Nonlinear association of serum copper-zinc ratio with overweight/obesity in children and adolescents

Abstract

Obesity in children and adolescents has become a serious global public health problem and is closely associated with a range of metabolic diseases. Copper and zinc, as important trace elements, play important roles in energy metabolism, oxidative stress, and lipid metabolism, respectively. An imbalance in the copper-to-zinc ratio may be associated with obesity, but its relationship with obesity in children and adolescents has not been fully explored. The present study explored and revealed the association between the two. This study used data from the National Health and Nutrition Examination Survey (NHANES) between 2011 and 2016, including 1730 children and adolescents aged 6 to 19 years. First, the baseline characteristics of all participants were compared. Subsequently, the association between serum copper-zinc ratio and overweight/obesity was explored by multivariate logistic regression modeling. Next, five subgroup analyses were conducted for gender, age, race, household income, and smoking. Finally, threshold effects were assessed using restricted cubic spline regression analysis and segmented regression modeling. Serum copper-zinc ratio was significantly and positively associated with overweight/obesity in children and adolescents, suggesting a potential relationship (OR = 3.06; 95% CI 2.41–3.92; P < 0.001). Subgroup analyses indicated that the correlations were generally stable and consistent, with significant interactions for race and family income. The prevalence of overweight/obesity increased progressively with increasing copper-zinc ratios. However, this trend was nonlinear, clarifying that a serum copper-zinc ratio of 1.561 was a significant inflection point in this nonlinear relationship. This study explores the association between serum copper-zinc ratio and overweight/obesity in children and adolescents, revealing a nonlinear relationship and its threshold effect. As a measure of oxidative stress and lipid metabolism imbalance, the serum copper-zinc ratio may become a new tool for obesity risk assessment. Given its close association with obesity, the serum copper-zinc ratio should be incorporated into child and adolescent health management for early intervention in the future.

Introduction

The problem of overweight obesity in children and adolescents has become a serious challenge in global public health¹. In recent years, with the development of the global economy and the improvement of living standards, obesity rates have risen significantly. According to statistics, about 340 million children and adolescents aged 5 to 19 years old were affected by overweight or obesity worldwide in 2016, which is more than a tenfold increase from 1976². The spread of obesity not only poses a threat to physical health but is also closely associated with a range of metabolic diseases (e.g., dyslipidemia, fatty liver, type 2 diabetes, and insulin resistance)^3–6. More seriously, obesity in childhood often continues into adulthood, increasing the risk of chronic diseases such as cardiovascular disease⁷. In addition to this, obesity has a profound impact on the mental health, social adjustment, and quality of life of children and adolescents, leading to mental and psychosocial disorders⁸. As the prevalence of obesity continues to rise, it is projected that by 2030, more than 250 million children and adolescents worldwide will be obese, further exacerbating the global public health burden⁹. Therefore, early monitoring and effective intervention are essential to reduce the health and social burden of obesity.

Zinc and copper, as essential trace elements, have attracted much attention due to their metabolic homeostasis associated with energy regulation, oxidative stress, and lipid metabolism. Zinc is involved in the insulin signaling pathway and the regulation of adipocyte differentiation. At the same time, copper plays a key role in free radical scavenging and energy metabolism as a cofactor for various enzymes. Existing study¹⁰ has shown that copper has a protective effect on cells catalyzed by superoxide dismutase. However, copper accumulation not only catalyzes reactive oxygen species and reactive nitrogen, leading to an imbalance in the oxidative/antioxidant system^11–13, but excessive copper-to-zinc substitution also increases oxidative stress¹⁴, which collectively promotes the development of overweight/obesity. Moreover, copper/zinc ratio imbalance may be a potential biomarker reflecting obesity-related complications. More notably, children and adolescents are at a critical stage of growth and development, and their micronutrient requirements and metabolic profiles are different from those of adults. However, the correlation between serum Cu/Zn ratios and overweight/obesity in this population is unknown.

The National Health and Nutrition Examination Survey (NHANES) is an ongoing research program designed to systematically assess and monitor different age groups’ health and nutritional status in the United States. The survey plays an important role in revealing health disparities trends and providing a scientific basis for public health policy development. Our study hypothesized that serum Cu/Zn ratios are associated with overweight/obesity in child and adolescent populations. This study aimed to validate the association between the two based on data from the NHANES database and to explore this correlation’s specific trends and key nodes.

Materials and methods

Survey description and study participants

The NHANES database survey was organized and administered by the National Center for Health Statistics, approved by the Ethics Review Board of the Centers for Disease Control and Prevention, and all participants provided written informed consent. Comprehensive information about the study design and data collection is available at https://www.cdc.gov/nchs/nhanes/irba98.htm and https://www.cdc.gov/nchs/nhanes/about_nhanes.htm. The survey collected a wide range of information, including detailed demographic statistics, physical examination results, and other health indicators. All study protocols adhered to the principles in the Declaration of Helsinki and followed the STROBE Initiative reporting guidelines for observational epidemiologic studies.

