Authors’ Reply

• cardiovascular
• Epidemiology and outcomes
• FGF-23

The published observational data show similarly sized epidemiologic associations between increased fibroblast growth factor-23 (FGF-23) concentration and risk of a range of cardiovascular (atherosclerotic and nonatherosclerotic) and noncardiovascular outcomes. There is also an absence of any clear exposure-response relationship.¹

Pelletier and Fouque² have hypothesized that Klotho modifies FGF-23 associations with cardiovascular outcomes, and confirmation in large-scale studies is of interest. Such analyses, however, are unlikely to provide an explanation for the nonspecificity of FGF-23 associations. Instead, as recommended by Zhou et al.,³ further mechanistic experiments are needed. We suggest that such studies include the full range of noncardiovascular outcomes identified as associated with FGF-23. In the absence of good mechanistic evidence clearly supporting FGF-23’s role in a much wider range of disease processes, residual confounding is currently the most plausible explanation for the nonspecific associations. The epidemiology is suggesting the relationship between FGF-23 and cardiovascular disease risk may not be one of cause and effect.

• Copyright © 2018 by the American Society of Nephrology