Ketone measurements using dipstick methodology in cats with diabetes mellitus

O bjectives : To compare the results of urine and plasma ketone dip test in a group of diabetic cats with possible ketosis or ketoacidosis, using laboratory plasma β-hydroxybutyrate measurements as the gold standard.
M ethods : According to clinical examinations, plasma β-hydroxybutyrate measurements and venous blood gas analysis, 54 cats with diabetes mellitus were classified as non-ketotic (n=3), ketotic (n=40) or ketoacidotic (n=11). Plasma and urine acetoacetate concentrations were determined using urine reagent strips.
R esults : Although there was a significant positive correlation between blood and urine ketone measurements (r=0·695, P<0·001), the results differed significantly (Z=?3·494, P<0·001). Using the differential positive rates, the best cut-off value to detect cats with ketoacidosis was 1·5 mmol/l for urine and 4 mmol/l for plasma. The sensitivity/specificity was 82/95 per cent for urine and 100/88 per cent for plasma, respectively.
C linical S ignificance : The urine and plasma ketone dip tests have a different diagnostic accuracy, and results have to be interpreted differently. Because of its high sensitivity, the plasma ketone dip test performs better than the urine ketone dip test to identify cats with impending or established ketoacidosis.