A decreased level of serum soluble Klotho is an independent biomarker associated with arterial stiffness in patients with chronic kidney disease

Abstract

Background: Klotho was originally identified in a mutant mouse strain unable to express the gene that consequently showed shortened life spans. In humans, low serum Klotho levels are related to the prevalence of cardiovascular diseases in community-dwelling adults. However, it is unclear whether the serum Klotho levels are associated with signs of vascular dysfunction such as arterial stiffness, a major determinant of prognosis, in human subjects with chronic kidney disease (CKD).

Methods: We determined the levels of serum soluble Klotho in 114 patients with CKD using ELISA and investigated the relationship between the level of Klotho and markers of CKD-mineral and bone disorder (CKD-MBD) and various types of vascular dysfunction, including flow-mediated dilatation, a marker of endothelial dysfunction, ankle-brachial pulse wave velocity (baPWV), a marker of arterial stiffness, intima-media thickness (IMT), a marker of atherosclerosis, and the aortic calcification index (ACI), a marker of vascular calcification.

Results: The serum Klotho level significantly correlated with the 1,25-dihydroxyvitamin D level and inversely correlated with the parathyroid hormone level and the fractional excretion of phosphate. There were significant decreases in serum Klotho in patients with arterial stiffness defined as baPWV≥1400 cm/sec, atherosclerosis defined as maximum IMT≥1.1 mm and vascular calcification scores of ACI>0%. The serum Klotho level was a significant determinant of arterial stiffness, but not endothelial dysfunction, atherosclerosis or vascular calcification, in the multivariate analysis in either metabolic model, the CKD model or the CKD-MBD model. The adjusted odds ratio of serum Klotho for the baPWV was 0.60 (p = 0.0075).

Conclusions: Decreases in the serum soluble Klotho levels are independently associated with signs of vascular dysfunction such as arterial stiffness in patients with CKD. Further research exploring whether therapeutic approaches to maintain or elevate the Klotho level could improve arterial stiffness in CKD patients is warranted.

Figure 1. Correlation between the serum Klotho levels (pg/mL) and various parameters.

The relationships between the serum Klotho levels and patient age (years) (A), estimated glomerular filtration rate (eGFR) (mL/min/1.73 m²) (B) and markers of chronic kidney disease-mineral and bone disorder (CKD-MBD), including 1,25-dihydroxyvitamin D (1,25D) (pg/mL) (C), log intact parathyroid hormone (PTH) (pg/mL) (D), fractional excretion of phosphate (FEPi) (%) (E) and fractional excretion of calcium (FECa) (%) (F) are shown. The serum Klotho levels were inversely correlated with age and positively correlated with eGFR (A, B). Regarding CKD-MBD markers, the serum Klotho levels were significantly correlated with 1,25D and negatively correlated with log intact PTH and FEPi; however, no significant correlation was observed with FECa (C–F). (A–F) N = 114.

Figure 2. Box and line plots showing the levels of serum Klotho (pg/mL) according to the stratified levels of vascular dysfunction.

They include flow-mediated dilatation (FMD) (%), a marker of endothelial dysfunction (A), ankle-brachial pulse wave velocity (baPWV) (cm/sec), a marker of arterial stiffness (B), maximum intima-media thickness (max IMT) (mm), a marker of atherosclerosis (C), and the aortic calcification index (ACI) (%), a marker of vascular calcification (D). The serum Klotho levels were significantly lower in patients with FMD<6.0%, PWV≥1400 cm/s, max IMT≥1.1 mm and ACI>0% compared to patients with FMD≥6.0%, PWV<1400 cm/s, max IMT<1.1 mm and ACI = 0%, respectively (A–D). (A) N = 70 and n = 40 in FMD<6.0% and FMD≥6.0%, respectively. (B) N = 60 and n = 45 in PWV<1400 cm/s and PWV≥1400 cm/s, respectively. (C) N = 82 and n = 29 in max IMT<1.1 mm and max IMT≥1.1 mm, respectively. (D) N = 28 and n = 75 in ACI = 0% and ACI>0%, respectively. The boxes denote the medians and 25th and 75th percentiles. The lines mark the 5th and 95th percentiles.

Figure 3. Multivariate odds ratio for ankle-brachial pulse wave velocity (baPWV) among patients with CKD displayed as the odds ratio (OR) (solid boxes) with 95% confidence intervals (CIs) (horizontal limit lines).

For continuous variables, the unit of change is given in parenthesis based on the multivariate model described in Table 2. MBP, mean blood pressure; eGFR, estimated glomerular filtration rate; PTH, parathyroid hormone; 1,25D, 1,25-dihydroxyvitamin D; FGF23, fibroblast growth factor 23.