We included participants from three cycles of the NHANES database from 2011 to 2016, during which 29,903 individuals participated in the NHANES survey, of which 8,053 were children and adolescents aged 6 to 19 years. Nonetheless, we excluded individuals with missing BMI classifications (n = 400), individuals with incomplete serum copper and zinc in blood indicators (n = 5550), and individuals with missing data on other important covariates (n = 373). Thus, our final study cohort comprised 1730 participants (see Fig. 1 for details).

Fig. 1.

NHANES participant selection flowchart.

Definition of overweight and obesity

Health professionals at the Mobile Health Screening Center use standardized tools to measure the height and weight of participants. We used the body mass index (BMI) to assess overweight/obesity among 6–19-year-olds, calculated as weight (kg) divided by height (m) squared (kg/m²). Critical criteria are based on the 2000 Centers for Disease Control and Prevention BMI growth charts for gender and age. They are categorized by percentile into four categories: underweight (BMI < 5th percentile), normal weight (5th percentile ≤ BMI < 85th percentile), overweight (85th percentile ≤ BMI < 95th percentile), and obese (BMI ≥ 95th percentile). This classification is recognized as the gold standard for accurately identifying overweight and obese children and adolescents.

Determination of serum copper-zinc ratio

Serum copper levels were determined from the LBXSCU in the NHANES database, and serum zinc levels from the LBXSZN. For analysis, collected serum samples were processed, stored, and transported to the Laboratory Sciences Division of the National Center for Environmental Health, Centers for Disease Control and Prevention, Atlanta, Georgia. Two elements were determined using inductively coupled plasma dynamic reaction cell mass spectrometry (ICP-DRC-MS), a multi-element analytical technique capable of trace element analysis. This study calculated the serum copper/zinc ratio as the quotient of serum copper level (ug/dL) and serum zinc level (ug/dL).

To ensure the accuracy and reliability of serum copper and zinc measurements, this study utilized data generated by the NHANES program, which employed standardized protocols for sample collection, storage, and analysis. Blood samples were collected using certified micro-metal contamination-free tubes, and serum was separated and frozen at −70 °C prior to analysis to minimize contamination and degradation. Serum copper and zinc concentrations were quantified using inductively coupled plasma dynamic reaction cell mass spectrometry (ICP-DRC-MS), a technique widely recognized for its high sensitivity and specificity in trace metal detection. The analytical process strictly adheres to internal quality control procedures, with each analytical batch including low, medium, and high concentration quality control materials, and repeated analyses are conducted. The entire batch is reanalyzed if quality control results exceed two standard deviations from the expected mean. Additionally, external quality control is ensured through participation in the Clinical Laboratory Improvement Amendments program of the US Centers for Disease Control and Prevention and proficiency testing under the College of American Pathologists or equivalent programs. These measures collectively ensure the stability of analytical precision (CV < 5%) and methodological consistency throughout the investigation, thereby supporting the validity of the serum copper data used in this study.

Covariate

Covariates in this study were selected based on their established correlations with serum copper-zinc ratios or obesity in previous studies to ensure the accuracy of the results. Demographic characteristics primarily included participants’ sex (male or female), age, race (Mexican American, other Hispanic, non-Hispanic white, non-Hispanic black, or other race), and household poverty-to-income ratio (PIR). Dietary intake data consisted primarily of energy, protein, carbohydrates, total sugars, and total fat in foods and beverages consumed 24 h before the interview. Laboratory data included serum cotinine, white blood cell (WBC) count, eosinophil percentage (EOS%), lymphocyte percentage (MID%), neutrophil percentage (NEUT%), hemoglobin (Hb), high-density lipoprotein cholesterol (HDL-C), and total cholesterol (TC). To explore the potential associations and justify the selection of covariates, we constructed a Directed Acyclic Graph (DAG) (Fig. 2) using the software DAGitty (http://www.dagitty.net/). The DAG systematically represents the assumed causal relationships among exposure (serum copper-zinc ratio), outcome (overweight/obesity), and potential confounders based on prior knowledge and relevant literature. In our DAG model, the serum copper-zinc ratio was designated as the exposure variable and overweight/obesity as the outcome variable. Covariates, including age, sex, race, poverty-income ratio (PIR), and household, were incorporated as common causes of both the exposure and the outcome and were considered confounding variables. These variables may influence trace element metabolism and obesity risk through demographic, socioeconomic, and genetic pathways. Dietary factors (e.g., energy, protein, carbohydrates, total sugars, total fat) were modeled upstream determinants of the serum copper-zinc ratio. In addition, blood indices such as serum cotinine, WBC count, EOS%, LYM%, NEUT%, Hb, and HDL-C were also included as possible mediators or shared causes influencing both the exposure and the outcome. The DAG indicates that to appropriately estimate the total effect of serum copper-zinc ratio on overweight/obesity while minimizing bias due to confounding; we must adjust for the minimally sufficient set of variables, including age, sex, race, PIR, dietary intake, and blood indices. This set blocks all backdoor paths from exposure to outcome, fulfilling the criteria for covariate adjustment under the causal inference framework.

