Introduction

Idiopathic hypercalciuria (IH) was first described by Albright et al. [1], who defined it as an excessive urinary calcium loss accompanied by normal serum calcium levels. IH is the most common metabolic abnormality in patients with nephrolithiasis, accounting for 30–50% of calcium-oxalate stone formers [2–4].

The pathogenesis of IH is not yet fully understood. However, it is generally considered that IH is caused by an alteration in calcium homeostasis at sites where large amounts of calcium must be precisely controlled [5]. Several studies have shown decreased bone mineral density (BMD) in patients with IH [6–18]. This progressive decrease in bone mineral content suggests that osteoclasts and osteoblasts might play a key role in the chain of events leading to hypercalciuria. The function of osteoblasts and osteoclasts and the resulting balance between bone formation and resorption are regulated by multiple mediators with the participation of cytokines [19–25]. In IH, cytokines may be responsible for triggering specific alterations to bone metabolism, which in turn contribute to the development of excessive bone remodeling, with the possible predominance of bone mass resorption over formation [19–25].

In the setting of IH, we hypothesized that the measurement of cytokines as non-invasive biomarkers could improve the diagnostic capability or help determine the risk of persistent hypercalciuria and of bone mineral loss. Therefore, we measured plasma and spot-urine levels of interleukin 1 beta (IL-1β), interleukin 6 (IL-6), interleukin 8 (IL-8), tumor necrosis factor α (TNF-α), transforming growth factor β1 (TGF-β1) and monocyte chemoattractant protein (MCP-1) in children and adolescents with IH and in sex- and age-matched healthy subjects.

Patients and methods

Study design

This cross-sectional study included patients with confirmed diagnosis of IH and a group of sex- and age-matched healthy subjects as controls.

Patients with IH

This group included a sample of children and adolescents with well-established IH, followed up at the Pediatric Nephrology Unit of our institution from 2009 to 2011, whose parents gave their consent to participate in the study protocol. Hypercalciuria was defined by serum calcium within normal limits and 24-h urinary excretion of calcium equal to or higher than 4 mg/kg per day for both genders in two nonconsecutive samples, on an unrestricted diet [26, 27]. In order to define the diagnosis of IH, all patients were submitted to a systematic protocol to investigate the possibility of hypercalciuria secondary to diseases and conditions that might affect urinary calcium excretion, as described elsewhere [26, 28]. Briefly, the protocol included: blood gas analysis, serum electrolytes, erithrogram, urea, creatinine, uric acid, PTH, TSH, T4, spot urine, and 24-h urinary concentrations of calcium, citrate, uric acid, oxalate, cystine, and creatinine [26, 28]. Patients with known diseases or use of medication that could affect calcium excretion, bone remodeling or monocyte function were excluded [6]. Therefore, a total of 81 patients with confirmed IH were invited to participate in the study. From this group, 11 patients refused to participate. The remaining 70 patients were then divided in two subgroups according to their urinary calcium excretion at the time of urine and blood sample collection. Patients with calcium excretion equal or superior to 4 mg/kg/day were allocated to the persistent IH group (n=27). Patients with calcium excretion below 4 mg/kg/day were allocated to the controlled IH group (n=43) [26].

Controls

The control group consisted of sex- and age-matched healthy subjects from our Pediatric Primary Care Center. Healthy status was determined through the subjects’ medical history and either a parental report or self-report to rule out the presence of chronic or acute diseases.

Study protocol

All participants were interviewed and underwent physical examination at the time of blood and urine collection. Blood and urine samples were obtained simultaneously in patients and controls. Age, gender, race, weight, height, body mass index, systolic and diastolic blood pressure, serum creatinine, calciuria, citraturia, phosphaturia, magnesuria, BMD, family history of nephrolithiasis, presence of calculus, past history of extracorporeal shock lithotripsy, and symptoms were analyzed in all IH patients. The study protocol did not interfere with medical prescriptions for IH. Treatment basically included the long-term administration of potassium citrate (0.5 to 1 mEq/kg/day) associated with hydrochlorothiazide (0.5 to 1 mg/kg/day) for 3 to 6 months to control high urinary calcium excretion. Medical prescriptions were also recorded for analysis.

Regarding the control group, clinical variables (age, gender, race, weight, height, body mass index, and systolic and diastolic blood pressure) were also obtained. In order to exclude from the control group all subjects with impaired renal function and with increased urinary calcium excretion, plasma creatinine and spot-urine levels of creatinine and of calcium were assayed in the same sample collected for cytokine measurements. Therefore, all controls included in the study exhibited plasma creatinine between reference values and spot-urine calcium:creatinine ratios lower than 0.20.

