Fibroblast Growth Factor-23, Mineral Metabolism, and Adiposity in Normal Renal Function: The Multi-Ethnic Study of Atherosclerosis

Presentation Number: SUN 358
Date of Presentation: April 2nd, 2017

Sarah Zaheer*1, Ian de Boer2, Matthew Allison3, Jenifer Michelle Brown4, Bruce Psaty2, Cassianne Robinson-Cohen2, Erin Michos5, Joachim Ix3, Bryan Kestenbaum6, David Siscovick7 and Anand Vaidya1
1Brigham and Women's Hospital, Harvard Medical School, Boston, MA, 2University of Washington, Seattle, WA, 3University of California, San Diego, 4Brigham and Women's Hospital, Harvard Medical School, MA, 5Johns Hopkins School of Medicine, Baltimore, MD, 6University of Washington, 7The New York Academy of Medicine


CONTEXT: Excess adiposity is associated with poor bone mineralization and increased risk of fracture. Fibroblast-growth factor 23 (FGF23) plays an important role in phosphate, vitamin D, and skeletal physiology, and may contribute to impaired skeletal mineralization. Increased expression of FGF family proteins has been described in adipocytes.

OBJECTIVE: We hypothesized that greater body adiposity would be associated with higher FGF23 levels among individuals with normal renal function. We investigated the association between adiposity and FGF23 within the context of other regulators of mineral metabolism.

METHODS: We conducted a cross-sectional analysis of participants in the Multi-Ethnic Study of Atherosclerosis who had eGFR >60 mL/min/1.73m2. Body adiposity was assessed using crude measures (BMI, waist circumference [WC], and waist-to-hip ratio [WHR]; n=5610) as well as more refined measures (abdominal adipose tissue area by computed tomography; n=1313) that were available in a smaller subset, to confirm our findings. FGF-23, and other regulators of mineral metabolism, were evaluated as a function of BMI, WC, WHR, and by depots of abdominal adipose tissue, including total (TAAT), subcutaneous (SAT), and visceral (VAT) adipose tissue areas. Multivariable linear regression was used to assess the independent association between adiposity and FGF23, adjusting for known and potential confounders of FGF23 (including parathyroid hormone, serum and urinary phosphate, serum and urinary calcium, vitamin D, and eGFR).

RESULTS: FGF23 was higher across BMI categories (BMI <25: 37.7 pg/mL, BMI 25-29.99: 38.7 pg/mL, BMI 30-39.99: 39.8 pg/mL, BMI ≥40: 40.9 pg/mL, unadjusted P-trend<0.0001). There was a significant and positive association between BMI (as a continuous variable) and FGF23, independent of known confounders of FGF23 (adjusted β=+7.2% higher FGF23 per 10 kg/m2of BMI, P<0.0001). Similar results were observed using WC and WHR (adjusted β=+4.3% and β=+2.8% higher FGF23 per standard deviation [SD] of WC and WHR, respectively, P<0.0001). TAAT was independently associated with higher FGF23 (adjusted β=+3.1% higher FGF23 per SD of adipose area, P<0.01), with similar results observed for SAT and VAT. Notably, the all of the positive associations between adiposity and FGF23 were despite the fact that eGFR did not decline and serum phosphate levels did not rise with higher adiposity.

CONCLUSION: In a large cohort with normal renal function, adiposity was associated with higher FGF23 levels independent of known regulators of FGF23, including eGFR and serum phosphate. FGF23 may be an additional factor contributing to the association of adiposity and poor skeletal health. Further studies are needed to evaluate the cause(s) of higher FGF23 in settings of greater adiposity and the potential impact on skeletal health.


Nothing to Disclose: SZ, ID, MA, JMB, BP, CR, EM, JI, BK, DS, AV