Fig. 2.

Directed acyclic graph of covariates.

Statistical methods

Descriptive analyses were performed on the baseline information of all participants in this study. The statistical analyses involved were performed using R software (version 4.1.3) and Empower Stats (version 2.0), with two-sided tests of significance and significance levels at P < 0.05. Mean (x̄), standard deviation (SD), or median interquartile range (IQR) were used to express the collected continuous variables, and t-test analysis was applied. Categorical variables were expressed as percentages (%) and analyzed using the chi-square test. Before regression modeling, the distributions of serum copper and zinc and their ratio were examined using histograms and the Shapiro-Wilk test. All three variables exhibited moderate right-skewness. A sensitivity analysis was performed using a natural logarithmic transformation of the serum copper/zinc ratio to account for potential non-normality. Logistic regression models using the log-transformed ratio yielded results consistent in direction and significance with the original models, indicating that skewness had minimal impact on the observed associations. For the primary analyses, we used the untransformed copper/zinc ratio to enhance the interpretability of the findings in clinical settings. Subsequently, multivariate logistic regression analyses were performed on the data to explore the correlation between serum copper-zinc ratio and overweight and obesity. Model 1 was not adjusted for any covariates; Model 2 took into account key demographic factors, including sex, age, and race; and Model 3 was a fully adjusted model with PIR, cotinine, WBC, LYM%, EOS%, NEUT%, Hb, HDL-C, TC, energy intake, protein intake, carbohydrate intake, total sugar intake, and total fat intake were added. Subgroup analyses were performed based on age, sex, race, passive smoking, and household income. Possible correlations between serum Cu/Zn ratios and overweight/obesity were assessed more precisely using restricted cubic spline curve (RCS) fitting. To further explore the potential nonlinear association between serum copper-zinc ratio and overweight/obesity, we constructed an RCS model using the package in R (version 4.1.3). The spline was fitted within a multivariable logistic regression framework, adjusting for all covariates included in Model 3. We selected four knots at the 5th, 35th, 65th, and 95th percentiles of the serum copper-zinc ratio distribution, a commonly recommended strategy to ensure sufficient model flexibility while preventing overfitting. The reference value was set at the median serum copper-zinc ratio. The nonlinearity of the association was evaluated by testing the significance of the spline terms using a likelihood ratio test, comparing the model with only the linear term against the model with both linear and spline terms. The fitted RCS curve was used to visualize the dose-response pattern and detect inflection points in the relationship. Finally, segmented models were designed using the R segmented software package, and inflection points were estimated based on model fitting and statistical tests.

Results

Clinical baseline characteristics of the subjects

Table 1 demonstrates the baseline characteristics of 1730 participants grouped by overweight/obesity. A total of 686 were overweight/obese, with a prevalence of 39.65%. The mean age of all participants was 12 years, with 49.02% males and 50.98% females. In terms of racial distribution, non-Hispanic whites had the highest number (27.05%), and Mexican Americans had the most significant percentage (26.09%) in the overweight/obese group.TC and protein intake did not differ significantly between the two groups (p > 0.05). In the overweight/obese group, WBC, NEUT%, Cotinine, and serum CuZn were higher, PIR, EOS%, LYM%, Hb, and HDL-C levels were lower, and Energy, Carbohydrate, Total sugars, and Total fat intake were lower (p < 0.05). Figure 3 visualizes the gender and age composition of the study population, which was roughly symmetrically distributed.

Table 1.

Baseline characteristics of study participants according to the presence of overweight/obesity.

