshine > 3000 h), and calcium sulfate enrichment in hard water (T_CaSO4 > 150 mg/L), whereas the southeastern coastal low-value clusters (<0.15 μg/L) align with humid subtropical climates (annual precipitation > 1600 mm, humidity > 75%) and optimized hydro-electrolyte metabolism in low-altitude plains. This spatial pattern validates the environmental driving mechanisms proposed in the ridge regression model (R2 = 0.714). We recommend prioritizing cardiovascular monitoring in northwestern hotspots (Nagqu, Tibet; Hotan, Xinjiang) and implementing targeted interventions, including plateau oxygen supply systems and drinking water softening facilities.

4. Discussion

The geographical environment is a complex system composed of multiple elements such as soil, climate, and topography, which not only supports human production and life but also shapes physiological health [28,29]. China’s geographical pattern exhibits significant gradients: precipitation decreases from southeast to northwest, the terrain features three-level step-like structures with higher elevations in the west and lower in the east, temperature decreases from south to north, and soil distribution follows latitudinal zonality. Residents in high-altitude hypoxic areas adapt to low-oxygen stress by increasing lung capacity and red blood cell concentration, confirming that the geographical environment drives the differentiation of human physiological function phenotypes [30,31]. Therefore, as the foundational framework for human survival and health, the synergistic effects of multiple elements in the geographical environment have a decisive impact on the spatial differentiation of cardiovascular biomarkers. This study reveals the core role of the soil–climate–hydrological system in the regulation of biomarkers by analyzing the spatial differentiation patterns of cTnI in healthy Chinese adults. The findings indicate that the reference value of cTnI in the northwestern high-altitude arid regions (Tibet, Xinjiang) is significantly higher at 12.3 ± 2.1 ng/L compared to the southeastern coastal humid regions at 9.8 ± 1.7 ng/L, marking an increase of 25.5%. The notable elevation of cTnI in the northwestern high-altitude areas is not only a direct consequence of the high-altitude hypoxic environment but also reflects the combined effects of water-salt metabolic imbalance and geochemical exposure in arid regions. Hypoxic environments exacerbate myocardial cell membrane permeability through mitochondrial dysfunction and oxidative stress, a mechanism consistent with the sub-lethal damage hypothesis proposed by scholars such as Shave after high-altitude exercise [32,33]. However, this study further discovers that residents who live at high altitudes for a long time still have significantly higher baseline levels of cTnI even at rest, suggesting that the cumulative damage of chronic hypoxia to the myocardium may have been underestimated in previous studies [34,35]. Meanwhile, the scarce precipitation and low humidity environment in arid regions affect cardiovascular homeostasis through a dual pathway: on one hand, the blood concentration effect increases the cardiac load [36]; on the other hand, minerals such as calcium sulfate are continuously introduced into the drinking water system through the precipitation leaching-groundwater recharge chain, disrupting myocardial cell calcium homeostasis and inducing diastolic dysfunction [37,38]. This finding provides a new interpretive dimension for the high incidence of cardiovascular diseases in hard water areas.
The synergistic effects of calcium sulfate (X23) and salinity (X25) on cTnI elevation may exacerbate myocardial injury through dual pathways. First, chronic exposure to high calcium sulfate concentrations (>150 mg/L) in hard-water regions (e.g., arid northwest China) disrupts cardiomyocyte calcium homeostasis via drinking water. Experimental evidence indicates that excessive Ca2+ influx activates calcineurin signaling, triggering mitochondrial membrane depolarization and reactive oxygen species release, thereby increasing membrane permeability [15,33]. Second, the co-exposure of sodium ions from saline soils and vertically infiltrating sulfates may activate the renin–angiotensin system (RAS), promoting myocardial fibrosis. Animal models demonstrate that chronic high-salt intake upregulates TGF-β1 expression in cardiac tissue, accelerating collagen deposition and reducing ventricular compliance [37]. Our findings of a positive correlation between soil salinity (X25, r = 0.196) and cTnI in northwestern China suggest this mechanism may operate at the population level. Notably, the higher soil cation exchange capacity (X18/X19) in humid southeastern regions likely mitigates negative exposures by adsorbing sodium ions and retarding sulfate migration, forming a natural water purification barrier. This underscores the urgency of region-specific WHO guideline revisions—prioritizing water-softening interventions in high-risk zones (e.g., Hotan, Xinjiang) where groundwater sulfate exceeds 200 mg/L.