Bone mineral density was assessed by dual energy X-ray absorptiometry at the lumbar spine (L1–L4) using a Lunar Prodigy Primo DXA System (GE Healthcare Lunar, Madison, WI, USA) in 46 patients with IH. Bone density was expressed in g/cm2 and was also stratified as Z-score >−1 SD and ≤−1 SD according to previous studies [12, 17, 18, 29–31]. Z scores were calculated in relation to a population of individuals of the same gender, age range, and ethnicity [12, 17, 18, 29–31].

Blood sampling

After informed consent, all subjects were subjected to blood collection. Blood sampling occurred on only one occasion, at the same time as other routine examinations. The samples were collected into sterile citrate tubes, which were immediately immersed in ice, and processed within 30 min of collection. Cells were sedimented by centrifugation at 700 × g for 10 min at 4°C. Then the supernatant was collected and re-spun for another 20 min at 1,300 g to sediment platelets. Cell-free plasma was aliquoted into 0.5-mL samples and stored at −80°C until measurement.

Urine sampling

A single urine sample was obtained from all patients on the same day as blood collection from 7.30 a.m. to 9.00 a.m. After homogenization, 10 mL of the collected urine were centrifuged at 4°C for 20 min at 1,300 g. Cell-free urine was aliquoted into 0.5-mL tubes and stored at −80°C until measurement.

Cytokines measurement

Plasma and urinary levels of IL-1β, IL-6, IL-8, TNF-α, TGF-β1, and MCP-1 were measured by specific enzyme-linked immunoassay (ELISA) kits (R&D Systems, Minneapolis, MN, USA), following the manufacturer’s instructions, as described elsewhere [32]. Urine cytokine levels were expressed as absolute concentrations (pg/mL) as well as concentrations standardized for urine creatinine measured in the same urine spot (pg/mg cr). All samples were assayed in duplicate in two separate assays with inter-assay variation below 5%. Our intra-assay variation for the ELISA measurements was below 3%. Specifically for the measurement of TGF-β1, we used a Quantikine kit (R&D Systems). Samples were activated before the TGF-β1 assay. Sample activation basically comprised biochemical steps (acidification followed by neutralization of the pH) in order to activate latent TGF-β1 to immunoreactive TGF-β1 detectable by the Quantikine TGF-β1 immunoassay, as recommended by the manufacturer and previously described [32]. The detection limits were 0.1 μg/mL (IL-1β), 0.039 pg/mL (IL-6), 6 pg/mL (IL-8), 0.106 pg/mL (TNF-α), 6 pg/mL (TGF-β1), and 8 pg/mL (MCP-1).

Statistical analysis

Values are expressed as medians and interquartile range (25th percentile, 75th percentile) or means and standard deviation (SD), when appropriate. The Mann–Whitney and the Kruskal–Wallis tests were used to compare nonparametric continuous variables. Means were compared using non-paired Student’s t test. Dichotomous variables were compared using the two-sided Fisher's exact test. Correlation among plasma cytokines, urinary cytokines, and BMD was performed using a nonparametric test (Spearman rank correlation test). The level of significance was set at p < 0.05.

Ethical aspects

The Ethics Committee of the Federal University of Minas Gerais approved the study. Informed consent was obtained from all parents and, when appropriate, also from the included patients and healthy controls. The research protocol did not interfere with any medical recommendations or prescriptions. Subject follow-up was guaranteed even in cases of refusal to participate in the study.

Results

General clinical characteristics

A total of 70 patients with IH and 37 healthy controls were included in the analysis. Clinical and laboratory characteristics were obtained at the same time as the cytokine measurements and summarized in Table 1. No differences were observed in general clinical characteristics among patients with persistent IH, controlled IH, and healthy controls (Table 1). Both subgroups of patients with IH and healthy controls were normotensive and had normal serum creatinine levels at the time of sample collections (Table 1). Except for the increased prescription of hydrochlorothiazide in patients with persistent IH (p<0.05), there were no other differences in clinical and laboratory variables between patients with persistent and those with controlled IH. Signs and symptoms at initial presentation of IH were also very similar in both subgroups of IH patients (Table 1).

Table 1 Clinical and biochemical characteristics in healthy subjects (control group) and in patients with idiopathic hypercalciuria (IH) divided according to the level of urinary calcium excretion into persistent IH (≥4 mg/kg/day) and controlled IH (<4 mg/kg/day) at the time of blood and urine sampling
Full size table

Bone mineral density measurements

A total of 46 patients underwent BMD measurements during the study period: 20 patients with persistent IH and 26 patients with controlled IH. No differences were detected in the number of patients with a Z-score ≤−1 SD and with a Z-score ≤−2 SD between these subgroups of patients with IH (Table 2).

