The rate of the Jaffé reaction depends on the concentration of sodium hydroxide; the pseudo-first-order rate constant of the reaction, at 37 degrees C in 10 mmol/L picrate solution, is 0.004 mmol/L. We formulated an automated method to determine urinary creatinine directly without manual sample dilution. The conditions are as follows: 10 mmol/L picrate and 60 mmol/L sodium hydroxide (final concentrations); ratio of sample to final volume, 1:41; temperature, 37 degrees C; wavelengths of measurement, 500 or 510 nm; interval of measurement, 30 to 90 s; and mode of measurement, kinetic. Determinations of creatinine in patients' samples by the new method compared favorably with those obtained with the AutoAnalyzer and aca. The run-to-run CVs were 3.6% or less, and the method was accurate for concentrations of creatinine up to 3000 mg/L. We recommend this method as a good replacement for the AutoAnalyzer or aca methods.

The concentration of creatinine in serum, which is used to estimate glomerular filtration rate, is measured by reaction with alkaline picrate, but this reaction is not specific for creatinine. Although several other cephalosporin antibiotics have been reported not to react with picrate, we reacted picrate with creatinine, cefoxitin, penicillin, and eight different cephalosporins, and found that all compounds reacted with picrate and showed superimposable spectrophotograms with absorption maxima at 485 nm. From these results we conclude that the color-absorbing moiety of the product is the picrate molecule. Further, the structure common to creatinine and the cephalosporins, cefoxitin, or penicillin is the carbonyl group attached to a nitrogen and a carbon atom. We postulate that the carbonyl group with the adjacent carbon and nitrogen atoms is probably the chemical moiety that reacts with picrate to absorb energy at 485 nm.

We compared results for urinary creatinine, serum creatinine, and creatinine clearance, as determined with the Ames Seralyzer, with results determined with the Beckman ASTRA, the DuPont aca, and Technicon's AutoAnalyzer and SMAC. Results for urinary creatinine from the Seralyzer differed significantly (p less than 0.05) from those obtained with the ASTRA and AutoAnalyzer, but not with the aca. The Seralyzer results for serum creatinine were at least 1.0 mg/L higher (p less than 0.05) than by the other three methods. Results for creatinine clearance from the Seralyzer were 8 to 11 mL/min lower (p less than 0.05) than results by the other three methods. These differences are related to the positive interference by bilirubin in the Seralyzer creatinine method. We also evaluated 23 other compounds for interference with these methods for creatinine.