Abstract
Introduction
Sclerostin is involved in chronic kidney disease-mineral and bone disorder (CKD-MBD), but its relation to renal magnesium handling remains unclear. We investigated whether urinary and serum sclerostin were associated with fractional excretion of magnesium (FeMg) in proteinuric nondiabetic chronic kidney disease (CKD).
Methods
In this cross-sectional study, 70 adults with proteinuric nondiabetic CKD stages 1–5 at a tertiary medical center were enrolled. Urinary sclerostin normalized to urine creatinine (Uscl/Ucre) and serum sclerostin were measured. Their associations with ln-transformed FeMg [ln(FeMg)] were examined using multivariable linear regression with sequential adjustment for age, sex, estimated glomerular filtration rate (eGFR), proteinuria, electrolytes, serum sclerostin, and CKD-MBD markers including plasma intact parathyroid hormone, serum fibroblast growth factor 23, and soluble alpha-Klotho.
Results
Urinary sclerostin (Uscl/Ucre) increased across CKD stages, and higher Uscl/Ucre tertiles were associated with higher FeMg. Uscl/Ucre was positively associated with ln(FeMg) in crude analysis (β 0.652; p < 0.001), and this association still existed after adjustment for age, sex, eGFR, serum magnesium, and CKD-MBD markers (β 0.366; p < 0.001). Compared with the lowest tertile, the highest urinary sclerostin tertile was significantly associated with higher ln(FeMg) (β 0.662; p = 0.007). In contrast, serum sclerostin was not associated with ln(FeMg) after adjustment.
Conclusion
In proteinuric nondiabetic CKD, urinary but not serum sclerostin is independently associated with fractional excretion of magnesium. Urinary sclerostin may reflect intrarenal processes relevant to tubular magnesium handling.
Keywords: Chronic kidney disease, Fractional excretion of magnesium, Renal magnesium handling, Urinary sclerostin
Plain Language Summary
Chronic kidney disease is a long-term condition in which the kidneys gradually lose their ability to filter blood and regulate minerals. Magnesium is one of these minerals, and the kidneys help control how much magnesium is kept in the body and how much is lost in urine. Sclerostin is a protein best known for its role in bone metabolism, but it can also be measured in blood and urine. This study looked at 70 adults with chronic kidney disease who did not have diabetes but had protein in the urine. The researchers examined whether sclerostin measured in urine or blood was related to how the kidneys handle magnesium. Urinary sclerostin was adjusted for urine concentration by comparing it with urine creatinine, a common urine marker. The study found that people with higher urinary sclerostin tended to lose more magnesium in the urine. This relationship remained even after considering kidney function, protein in the urine, blood magnesium, and other markers related to mineral metabolism. In contrast, sclerostin measured in blood was not clearly related to urinary magnesium loss after these factors were considered. These findings suggest that urinary sclerostin may provide information about kidney-related mineral handling that is not captured by blood sclerostin alone.
Introduction
Chronic kidney disease-mineral and bone disorder (CKD-MBD) is characterized by complex disturbances in mineral metabolism involving calcium, phosphate, parathyroid hormone, fibroblast growth factor-23 (FGF23), and vitamin D, and is associated with adverse skeletal and cardiovascular outcomes [1, 2]. Magnesium is also increasingly recognized as a relevant component of chronic kidney disease (CKD) mineral metabolism because lower magnesium levels have been linked to vascular calcification and adverse outcomes [3], and experimental data suggest that magnesium inhibits phosphate-driven vascular calcification [4]. In the kidney, magnesium balance is regulated mainly by tubular transport processes [5]. As glomerular filtration declines, fractional excretion of magnesium (FeMg) increases to compensate for reduced filtered magnesium, although this adaptive response may become insufficient in advanced CKD [6].
Sclerostin is a glycoprotein mainly produced by osteocytes and acts as an inhibitor of the Wnt/β-catenin pathway [7]. Circulating sclerostin levels increase in CKD and have been linked to low-turnover bone disease and vascular outcomes [8–10]. In addition, urinary sclerostin and fractional excretion of sclerostin increase with declining kidney function [11]. However, most studies have focused on serum sclerostin as a systemic biomarker, whereas its relevance to renal tubular mineral handling remains unclear.
In our previous study, urinary but not serum sclerostin was associated with FeMg in patients with type 2 diabetes with or without CKD [12]. Whether this association extends to nondiabetic CKD remains uncertain. We therefore investigated the associations of urinary and serum sclerostin with FeMg in patients with nondiabetic CKD, particularly those with proteinuria, a clinically relevant setting in which urinary biomarker assessment may be especially informative. We hypothesized that higher urinary sclerostin, rather than serum sclerostin, would be independently associated with higher FeMg after accounting for kidney function, proteinuria, and systemic CKD-MBD markers. This observational study was designed to evaluate this clinical association.
Methods
Study Population
This was a cross-sectional study in which participants were prospectively enrolled. For participant enrollment, initially patients visiting the Nephrology Outpatient Department of National Cheng Kung University Hospital, Tainan, from September 2019 to October 2021 were screened. Non-dialysis CKD patients aged ≥20 years with proteinuria (defined as a urine protein-to-creatinine ratio [UPCR] >300 mg/g), without a history of fracture or active bone disease, and without exposure to vitamin D or its analogs, steroids, calcium-containing medications, or osteoporosis treatments within the preceding 3 months, and without diuretic exposure within the preceding 2 weeks were eligible for enrollment.
Patients with malnutrition (defined as serum albumin of <3.5 g/dL), pregnancy, type 2 diabetes mellitus, cancer, liver disease, renal transplantation, or active infection were excluded. Accordingly, 83 participants were prospectively enrolled, and 70 with complete data were included in this cross-sectional analysis. Analyses were performed using complete-case data. The study size was determined by the number of eligible participants with complete clinical and laboratory data during the study period. Enrollment was conducted with the aim of achieving a broadly balanced distribution across CKD stages 1–2, 3, 4, and 5. Written informed consent was obtained from all participants.
Demographic Data and Specimen Collection
For each enrolled participant, clinical parameters including age and sex were recorded. Blood and random spot urine samples were collected concurrently at the central laboratory of National Cheng Kung University Hospital, either on the date of enrollment or at a subsequent visit during the study period. At the central laboratory, serum and urine supernatant were prepared according to routine procedures, and appropriate portions were used for the required clinical laboratory analyses. The remaining urine supernatants were temporarily stored at 4°C. Within 8 h after specimen collection, the remaining serum and urine supernatant specimens were retrieved and transported to the research laboratory, aliquoted, and stored at −80°C until subsequent analysis. No protease inhibitor was added to the urine supernatant. Each aliquot underwent a single freeze-thaw cycle for analysis and was discarded thereafter.
