NxStage Home Hemodialysis System
Privacy Policy


Follow Hemodialysis_co on Twitter
Hemodialysis Dialysis
Author Interviews


Author Interview: Yoshitaka Isaka, M.D Ph.D.

Department of Nephrology
Osaka University Graduate School of Medicine


Intact fibroblast growth factor 23 levels predict incident cardiovascular event before but not after the start of dialysis.

Nakano C, Hamano T, Fujii N, Obi Y, Matsui I, Tomida K, Mikami S, Inoue K, Shimomura A, Nagasawa Y, Okada N, Tsubakihara Y, Rakugi H, Isaka Y.
Department of Geriatric Medicine and Nephrology, Osaka University Graduate School of Medicine, Osaka, Japan.
Bone. 2012 Mar 6. [Epub ahead of print]

What are the main findings of the study?

In this study, we tried to identify a predictor of fatal/non-fatal cardiovascular events among biomarkers related to mineral bone disorders; 25-hydroxyvitamin D, bone-specific alkaline phosphatase, fibroblast growth factor 23 (FGF23), and parathyroid hormone (PTH). We measured biologically active “whole PTH” and “intact FGF23” instead of c-terminal FGF23 in predialysis patients.

The main findings are summarized as follows.

  1. In pre-dialysis patients with CKD, intact FGF23 predicted future cardiovascular events only before the initiation of dialysis.

  2. When we did not censor patients at the initiation of dialysis and continued follow-up after starting dialysis, intact FGF23 did not predict future cardiovascular events.

  3. The accuracy of risk stratification for predialysis cardiovascular events was significantly improved by adding FGF23 to a conventional model.

Were any of the findings unexpected?

No significant relationship was observed between FGF23 levels and incident cardiovascular disease (CVD) during the entire follow-up period including both pre- and post- dialysis initiation despite greater statistical power due to larger number of the patients achieving outcome.

What should clinicians and patients take away from this study?

In predialysis CKD, elevated FGF23 was an important risk factor of future cardiovascular events before the initiation of dialysis. Approximately 7 out of 100 patients were correctly reclassified by adding FGF23 to the model of age, sex, diabetes, prior CVD, pulse pressure, and eGFR.

However, intact FGF23 did not predict cardiovascular events during the entire follow-up period.

This might be due to a single measurement of FGF23 levels in the pre-dialysis phase. Another reason might be that the incidence of heart failure after the start of dialysis was reduced by the removal of fluid by dialysis, as shown in much lower percentage of heart failure among CVD in the dialysis phase compared to pre-dialysis.

What recommendations do you have for future studies as a result of your study?

In the course of CKD, FGF23 levels should be measured repeatedly (e.g. at the initiation of dialysis in addition to pre-dialysis phase) for risk stratification of future cardiovascular events, since we cannot predict the time course of FGF23 by a single measurement in time.

More Author Interviews from Hemodialysis.com
More on Heart Disease and Kidney Disease


Subscribe to the Free Hemodialysis.com Newsletter Privacy Policy Advertising Policy Hemodialysis.com Mission Statement