Characteristic	Overall N = 1730	Overweight/obese N = 686	Non-overweight/obese N = 1044	P-value
Race(%)				< 0.001
Mexican American	390 (22.54%)	179 (26.09%)	211 (20.21%)
Other Hispanic	195 (11.27%)	85 (12.39%)	110 (10.54%)
Non-Hispanic White	468 (27.05%)	164 (23.91%)	304 (29.12%)
Non-Hispanic Black	409 (23.64%)	172 (25.07%)	237 (22.70%)
Other Race	268 (15.49%)	86 (12.54%)	182 (17.43%)
Gender(%)				0.192
Male	848 (49.02%)	323 (47.08%)	525 (50.29%)
Female	882 (50.98%)	363 (52.92%)	519 (49.71%)
Age(%)	12.0 (9.0, 16.0)	13.0 (9.0, 16.0)	12.0 (9.0, 16.0)	0.169
PIR(%)	1.47 (0.77, 2.96)	1.38 (0.76, 2.62)	1.59 (0.80, 3.17)	< 0.001
TC (mg/dL)	155 (138, 174)	157 (139, 176)	155 (138, 173)	0.226
HDL-C(mg/dL)	52 (45, 62)	46 (41, 55)	56 (48, 65)	< 0.001
Cotinine (ng/mL)	0.0 (0.0, 0.2)	0.0 (0.0, 0.3)	0.0 (0.0, 0.2)	0.002
WBC(1000 cells/uL)	6.70 (5.50, 8.10)	7.20 (6.00, 8.60)	6.40 (5.30, 7.73)	< 0.001
NEUT(%)	52 (44, 58)	54 (46, 60)	50 (43, 56)	< 0.001
EOS(%)	2.60 (1.60, 4.40)	2.40 (1.50, 4.00)	2.70 (1.70, 4.63)	< 0.001
LYM(%)	36 (30, 42)	34 (29, 41)	37 (31, 43)	< 0.001
Hb (g/dL)	13.30 (12.70, 14.20)	13.30 (12.53, 14.10)	13.40 (12.70, 14.20)	0.041
Energy (kcal)	1,862 (1,403, 2,422)	1,769 (1,351, 2,279)	1,911 (1,441, 2,480)	< 0.001
Protein (gm)	65 (46, 87)	65 (46, 87)	66 (47, 86)	0.410
Carbohydrate(gm)	242 (182, 321)	231 (172, 300)	253 (190, 331)	< 0.001
Total sugars (gm)	109 (73, 153)	103 (69, 147)	114 (75, 158)	0.005
Total fat (gm)	69 (49, 95)	66 (46, 91)	71 (51, 97)	0.002
Serum copper-zinc ratio	1.41 (1.18, 1.69)	1.53 (1.29, 1.85)	1.34 (1.13, 1.58)	< 0.001

PIR poverty-to-income ratio, TC total cholesterol, HDL-C HDL-Cholesterol, WBC white blood cell, NEUT neutrophils, EOS eosinophils, LYM Lymphocyte, Hb hemoglobin.

Fig. 3.

Graph of gender and age distribution of participants.

Table 2 divides the subjects into four groups based on the quartiles of serum Cu/Zn ratios, with group 1 ranging from 0.609 to 1.18, group 2 ranging from 1.18 to 1.41, group 3 ranging from 1.41 to 1.69, and group 4 ranging from 1.69 to 4.8. We were surprised that sex, age, and protein intake differed significantly among the four subgroups (P < 0.05), which differs from the results in Table 1. The prevalence of overweight/obesity in children and adolescents increased significantly with increasing serum copper to zinc ratios of the subjects (Q1:23.33%; Q2:36.34%; Q3:43.98%; Q4:54.97%; P < 0.001).

Table 2.

Baseline characteristics of the copper-zinc ratio-based study population.

Characteristic	Serum copper-zinc ratio
	Q1 [0.609,1.18), N = 433	Q2 [1.18,1.41), N = 432	Q3 [1.41,1.69), N = 432	Q4 [1.69,4.8], N = 433	P-value
Race					< 0.001
Mexican American	102 (23.56%)	101 (23.38%)	97 (22.45%)	90 (20.79%)
Other Hispanic	43 (9.93%)	56 (12.96%)	46 (10.65%)	50 (11.55%)
Non-Hispanic White	123 (28.41%)	124 (28.70%)	124 (28.70%)	97 (22.40%)
Non-Hispanic Black	77 (17.78%)	85 (19.68%)	113 (26.16%)	134 (30.95%)
Other Race	88 (20.32%)	66 (15.28%)	52 (12.04%)	62 (14.32%)
Gender					< 0.001
Male	261 (60.28%)	206 (47.69%)	192 (44.44%)	189 (43.65%)
Female	172 (39.72%)	226 (52.31%)	240 (55.56%)	244 (56.35%)
Overweight/obese					< 0.001
Yes	101 (23.33%)	157 (36.34%)	190 (43.98%)	238 (54.97%)
No	332 (76.67%)	275 (63.66%)	242 (56.02%)	195 (45.03%)
Age	15.0 (12.0, 17.0)	13.0 (10.0, 15.0)	11.0 (8.0, 14.0)	11.0 (8.0, 15.0)	< 0.001
PIR	1.51 (0.89, 3.09)	1.58 (0.78, 2.82)	1.44 (0.75, 3.03)	1.41 (0.70, 2.83)	0.311
TC (mg/dL)	156 (138, 174)	155 (138, 171)	157 (136, 175)	155 (140, 175)	0.757
HDL-C(mg/dL)	53 (45, 61)	52 (45, 62)	53 (45, 62)	51 (43, 61)	0.029
Cotinine (ng/mL)	0.0 (0.0, 0.2)	0.0 (0.0, 0.3)	0.0 (0.0, 0.2)	0.0 (0.0, 0.3)	0.968
WBC(1000 cells/uL)	6.10 (5.10, 7.30)	6.60 (5.50, 7.93)	7.00 (5.80, 8.10)	7.40 (6.10, 8.90)	< 0.001
NEUT(%)	52 (45, 58)	51 (44, 57)	51 (44, 58)	53 (45, 60)	0.099
EOS(%)	2.60 (1.70, 4.20)	2.70 (1.70, 4.33)	2.60 (1.60, 4.60)	2.40 (1.40, 4.60)	0.097
LYM(%)	36 (30, 42)	37 (31, 43)	36 (31, 43)	35 (29, 42)	0.096
Hb (g/dL)	14.00 (13.20, 15.00)	13.50 (12.90, 14.30)	13.10 (12.60, 13.90)	12.90 (12.20, 13.50)	< 0.001
Energy (kcal)	1,974 (1,486, 2,668)	1,864 (1,401, 2,424)	1,821 (1,366, 2,402)	1,766 (1,367, 2,236)	< 0.001
Protein (gm)	73 (52, 102)	63 (46, 84)	63 (45, 86)	63 (45, 82)	< 0.001
Carbohydrate(gm)	258 (191, 349)	253 (183, 323)	233 (174, 308)	234 (178, 300)	0.001
Total sugars (gm)	111 (74, 159)	112 (75, 155)	107 (71, 147)	107 (74, 147)	0.492
Total fat (gm)	74 (52, 105)	68 (50, 92)	69 (48, 96)	65 (45, 89)	< 0.001

