Parathyroid hormones in relation to serum cadmium and lead: the NHANES 2003–2006

The toxic effects of cadmium and lead exposure on parathyroid function have been investigated extensively while the results were conflicting. We aimed to investigate the association of blood cadmium and lead with serum parathyroid hormone (PTH) based on the general population. We analyzed a sample of 9400 adults aged 18 years or older from the US National Health and Nutrition Examination Survey 2003–2004 to 2005–2006. We estimated the association using multivariable linear regression analysis by taking cadmium and lead as a continuous or quartile variable. Restricted cubic splines were utilized to explore the non-linear relationship between cadmium or lead and PTH. The median blood cadmium and lead levels were 0.34 µg/L and 1.43 µg/dL, respectively. Cadmium was significantly associated with PTH when taken as a continuous variable [Odds ratio (OR): 0.97; 95% confidence interval (95% CI): 0.95–0.99] or a quartile variable (fourth vs first quartile, OR: 0.94; 95% CI: 0.91–0.97). Lead was significantly associated with PTH when taken as a continuous variable (OR: 1.04; 95% CI: 1.02–1.07) or a quartile variable (fourth vs first quartile, OR: 1.06; 95% CI: 1.01–1.10). P values for non-linearity were all less than 0.05. The association between cadmium and PTH was not significant in females. Cadmium was negatively associated with PTH, and lead was positively associated with PTH both in a non-linear manner among the general population. However, there was a gender difference in the relationship between cadmium and PTH.


Introduction
Heavy metals have detrimental effects on human health, and exposure to these metals from polluted soil, water, and air is a global problem with rapid industrial development and modernization. Cadmium and lead are among the most studied heavy metals and their toxicity has been recognized for a long time (Okereafor et al. 2020;Vries et al. 2007). Due to their uses in a wide variety of products, such as leaded gasoline, batteries, pipes, paint, mining wastes, and household items, cadmium and lead have been increasingly presented in the human environment. Unlike organic contaminants, they are not biodegradable. They can enter the human body through digestive, respiratory, and dermal routes and easily accumulate in body organs and living organisms (Balali-Mood et al. 2021;Ebrahimi et al. 2020). Exposure to cadmium and lead has been linked to a variety of detrimental effects on human health. For example, there are toxic effects of cadmium and lead on the immune system and cancer progression and both of them have potent endocrine-disrupting activity (Ebrahimi et al. 2020;Iavicoli et al. 2009;Kortenkamp 2011).
Parathyroid hormone (PTH) is a polypeptide that is synthesized and cleaved into an active form within the parathyroid gland. It is an important hormone that regulates serum calcium concentration. Cadmium and lead were reported to affect the blood level of PTH in different studies. For example, Ibrahim et al. (2016) investigated the possible effect of occupational cadmium exposure on the parathyroid gland in workers and found that biological cadmium exposure indices correlated negatively with plasma PTH levels. Similarly, Kristal-Boneh et al. (1998) sought to clarify the association between occupational exposure to lead and serum concentration of PTH in a cross-sectional study with 146 individuals and found that subjects occupationally exposed to lead showed a substantial compensatory increase in PTH. However, most of these studies are limited by small sample size and the results are conflicting (Babić Leko et al. 2022). Earlier studies proved that cadmium was not associated with PTH (Järup et al. 1995;Kido et al. 1991;Nogawa et al. 1984;Tsuritani et al. 1992) or was positively associated with PTH (Koji Nogawa et al. 1987). However, later studies demonstrated that cadmium was negatively associated with PTH (Åkesson et al. 2006;Engström et al. 2009;Ibrahim et al. 2016;Rignell-Hydbom et al. 2009;Schutte et al. 2008). A recent study showed that the association of cadmium in erythrocytes with PTH was totally different in boys than that in girls (Malin Igra et al. 2019). In addition, little information is available about the possible effects of cadmium and lead on PTH in the general population having long-term low levels of exposure.
To address these knowledge gaps, we aimed to explore whether blood cadmium and lead are associated with serum PTH based on the participants from a nationally representative survey, the National Health and Nutrition Examination Survey (NHANES). We used NHANES 2003-2006, which was a cross-sectional, complex sample survey that included an inhome interview, physical examination, and laboratory tests to collect data on demographics, health, and nutritional status of the US civilian noninstitutionalized population. The NHANES study protocols were approved by the National Center for Health Statistics Institutional Review Board, and written informed consent was obtained from all participants.

Study population
We limited our participants to those aged 18 years or older having information on the measurement of blood PTH, cadmium, and lead level. There were 9965 participants who met these inclusion criteria. We further excluded 565 women who were pregnant. This resulted in a total of 9400 subjects available for analysis.