Assays
Hemoglobin, serum blood urea nitrogen, creatinine, albumin, calcium (Ca), phosphate (P), magnesium (Mg), and high-sensitivity C-reactive protein, as well as plasma intact parathyroid hormone (iPTH), urine creatinine, urine Ca, urine P, and urine Mg, UPCR, and urine albumin-to-creatinine ratio (ACR) were measured at the Department of Laboratory Medicine of National Cheng Kung University Hospital using standard automated laboratory methods.
Estimated glomerular filtration rate (eGFR) was calculated using the four-variable Modification of Diet in Renal Disease equation. Besides, fractional excretion of electrolyte (X) was calculated as FeX (%) = (Ux × Pcr×100)/(Px × Ucr), where Ux and Px represent urine and serum electrolyte concentrations, respectively, and Ucr and Pcr represent urine and serum creatinine concentrations. For magnesium, serum magnesium was multiplied by 0.7 to account for the nonfilterable protein-bound fraction.
Urinary sclerostin concentrations were normalized to urine creatinine and expressed as the urinary sclerostin-to-creatinine ratio (Uscl/Ucre). This approach is commonly used in urinary biomarker studies to minimize the influence of urine dilution and variability in spot urine samples.
Serum and urine sclerostin concentrations were measured using ELISA (BI-20492, Biomedica, Austria), with inter- and intra-assay coefficients of variation ≤10% and ≤7%, respectively. This assay uses a polyclonal goat anti-human sclerostin antibody as the capture antibody and a monoclonal mouse anti-sclerostin antibody as the detection antibody. The exact epitopes recognized and the specific molecular forms detected by this assay have not been publicly defined. Accordingly, values obtained with BI-20492 were interpreted as immunoreactive sclerostin rather than a defined molecular form. The assay was not designed to specifically quantify receptor-binding bioactive sclerostin. The lower limit of detection was 3.2 pmol/L. Urinary sclerostin was detectable in 60 of the 70 participants.
Serum bone-specific alkaline phosphatase (BAP) was measured using the Ostase® BAP immunoenzymetric assay (AC-20F1, IDS, UK) with both inter- and intra-assay coefficients of variation <7%. The detection limit was 0.7 μg/L.
Serum collagen type 1 cross-linked C-telopeptide was measured using ELISA (SL0540Hu, SunLong Biotech, China), with inter- and intra-assay coefficients of variation <12% and <10%, respectively. The detection limit was 12.5 pg/mL.
Serum 25(OH) vitamin D (25OHD) was measured using ELISA (K2108, Immundiagnostik AG, Germany), with inter- and intra-assay coefficients of variation <14% and <13%, respectively. The detection limit was 11.9 nmol/L.
Serum C-terminal FGF23 was measured using a commercially available ELISA kit (60–6,100, Immutopics, USA) with inter- and intra-assay coefficients of variation <5%. The lower limit of detection was 1.5 RU/mL.
Serum soluble alpha-Klotho (sKlotho) was measured using ELISA (27988, IBL, Japan), with within- and between-run coefficients of variation <5% and <8%, respectively. The lower limit of detection was 6.15 pg/mL.
Statistical Analysis
The normality of continuous variables was assessed using the Kolmogorov-Smirnov test. Most variables were not normally distributed except hemoglobin, serum calcium, and sKlotho. Therefore, nonparametric statistical methods were used. Continuous variables were expressed as median (Q1, Q3), and categorical variables were expressed as frequency (percentage). Urinary sclerostin values below the detection limit were treated as zero in the analysis.
For group comparisons, the Kruskal-Wallis test or chi-square test was used as appropriate. Spearman’s rank correlation was used to evaluate correlations between variables.
For multivariable analyses, ln-transformation was not applied to urinary sclerostin because urinary sclerostin concentrations included values below the detection limit (n = 10). FeMg and UPCR were ln-transformed to improve distributional properties and facilitate interpretation of regression coefficients in multivariable linear regression analyses. Multivariable linear regression models were constructed to evaluate the association between urinary sclerostin and ln-transformed FeMg after adjustment for demographic variables, kidney function, and mineral metabolism parameters. Multicollinearity was assessed using the variance inflation factor, with a variance inflation factor of <5 considered acceptable.
All statistical analyses were performed using SPSS version 26 (IBM Corp., Armonk, NY, USA). A two-sided p value of <0.05 was considered statistically significant.
Results
Clinical Characteristics of the Study Population according to CKD Stage
Among these 83 enrolled participants, 70 with complete data were included in the final analysis, and urinary sclerostin measured by BI-20492 was detectable in 60 of them. As shown in Table 1, the median age of the study population was 60.0 years, 34 patients (48.6%) were male, and the median UPCR and ACR were 1,149 and 957 mg/g, respectively. The whole cohort was initially categorized according to CKD stage, with 18 patients in stages 1 and 2, 16 patients in stage 3, 16 patients in stage 4, and 20 patients in stage 5. Serum potassium, phosphate, magnesium, and sclerostin all increased with CKD severity. Among CKD-MBD markers in blood, FGF23 and iPTH increased, whereas sKlotho decreased significantly across CKD stages. In parallel, fractional excretion indices of sodium, calcium, phosphate, and magnesium all differed significantly by CKD stage, with the highest FeMg observed in stage 5 CKD (19.4%) compared with stages 1–2 (3.69%). BI-20492-measured urinary immunoreactive sclerostin normalized to urine creatinine (Uscl/Ucre) also increased progressively from 0.06 pmol/g in CKD stages 1–2 to 1.94 pmol/g in stage 5. By contrast, UPCR and ACR did not differ significantly across CKD stages.
Table 1.
