Reply to “Comment on: Difference between kidney function by cystatin C versus creatinine and association with muscle mass and frailty”

We are grateful for the opportunity to reply to Du and Hou's letter to the Editor commenting on our recent study on the difference in estimated glomerular filtration rates (eGFRDiff) by cystatin C versus creatinine and muscle mass and frailty in the MrOS cohort.¹

We fully agree with Du et al. that eGFR formulas² are based on assumptions that may not be honored depending on the population studied. Both creatinine and cystatin C are imperfect markers of kidney function because of all the non-GFR determinants³ that are rightfully mentioned by the authors. Inflammation⁴ is indeed a relevant confounder as it may affect both creatinine and cystatin C level as well as physical activity. The models in our study were adjusted for diabetes, hypertension, kidney disease, and smoking status.

It is precisely to highlight these known and unknown confounders that we opted to investigate the difference in eGFR by cystatin C versus creatinine. The eGFR equations only “adjust” for age, sex, and body size but there are many more non-GFR determinants to both creatinine and cystatin C. eGFRDiff can be thought of as a proxy for those non-GFR determinants.

The main goal of the study was to investigate whether muscle mass, as defined by deuterated creatine (D3Cr) dilution, may explain the relationship between eGFRDiff and frailty. This is why we did not include clinical outcomes such as activities of daily living or disability and falls. However, as those functional measures tend to be affected in people who are frail,⁵ we concur with Du et al. that they are clinically relevant and should be evaluated in future research.

Various conditions may make the kidney function vary, and we recognize that repeated or longitudinal data would be more informative than a single data point. We opted to perform a cross-sectional analysis of eGFRDiff and D3Cr dilution measurements as an initial study, but we agree with the authors that future studies should evaluate changes over time to confirm and strengthen our findings.

In conclusion, to our knowledge, our study was the first to assess the relationship between eGFRDiff and frailty while accounting for muscle mass as defined by D3Cr dilution. We found that D3Cr at least in part explains the association of eGFRDiff and frailty. These findings would be strengthened by longitudinal data and will need to be repeated and validated in other populations.

OAP and DER drafted the letter.

The authors declare no conflicts of interest.

The funders had no role.