Measurement of PTH, cadmium, and lead
Blood samples were collected during the physical examination, stored frozen (− 20 °C) and sent to the central laboratory for the test. Measurement of serum PTH was based on immunological tests as described on the NHANES website (National Health and Nutrition Examination Survey 2005-2006a Data Documentation, Codebook, and Frequencies, Parathyroid Hormone (PTH_D)). Whole blood cadmium and lead concentrations are determined using inductively coupled plasma mass spectrometry. The detection limit for cadmium was constant in the data set. In cases where the result was below the limit of detection, the value for that variable is the detection limit divided by the square root of two (National Health andNutrition Examination Survey 2005-2006b).

Covariates
Important covariates were identified in the literature on PTH, and included in multivariate analyses were age, gender, race/ ethnicity, body mass index (BMI), excessive alcohol use, estimated glomerular filtration rate, physical activity, and serum cotinine, calcium, and vitamin D level. Demographic and health-related information was collected by standardized questionnaires during the in-home interview. Race/ethnicity was self-reported and categorized as Mexican American, non-Hispanic white, non-Hispanic black, and others. BMI was determined as the weight in kilograms divided by height in meters squared. Excessive alcohol use was defined as drinking, on average, ≥ 5 drinks/day in the 12 months preceding the survey interview. The estimated glomerular filtration rate (eGFR) was calculated based on serum creatinine by using the Chronic Kidney Disease Epidemiology Collaboration equation (Levey et al. 2009). Calibrations for the measurements of serum creatinine in different surveys were conducted according to the recommendation of NHANES if necessary. Physical activity during NHANES 2003NHANES -2004NHANES and 2005NHANES -2006 was measured with the ActiGraph AM-7164 accelerometer (ActiGraph, LLC, Fort Walton Beach, FL). The accelerometers measure the movement and its intensity in activity counts per minute. We calculated the average minutes of moderate to vigorous physical activity (MVPA) per week in the current study. The MVPA was defined as counts per minute of 2020 or higher (Fishman et al. 2016). Serum cotinine, calcium, and vitamin D levels were measured in the central laboratory with blood samples collected during the physical examination.

Statistical analysis
To account for oversampled population, appropriate 4-year sampling weights were constructed according to the NHANES recommendation. All statistical analysis was conducted in R version 4.1.3 with the "Survey" package after accounting for the complex sampling design. All statistical tests were 2-sided, and P < 0.05 was considered statistically significant. PTA, cadmium, and lead were log-transformed as they were right-skewed. Continuous variables were expressed as means (standard error) and categorical variables as percentages (standard error).
Linear regression was applied to evaluate the relationship of PTH with cadmium and lead. First, cadmium and lead were modeled as continuous exposures after log transformation. Second, they were divided into quartiles based on the weighted distributions and were modeled as ordinal variables with the first quartile as the reference. Third, we estimated the non-linear relationship between cadmium or lead and PTH using restricted cubic splines (RCS) with 4 knots in the regression model. The locations are the 5th, 35th, 65th, and 95th percentiles of the level of cadmium and lead according to the recommendation (Austin et al. 2022;Desquilbet and Mariotti 2010). The actual values of cadmium corresponding to these four knots were 0.1, 0.28, 0.50, and 1.72 µg/L, respectively. The actual values of lead corresponding to these four knots were 0.53, 1.10, 1.82, and 4.20 µg/dL, respectively.
P-values for non-linear trends were obtained by Wald tests for RCS coefficients. We developed three models for the regression analysis. In model 1, no covariates were adjusted. In model 2, age, gender, and race were adjusted. In model 3, BMI, excessive alcohol use, eGFR, minutes of MVPA per week, and level of cotinine, calcium, and vitamin D were further adjusted on the basis of model 2.

