1. Introduction
Exposure to heavy metals, such as lead, cadmium, and mercury, occurs through contaminated food sources and water; furthermore, they can be inhaled from cigarette smoke and other sources of air pollution [
1]. Cadmium exposure among non-occupationally exposed people occurs primarily via smoking tobacco and secondarily by eating foods containing cadmium [
2]. Once inhaled or ingested, cadmium, lead, and mercury are distributed in different tissues and organs [
3,
4]. Cadmium is inefficiently excreted and accumulates primarily in the liver and kidneys [
2]. Tissue stores of cadmium and mercury are slowly excreted from the body via urine and feces at an approximately equal rate, whereas lead is rapidly excreted via urine [
3,
5].
Cadmium affects the absorption of key divalent cations for bone metabolism such as calcium [
1]. There have also been relationships described between several essential elements such as Ca, Fe and Zn that affect their absorption, excretion and tissue retention [
1]. Cadmium is also able to interfere with the parathyroid hormone stimulation of vitamin D activation in kidney cells, to increase urinary exretion of Ca, reduce its absorption from the intestines, and to interfere with Ca incorporation into bone cells [
6]. Dermience et al. [
7] recently reviewed and summarized the toxic effects of lead and cadmium on bone metabolism; their study also highlighted the currently unknown effect of mercury on human bone metabolism and the need for further investigation about the possible effects of mercury on bone metabolism. Lead has been related to increased bone turnover and reduced mineralization, a decrease in bone mineral density (BMD) and mass as well as a cause of osteoporosis in the most severe cases [
7]. Lead (Pb
2+) can substitute to Ca (2+) in hydroxyapatite crystal and additionally lead has a higher affinity for osteocalcin than calcium [
7,
8]. Cadmium has been associated with a decrease in BMD, cadmium interacts with calcium metabolism and causes hypercalciuria, increased risk of fracture, osteomalacia and osteoporosis and chronic cadmium exposure causes Itai-itai disease [
7,
9], which is associated with weak and brittle bones. There is also evidence that cadmium disturbs calcium metabolism and calciotropic homones; cadmium decreases liver concentration of other elements such as iron, magnesium, and selenium, and increases levels of copper, zinc and manganese [
7].
Previous studies have shown that femur T-scores are associated with the accumulation of cadmium, and this association is gender-specific [
5,
10,
11,
12]. Urine cadmium levels have also been associated with osteoporosis [
13,
14] and dietary intake [
15,
16]. Even low-level cadmium exposure from food has been associated with low BMD and an increased risk of osteoporosis and fractures [
17]. Blood levels of lead, mercury and cadmium are negatively associated with BMD, and this association is gender-specific [
18].
Few studies have investigated the extent to which dietary heavy metals are associated with low BMD [
17].
We aimed to characterize the dietary intake of the heavy metals lead, cadmium and mercury among healthy, non-smoking postmenopausal women (the population stratum with high cadmium retention) [
19] in Spain. Furthermore, we sought to establish a putative relationship between bone health and the intake of these heavy metals in similar groups of women.
2. Materials and Methods
2.1. Participants
Healthy postmenopausal women were recruited from the local area via internet advertising and primary care consults. To be eligible for this study, all women had to be healthy, reside in the community, be of white European origin and have no mental or physical functional impairments.
The University of Extremadura Ethical Advisory Committee approved this study. All participants provided written informed consent in accordance with the 1975 Declaration of Helsinki.
We aimed to have enough power to detect a clinically significant 5.4% change in lumbar spine BMD [
20]. A sample size of at least
n = 228 (two groups of 114) was required [
21] to achieve a statistical power of 80% and
p < 0.05. A total of 281 postmenopausal women were included in this study.
All of the women resided in the urban area of the health district of Caceres, Spain. These women underwent primary or secondary examinations. Most of them were married and had children, and their social status was average. None of the participants had dietary restrictions, neurological impairments, or physical disabilities, and their medical histories showed no presence of low-trauma fractures.