Relationship between serum copper-zinc ratio and overweight/obesity

Table 3 demonstrates the results of univariate and multivariate logistic regression analyses by comparing effect estimates under different models. In model 1, which was not adjusted for any covariates, there was a significant positive association between serum copper-zinc ratio and the prevalence of overweight/obesity (OR = 3.06; 95% CI 2.41–3.92; P < 0.001). This positive association remained significant in models adjusted for major demographic factors (age, sex, and race) and adjusted for all confounders. We transformed the continuous variable serum copper-zinc ratio into quartiles and subsequently performed sensitivity analyses, using the first set of quartiles as the reference control. The results showed that in model 1, participants in the highest quartile of serum copper-zinc ratios had a 4.01-fold higher prevalence of overweight/obesity compared with the reference group (OR = 4.01; 95% CI 3.00–5.39.00.39; P < 0.001), and the risk of overweight/obesity was elevated in the group with the highest quartile, with a significant difference between the groups (P for trend < 0.001). In Model 2 and Model 3, the trend of positive correlation persisted, and differences and trends between groups remained significant (P < 0.001), with the highest risk of overweight/obesity in the fourth group of Model 2 (OR = 4.92; 95% CI 3.60–6.76; P < 0.001).

Table 3.

Correlation between copper-zinc ratio and overweight/obesity based on logistic regression analysis.

Exposure	OR(95% CI), p-value
	Model 1	Model 2	Model 3
Per 1 unit incrementa	3.06 (2.41, 3.92) < 0.001	3.50 (2.70, 4.56) < 0.001	2.61 (1.96, 3.51) < 0.001
copper-zinc ratio(quartile)
Quartile 1	Reference	Reference	Reference
Quartile 2	1.88(1.40, 2.53) < 0.001	2.09 (1.54, 2.85) < 0.001	2.07(1.49, 2.90) < 0.001
Quartile 3	2.58 (1.93, 3.47) < 0.001	3.15 (2.31,4.32) < 0.001	2.86 (2.03, 4.05) < 0.001
Quartile 4	4.01 (3.00,5.39) < 0.001	4.92 (3.60, 6.76) < 0.001	3.75 (2.62, 5.40) < 0.001
P for trend	< 0.001	< 0.001	< 0.001

Model 1: No covariates were adjusted. Model 2: Adjusted for gender, age, and race. Model 3: Adjusted for gender, year, race, PIR, TC, HDL-C, Cotinine, WBC, NEUT%, EOS%, LYM%, Hb, Energy, Protein, Carbohydrate, Total sugars and Total fat.^aIn the sensitivity analysis, copper-zinc ratio was converted from a continuous variable to a categorical variable (quartiles).

In the unadjusted model (Model 1), each one-unit increase in the serum copper-zinc ratio was associated with a 3.06-fold increase in the odds of overweight or obesity (OR = 3.06, 95% CI: 2.41–3.92, P < 0.001). After adjusting for age, sex, and race in Model 2, the association remained strong (OR = 3.50, 95% CI: 2.70–4.56, P < 0.001). In the fully adjusted model (Model 3), which included additional covariates such as socioeconomic status, inflammation markers, and dietary intake, each one-unit increase in the copper-zinc ratio was still significantly associated with a 2.61-fold higher odds of overweight/obesity (95% CI: 1.96–3.51, P < 0.001), indicating a robust and independent relationship.

