Changes in Mortality According to Creatinine/Cystatin C Ratio in Chronic Kidney Disease and Non-chronic Kidney Disease Patients

Abstract

Background: Serum creatinine and cystatin C are not only good indicators of renal function but have also been confirmed to be related to disease prognosis and mortality in various diseases via creatinine/cystatin C ratio (CCR). However, although they are biomarkers of renal function, there is no study regarding renal impairment as a confounding variable in the relationship between CCR and all-cause mortality.

Methods: Patients who had simultaneous measurements of serum creatinine and cystatin C between 2003 and 2020 were enrolled. The patients with chronic kidney disease (CKD) were defined as having an estimated glomerular filtration rate (eGFR) CKD-EPI Cr-Cystatin C < 60 ml/min/1.73 m². CCR was calculated by dividing the serum creatinine level by the cystatin C level measured on the same day. The main outcome assessed was all-cause mortality according to CCR in CKD or non-CKD groups.

Results: Among the 8,680 patients in whom creatinine and cystatin C levels were measured simultaneously, 4,301 were included in the CKD group, and 4,379 were included in the non-CKD group, respectively. CCR was 1.4 ± 0.6 in total participants. The non-CKD group showed higher mean CCR, (1.5 ± 0.7 vs. 1.3 ± 0.5) as well as a wider distribution of CCR (p < 0.001) when compared to the CKD group. In non-CKD group, 1^st, 4^th and 5^th quintiles of CCR significantly increased the all-cause mortality risk compared to 2^nd quintile of CCR, suggesting U-shaped mortality risk according to CCR in non-CKD. On the other hand, in CKD group, the risk of all-cause mortality linearly increased and 5^th quintile of CCR showed 1.82 times risk of mortality compared to 2^nd quintile of CCR. In the subgroup analysis of mortality by age and sex, the mortality difference according to CCR were diminished in old age and female sex subgroups.

Conclusion: We discovered a U-shaped relationship between mortality and CCR levels in normal renal function, and an increased risk of mortality in CKD with elevated CCR.

Keywords: creatinine; creatinine/cystatin C ratio; cystatin C; mortality; renal dysfunction.

Figure 1

Density plot (A) and Q-plot (B) for the distribution of CCR between CKD and non-CKD participants.

Figure 2

Cubic spline curves for all-cause mortality in non-CKD (A) and CKD (B) participants. Solid curves represent the hazard ratio for all-cause mortality, and dashed lines represent the upper and lower confidence interval of the hazard ratio.