Automated and visual analysis of commercial urinary dipsticks in dogs, cats and cattle

Two dipsticks developed for human use were evaluated for routine urinalysis and for detection of proteinuria in dogs (n=101), cats (n=50) and cattle (n=100). The aims were to determine their diagnostic usefulness in dogs, cats and cattle and to compare automated versus visual methods of reading. Results obtained with automated reading correlated better with reference methods than visual reading. Correlation with the reference methods was good to excellent for automated estimation of creatinine (dog: r(s)=0.86, cat: r(s)=0.83, cattle: r(s)=0.87) and pH (dog: r(s)=0.96, cat: r(s)=0.91, cattle: rs=0.94). The correlation was good for protein (dog: r(s)=0.88, cat: r(s)=0.91), glucose (cat: r(s)=0.83) and urine protein:creatinine (UPC) ratio (dog: r(s)=0.75, cat: r(s)=0.89). Estimation of proteinuria in cattle and pyuria in cats lacked specificity and detection of isosthenuria lacked sensitivity in all species. Semiquantitative estimation of UPC ratio was specific (100% and 91.2% at a cut-off of 0.2 in cats and 0.4 in dogs, respectively).