When the serum copper-zinc ratio was categorized into quartiles, individuals in the highest quartile (Q4: 1.69–4.80) had 3.75 times higher odds of being overweight or obese compared to those in the lowest quartile (Q1: 0.609–1.18) in the fully adjusted model (OR = 3.75, 95% CI: 2.62–5.40, P < 0.001). The second and third quartiles also showed significantly elevated risks, with a precise dose-response trend (P for trend < 0.001). These results suggest that higher serum copper-zinc ratios are firmly and progressively associated with increased risk of overweight/obesity among children and adolescents.

Subgroup analysis

To further explore the consistency and potential moderators of the correlation between serum copper-zinc ratio and overweight/obesity, we conducted subgroup analyses using logistic regression modeling across five dimensions: sex, age, race, household income, and passive smoking. As shown in Fig. 4, the results showed a significant positive association between serum copper-zinc ratio and overweight/obesity in the total population (OR = 2.62; 95% CI 1.96–3.49; P < 0.001). The interaction P values for sex, age, and passive smoking were all greater than 0.05, indicating stable and consistent findings across these subgroups. Interestingly, we also found a significant interaction between race and household income, where the serum copper-zinc ratio was not associated with overweight/obesity in the Non-Hispanic White population group (p = 0.329) and the high household income group (p = 0.31).

Fig. 4.

Subgroup analysis of the association between serum copper-zinc ratio and overweight/obesity.

Nonlinear relationship between serum copper-zinc ratio and overweight/obesity

We performed curve fitting using the RCS model, adjusting for all covariates, to explore the possible dose-response relationship between changes in serum copper-zinc ratio and overweight/obesity. The results, as shown in Fig. 5, showed a positive correlation between serum copper-zinc ratio and overweight/obesity on a macroscopic scale. This relationship could be seen from the red curve in the figure, which belonged to a nonlinear pattern, with the curve’s trend tending to rise linearly at first, then slowing down after a particular node (P-nonlinear = < 0.001). Thus, we used a segmented logistic regression model to explore the key turning points. The results showed (Table 4) that when the serum copper-zinc ratio was less than 1.561, OR = 7.47 (3.55, 15.74), P < 0.001, suggesting that there was a significant positive correlation between serum copper-zinc ratio and overweight/obesity, and when it was greater than or equal to 1.561, OR = 1.38 (0.86, 2.23), P = 0.182, suggesting that there was no association between the two. Therefore, we regarded the serum copper-zinc ratio of 1.561 as an inflection point in the nonlinear relationship.

Fig. 5.

Dose-response relationship between serum copper-zinc ratio and overweight/obesity.

Table 4.

Threshold effect analysis of copper-zinc ratio on overweight/obesity.

	Adjusted OR (95% CI)*	P-value
Fitting by standard Logistic regression model
Total	2.61 (1.95, 3.49)	< 0.001
Fitting by piecewise Logistic regression model (Break-Point = 1.561)
copper-zinc ratio < 1.561	7.47 (3.55, 15.74)	< 0.001
copper-zinc ratio ≥ 1.561	1.38(0.86, 2.23)	0.182
Log likelihood ratio		< 0.001

^*Adjusted for: gender, year, race, PIR, TC, HDL-C, Cotinine, WBC, NEUT%, EOS%, LYM%, Hb, Energy, Protein, Carbohydrate, Total sugars and Total fat.

Discussion

This study is the first to examine the relationship between serum copper-zinc ratios and overweight/obesity in children and adolescents in the United States. The results showed a significant nonlinear positive association between serum copper-zinc ratio and overweight/obesity. Through restricted cubic spline curve fitting analysis, we found that this relationship was strongly positively correlated before the inflection point (serum copper-zinc ratio of 1.561) and not significantly associated after the inflection point. This finding provides a new perspective for understanding the complex relationship between obesity and micronutrient metabolism, suggesting that the copper-to-zinc ratio may serve as a potential biomarker for future obesity monitoring and intervention strategies. This study found that the positive correlation between overweight/obesity and serum copper-to-zinc ratio in children and adolescents remained statistically significant after multivariate adjustment. Therefore, we hypothesize that with further research and validation, serum copper-to-zinc ratio may be associated with an increased risk of obesity. Copper and zinc, as important trace elements, are involved in various physiological processes, including redox reactions, immunomodulation, and lipid metabolism, and our following discussion focuses on their association with obesity.