Results
The characteristics of participants by quartiles of blood cadmium and lead are presented in Tables 1 and 2. The median level of blood cadmium was 0.34 µg/L (interquartile range: 0.2-0.61 µg/L). Compared to participants with the lowest level of cadmium, those with the highest level were older and less likely to be male, had lower BMI, were more likely to be excessive alcohol users, and had lower eGFR, minutes of MVPA per week, and higher cotinine (Table 1). The median level of blood lead was 1.43 µg/dl (interquartile range: 0.93-2.20 µg/dL). Compared to participants with the lowest blood lead level, those with the highest level were older and more likely to be male, had lower BMI, were more likely to be excessive alcohol users, and had lower eGFR and higher cotinine and calcium ( When the blood cadmium level was divided into quartiles, participants with the highest quartile of cadmium level had lower PTH compared to those with the lowest quartile in model 1 (OR: 0.93; 95% CI: 0.91-0.96), model 2 (OR: 0.87; 95% CI: 0.85-0.89), and model 3 (OR: 0.94; 95% CI: 0.91-0.97). Similarly, when the blood lead level was divided into quartiles, participants with the highest quartile of lead level had higher PTH compared to those with the lowest quartile in model 1 (OR: 1.12; 95% CI: 1.08-1.17) and model 3 (OR: 1.06; 95% CI: 1.01-1.10) (Table 3).
Generally, RCS showed that cadmium was negatively associated with PTH in all three models and the P values for non-linearity were significant (Fig. 1). Conversely, RCS showed that lead was positively associated with PTH in all three models and the P values for non-linearity were all less than 0.05 (Fig. 1).
Women had a higher level of blood and urine cadmium and blood PTH than men and they had a lower level of blood and urine lead than men (Table I in Supplemental Material). We stratified our analysis by gender. For the male, the association between cadmium and PTH was significant in model 3 when taking cadmium as a continuous or quartile variable or using the RCS in the regression analysis. For the female, the association between cadmium and PTH was not significant in model 3 in all the aforementioned regression analyses (Fig. 2). For the male, the association between lead and PTH was significant in model 3 when taking lead as a continuous variable in the regression analysis. For the female, the association between lead and PTH was significant in model 3 in all the aforementioned regression analyses (Fig. 2).

Discussion
The current study used data from a representative sample of US adults to investigate the association of low blood cadmium and lead levels with serum PTH. The results suggested that cadmium was negatively associated with PTH and lead was positively associated with PTH both in a nonlinear manner. Their relationships remained significant after adjustment for calcium and vitamin D, as well as confounding factors for PTH estimation. However, the potential role of cadmium exposure on parathyroid function may differ by sex.
The effects of cadmium by occupational or high environmental pollution exposures on parathyroid function have been investigated by several studies with conflicting results (Babić Leko et al. 2022). Some studies also focused on the effects of a low level of cadmium on PTH among healthy adults, and the results were consistent that an increased cadmium level was related to a decreased PTH level (Åkesson et al. 2006;Engström et al. 2009;Schutte et al. 2008). However, these studies were limited to female participants and the effects among the general male participants were not investigated. Contrary to the above results from the general population, we found that the relationship between cadmium and PTH was not significant among females after adjustment of all confounding factors. This result was consistent with one study that the association of cadmium with PTH was not significant after inclusion of smoking in the model (Rignell-Hydbom et al. 2009). The cause of these differences might be different characteristics of participants or different included covariables. One meta-analysis reported changes in PTH levels after occupational lead exposure (Upadhyay et al. 2022). All the included studies but one observed lower serum PTH levels among lead exposed as compared to the control group. Pooled results showed that lead exposed group revealed a mean 37.97 pg/ml lower PTH level compared to the control Fig. 1 Odds ratio (95% confidence interval) of serum parathyroid hormone level by blood cadmium and lead level. Odds ratio (solid lines) and 95% confidence interval (curved lines) were based on the restricted cubic splines for log-transformed cadmium and lead level with 4 knots. No variates were adjusted in model 1. Age, gender, and race were adjusted in model 2. Body mass index, excessive alcohol use, estimated glomerular filtration rate, minutes of moderate to vigorous physical activity per week, and level of cotinine, calcium, and vitamin D on the basis of model 2 were further adjusted in model 3