Clinical characteristics of the study population according to CKD stage
| Variables | Total (n = 70) | CKD stages 1–2 (n = 18) | CKD stage 3 (n = 16) | CKD stage 4 (n = 16) | CKD stage 5 (n = 20) | p value |
|---|---|---|---|---|---|---|
| Age | 60.0 (46.0, 68.3) | 54.0 (41.0, 62.0) | 65.0 (46.0, 69.0) | 64.0 (56.0, 71.0) | 61.0 (50.0, 68.0) | 0.176 |
| Male, % | 34 (48.6) | 7 (38.9) | 8 (50.0) | 11 (68.8) | 8 (40.0) | 0.274 |
| BUN, mg/dL | 29.5 (21.0, 45.3) | 18.0 (12.0, 21.0) | 25.5 (21.5, 31.0) | 35.5 (28.5, 41.0) | 64.0 (46.0, 79.0) | <0.001 |
| Creatinine, mg/dL | 2.23 (1.09, 4.47) | 0.81 (0.64, 1.01) | 1.40 (1.21, 1.94) | 2.47 (2.27, 2.97) | 6.01 (4.83, 7.88) | <0.001 |
| eGFR, mL/min/1.73 m2 | 29.1 (12.1, 61.4) | 79.9 (67.8, 112.9) | 41.3 (35.0, 50.2) | 23.7 (20.1, 27.4) | 7.30 (6.05, 10.5) | <0.001 |
| Hemoglobin, g/dL | 12.5 (10.8, 13.8) | 13.7 (13.2, 15.1) | 12.9 (12.0, 14.6) | 12.7 (12.1, 13.4) | 9.70 (9.30, 10.9) | <0.001 |
| Albumin, g/dL | 4.30 (4.10, 4.60) | 4.30 (4.20, 4.60) | 4.20 (4.05, 4.40) | 4.45 (4.20, 4.65) | 4.35 (4.10, 4.60) | 0.192 |
| Na, mmol/L | 140 (139, 142) | 142 (139, 143) | 141 (139, 141) | 141 (139, 142) | 140 (137, 141) | 0.208 |
| K, mmol/L | 4.40 (4.10, 4.70) | 4.30 (3.98, 4.50) | 4.30 (4.13, 4.48) | 4.50 (4.23, 4.70) | 4.90 (4.33, 5.18) | 0.014 |
| Ca, mg/dL | 9.40 (9.10, 9.60) | 9.45 (9.20, 9.60) | 9.25 (9.10, 9.60) | 9.50 (9.15, 9.75) | 9.30 (9.00, 9.70) | 0.653 |
| P, mg/dL | 3.80 (3.50, 4.10) | 3.65 (3.30, 4.00) | 3.75 (3.50, 3.80) | 3.75 (3.50, 3.95) | 4.30 (3.95, 4.95) | <0.001 |
| Mg, mg/dL | 2.25 (2.10, 2.40) | 2.10 (2.00, 2.20) | 2.20 (2.20, 2.35) | 2.35 (2.25, 2.45) | 2.40 (2.20, 2.85) | <0.001 |
| Sclerostin, pmol/L | 73.4 (43.0, 102) | 38.2 (32.3, 45.0) | 57.7 (35.3, 83.6) | 89.0 (69.5, 110) | 102 (82.6, 151) | <0.001 |
| ALP, U/L | 71.0 (56.0, 81.5) | 71.5 (55.5, 76.3) | 71.0 (54.0, 80.5) | 65.0 (59.5, 78.8) | 76.5 (48.5, 109) | 0.735 |
| BAP, μg/L | 15.5 (11.4, 21.0) | 15.1 (10.1, 18.6) | 16.2 (12.5, 21.7) | 15.6 (13.2, 20.4) | 15.6 (9.8, 30.6) | 0.615 |
| CTX, pg/mL | 232 (183, 422) | 208 (152, 276) | 239 (169, 403) | 378 (265, 426) | 209 (152, 679) | 0.086 |
| FGF23, RU/mL | 22.2 (15.5, 55.8) | 14.1 (11.5, 17.2) | 19.7 (11.8, 24.8) | 19.4 (16.9, 36.9) | 83.6 (46.2, 521) | <0.001 |
| sKlotho, pg/mL | 285 (167, 458) | 399 (258, 550) | 407 (280, 503) | 202 (150, 274) | 179 (114, 408) | 0.016 |
| iPTH, pg/mL | 68.8 (41.9, 128) | 44.9 (30.7, 50.8) | 66.1 (31.8, 83.1) | 80.3 (53.1, 153) | 164 (107, 292) | <0.001 |
| 25OHD, nmol/L | 54.4 (35.5, 71.2) | 53.5 (34.2, 65.4) | 55.5 (38.0, 85.5) | 56.4 (29.7, 71.1) | 55.8 (38.1, 70.6) | 0.893 |
| hsCRP, mg/L | 0.96 (0.39, 2.44) | 0.90 (0.50, 1.73) | 2.38 (0.53, 6.68) | 0.52 (0.25, 1.23) | 1.08 (0.35, 1.69) | 0.030 |
| UPCR, mg/g | 1,149 (665, 1,840) | 1,020 (673, 1,735) | 1,386 (914, 1,997) | 623 (381, 1,595) | 1,238 (968, 2,161) | 0.168 |
| ACR, mg/g | 957 (447, 1,586) | 836 (480, 1,412) | 1,155 (748, 1,745) | 470 (223, 1,411) | 1,026 (721, 1,652) | 0.285 |
| FeNa, % | 0.99 (0.59, 2.52) | 0.53 (0.31, 0.74) | 0.74 (0.50, 1.01) | 1.12 (0.72, 2.36) | 3.15 (2.07, 6.03) | <0.001 |
| FeCa, % | 0.67 (0.31, 1.49) | 0.57 (0.24, 1.12) | 0.54 (0.26, 0.88) | 0.36 (0.28, 0.71) | 1.56 (0.80, 3.11) | 0.001 |
| FeP, % | 27.7 (13.6, 41.3) | 10.3 (7.89, 15.9) | 18.3 (13.5, 29.7) | 32.6 (20.4, 38.7) | 49.5 (39.7, 55.0) | <0.001 |
| FeMg, % | 6.94 (4.38, 17.2) | 3.69 (1.86, 5.06) | 6.07 (3.85, 6.95) | 8.38 (6.23, 11.7) | 19.4 (17.2, 32.6) | <0.001 |
| Uscl/Ucre, pmol/g | 0.67 (0.13, 1.56) | 0.06 (0.00, 0.16) | 0.33 (0.12, 0.77) | 1.00 (0.35, 1.60) | 1.94 (1.27, 2.32) | <0.001 |
Baseline demographic, biochemical, and renal tubular handling parameters of the study population (n = 70) stratified by CKD stage. Continuous variables are expressed as median (interquartile range) and categorical variables as number (%). Bold values indicate variables with statistically significant differences across groups (p < 0.05).