Serum zinc may influence glucolipid metabolism by modulating insulin signaling pathways (e.g., PI3K/AKT cascade response)¹⁵. As a key component of pancreatic β-cell secretory granules, zinc is involved in insulin synthesis, storage, and release¹⁶. Animal models have shown that zinc deficiency reduces GLUT4 transporter expression in adipose tissue and inhibits glucose uptake, leading to a skewed energy metabolism toward fat accumulation¹⁷. Serum copper is involved in adipogenesis by regulating the activity of lipid synthesis-related enzymes (e.g., fatty acid synthase)¹⁸. Copper ions may activate the transcription factor SREBP-1c to promote hepatic lipogenesis, but its specific target in adipocyte differentiation has not been clarified¹⁹. An imbalance in the ratio of copper to zinc may indirectly promote adipocyte hypertrophy by competitively inhibiting zinc-dependent enzymes (e.g., matrix metalloproteinases) and affecting extracellular matrix remodeling²⁰.

The role of zinc on the leptin-hypothalamic axis may be a key node in regulating obesity. Zinc ions enhance leptin receptor sensitivity by modulating JAK2/STAT3 phosphorylation, whereas a low-zinc state may induce leptin resistance and contribute to abnormal feeding behavior^21,22. In addition, zinc deficiency reduces adipose tissue lipocalin secretion and exacerbates insulin resistance²³. The potential role of copper ions on thyroid hormone metabolism is interesting, as copper-dependent enzymes (e.g., thyroid peroxidase) are required for T4 synthesis²⁴. However, whether the high prevalence of hypothyroidism in obese patients is associated with fluctuating serum copper levels still needs to be verified by molecular mechanistic studies, and there is no direct evidence that changes in the copper-to-zinc ratio independently modulate growth hormone or cortisol secretion.

Regulation of copper-zinc superoxide dismutase activity is particularly critical in the chronic oxidative stress associated with obesity²⁵. Visceral fat accumulation induces activation of NADPH oxidase, leading to overproduction of superoxide anion²⁶. Zinc deficiency may lead to structural destabilization of copper-zinc superoxide dismutase, making it less efficient at scavenging superoxide anions, which in turn triggers mitochondrial dysfunction^27,28. In addition, copper may catalyze the generation of hydroxyl radicals in specific microenvironments through the Fenton reaction, directly attacking adipocyte membrane phospholipid structures^29,30. Zinc antagonizes the pro-oxidant effect of copper through metallothionein synthesis, and an imbalance in the ratio of the two may accelerate oxidative damage in adipose tissue³¹. Notably, brown adipose tissue contains higher levels of antioxidant enzymes; however, whether the copper-zinc ratio affects the thermogenesis of brown adipose tissue remains to be explored.

Zinc plays a pivotal role in obesity-associated inflammation through dual regulation of the NF-κB pathway^32,33: on the one hand, zinc inhibits IKKβ kinase activity, blocks IκBα degradation, and reduces pro-inflammatory factor (e.g., TNF-α, IL-1β) release; on the other hand, zinc inhibits TLR4 signaling through upregulation of the A20 protein (ubiquitin-modifying enzyme). Conversely, free copper ions may activate NLRP3 inflammatory vesicles, promote IL-18 secretion, and enhance M1-type polarization of adipose tissue macrophages³⁴. Clinical studies^35,36 have shown a negative correlation between serum zinc levels and adipose tissue zinc content in obese patients. This suggests that zinc may be transferred from the circulatory system to adipocytes to inhibit inflammation locally. However, the temporal and spatial specificity of zinc transport proteins (e.g., ZIP14) in this process is unclear.

Serum copper-zinc ratio, as a biomarker, is expected to provide a new clinical screening tool to assist physicians in the early identification of overweight/obese populations at high risk, especially in children and adolescents during the growth and developmental stages. By monitoring changes in the copper-zinc ratio, physicians can further assess the metabolic risk of obesity and provide a basis for individualized intervention. An imbalance in the serum copper-zinc ratio may indicate increased levels of oxidative stress and disturbances in fat metabolism. These changes provide a physiologic basis for the development of obesity. Based on this physiological mechanism, serum copper-zinc ratios may not only help predict the onset of obesity but may also provide new targets for obesity intervention. For example, in obesity management, adjusting the balance of the copper-zinc ratio may become a new therapeutic idea. Adjusting the copper-zinc ratio through dietary intervention, micronutrient supplementation, or medication may reduce metabolic disorders triggered by oxidative stress, reducing the risk of obesity and its associated complications. The present study demonstrated that the relationship between serum copper-zinc ratios and obesity showed different effects at different copper-zinc ratio levels, especially at copper-zinc ratios lower than 1.561, where the relationship with obesity was particularly significant. This finding suggests that intervention strategies for obesity should be tailored to the individual copper-zinc ratio; for high-risk groups with low copper-zinc ratios, zinc supplementation and modulation of copper intake may be needed to improve the metabolic environment of obesity and reduce the occurrence of obesity-related complications. Individuals with high serum copper-zinc ratios may require more comprehensive nutritional and lifestyle interventions to avoid oxidative stress and metabolic abnormalities associated with high copper-zinc ratios.

