Single FGF-23 measurement and time-averaged plasma phosphate levels in hemodialysis patients

Abstract

Background and objectives: Plasma phosphate levels display considerable intraindividual variability. The phosphatonin fibroblast growth factor 23 is a central regulator of plasma phosphate levels, and it has been postulated to be a more stable marker than conventional CKD-mineral and bone disorder parameters. Thus, fibroblast growth factor 23 has been hypothesized to reflect time-averaged plasma phosphate levels in CKD patients.

Design, setting, participants, & measurements: Among 40 patients from the outpatient dialysis center, serial measurements of plasma calcium and phosphate (before every dialysis session) as well as C-terminal fibroblast growth factor 23, parathyroid hormone, and alkaline phosphatase (one time weekly) were performed over a study period of 4 weeks in November and December of 2011. Intraindividual variability of repeated plasma fibroblast growth factor 23 measurements compared with other CKD-mineral and bone disorder markers was tested, and the association of a single plasma fibroblast growth factor 23 measurement with time-averaged plasma phosphate levels was analyzed.

Results: Against expectations, intraindividual variability of fibroblast growth factor 23 (median coefficient of variation=27%; interquartile range=20-35) was not lower than variability of plasma phosphate (median coefficient of variation=15%; interquartile range=10-20), parathyroid hormone (median coefficient of variation=24%; interquartile range=15-39), plasma calcium (median coefficient of variation=3%; interquartile range=2-4), or alkaline phosphatase (median coefficient of variation=5%; interquartile range=3-10). Moreover, the correlation between the last fibroblast growth factor 23 measurement after 4 weeks and time-averaged plasma phosphate did not surpass the correlation between the last fibroblast growth factor 23 measurement and a single plasma phosphate value (r=0.67, P<0.001; r=0.76, P<0.001, respectively).

Conclusions: Surprisingly, fibroblast growth factor 23 was not more closely associated to time-averaged plasma phosphate levels than a single plasma phosphate value, and it did not show a lower intraindividual variability than other tested markers of CKD-mineral and bone disorder. Thus, fibroblast growth factor 23 should not be used in clinical practice as a reflector of time-averaged plasma phosphate levels.

Figure 1.

Plasma phosphate and calcium levels throughout the study period. Note that measurements 1, 4, 7, and 10 refer to the first measurements of each week; the 4 study weeks are separated by dashed lines. Phosphate and calcium levels are depicted as means and SDs.

Figure 2.

Range of variation in CKD-MBD parameters. Depicted are medians, interquartile ranges, and minimum and maximum blood values of (A) calcium, (B) phosphate, (C) parathyroid hormone, (D) alkaline phosphatase, (E) C-terminal FGF-23 on a linear scale, and (F) C-terminal FGF-23 on a logarithmic scale. Values for each individual patient are plotted in rank order of their individual means. AP, alkaline phosphatase; CKD-MBD, CKD–mineral and bone disorder; FGF-23, fibroblast growth factor 23; PTH, parathyroid hormone.

Figure 3.

Intraindividual variation of CKD-MBD parameters. For each individual patient, coefficients of variation were calculated from four measurements (at the end of the first, second, third, and fourth weeks). Depicted are medians, interquartile ranges, and outliers from individual coefficients of variation across the study cohort of 40 hemodialysis patients. AP, alkaline phosphatase; CKD-MBD, CKD–mineral and bone disorder; FGF-23, fibroblast growth factor 23; PTH, parathyroid hormone.

Figure 4.

Correlation of FGF-23 measured at study end with (A) a single plasma phosphate measurement obtained at study end or (B) plasma phosphate measurements averaged across the 4-week study period. As a representative example illustrating wide ranges of individual plasma phosphate measurements in patients with similar FGF-23 levels, lines are drawn for patients with FGF-23 of ∼8.000 rU/ml. Note that FGF-23 levels are depicted on a logarithmic scale. FGF-23, fibroblast growth factor 23.