ACR, urine albumin-to-creatinine ratio; ALP, alkaline phosphatase; BAP, bone-specific alkaline phosphatase; BUN, blood urea nitrogen; Ca, calcium; CTX, collagen type 1 cross-linked C-telopeptide; eGFR, estimated glomerular filtration rate; FeCa, fractional excretion of calcium; FeMg, fractional excretion of magnesium; FeNa, fractional excretion of sodium; FeP, fractional excretion of phosphate; FGF23, fibroblast growth factor 23; HbA1c, glycated hemoglobin; hsCRP, high sensitivity C reactive protein; iPTH, intact parathyroid hormone; K, potassium; Mg, magnesium; Na, sodium; 25OHD, 25-hydroxy vitamin D; P, phosphate; sKlotho, soluble alpha-Klotho; UPCR, urine protein-to-creatinine ratio; Uscl/Ucre, urine sclerostin-to-creatinine ratio.
Clinical Characteristics of the Study Population according to Tertiles of Urinary Sclerostin
To evaluate the association of urinary sclerostin with other parameters, participants were further categorized into tertiles of Uscl/Ucre: tertile 1 (T1), ≤0.24 pmol/g (n = 23); tertile 2 (T2), 0.24–1.21 pmol/g (n = 24); and tertile 3 (T3), >1.21 pmol/g (n = 23). As shown in Table 2, patients in the higher urinary sclerostin tertiles had lower eGFR and hemoglobin as well as higher serum potassium, phosphate, magnesium, and serum sclerostin levels. Among mineral metabolism markers, BAP, FGF23, and iPTH increased significantly across Uscl/Ucre tertiles, whereas sKlotho and 25(OH) vitamin D did not differ significantly. Notably, indices of renal electrolyte excretion showed marked differences across tertiles: compared with the lowest tertile, the highest tertile had substantially higher FeNa (2.72% [1.95, 6.36] vs. 0.59% [0.31, 0.74]), FeCa (1.49% [0.69, 3.74] vs. 0.46% [0.19, 0.84]), FeP (48.5% [40.2, 54.6] vs. 12.3% [8.37, 16.2]), and FeMg (18.2% [11.9, 30.7] vs. 4.39% [2.19, 5.65]) (all p < 0.001). In contrast, UPCR and ACR were not significantly different across urinary sclerostin tertiles.
Table 2.
Clinical characteristics of study cohort according to tertiles of urinary sclerostin
| Variables | Total (n = 70) | T1, lowest Uscl/Ucre (n = 23) | T2, middle Uscl/Ucre (n = 24) | T3, highest Uscl/Ucre (n = 23) | p value |
|---|---|---|---|---|---|
| Age | 60.0 (46.0, 68.3) | 57.0 (41.0, 66.0) | 62.0 (46.3, 67.8) | 63.0 (54.0, 72.0) | 0.106 |
| Male, % | 34 (48.6) | 11 (47.8) | 12 (50.0) | 11 (47.8) | 0.985 |
| BUN, mg/dL | 29.5 (21.0, 45.3) | 21.0 (15.0, 25.0) | 30.0 (23.3, 41.5) | 46.0 (35.0, 74.0) | <0.001 |
| Creatinine, mg/dL | 2.23 (1.09, 4.47) | 0.98 (0.75, 1.40) | 1.95 (1.20, 3.77) | 4.79 (2.76, 7.87) | <0.001 |
| eGFR, mL/min/1.73 m2 | 29.1 (12.1, 61.4) | 67.8 (48.2, 97.1) | 32.1 (17.0, 48.1) | 9.90 (6.40, 20.3) | <0.001 |
| Hemoglobin, g/dL | 12.5 (10.8, 13.8) | 13.6 (12.4, 15.1) | 12.8 (11.0, 14.2) | 10.8 (9.60, 12.1) | <0.001 |
| Albumin, g/dL | 4.30 (4.10, 4.60) | 4.30 (4.20, 4.50) | 4.25 (4.13, 4.60) | 4.40 (4.10, 4.70) | 0.639 |
| Na, mmol/L | 140 (139, 142) | 141 (139, 142) | 141 (139, 142) | 140 (137, 141) | 0.185 |
| K, mmol/L | 4.40 (4.10, 4.70) | 4.30 (4.00, 4.50) | 4.40 (4.10, 4.58) | 4.90 (4.30, 5.00) | 0.035 |
| Ca, mg/dL | 9.40 (9.10, 9.60) | 9.50 (9.10, 9.60) | 9.25 (9.10, 9.75) | 9.40 (9.00, 9.60) | 0.491 |
| P, mg/dL | 3.80 (3.50, 4.10) | 3.70 (3.30, 4.00) | 3.80 (3.40, 3.88) | 4.00 (3.80, 4.90) | 0.003 |
| Mg, mg/dL | 2.25 (2.10, 2.40) | 2.10 (2.00, 2.20) | 2.30 (2.20, 2.40) | 2.40 (2.30, 2.80) | <0.001 |
| Sclerostin, pmol/L | 73.4 (43.0, 102) | 43.2 (32.0, 73.1) | 70.5 (45.7, 98.4) | 102 (82.1, 142) | <0.001 |
| ALP, U/L | 71.0 (56.0, 81.5) | 63.0 (52.0, 76.0) | 71.5 (59.5, 80.3) | 78.0 (53.0, 110) | 0.074 |
| BAP, μg/L | 15.5 (11.4, 21.0) | 13.1 (10.4, 15.5) | 16.7 (13.5, 22.5) | 15.7 (12.3, 33.7) | 0.023 |
| CTX, pg/mL | 232 (183, 422) | 276 (189, 449) | 220 (141, 330) | 255 (200, 721) | 0.150 |
| FGF23, RU/mL | 22.2 (15.5, 55.8) | 17.0 (11.6, 26.9) | 19.6 (14.9, 35.3) | 64.9 (24.0, 101) | <0.001 |
| sKlotho, pg/mL | 285 (167, 458) | 348 (216, 495) | 281 (189, 425) | 187 (135, 451) | 0.348 |
| iPTH, pg/mL | 68.8 (41.9, 128) | 35.4 (29.2, 53.6) | 74.9 (45.2, 114) | 169 (106, 258) | <0.001 |
| 25OHD, nmol/L | 54.4 (35.5, 71.2) | 55.7 (34.4, 67.7) | 52.0 (40.4, 83.8) | 57.8 (29.0, 69.8) | 0.843 |
| hsCRP, mg/L | 0.96 (0.39, 2.44) | 0.90 (0.39, 4.46) | 1.27 (0.48, 3.16) | 0.65 (0.27, 1.52) | 0.248 |
| UPCR, mg/g | 1,149 (665, 1,840) | 905 (398, 1,241) | 1,336 (943, 1,942) | 1,145 (607, 2,258) | 0.066 |
| ACR, mg/g | 957 (447, 1,586) | 745 (314, 1,030) | 1,059 (771, 1,726) | 955 (383, 1,788) | 0.106 |
| FeNa, % | 0.99 (0.59, 2.52) | 0.59 (0.31, 0.74) | 0.87 (0.62, 1.37) | 2.72 (1.95, 6.36) | <0.001 |
| FeCa, % | 0.67 (0.31, 1.49) | 0.46 (0.19, 0.84) | 0.49 (0.30, 0.99) | 1.49 (0.69, 3.74) | <0.001 |
| FeP, % | 27.7 (13.6, 41.3) | 12.3 (8.37, 16.2) | 24.9 (18.3, 32.5) | 48.5 (40.2, 54.6) | <0.001 |
| FeMg, % | 6.94 (4.38, 17.2) | 4.39 (2.19, 5.65) | 6.72 (4.46, 9.15) | 18.2 (11.9, 30.7) | <0.001 |
| Uscl/Ucre, pmol/g | 0.67 (0.13, 1.56) | 0.03 (0.00, 0.14) | 0.67 (0.40, 0.96) | 2.13 (1.55, 2.44) | <0.001 |
Participants were categorized into tertiles of urinary sclerostin-to-creatinine ratio (Uscl/Ucre) as follows: T1, ≤ 0.24 pmol/g (n = 23); T2, 0.24–1.21 pmol/g (n = 24); and T3, >1.21 pmol/g (n = 23). Continuous variables are expressed as median (interquartile range) and categorical variables as number (%). Urinary sclerostin values below the lower limit of detection were treated as zero in the analysis. Bold values indicate variables with statistically significant differences across urinary sclerostin tertiles (p < 0.05).
