An enzymatic kit method for the determination of plasma creatinine was optimised for use with a centrifugal analyser and its performance characteristics and practicability compared with an end point and a kinetic Jaffé-based method. The enzymatic method exhibited several advantages over Jaffé-based methods--namely, smaller sample size, rapid sample throughput (200 per hour), and improved specificity. Glucose, acetoacetate, and cefoxitin did not interfere with the enzymatic method, although bilirubin did cause a negative interference which depended on both creatinine and bilirubin concentrations. The enzymatic method has particular clinical application in neonates, diabetic ketotic patients, and those receiving cephalosporins.