Table 2 Bone mineral density (BMD) Z-score in 46 patients with idiopathic hypercalciuria (IH) divided according to the level of urinary calcium excretion into persistent IH (≥4 mg/kg/day) and controlled IH (<4 mg/kg/day)
Full size table

Association of plasma and urinary cytokine concentrations with urinary calcium excretion

In IH subgroups (persistent and controlled) and in healthy controls, plasma and urinary concentrations of IL-1β, IL-6, IL-8, and TNF-α were below the detection limits of the ELISA kits. Plasma and spot-urine concentrations of MCP-1 and TGF-β1 were detectable in both subgroups of patients with IH. However, the median value for plasma and spot-urine concentrations of TGF-β1 was zero in both IH subgroups (data not shown). In the control group, TGF-β1 levels were undetectable in the majority of samples, while MCP-1 concentrations were measurable in plasma and spot-urine. As shown in Table 3, no significant differences were verified between controlled and persistent IH patients for plasma and spot-urine (absolute and standardized for creatinine) levels of MCP-1. On the other hand, plasma and urinary levels of MCP-1 were significantly higher in both groups of IH patients in comparison to healthy controls (Table 3).

Table 3 Median and interquartile range (25th percentile, p25, and 75th percentile, p75) of plasma and urinary (absolute and standardized to creatinine) levels of monocyte chemoattractant protein (MCP-1) in healthy subjects (control group) and in patients with persistent and controlled idiopathic hypercalciuria (IH group)
Full size table

There was a trend toward a positive correlation between plasma and spot-urine levels of MCP-1 standardized to creatinine in patients with IH (r=0.24, p=0.08).

Association of plasma and urinary cytokine concentrations with bone mineral density

Patients were stratified according to their BMD Z-score into two groups: >−1 SD, n=28; ≤−1SD; n=18, as shown in Table 4. The comparison between these groups did not reveal differences in general clinical findings or in 24-h urinary calcium excretion.

Table 4 Patients characteristics according to bone mineral density (BMD) IH idiopathic hypercalciuria Z-score
Full size table

The comparison of plasma and urinary concentrations of MCP-1 and TGF-β1 in patients with BMD Z-score >−1 SD and ≤−1 SD did not reveal significant differences (Table 5). However, there was a positive correlation between urinary levels of MCP-1 and bone mineral content (r=0.379, p=0.013).

Table 5 Median and interquartile range (25th percentile, p25, and 75th percentile, p75) of plasma and urinary (absolute and standardized to creatinine) levels of monocyte chemoattractant protein (MCP-1) and TGF-β1 in patients according to bone mineral density (BMD) Z-score
Full size table

Association of plasma and urinary cytokine concentrations with age groups

In order to detect possible changes in cytokine levels related to age, the patients with IH were stratified into the following age groups: school (age ≤ 12 years, n=18) and adolescent (age > 12 years, n=52). The absolute levels of MCP-1 (pg/mL) were significantly higher in adolescents than in school age children (p=0.02). However, this difference was not observed when values were standardized to creatinine (p=0.61, Table 6). Healthy controls were also stratified into the same age groups: school (age ≤12 years, n=10) and adolescent (age >12 years, n=27). There were no differences in the comparison between plasma and spot-urine levels of MCP-1 (absolute and standardized for creatinine) in these age groups among healthy subjects (data not shown).

Table 6 Median and interquartile range (25th percentile, p25, and 75th percentile, p75) of plasma and urinary (absolute and standardized to creatinine) levels of monocyte chemoattractant protein (MCP-1) and TGF-β1 in patients equal to or under 12 years old (school, n=18) and above 18 years old (adolescent, n=52)
Full size table

Association of plasma and urinary cytokine concentrations with hydrochlorothiazide prescription

In order to verify if the use of hydrochlorothiazide influenced MCP-1 levels, IH patients were also divided according to the prescription (n=20) or not (n=50) of this medication at the time of blood and urine sampling. As shown in Table 7, no significant differences were detected in plasma and spot-urine (absolute and standardized to creatinine) levels of MCP-1 in relation to the prescription or not of hydrochlorothiazide.

Table 7 Median and interquartile range (25th percentile, p25, and 75th percentile, p75) of plasma and urinary (absolute and standardized to creatinine) levels of monocyte chemoattractant protein (MCP-1) in patients with idiopathic hypercalciuria (IH) in the use of hydrochrorothiazide (n = 20) and those not under this treatment (n = 50)
Full size table

Discussion

To our knowledge, this is the first study that simultaneously measure bone turnover cytokines in plasma and spot-urine in pediatric patients with IH. Our results showed that single measurements of bone turnover cytokines seem not to be useful in distinguishing patients with persistent hypercalciuria or reduced BMD Z-score. Indeed, many cytokines were below the detection limits for ELISA kits in patients with IH and in healthy controls. On the other hand, the chemokine MCP-1 is significantly higher in plasma and spot-urine samples of patients with IH compared with healthy controls. In addition, there was a positive correlation between urinary MCP-1 levels and bone mineral content.