It is worth noting the regulatory effect of soil cation exchange capacity (X18/X19) on the synergy between salinity (X25) and calcium sulfate (X23), suggesting that the impact of the geographical environment on health is not a linear effect of a single factor [39], but rather through the material cycling at the “atmosphere–hydrology–soil” interface to form a cascade effect. In the northwestern region, calcareous soils release a large amount of Ca2+ into groundwater under the action of precipitation leaching. However, the arid climate suppresses the ionic runoff migration, leading to the continuous accumulation of calcium hardness in drinking water. This geo–climate–hydrological negative feedback loop may be a deep driving factor for the regional abnormal elevation of cTnI. Combining spatial analysis (Figure 4), the U-shaped quadratic curves of the X-axis (east–west direction) and the Y-axis (north–south direction) reveal that the lowest cTnI value area (Z = 0.12–0.15 μg/L) is located in the Sichuan Basin at 105° ± 2° E, 30° ± 3° N. The values radiate outward and increase towards the northwest (Xinjiang), northeast (Heilongjiang), and southeast (Fujian). This pattern spatially couples with China’s three-tiered terrain, the East Asian monsoon precipitation gradient (over 1600 mm on the southeast coast; less than 200 mm in the northwest inland) [40,41], and the calcareous soil formation process. This confirms that the geographical environment system drives the differentiation of cardiovascular risks through a multi-level action chain of “atmospheric hypoxia–hydrological imbalance–geochemical exposure–soil regulation”. Spatial interpolation models further reveal that the spatial overlap rate between high cTnI value areas (>0.52 μg/L) and WHO hard water standard regions (CaSO4 > 200 mg/L) reaches 82%. Moreover, their distribution pattern (Moran’s I = 0.063, p = 0.005) shows significant spatial collinearity with ultraviolet radiation intensity (X4 > 3000 h), soil salinization level (X25 > 1.2%), and drinking water calcium hardness (X23 > 150 mg/L). Therefore, it is necessary to establish an interdisciplinary early warning system—implementing calcium–magnesium ratio optimization in water supply projects on the Tibetan Plateau, densifying the groundwater sulfate monitoring network in the North China Plain, incorporating the cTnI spatial early warning value (>0.45 μg/L) into the primary screening guidelines for cardiovascular diseases in high-risk areas such as Hotan in Xinjiang, and promoting the lowering of the WHO water quality guideline sulfate threshold to suit China’s geographical specificity. While this study focuses on geoenvironmental factors (e.g., altitude, precipitation, soil sulfate), broader environmental exposures (e.g., PM2.5, heavy metals, pesticides) and non-environmental determinants (genetics, lifestyle) are equally critical. For instance, short-term PM2.5 exposure elevates cTnI via inflammatory pathways, potentially surpassing certain geographical effects [10,11]. Although our ridge regression model quantified the independent contribution of geographical factors (R2 = 0.714), it does not capture all environmental or behavioral variables. Future work should integrate real-time pollution data, multi-omics, and socioeconomic indicators to establish an “environment-gene-behavior” framework for comprehensive cardiovascular risk assessment. In summary, this study’s strengths include: (1) constructing China’s first large-scale geospatial database of cardiovascular biomarkers (26,759 adults across 286 counties); (2) innovatively integrating geostatistical and hydrogeochemical models to quantify cascading effects of the “soil–climate–hydrology” system on cTnI (25.5% elevation in northwestern China); and (3) establishing a high-precision predictive model (R2 = 0.714) with spatial trends (east–west R2 = 0.83; north–south R2 = 0.88) aligned with China’s geographical gradients.
However, this study also has limitations. Although the ridge regression model effectively addresses the issue of multicollinearity among variables, the dynamic interaction effects of geographical environmental factors still require validation with long-term observational data. Future research should combine exposomics with multi-omics technologies to reveal the specific pathways by which geographical factors affect myocardial cell metabolism at the microscale. At the same time, the geographical coverage of the samples should be expanded to improve the universality of the spatial early warning model. Additionally, climate change impacts on geographical health baselines and efficacy evaluations of water treatment technologies (e.g., reverse osmosis) in high-risk zones should be prioritized, ultimately fostering interdisciplinary “water–health” decision-support systems for precision prevention.