Fig. 2
Odds ratio (95% confidence interval) of serum parathyroid hormone level by blood cadmium and lead level in stratified analysis by gender. Odds ratio (solid lines) and 95% confidence interval (curved lines) were based on the restricted cubic splines for log-transformed cadmium and lead level with 4 knots. Age, gender, race, body mass index, excessive alcohol use, estimated glomerular filtration rate, minutes of moderate to vigorous physical activity per week, and level of cotinine, calcium, and vitamin D were adjusted group. However, the results were statistically not significant and there were unacceptable levels of heterogeneity. The relationship between low lead level and PTH has also been reported among the general population and dialysis patients. While some showed no significant relationship (Åkesson et al. 2006;Osterloh, 1991), most of them supported that there was a positive relationship between them (Lin et al. 2010;Mazumdar et al. 2017). It is also interesting to notice that the relationship was significant in men but not significant in women (Åkesson et al. 2006;Kristal-Boneh et al. 1998).
An experimental animal study observed increases in PTH levels after cadmium exposure (Brzóska and Moniuszko-Jakoniuk 2005). In this study, young male rats received drinking water containing cadmium for 12 months. Results showed that rats' exposure to cadmium disturbed the process of bone turn over and bone mass accumulation, which led to the formation of less dense than normal bone tissue. The effects were accompanied by an increase in the serum concentration of PTH. Based on a population study, Schutte et al. explained serum PTH levels decreased with higher cadmium exposure as a consequence of the direct osteotoxic effect of cadmium. This decrease might be expected when a toxic substance such as the cadmium-induced release of calcium from bone tissue (Schutte et al. 2008). Notwithstanding, the opposite results between animal study and population study suggest that there might be another factor or mechanism mediating the relationship between cadmium and PTH.
Our study indicated a gender difference in the relationship of cadmium with PTH. Consistent with our results, it has been known for a long time that women, in general, have higher concentrations of cadmium in blood and urine than men (Vahter et al. 2002). Epidemiological studies showed that women were at higher risk for cadmium-induced bone effects than men (Staessen et al. 1999). There are also indications that cadmium-related health effects are more common among women than among men. Nishijo et al. (2004) suggested that these gender differences might be due to the difference in renal tubular dysfunction, calcium metabolism and its regulatory hormones, kidney sensitivity, body iron store status, and genetic factors between women and men. Pregnancy might also be an important factor (Nishijo et al. 2004). However, our study found that the relationship between cadmium and PTH was not significant after adjustment for confounding factors among the female participants. Further studies might be warranted to explore the potential mechanisms or reasons.
PTH regulates the serum calcium concentration by its effects on bone, kidney, and intestine, while the PTH levels are controlled by the calcium levels in a feedback manner (Goltzman 2018). PTH stimulates the release of calcium into the blood from the large reservoir contained in the bones (Wojda and Donahue 2018). PTH also influences the reabsorption of calcium so that the kidney retains more calcium. Moreover, PTH enables the production of active vitamin D in the kidney, and this active form of vitamin D increases the absorption of calcium by the intestine via calbindin. Cadmium could increase the serum level of calcium as both epidemiological and animal studies demonstrate that cadmium has a direct osteotoxic effect and bone demineralization begins after the start of cadmium exposure (Järup and Alfvén 2004;Staessen et al. 1999;Wilson and Bhattacharyya 1997). Thus, the serum PTH levels decrease with higher cadmium exposure when cadmium induces the release of calcium from bone tissue. Conversely, lead decreases the serum calcium levels by impairing the activation of vitamin D as it can inhibit the 1-α-hydroxylase enzyme in renal tubules, which is required for the synthesis of vitamin D (Anetor et al. 2005;Batra et al. 2020).
However, our study showed that cadmium level was not associated with serum calcium level and lead level was positively associated with serum calcium level. These results were contrary to the aforementioned mechanism which cadmium and lead affect the PTH by the serum calcium. Factors other than calcium might have been involved in the association of cadmium and lead with PTH. For example, it has been established that phosphate regulates the PTH, independent of its effect on serum calcium and vitamin D (Marks et al. 1996). Low phosphate can lead to a marked decrease in PTH mRNA and PTH levels (Kilav et al. 1995) and a higher blood level of cadmium was associated with a lower level of phosphate (Hsu et al. 2014). Notwithstanding, we are not able to verify this mechanism as the serum phosphate level was not available in NHANES and more studies are needed to explore the effects of long-term low-level exposure of cadmium and lead on PTH.
This study has several limitations that should be noted. First, this is a cross-sectional study and any causal interferences of cadmium or lead on PTH could not be made. Second, the information on cadmium, lead, and PTH was collected from only one blood test. Several blood tests might be more reliable than a single test. What's more, the blood lead might not be a better indicator than bone lead as to the long-term lead exposure since bone lead has a half-life of decades, while that of blood is of only 30 days (Bouchard et al. 2009). Another key limitation of this study was the lack of data on the duration of exposure, which has been demonstrated as an important mediator of the serum PTH level. Nonetheless, this study has several strengths, including large sample size, reliability of NHANES data, and adjustment for important covariables. Moreover, we detected a non-linear relationship between cadmium or lead and PTH, which was not reported in the previous studies.
In conclusion, the current study showed that blood cadmium was negatively associated with serum PTH and blood lead was positively associated with serum PTH both in a non-linear manner among the general population. However, there was a gender difference in the relationship between cadmium and PTH.
Author contribution Sangzi Jiang designed the study. Huan Zhang collected the data and performed the statistical analysis. Wenbiao Ma and Yan Dong searched the literature and prepared the manuscript. Bo Shi reviewed the results and manuscript.
Data availability All data are from NHANES and are openly available on https:// www. cdc. gov/ nchs/ nhanes/ index. htm.

Declarations
Ethical approval The NHANES study protocols were approved by the National Center for Health Statistics Institutional Review Board.
Consent to participate Written informed consent was obtained from all participants.

Consent for publication
All authors agreed on the publication of this manuscript.

Competing interests
The authors declare no competing interests.