Our findings also provide new ideas for developing global public health strategies. In particular, in terms of nutritional interventions, public health authorities can provide targeted nutritional supplementation and health guidance to children and adolescents in different regions and age groups based on the copper-to-zinc ratio level. Encouraging a balanced diet and reducing excessive intake of copper and zinc by popularizing knowledge about copper-zinc balance may help to radically reduce the incidence of obesity and the risk of its associated complications. In recent years, many studies have shown a strong relationship between micronutrient imbalance and obesity. Public health policies can further integrate micronutrient management with obesity control strategies. For example, for overweight and obese children groups, obesity management and micronutrient monitoring can be combined to prevent obesity-induced metabolic diseases through precise nutritional interventions that not only control body weight but also improve their micronutrient metabolism. This integrated intervention strategy improves the effectiveness of intervention and provides a more comprehensive public health program for future obesity prevention and treatment.

In our subgroup analyses, the positive association between serum copper-zinc ratio and overweight/obesity remained stable across sex, age, and passive smoking status, suggesting that this relationship is broadly applicable among diverse pediatric subpopulations. However, we observed significant race and household income interactions, which warrant further discussion. Specifically, the association was not statistically significant in non-Hispanic White children and adolescents or those from higher-income households. This may reflect differences in underlying nutritional status, access to healthcare, dietary patterns, or environmental exposures across socioeconomic and racial groups. For example, higher-income families may have more balanced diets and better trace element regulation through supplementation or food diversity, potentially buffering the metabolic effects of an elevated copper-zinc ratio. Additionally, racial differences in copper and zinc metabolism, inflammatory responses, and obesity susceptibility have been reported, suggesting that genetic and epigenetic factors may modulate the observed associations. These findings highlight the importance of considering socioeconomic and ethnic heterogeneity when interpreting trace element–related metabolic risk. From a public health perspective, it suggests that interventions targeting trace element balance may need to be tailored by population subgroup rather than adopting a one-size-fits-all approach. Future longitudinal studies should explore whether modifying copper and zinc status differentially impacts obesity outcomes across demographic strata.

Although this study revealed the relationship between serum copper-zinc ratio and overweight/obesity through rigorous statistical analysis, we could not determine a causal relationship between the two because the study used cross-sectional data. On the other hand, we used the serum copper-zinc ratio to measure trace element balance in the body. However, the representative failed to directly reflect the intracellular concentrations of copper and zinc. Future longitudinal studies could further explore the causal relationship between changes in copper-zinc ratios and the occurrence of overweight/obesity and incorporate more lifestyle variables to understand this association comprehensively. It could also incorporate cellular-level analyses to reveal further the specific physiological mechanisms underlying the imbalance in the copper-to-zinc ratio. In this study, we did not use NHANES’s complex sampling design. However, this seemingly limited analytical method was scientifically valid and appropriate for the research question after clearly defining the research objectives and fully adjusting for covariates.

Conclusion

For the first time, this study reveals a nonlinear relationship between serum copper-zinc ratio and overweight/obesity in children and adolescents. It confirms the potential threshold effect of serum copper-zinc ratio in the development of obesity. The results showed that elevated serum copper-zinc ratios were positively associated with the risk of overweight/obesity, and the relationship exhibited a significant turnaround at a serum copper-zinc ratio of 1.561, suggesting a potential role in obesity risk assessment. This nonlinear association provides a new perspective on understanding the role of micronutrient imbalance in the development of obesity. In addition, the serum copper-zinc ratio can potentially be a tool for obesity risk assessment as a biomarker of oxidative stress and lipid metabolism imbalance. This study provides theoretical support for developing personalized obesity intervention strategies based on serum copper-zinc ratio levels. It provides new directions for early screening and intervention of obesity in public health policies. Future studies can further explore the mechanistic role of the serum copper-zinc ratio and validate its generalizability to other populations.

Abbreviations

NHANES

National Health and Nutrition Examination Survey

BMI

body mass index

PIR

poverty-to-income ratio

WBC

serum cotinine, white blood cell count

EOS

eosinophil

LYM

lymphocyte

NEUT

neutrophil percentage

Hb

haemoglobin

HDL-C

high-density lipoprotein cholesterol

TC

total cholesterol

Author contributions

C.Z.J. designed the study and drafted the manuscript. S.J.Y. and Z.Z.Y. searched strategy. C.Z.J. and W.Y.Y. completed the statistical analysis. Y.B. and H.C.C. reviewed and modified the manuscript. All authors contributed to the article and approved the submitted version.

Data availability

These survey data are free and publicly available, and can be downloaded directly from the NHANES website (http://www.cdc.gov/nchs/nhanes.htm) by users and researchers worldwide.

Declarations

Competing interests

The authors declare no competing interests.

Ethics and consent to participate

All participants provided written informed consent and study procedures were approved by the National Center for Health Statistics Research Ethics Review Board.