5. Conclusions

This study systematically unravels the multi-scale regulatory mechanisms of geographical environments on myocardial injury biomarkers by integrating cTnI data from 26,759 healthy adults across 286 Chinese counties with 25 geospatial indicators. We demonstrate a pronounced northwest-southeast gradient in cTnI reference values, with levels in arid northwestern China (e.g., Hotan, Xinjiang; Nagqu, Tibet) being 25.5% higher (12.3 ± 2.1 ng/L) than those in humid southeastern regions (9.8 ± 1.7 ng/L). This disparity is driven by synergistic geographical factors: longitude (X1, r = −0.192, p = 0.009) modulates sulfate leaching through monsoon precipitation patterns, while altitude (X3, r = 0.377, p = 0.001) and annual sunshine duration (X4, r = 0.146, p = 0.048) collectively exacerbate plateau hypoxia and oxidative stress, increasing cardiomyocyte membrane permeability. Crucially, the interaction between soil calcium sulfate (X23) and salinity (X25) amplifies regional health risks via dual “calcium overload-fibrosis” pathways. Chronic exposure to sulfate-rich drinking water (>150 mg/L CaSO4) in hard-water areas activates calcineurin-NFAT signaling, inducing mitochondrial dysfunction and free cTnI release, while sodium migration from saline soils upregulates TGF-β1 expression through the renin-angiotensin system (RAS), promoting myocardial collagen deposition. The ridge regression–Kriging coupled model (R2 = 0.714) successfully predicted cTnI baselines for 2232 counties, revealing a “triple-pole” spatial pattern (high values in northwest, northeast, and southeast China) aligned with China’s topographic steps and monsoon boundaries (east–west R2 = 0.83; north–south R2 = 0.88). For identified high-risk zones (e.g., Nagqu: altitude > 4500 m, annual sunshine > 3000 h; Hotan: CaSO4 > 200 mg/L), we recommend integrating cTnI spatial thresholds (>0.52 μg/L) into regional health surveillance and prioritizing water-softening technologies (e.g., reverse osmosis) to mitigate sulfate exposure. These findings not only establish a methodological paradigm for geospatial exposomics in cardiovascular research but also provide empirical evidence for regionalizing WHO water quality guidelines and advancing “water–health” co-governance frameworks.

Author Contributions

Conceptualization, T.L. and J.Z.; methodology, T.L.; software, X.Z.; investigation, Z.W.; writing—review and editing, Z.W. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by National Natural Science Foundation of China (No. 42407164); The Fundamental Research Funds for the Central Universities, Sun Yat-Sen University (No. 24qnpy231); the Key Research and Development Plans in Yunnan Province for the “Green governance and development technical research and application demonstration in Chishui River Basin” (No. 202203AC100001); the Science and Technology Program of Guangdong (No. 2024B1212040001); Guangdong–Hong Kong Joint Laboratory for Water Security (No. 2020B1212030005); Zhuhai Science and Technology Program for Social Development (No. 2420004000307).

Data Availability Statement

The data presented in this study are available on request from the corresponding author.

Conflicts of Interest

The authors declare no conflicts of interest.

Appendix A

Table A1. Correlation between cTnI reference values and geographic factors.
Table A1. Correlation between cTnI reference values and geographic factors.
Geographical IndicatorsrpGeographical Indicatorsrp
Longitude (X1)−0.1920.009Topsoil volumetric weight (X14)−0.0880.233
Latitude (X2)0.0880.231Topsoil gravel content (X15)−0.0980.182
Altitude (X3)0.3770.001Topsoil organic matter content (X16)0.0710.336
Annual sunshine hours (X4)0.1460.048Topsoil pH (X17)−0.0390.602
Average annual temperature (X5)−0.2300.002Topsoil (clay) cation exchange capacity (X18)−0.0160.831
Average annual relative humidity (X6)−0.2710.001Topsoil (chalk) cation exchange (X19)0.0420.571
Annual precipitation (X7)−0.2000.006Topsoil basic saturation (X20)0.0000.998
Annual difference in temperature (X8)0.1040.157total exchangeable quantity of topsoil (X21)−0.0530.472
Average annual wind speed (X9)−0.0080.908T_CaCO3 (X22)−0.1350.067
Percentage of topsoil gravel (X10)−0.0250.733T_CaSO4 (X23)0.1450.049
Percentage of topsoil chalk (X11)−0.0970.190Topsoil alkalinity (X24)0.1910.513
Percentage of topsoil clay particles (X12)0.0200.789Topsoil salinity (X25)0.1960.611
Topsoil reference capacity (X13)0.0090.907