ACR, urine albumin-to-creatinine ratio; ALP, alkaline phosphatase; BAP, bone-specific alkaline phosphatase; BUN, blood urea nitrogen; Ca, calcium; CTX, collagen type 1 cross-linked C-telopeptide; eGFR, estimated glomerular filtration rate; FeCa, fractional excretion of calcium; FeMg, fractional excretion of magnesium; FeNa, fractional excretion of sodium; FeP, fractional excretion of phosphate; FGF23, fibroblast growth factor 23; HbA1c, glycated hemoglobin; hsCRP, high sensitivity C reactive protein; iPTH, intact parathyroid hormone; K, potassium; Mg, magnesium; Na, sodium; 25OHD, 25-hydroxy vitamin D; P, phosphate; sKlotho, soluble alpha-Klotho; UPCR, urine protein-to-creatinine ratio; Uscl/Ucre, urine sclerostin-to-creatinine ratio.
Association between Urinary Sclerostin and FeMg
Spearman correlation analysis of baseline variables showed that both urinary sclerostin (Uscl/Ucre) and serum sclerostin were correlated with several fractional excretion indices, including FeMg (online suppl. Fig. S1; for all online suppl. material, see https://doi.org/10.1159/kbr/adjag003). Focusing on renal magnesium handling, a positive relationship between Uscl/Ucre and ln-transformed FeMg was shown in Figure 1 (crude β = 0.652, 95% CI 0.498–0.807; p < 0.001). In multivariable linear regression analyses shown in Table 3, this association remained robust after adjustment for basic variables including age, sex, eGFR, and serum magnesium (model 1), basic variables and additional proteinuria (model 2), electrolytes (model 3), serum sclerostin (model 4), or CKD-MBD markers including iPTH, FGF23, and sKlotho (model 5), with β coefficients ranging from 0.334 to 0.376 (all p < 0.001).
Fig. 1.
![A positive linear association between urinary sclerostin to creatinine ratio (Uscl/Ucre) and ln-transformed fractional excretion of magnesium [ln(FeMg)] is shown in this scatter plot. Each participant is represented by one point, and the fitted regression line is shown with its 95% confidence band.](https://cdn.ncbi.nlm.nih.gov/pmc/blobs/59af/13688352/068e6c7f9f7f/kbr-2026-0051-0001-554156_F01.webp)
The association between urinary sclerostin and ln(FeMg). The relationship between urinary sclerostin (Uscl/Ucre) and ln-transformed fractional excretion of magnesium [ln(FeMg)] was shown by a scatterplot. The solid line represents the unadjusted linear regression fit with 95% confidence band.
Table 3.
The association between urine sclerostin (Uscl/Ucre) and ln(FeMg)
| | β | 95% CI | Adjusted R2 | p value |
|---|---|---|---|---|
| Crude | 0.652 | 0.498–0.807 | 0.504 | <0.001 |
| Model 1 | 0.359 | 0.199–0.519 | 0.688 | <0.001 |
| Model 2 | 0.334 | 0.165–0.502 | 0.688 | <0.001 |
| Model 3 | 0.357 | 0.195–0.520 | 0.696 | <0.001 |
| Model 4 | 0.376 | 0.197–0.556 | 0.685 | <0.001 |
| Model 5 | 0.366 | 0.188–0.543 | 0.679 | <0.001 |
By multivariable linear regression analysis, the association between urinary sclerostin (Uscl/Ucre) and ln-transformed fractional excretion of magnesium (lnFeMg) was examined. Five regression models were constructed, adjusting sequentially for demographic factors, proteinuria, electrolytes, serum sclerostin, and CKD-mineral bone disorder markers. Regression coefficients are presented as unstandardized β with 95% confidence interval (CI). Model 1 (basic): adjusted for age, sex, eGFR, Mg. Model 2 (proteinuria-adjusted): adjusted for age, sex, eGFR, Mg, ln(UPCR). Model 3 (electrolyte-adjusted): adjusted for age, sex, eGFR, Mg, K, Ca, P. Model 4 (serum sclerostin-adjusted): adjusted for age, sex, eGFR, Mg, sclerostin. Model 5 (CKD-MBD-adjusted): adjusted for age, sex, eGFR, Mg, iPTH, FGF23, sKlotho.
