LAUSR

Dear Editor I read the recent article ‘Serum cystatin C level is associated with locomotive syndrome risk and can be an early predictor in community-living people: The Yakumo study’ [1] with great interest. This study evaluated the relationship between the 25-question geriatric locomotive function score (Locomo25) [2] and serum cystatin C level and reported a significant correlation between the two in patients aged 60 years, concluding that cystatin C level is an early predictor of locomotive syndrome (LS). While this is timely and an innovative study, I am afraid that its conclusion got ahead of itself for the reason as follows. Cystatin C is a protein encoded by the CST3 gene [3] that is widely used as a biomarker of renal function. It is well known that cystatin C correlates with aging, gender, and weight as well as serum creatinine level, and it is more closely correlated with aging [4]. Moreover, even in elderly people aged 64–100, its level has been shown to increase with age [5]. The authors tried to evaluate the predictive value of serum cystatin C level to assess the presence or absence of LS using receiver operating characteristics (ROC) analysis; however, there is a risk of confounding factors with this method. As age and serum cystatin C level are so strongly correlated, and these parameters are jointly and negatively correlated with renal function, the ROC analysis is at risk of predicting age or renal function for LS. Serum creatinine level was also evaluated as a biomarker of renal function. However, serum creatinine level is affected by muscle volume [6]. The reason for the lack of a statistically significant difference in the subgroup aged 60 years is lying on gender distribution. As the gender distribution in this study was 29 males and 26 females, consequently gender difference was mixed. The analysis should be separated by gender. Alternatively, the authors should include estimated glomerular filtration rate (eGFR) with cystatin C to evaluate LS because eGFR with cystatin C diminishes the influence of gender. If not, the influence of gender, age, and body weight should be corrected in the evaluation of the effect of cystatin C on LS. I think it would be interesting to evaluate the correlation between LS and the serum creatinine to cystatin C ratio (Cr/CysC). Cr/CysC is presently in the spotlight as it reflects muscle volume, specifically the sarcopenia index [7–13]. In contrast to cystatin C, which is excreted by all nucleated cells, serum creatinine is a derivative of the skeletal muscle protein. Therefore, the relative value of creatinine to cystatin C reflects skeletal muscle mass volume. As Cr/CysC reflects total muscular volume, it is reasonable that Cr/CysC correlates with Locomo25 scores because Cr/CysC reflects total skeletal muscular volume including the influences of gender, age, and body mass. The influence of Cr/CysC on LS is more likely to make sense. Therefore, the ROC may have demonstrated a significant level of Cr/CysC to predict the presence or absence of LS.