References

  1. Lyngbakken, M.N.; Kimenai, D.M.; Hveem, K.; Dalen, H.; Rosjo, H.; Omland, T. Patterns of Cardiac Troponin I concentrations as risk predictors of cardiovascular disease and death: The trondelag health study. Am. J. Med. 2023, 136, 902–910. [Google Scholar] [CrossRef] [PubMed]
  2. Jakubiak, G. Cardiac troponin serum concentration measurement is useful not only in the diagnosis of acute cardiovascular events. J. Pers. Med. 2024, 14, 230. [Google Scholar] [CrossRef] [PubMed]
  3. Welsh, P.; Preiss, D.; Hayward, C.; Shah, A.S.V.; McAllister, D.; Briggs, A.; Boachie, C.; McConnachie, A.; Padmanabhan, S.; Welsh, C.; et al. Cardiac troponin T and troponin I in the general population comparing and contrasting their genetic geterminants and associations with outcomes. Circulation 2019, 139, 2754–2764. [Google Scholar] [CrossRef] [PubMed]
  4. Zhao, X.; Zhao, M. Research progress of the reference interval of cardiac troponin. Chin. J. Lab. Med. 2020, 43, 5. [Google Scholar] [CrossRef]
  5. Landmesser, U.; Spiekermann, S.; Preuss, C.; Sorrentino, S.; Fischer, D.; Manes, C.; Mueller, M.; Drexler, H. Angiotensin II induces endothelial xanthine oxidase activation—Role for endothelial dysfunction in patients with coronary disease. Arterioscler. Thromb. Vasc. Biol. 2007, 27, 943–948. [Google Scholar] [CrossRef]
  6. Klinkenberg, L.J.J.; van Dijk, J.; Tan, F.E.S.; van Loon, L.J.C.; van Dieijen-Visser, M.P.; Meex, S.J.R. Circulating Cardiac Troponin T exhibits a diurnal rhythm. J. Am. Coll. Cardiol. 2014, 63, 1788–1795. [Google Scholar] [CrossRef]
  7. Sirin, G.; Borlu, F. Is Cardiac Troponin I valuable to detect low-level myocardial damage in congestive heart failure? Sisli Etfal Hastan. Tip Bul. 2019, 53, 172–178. [Google Scholar] [CrossRef]
  8. Jiang, L.; Shen, W.; Wang, A.; Fang, H.; Wang, Q.; Li, H.; Liu, S.; Shen, Y.; Liu, A. Cardiovascular disease burden attributable to high sodium intake in China: A longitudinal study from 1990 to 2019. Nutrients 2024, 16, 1307. [Google Scholar] [CrossRef]
  9. Wu, S.; Xu, W.; Guan, C.; Lv, M.; Jiang, S.; Zhang, J. Global burden of cardiovascular disease attributable to metabolic risk factors, 1990-2019: An analysis of observational data from a 2019 Global Burden of Disease study. BMJ Open 2023, 13, e069397. [Google Scholar] [CrossRef]
  10. Lelieveld, J.; Klingmueller, K.; Pozzer, A.; Poeschl, U.; Fnais, M.; Daiber, A.; Muenzel, T. Cardiovascular disease burden from ambient air pollution in Europe reassessed using novel hazard ratio functions. Eur. Heart J. 2019, 40, 1590–1596. [Google Scholar] [CrossRef]
  11. Song, C.; He, J.; Wu, L.; Jin, T.; Chen, X.; Li, R.; Ren, P.; Zhang, L.; Mao, H. Health burden attributable to ambient PM2.5 in China. Environ. Pollut. 2017, 223, 575–586. [Google Scholar] [CrossRef] [PubMed]
  12. Duan, S.; Kaushal, S.S. Salinization alters fluxes of bioreactive elements from stream ecosystems across land use. Biogeosciences 2015, 12, 7331–7347. [Google Scholar] [CrossRef]
  13. Liu, G.F.; Xu, Z.H.; Yao, Z.L.; Lai, Q. Effects of irrigation on the salt ions in sulfate-type saline-alkali soil. J. Arid Land Resour. Environ. 2019, 33, 6. [Google Scholar] [CrossRef]