We recorded participants’ complete medical histories and physically examined each woman before enrollment in the study. None of the women were taking medications that could interfere with calcium metabolism (e.g., corticoids, oral anticoagulants, antipsychotics, etc.). All of the women led active lives but did not regularly exercise. Alcohol intake was sporadic and did not exceed 100 mL/day in any case. None of the women smoked. Height was measured using a Harpenden stadiometer with a mandible plane parallel to the floor, and weight was measured using a biomedical precision balance scale. Both measurements were determined when the participants were wearing only light clothing and no shoes. Body mass index (BMI) was calculated as the weight in kilograms divided by the square of the height in meters (kg/m2).
2.2. Bone Measurements
An ultrasound was performed on the 2nd to the 5th proximal phalanx of the non-dominant hand using a DBM Sonic Bone Profiler (IGEA, Capri, Italy).
The femoral neck and L2–L4 spine BMDs were measured via dual-energy X-ray absorptiometry DXA (Norland XR-800, Norland Inc., Fort Atkinson, WI, USA) and expressed as the quantity of mineral (g) divided by the area scanned (cm2).
pQCT measurements were performed on the non-dominant distal forearm using a Stratec XCT-2000 device (Stratec Medizintechnik, Pforzheim, Germany).
2.3. Assessment of Diet and Covariates
According to Food and Agriculture Organization/World Health Organization (FAO/WHO) recommendations (WHO, 1985), three basic approaches are employed to assess the intake of food contaminants or other dietary elements: (a) total diet studies (TDSs), (b) duplicate diet studies, and (c) diary studies that combine the data for specific contaminants with individual (or household) consumption records (Perello et al., 2014). Women enrolled in this study completed a 131-item food frequency questionnaire (FFQ). This FFQ was previously validated and involves 24-h recall performed over seven days [
22,
23,
24,
25,
26]. A food cadmium, lead and mercury database was constructed based on the cadmium, lead and mercury contents previously reported with regard to the Spanish market [
27]. Significant differences were not observed in the dietary patterns across geographical areas of Spain; however, the quantities consumed differed greatly [
28]. The daily intake of the elements considered for the different food groups was calculated by accounting for food content and consumption. The toxic element concentrations of the different food groups were taken from the literature [
28]. Using the FFQ, we also assessed the dietary intake of calcium and vitamin D. Information regarding calcium and vitamin D originated from the Spanish Food Composition database [
29].
2.4. Statistical Analyses
Medians and the interquartile ranges were used to describe the sample.
Because of the asymmetric distribution of many of the studied variables (i.e., age, Ad-SoS, FN BMD, WT BMD, L2 BMD, L3 BMD, L2–L4 BMD, total area mm2, cortical area mm2, daily Cd intake, vitamin D intake, Ca intake, Fe intake, Mg, intake and Kcal intake) and the presence of atypical values (i.e., outliers), the non-parametric Wilcoxon test was used to evaluate the differences between groups with regard to the studied variables. To adjust for potential confounds, we used a non-parametric rank analysis of covariance model, where heavy metal intake was considered a factor, and kcal intake was considered a covariate. All statistical tests were conducted in SPSS version 22.0 (IBM Corp., Armonk, NY, USA).
4. Discussion
Our sample of postmenopausal women had a dietary cadmium intake that was notably lower than the provisional tolerable weekly intake (PTWI) established by the Joint FAO/WHO Expert Committee on Food Additives (JECFA; 7 μg/kg b.w./week) [
30]. In 2009, however, the European Food Safety Authority (EFSA) [
31,
32] reevaluated data on dietary cadmium intake and set a new PTWI of 2.5 μg/kg b.w. (0.357 μg/kg b.w./day) [
33]. Approximately 66% of the sample exceeded the threshold for this element.
JECFA has established a PTWI for lead of 25 μg/kg b.w.; however, the EFSA [
34] concluded that the former PTWI (given as μg/kg b.w.) was not appropriate because no evidence of a critical threshold for lead-induced effects exists. In 2011, the JECFA concluded that because of prior analyses, a critical threshold would be considered health protective [
33]. Nevertheless, all of the women had values well below the PTWI.
With respect to mercury, the EFSA recommends a maximum intake of 4 µg/kg b.w./week [
33], and only 0.7% of the sample exceeded this threshold. Previous studies [
27,
35] in different areas of Spain studied the intake of heavy metals and other potentially toxic materials. The results obtained regarding the average dietary intake of lead ranged from 4 µg/kg b.w./week to 56 µg/kg b.w./week among the four areas studied. These results are similar to the dietary intakes assessed in our area (4.8 µg/kg b.w./week).