Association between Serum Sclerostin and FeMg
In contrast to urinary sclerostin, serum sclerostin did not show an independent association with ln(FeMg) in multivariable analyses (online suppl. Table S1). Although serum sclerostin was significantly associated with ln(FeMg) in crude analysis (β = 0.010, 95% CI: 0.005–0.014; p < 0.001), the association was attenuated and no longer significant after adjustment for age, sex, eGFR, and serum magnesium (model 1: β = 0.003, 95% CI: −0.001 to 0.008; p = 0.163). No association between serum sclerostin and ln(FeMg) remained consistent in subsequent models additionally adjusted for proteinuria, electrolytes, urinary sclerostin, or CKD-MBD markers. Notably, after additional adjustment for urinary sclerostin, serum sclerostin was not associated with ln(FeMg) (model 4: β = −0.001, 95% CI: −0.005 to 0.003; p = 0.658). Thus, in the adjusted models, an association with ln(FeMg) was observed for urinary but not serum sclerostin.
CKD Group-Stratified Analysis
Because both urinary sclerostin and FeMg differed across CKD stages, we performed a stratified analysis to explore whether their association varied according to CKD severity. Patients were then categorized into an early CKD group (stages 1–3, n = 34) and a late CKD group (stages 4–5, n = 36). As shown in online supplementary Table S2, in the early CKD group, Uscl/Ucre was not significantly associated with ln(FeMg) after adjustment for age, sex, eGFR, and serum magnesium (β = 0.374, 95% CI: −0.256 to 1.005; p = 0.234). In contrast, in the late CKD group, higher Uscl/Ucre remained significantly associated with higher ln(FeMg) (β = 0.267, 95% CI: 0.084–0.449; p = 0.006). However, the interaction between urinary sclerostin and CKD group was not significant (p for interaction = 0.818), suggesting that the association was not significantly modified by CKD group.
Association between Urinary Sclerostin Tertiles and FeMg
To further assess the association between urinary sclerostin tertiles and ln(FeMg), multivariable linear regression analysis was performed with the lowest tertile (T1) as the reference group (Table 4). After adjustment for age, sex, eGFR, and serum magnesium, T2 was not significantly associated with ln(FeMg) (β = 0.020, 95% CI: −0.357 to 0.397; p = 0.915), whereas T3 remained significantly associated with higher ln(FeMg) (β = 0.662, 95% CI: 0.192–1.133; p = 0.007). These findings indicate that the adjusted association between Uscl/Ucre and ln(FeMg) was evident for participants in the highest Uscl/Ucre tertile compared with those in the lowest tertile. The distribution of ln(FeMg) across tertiles is shown in Figure 2.
Table 4.
The association between urinary sclerostin tertiles and ln(FeMg)
| Variable | Β | 95% CI | p value |
|---|---|---|---|
| T1 (reference) | – | – | – |
| T2 vs. T1 | 0.020 | −0.357 to 0.397 | 0.915 |
| T3 vs. T1 | 0.662 | 0.192 to 1.133 | 0.007 |
Participants were categorized into tertiles according to urinary sclerostin normalized to urine creatinine (Uscl/Ucre): T1, ≤0.24 pmol/g (n = 23); T2, 0.24–1.21 pmol/g (n = 24); and T3, >1.21 pmol/g (n = 23). By multivariable linear regression analysis, the association between urinary sclerostin tertiles and ln-transformed fractional excretion of magnesium [ln(FeMg)] was evaluated, with the lowest tertile as reference. Models were adjusted for age, sex, eGFR, and serum magnesium. Regression coefficients are presented as unstandardized β with 95% confidence interval (CI).
Fig. 2.
![The distribution of ln-transformed fractional excretion of magnesium [ln(FeMg)] across tertiles of urinary sclerostin (Uscl/Ucre) is shown in this violin plot. Participants are grouped as T1, ≤0.24 pmol/g; T2, 0.24–1.21 pmol/g; and T3, >1.21 pmol/g, with each participant represented by one dot. Higher ln(FeMg) values are observed across increasing urinary sclerostin tertiles, with significantly higher ln(FeMg) in T3 than in T1 (p = 0.007).](https://cdn.ncbi.nlm.nih.gov/pmc/blobs/59af/13688352/7cc1909ba9c2/kbr-2026-0051-0001-554156_F02.webp)
The distribution of ln(FeMg) across urinary sclerostin tertiles. Participants were categorized into tertiles according to urinary sclerostin normalized to urine creatinine (Uscl/Ucre): T1, ≤0.24 pmol/g (n = 23); T2, 0.24–1.21 pmol/g (n = 24); and T3, >1.21 pmol/g (n = 23). The distribution of ln-transformed fractional excretion of magnesium [ln(FeMg)] in each tertile was shown by this violin plot. Each dot represented one participant. The highest urinary sclerostin tertile showed significantly higher ln(FeMg) than the lowest tertile (T3 vs. T1, p = 0.007).
Discussion
In this cross-sectional study of proteinuric nondiabetic CKD, urinary sclerostin (Uscl/Ucre) was independently associated with FeMg, a marker of renal Mg handling. This association remained consistent by multivariable linear regression and tertile-based analysis. In contrast, serum sclerostin was not independently associated with ln(FeMg). These findings suggest that urinary sclerostin may capture intrarenal processes related to tubular electrolyte handling rather than merely reflecting systemic mineral metabolism.
Compared with our 2021 study [12] in patients with T2DM with or without CKD, the current study focused on proteinuric nondiabetic CKD and used a cross-sectional multivariable regression framework. In this cohort, the association between urinary sclerostin and ln(FeMg) remained robust after adjustment for sex, age, eGFR, proteinuria, electrolytes, serum sclerostin, and CKD-MBD markers, whereas serum sclerostin was not independently associated with ln(FeMg). These findings extend our previous observations and further support that urinary, rather than circulating, sclerostin is linked to renal magnesium handling in CKD.
An important assay-related consideration is that sclerostin in both our previous [12] and current studies was measured using the Biomedica Sclerostin ELISA (BI-20492), rather than the Bioactive Sclerostin ELISA (BI-20472). Manufacturer-provided technical information indicates that BI-20472 is designed to recognize an epitope within the low-density lipoprotein-related receptor 5/6 interaction region of sclerostin, whereas the epitopes recognized by BI-20492 have not been defined [13]. In paired serum measurements, BI-20492 and BI-20472 showed a strong but incomplete correlation (r = 0.87; R2 = 0.75), indicating that they provide relevant but assay-dependent, noninterchangeable estimates of sclerostin [14]. Importantly, although BI-20472 is designed to detect bioactive sclerostin, no published data have confirmed that sclerostin detected by this assay is intact or retains lipoprotein-related receptor 5/6-binding activity. Accordingly, serum and urinary values obtained with BI-20492 in the present study were interpreted as immunoreactive sclerostin rather than as specific measures of intact or receptor-binding-competent sclerostin. The detected signal may include intact sclerostin and/or immunoreactive fragments [15]. Moreover, use of the same assay does not indicate that serum and urinary signals represent identical molecular forms, because the relative contributions of intact protein, proteolytic fragments, and matrix-related effects may differ between these two biological compartments. In addition, available validation data for BI-20472 [13] indicate that the assay is validated for serum, EDTA plasma, and citrate plasma, whereas urine is not listed as a validated sample matrix. Because no residual stored urine specimens remained available for post hoc analysis, BI-20472 could not be evaluated in our cohort. Future studies are needed to assess paired serum and urine samples using both BI-20492 and BI-20472, with appropriate urine-matrix validation, to determine whether the association with FeMg is reproducible across assays that recognize different sclerostin forms.