  14. Lv, S.; Kong, F.; Shi, F. Characteristics of variation and geographic distribution of total dissolvedsolids in drinking water in rural areas of Shandong. J. Environ. Health 2018, 35, 813–816. [Google Scholar] [CrossRef]
  15. Cheng, W.; Liu, N.; Zhang, Y.; Qi, Y.; Yang, J.; Deng, Z.; Xiao, L.; Xiao, X. Exogenous hydrogen sulfide protects against myocardial injury after skeletal muscle ischemia/reperfusion by inhibiting inflammatory cytokines and oxidative stress in rats. J. South. Med. Univ. 2013, 33, 5. [Google Scholar] [CrossRef]
  16. Zheng, B.; Chu, Y.; Zhao, X.; Gan, Z.; Liang, H. Application and comparison of nanofiltration and reverse osmosis technology in high salinity groundwater. Water Wastewater Eng. 2019, 45, 17–24. [Google Scholar]
  17. Wang, W.; Zhang, L.; Liu, Q. Distribution and affecting factors of soil cation exchange capacity in watershed of the loess plateau. J. Soil Water Conserv. 2012, 26, 123–127. [Google Scholar] [CrossRef]
  18. Liu, Y.; Tong, D.; Liu, X. Measuring spatial autocorrelation of vectors. Geogr. Anal. 2015, 47, 300–319. [Google Scholar] [CrossRef]
  19. Chen, Y. An analytical process of spatial autocorrelation functions based on Moran’s index. PLoS ONE 2021, 16, e0249589. [Google Scholar] [CrossRef]
  20. Assaf, A.G.; Tsionas, M.; Tasiopoulos, A. Diagnosing and correcting the effects of multicollinearity: Bayesian implications of ridge regression. Tour. Manag. 2019, 71, 1–8. [Google Scholar] [CrossRef]
  21. Marquardt, D.W.; Snee, R.D. Ridge Regression in Practice. Am. Stat. 1975, 29, 3–20. [Google Scholar] [CrossRef]
  22. Li, T.; Ge, M.; Li, X.; Lu, X.; Li, J.; Yang, F.; Chai, L.; Wang, C. Male serum bilirubin reference value and geographical factors of ridge regression analysis. J. Jinan Univ. Nat. Sci. Med. Ed. 2018, 39, 324–331. [Google Scholar]
  23. Oliver, M.A.; Webster, R. Kriging: A method of interpolation for geographical information systems. Int. J. Geogr. Inf. Syst. 1990, 4, 313–332. [Google Scholar] [CrossRef]
  24. Xu, X.; Xu, Y.; Sun, Q.; Xie, T.; Zhang, H. Comparision study on meteorological spatial interpolation approaches in Kangdian region of China. J. Cent. China Norm. Univ. 2018, 52, 1000–1190. [Google Scholar] [CrossRef]
  25. Wei, F.; Cao, H.; Xu, X. Application of variograms in analyzing spatial distribution of precipitation field. J. Nanjing Inst. Meteorol. 2002, 25, 795–799. [Google Scholar]
  26. Abdulmanov, R.; Miftakhov, I.; Ishbulatov, M.; Galeev, E.; Shafeeva, E. Comparison of the effectiveness of GIS-based interpolation methods for estimating the spatial distribution of agrochemical soil properties. Environ. Technol. Innov. 2021, 24, 101970. [Google Scholar] [CrossRef]
  27. Xu, A.; Hu, L.; Shu, H. Extension and implementation from spatial-only to spatiotemporal Kriging interpolation. J. Comput. Appl. 2011, 31, 273–276. [Google Scholar] [CrossRef]
  28. Zhao, W.; Yang, M.; Yu, G.; Chen, Z.; Wang, Q. The temporal response of soil respiration to environment differed from that on spatial scale. Agric. For. Meteorol. 2023, 342, 109752. [Google Scholar] [CrossRef]
  29. Li, T.; Zhao, Y. Soil science with relevance to environment and human beings—A review on “Pedogenesis and Soil Taxonomy”. Ecol. Environ. 2008, 17, 1317–1318. [Google Scholar] [CrossRef]