There is very little information concerning the role of MCP-1 in bone metabolism. The principal function of MCP-1 is the recruitment of monocytes [33, 34]. In vitro and in vivo studies indicate that MCP-1 induces the recruitment of monocytes to bone, which, in turn, is associated with an increase in osteoblast number [35, 36]. MCP-1 is typically not expressed in normal bone or by normal osteoblasts. Upon stimulation by inflammatory mediators and growth factors [37–39], the expression of MCP-1 and the recruitment of monocytes are increased in both osseous inflammation and during bone remodeling. Indeed, monocytes seem to have different functional roles in areas of bone formation and resorption [33, 35]. The recruitment of monocytes in areas of bone formation is associated with a decrease in the number of osteoclasts, while in bone-resorbing areas, recruitment of cells of the monocytic lineage is associated with formation of osteoclasts [33, 35]. In this context, the increased levels of MCP-1 in IH patients compared with healthy subjects and the positive correlation between urinary MCP-1 and BMD may indicate that this chemokine might play a role in bone remodeling of patients with IH. Alternatively, no differences in MCP-1 levels were found when patients were stratified according to the persistence of high urinary calcium excretion and according to the prescription of hydrochlorothiazide.

In addition, the frequency of reduced BMD Z-scores was similar in both subgroups of IH patients, also suggesting that the persistence of hypercalciuria might not be directly linked to high bone resorption in our group of patients. We believe that the process of bone remodeling in IH is very complex and different mechanisms might be activated in spite of the levels of urinary calcium excretion. Therefore, MCP-1 could be locally produced by osteoblasts and signaling toward bone formation or bone resorption depending on the area of expression and on the interactions with other mediators.

In our study, plasma and urinary levels of IL-1β, IL-6, IL-8, and TNF-α were below detectable limits. Other authors, by using different methodologies, were able to evaluate these cytokines in patients with IH [18, 20, 22, 23, 39]. Freundlich et al. showed that the mRNA expression of IL-1α in peripheral blood mononuclear cells of children with IH did not differ from that of healthy controls [23]. Pacifici et al. described an association between IL-1β activity and bone resorption [39]. Weisinger et al. used unstimulated blood monocytes to show increased expression of IL-1α, IL-6, and TNF-α mRNA in patients with IH [18]. These authors also described a correlation between basal production of IL-1α, but not IL-1β, and decreased trabecular bone [20]. Indeed, these more refined methodologies have allowed the evaluation of cytokine expression in bone tissue or in peripheral mononuclear cells [18, 20, 22, 23, 39]. However, we opted to use conventional plasma and spot urine samples in order to ease collection and to evaluate the utility of these tests for patients with IH in clinical practice.

We are aware of the limitations of our study. The main possible weakness is the cross-sectional design. Once patients with IH probably decrease their calcium bone mineral content progressively, serial densitometries might be necessary to accurately evaluate bone mineral loss. Another weakness was the fact that our patients were not on standard diets during sample collections. Diets rich in protein and salt can significantly affect calciuria by different mechanisms from those involved in bone remodeling in IH [40]. Parathyroid hormone and vitamin D are also important variables not concomitantly measured with cytokines in our research protocol [41, 42]. In this study, age was addressed as a possible confounder. Bone remodeling regulatory mechanisms may vary according to age, once children and adolescents experience different stages in skeletal development [43]. In spite of that, no significant difference in cytokine measurements was found in the comparison between school age and adolescents in the control group and in patients with IH.

Nevertheless, some aspects of the study may increase the strength of our findings, such as the sample size, strict inclusion criteria, well-established protocols for cytokine measurements, and the homogeneity among groups. Our sample size was considerably larger than those of previous studies on cytokines in IH [20, 22, 39, 44]. Except for calciuria and the prescription of hydrochlorothiazide, there were no differences between the controlled and persistent IH groups. Indeed, the increased prescription of hydrochlorothiazide in the persistent group seemed not to influence cytokine measurements.

In conclusion, single bone turnover cytokine measurements were not useful in differentiating persistent and controlled IH. However, we found that MCP-1 levels were significantly higher in IH compared with healthy subjects and that spot-urine MCP-1 concentrations and BMC correlated positively. Future studies are necessary to evaluate whether this chemokine plays a role in bone remodeling in children with IH.