An additional consideration is that all participants had proteinuric CKD by design. Although the association between urinary sclerostin and ln(FeMg) remained significant after adjustment for ln(UPCR), the present study could not distinguish whether this association reflected altered glomerular permeability, the underlying proteinuric kidney disease phenotype, tubulointerstitial injury, or more generalized tubular dysfunction. Therefore, the observed findings should be interpreted within the context of proteinuric CKD and cannot be directly generalized to patients with non-proteinuric CKD.
Renal tissue sclerostin expression and urinary BI-20492-measured immunoreactive sclerostin should not be considered equivalent. A previous study has detected sclerostin immunoreactivity in renal tubular cells and has shown that urinary sclerostin excretion increases as kidney function declines [11]. However, whether urinary BI-20492-measured immunoreactive sclerostin reflects renal tissue SOST expression, altered filtration, impaired tubular handling, or proteolytic processing remains uncertain. Besides, a previous experimental study also suggested that endogenous renal sclerostin expression might participate in the response to chronic tubular injury and fibrogenesis. In an obstructive nephropathy model, Sost deletion aggravated renal interstitial fibrosis [16]. These findings support a potential compartment-specific role of renal sclerostin expression in tubular injury and fibrogenesis. Moreover, these findings do not indicate that higher urinary BI-20492-measured immunoreactive sclerostin reflects increased intrarenal sclerostin expression.
Several intrarenal processes may underlie the positive association between urinary immunoreactive sclerostin and ln(FeMg). Higher urinary sclerostin may reflect altered proximal tubular uptake or degradation of filtered sclerostin, proteinuria-associated tubular stress, or altered release and processing of intact sclerostin and/or immunoreactive fragments from injured tubular cells. These processes may coexist with tubulointerstitial injury, fibrosis, and generalized tubular dysfunction, thereby contributing to impaired magnesium reabsorption and higher FeMg. In addition, a distal convoluted tubule-centered hypothesis, in which altered Wnt signaling may influence transient receptor potential melastatin 6-mediated magnesium reabsorption, as proposed in our previous study [12], remains biologically plausible but untested. Future mechanistic studies incorporating markers of proximal tubular injury, tubulointerstitial fibrosis, and distal tubular magnesium transport are needed to test these possibilities.
A possible explanation for the discrepancy in association of FeMg between urinary and serum sclerostin is that these two measures may reflect different biological compartments and analytical signals. Serum sclerostin appears to reflect systemic CKD-MBD-related changes as kidney function declines [17, 18], whereas urinary BI-20492-measured immunoreactive sclerostin may reflect a mixture of filtered sclerostin, altered tubular uptake or degradation, proteinuria-associated tubular stress, and/or immunoreactive fragments generated within the urinary compartment. In our study, urinary sclerostin (Uscl/Ucre) was associated with serum sclerostin and UPCR but not with ACR (shown in online suppl. Fig. S1), suggesting that urinary sclerostin cannot be explained by albuminuria alone. However, these data have not established local renal production or direct tubular action of sclerostin yet.
Recent evidence has also suggested that FGF23 is independently associated with FeMg in non-dialysis CKD [19]. In our cohort, FGF23 similarly showed a correlation with FeMg in the overall correlation matrix (shown in online suppl. Fig. S1), but this relationship was not retained in the fully adjusted model, whereas urinary sclerostin remained significantly associated with ln(FeMg) even after adjustment for FGF23, iPTH, and sKlotho. Taken together, these findings suggest that urinary sclerostin may reflect aspects of renal magnesium handling that are not fully explained by systemic CKD-MBD signals alone.
The present study has several strengths. First, both urinary and serum sclerostin were measured in the same cohort, allowing direct comparison of their associations with renal magnesium handling and supporting the distinct finding that only urinary sclerostin was independently associated with ln(FeMg). Second, this association remained robust across multivariable regression models sequentially adjusting for kidney function, proteinuria, electrolytes, serum sclerostin, and CKD-MBD markers. Third, the main finding was supported by complementary tertile-based and CKD group-stratified analyses.
Several limitations should be acknowledged. First, urinary sclerostin was measured only by the BI-20492 assay, and urinary bioactive sclerostin was not assessed. Therefore, the assay provides an estimate of immunoreactive sclerostin and does not distinguish intact or receptor-binding-competent sclerostin from immunoreactive fragments. Moreover, BI-20472 has not been validated for urine as a standard sample matrix; therefore, urinary bioactive sclerostin measurement would require urine-matrix validation before biological interpretation. Second, all participants had proteinuric CKD. Therefore, the findings cannot be directly generalized to non-proteinuric CKD. Although the association remained significant after adjustment for ln(UPCR), the present data cannot distinguish contributions from altered glomerular permeability, the proteinuric CKD phenotype, or tubulointerstitial injury. Third, biomarkers of tubular injury were not measured; thus, we could not differentiate a distal convoluted tubule-specific mechanism from generalized proximal tubular or tubulointerstitial injury and dysfunction. Finally, the cross-sectional design precludes causal inference and does not establish whether urinary immunoreactive sclerostin directly influences renal magnesium handling or reflects kidney injury processes associated with magnesium wasting.
In conclusion, urinary, but not serum, sclerostin was independently associated with renal magnesium handling in proteinuric nondiabetic CKD. Further mechanistic and longitudinal studies are needed to clarify its biological and clinical significance.
Acknowledgment
The authors thank Ms. Yen-Hsiu Ko for technical assistance.
Statement of Ethics
All procedures performed in the study involving human participants were in accordance with the Ethics of Clinical Research, Declaration of Helsinki. This research protocol was approved by the Institutional Review Board of National Cheng Kung University Hospital (approval no.: B-ER-108-028). Written informed consent was obtained from each participant.
Conflict of Interest Statement
The authors have no conflicts of interest to declare.