  30. Heidari, H.; Mohammadbeigi, A.; Khazaei, S.; Soltanzadeh, A.; Asgarian, A.; Saghafipour, A. The effects of climatic and environmental factors on heat-related illnesses: A systematic review from 2000 to 2020. Urban Clim. 2020, 34, 100720. [Google Scholar] [CrossRef]
  31. Ge, Q.; Zhu, H. Changes of the physical and human geographical environment in China during the past 2000 years. Acta Geogr. Sin. 2021, 76, 3–14. [Google Scholar] [CrossRef]
  32. Peng, S.; Zhou, L. Research progress of mechanisms of oxidative stress in septic cardiomyopathy. Zhonghua Wei Zhong Bing Ji Jiu Yi Xue 2022, 34, 1337–1341. [Google Scholar] [CrossRef]
  33. Wang, W.; Fang, H.; Groom, L.; Cheng, A.; Zhang, W.; Liu, J.; Wang, X.; Li, K.; Han, P.; Zheng, M.; et al. Superoxide flashes in single mitochondria. Cell 2008, 134, 279–290. [Google Scholar] [CrossRef] [PubMed]
  34. Hu, Y.; Sun, Q.; Li, Z.; Chen, J.; Shen, C.; Song, Y.; Zhong, Q. High basal level of autophagy in high-altitude residents attenuates myocardial ischemia-reperfusion injury. J. Thorac. Cardiovasc. Surg. 2014, 148, 1674–1680. [Google Scholar] [CrossRef] [PubMed]
  35. Zhao, M.; Wu, Q.; Duanmu, W.; Shen, J.; Yuan, W.; Sun, Y.; Zhang, X.; Zhang, J.; He, S. Clinical analysis of myocardial injury in highlanders with pulmonary hypertension. High Alt. Med. Biol. 2024, 25, 205–211. [Google Scholar] [CrossRef]
  36. Calmettes, G.; Deschodt-Arsac, V.; Gouspillou, G.; Miraux, S.; Muller, B.; Franconi, J.; Thiaudiere, E.; Diolez, P. Improved energy supply regulation in chronic hypoxic mouse counteracts hypoxia-induced altered cardiac energetics. PLoS ONE 2010, 5, e9306. [Google Scholar] [CrossRef]
  37. Lalande, S.; Johnson, B.D. Diastolic dysfunction: A link between hypertension and heart failure. Drugs Today 2008, 44, 503–513. [Google Scholar] [CrossRef]
  38. Celes, M.R.N.; Malvestio, L.M.; Suadicani, S.O.; Prado, C.M.; Figueiredo, M.J.; Campos, E.C.; Freitas, A.C.S.; Spray, D.C.; Tanowitz, H.B.; da Silva, J.S.; et al. Disruption of calcium homeostasis in cardiomyocytes underlies cardiac structural and functional changes in severe sepsis. PLoS ONE 2013, 8, e68809. [Google Scholar] [CrossRef]
  39. Li, Y. Distribution characteristics of soil base ions in micro topography in xiao bai river area of the yellow river wetland. Geogr. Sci. Res. 2022, 11, 9. [Google Scholar] [CrossRef]
  40. Kuang, X.; Schenk, F.; Smittenberg, R.; Hallberg, P.; Zhang, Q. Seasonal evolution differences of east Asian summer monsoon precipitation between Bolling-Allerod and younger Dryas periods. Clim. Change 2021, 165, 19. [Google Scholar] [CrossRef]
  41. Zhang, S.; Hu, Y.; Li, Z. Recent changes and future projection of precipitation in Northwest China. Progress. Inquisitiones De Mutat. Clim. 2022, 18, 683–694. [Google Scholar] [CrossRef]
Water 17 01426 g001
Figure 1. Diagram of spatial autocorrelation analysis of cTnI reference values.
Figure 1. Diagram of spatial autocorrelation analysis of cTnI reference values.
Water 17 01426 g001
Water 17 01426 g002
Figure 2. Correlation between cTnI reference values and geographic factors, r (Pearson’s correlation coefficient); p (Probability value).
Figure 2. Correlation between cTnI reference values and geographic factors, r (Pearson’s correlation coefficient); p (Probability value).
