Assessing the use of urinary creatinine excretion rate to estimate urine output in lactating cows

Creatinine concentration in urine is often used to estimate urine output in cows. Studies have commonly used a fixed coefficient for the creatinine excretion rate, although this coefficient varies among studies. However, 29 mg/kg BW·d^-1 is the most popular value. The current study determined the coefficient for the creatinine excretion rate of individual cows and validated the use of a fixed coefficient to estimate urine output. Two experiments were conducted with 8 cows in each experiment. The experiments were conducted in a replicated 4 × 4 Latin square design. In experiment 1, dietary treatments were 2 levels of RDP with or without isoacids supplementation in 2 × 2 factorial arrangement. In experiment 2, diets at 2 levels of DCAD supplemented with saturated fatty acids or a Met analog were fed to cows as dietary treatments. Total urine was collected over multiple days in each period, followed by BW measurements before feeding on 2 consecutive days. Daily subsamples of urine were composited by cow and period and assayed for creatinine concentration. The coefficient for the creatinine excretion rate was calculated by daily urine creatinine excretion (mg/d) divided by BW for individual cows. Data from each experiment were analyzed using the MIXED procedure of SAS, where fixed effects of period, diet, and their interaction and random effects of cow within square were included. Additionally, data from both experiments were combined and analyzed using the MIXED procedure of SAS, where cow within trial and period within trial were random effects and trial was the fixed effect. The experimental period and dietary treatments did not affect total urinary creatinine excretion or the coefficient within each experiment, although the periods influenced milk yield and BW. There was no difference in DMI and milk yield between the experiments, but BW differed (672 vs. 746 kg). In addition, urinary creatinine excretion (12.8 vs. 17.8 g/d) and the coefficient (19.1 vs. 23.9 mg/kg BW·d^-1) were greater in experiment 2 compared with experiment 1. A linear regression analysis between BW and urinary creatinine excretion was conducted using the MIXED procedure of SAS (experiment as a random effect). The coefficient of 22.0 mg/kg BW·d^-1 was obtained from the analysis and this was lower than the 29 mg/kg BW·d^-1 value that has been commonly used in the literature. In conclusion, the coefficient varied among cows, likely because BW does not represent muscle mass. Therefore, caution is needed when urine output is estimated using a fixed coefficient in an experiment.