Background: Our multicenter clinical trial study for stage 4 chronic kidney disease (CKD) populations was conducted at 21 centers in China during the period 2011 to 2016. The CKD definition is based on glomerular filtration rate (GFR) values which can be estimated by creatinine-based predictive formulas. The validity and reliability of GFR estimation is thus largely dependent on the accurate and precise serum creatinine (SCr) measurements. As an integral part of this multicenter study, it is important to ensure the precision, accuracy, and center-to-center comparability of the SCr results. Methods: Prior to initiating the study, we unified the measurement method of SCr determination as an enzymatic method and standardized the procedure in all of the laboratories. Then, the analytical performance of each analyzer at each laboratory was evaluated, including precision, accuracy, and comparability. Results: All within-run and total CVs of the low and high level internal quality control (IQC) were comprised between 0.2% and 4.1% (< 1/3 CLIA'88). Total error of the IQC fall within the maximum 12% at all centers. The analytical bias against the Standard Reference material 967a target was less than +/- 0.5% at Central Laboratory, indicating good accuracy. Correlation between the analyzers and the reference method were very high (r > 0.99). Passing-Bablok regression showed no significant deviation from linearity (p > 0.05). Bland-Altman analysis also showed good agreement (>= 95% of results fell within the 95% limits of agreement). Conclusions: Performance evaluation helped in addressing pre-analytical variations in measurement and gave optimal quality assurance of laboratory measurement in the context of a multicenter clinical trial study.