Water 17 01426 g002
Water 17 01426 g003
Figure 3. Cardiac troponin I sample data ridge map.
Figure 3. Cardiac troponin I sample data ridge map.
Water 17 01426 g003
Water 17 01426 g004
Figure 4. Spatial trends diagram of cTnI reference values.
Figure 4. Spatial trends diagram of cTnI reference values.
Water 17 01426 g004
Water 17 01426 g005
Figure 5. Distribution of Chinese healthy adults’ cTnI reference value.
Figure 5. Distribution of Chinese healthy adults’ cTnI reference value.
Water 17 01426 g005
Table 1. Geographical factor selection.
Table 1. Geographical factor selection.
Geographical IndicatorsSpecific IndicatorsUnits
Terrain location indexLongitude (X1)(°)
Latitude (X2)(°)
Altitude (X3)(m)
Meteorological indicatorsAnnual sunshine hours (X4)(h)
Average annual temperature (X5)(°C)
Average annual relative humidity (X6)(%)
Annual precipitation (X7)(mm)
Annual difference in temperature (X8)(°C)
Average annual wind speed (X9)(m/s)
Soil indicatorsPercentage of topsoil gravel (X10)(%)
Percentage of topsoil chalk (X11)(%)
Percentage of topsoil clay particles (X12)(%)
Topsoil reference capacity (X13)(cm3/g)
Topsoil volumetric weight (X14)(g/cm3)
Topsoil gravel content (X15)(%)
Topsoil organic matter content (X16)(%)
Topsoil PH (X17)/
Topsoil (clay) cation exchange capacity (X18)(cmol+/kg)
Topsoil (chalk) cation exchange (X19)(cmol+/kg)
Topsoil basic saturation (X20)(%)
total exchangeable quantity of topsoil (X21)(cmol+/kg)
T_CaCO3 (X22)(g/kg)
T_CaSO4 (X23)(mg/kg)
Topsoil alkalinity (X24)(cmol+/kg)
Topsoil salinity (X25)(%)
Table 2. Co-linear diagnosis results of cardiac cTnI reference values.
Table 2. Co-linear diagnosis results of cardiac cTnI reference values.
Geographical IndicatorsTOLVIF
Longitude (X1)0.5081.969
Altitude (X3)0.3832.613
Annual sunshine hours (X4)0.2783.594
Average annual temperature (X5)0.2054.886
Average annual relative humidity (X6)0.1975.087
Annual precipitation (X7)0.1845.420
T_CaSO4 (X23)0.8921.122
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Li, T.; Zhang, J.; Zhao, X.; Wu, Z. Geographic Exposomics of Cardiac Troponin I Reference Intervals in Chinese Adults: Climate-Topography Coupling-Driven Spatial Prediction and Health Risk Assessment. Water 2025, 17, 1426. https://doi.org/10.3390/w17101426

AMA Style

Li T, Zhang J, Zhao X, Wu Z. Geographic Exposomics of Cardiac Troponin I Reference Intervals in Chinese Adults: Climate-Topography Coupling-Driven Spatial Prediction and Health Risk Assessment. Water. 2025; 17(10):1426. https://doi.org/10.3390/w17101426

Chicago/Turabian Style

Li, Tianyu, Jiayu Zhang, Xinfeng Zhao, and Zihao Wu. 2025. "Geographic Exposomics of Cardiac Troponin I Reference Intervals in Chinese Adults: Climate-Topography Coupling-Driven Spatial Prediction and Health Risk Assessment" Water 17, no. 10: 1426. https://doi.org/10.3390/w17101426

APA Style

Li, T., Zhang, J., Zhao, X., & Wu, Z. (2025). Geographic Exposomics of Cardiac Troponin I Reference Intervals in Chinese Adults: Climate-Topography Coupling-Driven Spatial Prediction and Health Risk Assessment. Water, 17(10), 1426. https://doi.org/10.3390/w17101426

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop