Regarding our analyses examining cystatin C as a risk factor for outcomes in chronic kidney disease, we do not make the assertion that cystatin C is better than creatinine in diagnosing stage 3 kidney disease. Regarding the question about our methods, cystatin C was measured in frozen samples collected at baseline from participants of the randomized cohort of the Modification of Diet in Renal Disease Study, as stated in the Methods section. There is a precedent for using frozen stored samples to analyze cystatin C (1–3). Serum and plasma samples stored at temperatures ranging from room temperature to –20 °C and from 2 days to more than 1 month were found to be stable (4, 5). Our samples were stored at –70 °C.

Moreover, the associations of cystatin C with variables, such as creatinine, estimated GFR, and body mass index, are similar to those seen in other studies. Although other factors, such as body mass index, may be associated with cystatin C, the strong correlation of cystatin with iothalamate GFR (r = 0.85) suggests that kidney function is the primary determinant of cystatin C.

Studying a cystatin-based equation, such as the one mentioned by Dr. Delanaye and colleagues, was outside the scope of our paper. Our primary focus was to compare cystatin C, creatinine, and iothalamate GFR as risk factors for outcomes in chronic kidney disease, given that only estimated GFR was available for comparison in past reports. Equations that use cystatin C alone are useful in showing the level of estimated GFR, which corresponds to a given level of cystatin C but will fundamentally have the same strength as risk factors. The question is more complex and needs to be addressed separately for equations that include multiple factors.

As noted, the correlation between iothalamate GFR and cystatin C in our study was 0.85; therefore, Dr. Delanaye and colleagues' hypothesis that basal metabolic rate is the underlying mechanism for the prognostic utility of cystatin C assumes an overwhelming correlation between basal metabolic rate and GFR by proxy. Although basal metabolic rate may influence GFR to some extent, we believe that it is not the sole or primary determinant of GFR. Furthermore, this postulated mechanism would not be limited to the literature of cystatin C, but rather to all studies of chronic kidney disease as a risk factor for cardiovascular disease. Without reliable measures of basal metabolic rate in large cohort studies, its importance cannot be evaluated directly. In our opinion, the basal metabolic rate is unlikely to be the major mechanism to explain the cardiovascular disease risk for chronic kidney disease.