N-terminal pro-B-type natriuretic peptide (NT-proBNP) concentration has been used to identify cardiovascular causes of dyspnea in dogs. However, the day-to-day variability of circulating NT-proBNP in dogs is not known. The purpose of this study was to examine the weekly variability of NT-proBNP concentration in normal dogs.
Thirty-seven clinically normal dogs from 2 study sites were examined prospectively. All dogs had normal cardiac auscultation and no significant abnormalities on 2-D, M-mode and Doppler echocardiographic examinations. Serum samples from site A (n=19) and plasma samples from site B (n=18) for NT-proBNP assay (Canine CardioCare, Veterinary Diagnostics Institute, Irvine, CA) were obtained from dogs at one week intervals for a total of 3 samples.
The median age of the patient population was 5 years (range: 0.8-10 years), whereas the median weight was 22 kilograms (range: 5-50 kilograms). There was no significant difference in median NT-proBNP for all dogs between week 1 (378 pmol/L [25-75% interquartile range: 219-485 pmol/L], week 2 (366 pmol/L [231-496]), and week 3 (367 pmol/L [223-469]) (p=0.99). However, the coefficients of variation (CV=standard deviation/average) for individual dogs ranged from 0-51% (median CV=12%). Eighteen of 37 dogs (49%) had CV <12%, 15 of 37 dogs (41%) had CV of 12-30%, and 4 of 37 dogs (11%) had CV >30%. Six of 37 dogs (16%) possessed one NT-proBNP reading that exceeded the manufacturer's current upper normal value of 566 pmol/L. Two of 37 dogs (5%) possessed two readings >566 pmol/L, and in 2 of 37 dogs (5%) all three weekly results were >566 pmol/L. The median serum NT-proBNP concentration of dogs from study site A was significantly higher than plasma concentrations of dogs from study site B, (Site A, 420 pmol/L [25-75% interquartile range: 296-542 pmol/L], vs. Site B, 257 pmol/L, [162-432]; p=<0.001), suggesting either a difference between serum and plasma concentrations vs. a regional difference in the study population.
Weekly variability of NT-proBNP measurement is generally low, however, in a small percentage of dogs, variability has the potential to confound a single result and serial measurements are recommended.