In 2014, the temporal trends of the dietary intakes of cadmium, lead and mercury (2000, 2005, 2008 and 2012) were estimated for Catalonia, Spain [
33]. The dietary intakes for cadmium, mercury, and lead in this region were 0.87 µg/kg b.w./week, 1.1 µg/kg b.w./week and 0.84 µg/kg b.w./week, respectively; these figures are below the values observed in our sample. The data from our study confirm previous studies in different areas of Spain showing that the dietary intakes of heavy metals in the Spanish diet are generally within the recommended limits [
36,
37,
38,
39,
40,
41,
42,
43]. Moreiras and Cuadrado [
28] examined the estimated intake of heavy metals in the diets of people from our specific area of Spain in 1992 and indicated that Extremadura showed the theoretically lowest intake of cadmium but the highest intake of mercury in all of Spain. It is possible that the dietary habits of our area have changed since then, thereby changing the trend in the dietary intake of the studied heavy metals.
According to Perelló and colleagues [
33], the major dietary sources of cadmium are cereals and fish; these conclusions corroborate our results. Similar results were found for the intake of mercury: fish was the major dietary contributor. Cereals made the greatest contribution to total dietary lead intake in three of the four areas studied; this result supports our observation about the contribution of cereals to the dietary intake of lead in the current sample. A previous study also showed that cereals are a major source of dietary lead and cadmium in Spain [
44].
An association exists between dietary cadmium exposure and higher rates of bone fracture (including hip fracture) independent of tobacco smoking in men [
45]. Similarly, even low-level exposure to dietary cadmium has been associated with bone fragility among postmenopausal women [
17]. Recently, positive associations with osteoporosis-related incident fractures were described in a cohort of elderly Swedish men, and these associations were also described in men who had never smoked but who were exposed primarily through their diets [
46]. These results show that older men with relatively low dietary cadmium exposure are also at an increased risk of low BMD and fracture associated with cadmium.
Although high cadmium exposure causes bone damage, the association between low-level cadmium exposure (i.e., dietary exposure) and bone health must be clarified, especially in women [
47]. Little evidence exists on the associations between BMD status and cadmium, lead or mercury intake among osteopenic or osteoporotic Korean adults [
48]. However, the negative effects of low-level cadmium exposure on bone, possibly exerted via increased bone resorption, seem to intensify after menopause [
47]. We did not observe any association between BMD and dietary cadmium intake in our cohort of Spanish women. Recent research has focused on the putative roles that cadmium plays in volumetric BMD and bone morphometry. In vivo studies of Sprague-Dawley male rats have shown that cadmium exposure can induce low vBMD [
49] by decreasing the trabecular number, thereby reducing the quantity of mineralized bone tissue.
Similarly, conflicting evidence suggests that bone lead or blood lead reduces areal BMD. Recent data have shown that bone lead accumulated from exposure over time can detrimentally affect bone by reducing cortical thickness and integral volumetric bone density [
50] in postmenopausal women. We observed a non-significant difference in the trabecular density between the groups of women with either high or low dietary intakes of lead; the trabecular density of the higher dietary cadmium intake group was greater. We believe that this result deserves additional study because it might support the hypothesis that lead affects volumetric BMD.
The strengths of our study include its use of a heavy metals database based on concentrations in foods sold on the Spanish market as well as the assessment of BMD via DXA (the reference standard for BMD measurements), volumetric BMD and bone morphometry via pQCT and bone quality via QUS.
We recognize that our study has several limitations. First, its cross-sectional design does not allow us to establish causal relationships. Second, although we adjusted for calorie intake and no differences were observed in the major determinants of bone density between the studied groups (age, BMI, calcium intake, vitamin D intake, or smoking status), we cannot completely exclude the possibility that our findings are biased by unmeasured or residual confounds. Additionally, we aimed to have enough power to detect a least significant change in spine BMD of 5.4% that was established for intragroup comparisons and this could be conservative. Therefore, our study might have a moderate risk of finding no difference when in fact a difference might exist below the indicated threshold. Finally, dietary estimation of cadmium exposure should be used with caution in our study because of the lack of association between estimated dietary cadmium and measured urinary cadmium exposure.