Funding Sources
This work was supported by E-Da Hospital and E-Da Cancer Hospital (project EDCHP111001, EDCHP112007, EDCHP113006, NCKUEDA10609, and NCKUEDA10804). The funders had no role in study design, data collection, analysis, interpretation, manuscript preparation, or the decision to submit the article.
Author Contributions
H.-H.L. and M.-C.W. designed and directed the project. T.-H.K., A.-B.W., W.-H.L., J.-Y.C., and C.-C.T. carried out data collection. C.-F.W., T.-H.K., and S.-Y.H. participated in data analysis. C.-F.W. wrote the manuscript. All authors discussed the results and approved the manuscript as submitted. All authors agree to be accountable for the work and to ensure that questions related to the accuracy or integrity of the work are appropriately investigated and resolved.
Funding Statement
This work was supported by E-Da Hospital and E-Da Cancer Hospital (project EDCHP111001, EDCHP112007, EDCHP113006, NCKUEDA10609, and NCKUEDA10804). The funders had no role in study design, data collection, analysis, interpretation, manuscript preparation, or the decision to submit the article.
Data Availability Statement
All data supporting the findings of this study are presented in the manuscript and are available from the corresponding authors upon reasonable request.
References
- 1. Kidney Disease Improving Global Outcomes KDIGO CKD-MBD Update Work Group . KDIGO 2017 clinical practice guideline update for the diagnosis, evaluation, prevention, and treatment of chronic kidney disease-mineral and bone disorder (CKD-MBD). Kidney Int Suppl. 2017;7(1):1–59. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Aguilar A, Gifre L, Urena-Torres P, Carrillo-Lopez N, Rodriguez-Garcia M, Masso E, et al. Pathophysiology of bone disease in chronic kidney disease: from basics to renal osteodystrophy and osteoporosis. Front Physiol. 2023;14:1177829. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Sakaguchi Y, Fujii N, Shoji T, Hayashi T, Rakugi H, Isaka Y. Hypomagnesemia is a significant predictor of cardiovascular and non-cardiovascular mortality in patients undergoing hemodialysis. Kidney Int. 2014;85(1):174–81. [DOI] [PubMed] [Google Scholar]
- 4. Montezano AC, Zimmerman D, Yusuf H, Burger D, Chignalia AZ, Wadhera V, et al. Vascular smooth muscle cell differentiation to an osteogenic phenotype involves TRPM7 modulation by magnesium. Hypertension. 2010;56(3):453–62. [DOI] [PubMed] [Google Scholar]
- 5. Curry JN, Yu ASL. Magnesium handling in the kidney. Adv Chronic Kidney Dis. 2018;25(3):236–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Oliveira B, Cunningham J, Walsh SB. Magnesium balance in chronic and end-stage kidney disease. Adv Chronic Kidney Dis. 2018;25(3):291–5. [DOI] [PubMed] [Google Scholar]
- 7. Figurek A, Rroji M, Spasovski G. Sclerostin: a new biomarker of CKD-MBD. Int Urol Nephrol. 2020;52(1):107–13. [DOI] [PubMed] [Google Scholar]
- 8. Lin Y, Mao L, Chen S, Zhou C. Serum sclerostin in vascular calcification in CKD: a meta-analysis. Ren Fail. 2023;45(1):2186151. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Neto R, Pereira L, Magalhaes J, Quelhas-Santos J, Martins S, Carvalho C, et al. Sclerostin and DKK1 circulating levels associate with low bone turnover in patients with chronic kidney disease stages 3 and 4. Clin Kidney J. 2021;14(11):2401–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Morena M, Jaussent I, Dupuy AM, Bargnoux AS, Kuster N, Chenine L, et al. Osteoprotegerin and sclerostin in chronic kidney disease prior to dialysis: potential partners in vascular calcifications. Nephrol Dial Transpl. 2015;30(8):1345–56. [DOI] [PubMed] [Google Scholar]
- 11. Cejka D, Marculescu R, Kozakowski N, Plischke M, Reiter T, Gessl A, et al. Renal elimination of sclerostin increases with declining kidney function. J Clin Endocrinol Metab. 2014;99(1):248–55. [DOI] [PubMed] [Google Scholar]
- 12. Wu CF, Liou HH, Kuo CC, Tsai MH, Chang MY, Lee YC, et al. The association of urinary sclerostin and renal magnesium handling in type 2 diabetic patients with chronic kidney disease. Kidney Blood Press Res. 2021;46(4):514–22. [DOI] [PubMed] [Google Scholar]
- 13. Biomedica Medizinprodukte GmbH . Bioactive sclerostin ELISA, BI-20472—validation data [technical validation report]. [date unknown]. Available from: https://bmgrp.com/product/cardiovascular/biomedica-bioactive-sclerostin-elisa-human-sost/ (accessed July 3, 2026).
- 14. Traechslin C, Sewing L, Baumann S, Grize L, Vavanikunnel J, Kraenzlin M, et al. Association of total and bioactive serum sclerostin levels with bone metabolism in type 2 diabetes mellitus. J Clin Transl Endocrinol. 2025;40:100393. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Delanaye P, Paquot F, Bouquegneau A, Blocki F, Krzesinski JM, Evenepoel P, et al. Sclerostin and chronic kidney disease: the assay impacts what we (thought to) know. Nephrol Dial Transpl. 2018;33(8):1404–10. [DOI] [PubMed] [Google Scholar]
- 16. Wang LF, Wu H, Xu Y, Deng M, Han XL, Bai D. Effect of SOST gene deletion on the progression of renal interstitial fibrosis in obstructive kidney injury. Ren Fail. 2015;37(9):1514–7. [DOI] [PubMed] [Google Scholar]
- 17. Lima F, Monier-Faugere MC, Mawad H, David V, Malluche HH. FGF-23 and sclerostin in serum and bone of CKD patients. Clin Nephrol. 2023;99(5):209–18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Cejka D, Herberth J, Branscum AJ, Fardo DW, Monier-Faugere MC, Diarra D, et al. Sclerostin and Dickkopf-1 in renal osteodystrophy. Clin J Am Soc Nephrol. 2011;6(4):877–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Grigore TV, Zuidscherwoude M, Witasp A, Barany P, Wernerson A, Bruchfeld A, et al. Fibroblast growth factor 23 is independently associated with renal magnesium handling in patients with chronic kidney disease. Front Endocrinol. 2022;13:1046392. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
All data supporting the findings of this study are presented in the manuscript and are available from the corresponding authors upon reasonable request.
