Canine bone blood flow measurements using serial microsphere injections

The objective of this study was to determine the reproducibility of serial bone blood flow (BBF) quantitation using multiple microsphere injections. Three consecutive estimates of BBF were obtained, using 15-mu radionuclide-labeled microspheres from 21 anesthetized adult dogs. A dose of 3 million spheres/kg was used in seven dogs (Cohort 1); a dose of 0.5 million spheres/kg was used in the remaining 14 dogs (Cohorts 2 and 3). Estimates of BBF were made at an average of 129, 153, and 175 minutes after the animals had been anesthetized in the first two cohorts and 179, 203, and 225 minutes in Cohort 3. The dogs in Cohort 1 had no surgical intervention; the dogs in Cohorts 2 and 3 had increasingly complex surgical interventions. Despite stabilization of cardiovascular status, BBF was found to vary by 33.4% in Cohort 1, 25.7% in Cohort 2, and 42.5% in Cohort 3 over the three injections. Cortical BBF fell by 13.9%, 12.1%, and 12.4% between the first and second, and by 31.0%, 11.2%, and 29.9% between the second and third estimates for Cohorts 1, 2, and 3, respectively. Variation in right to left blood flow was used as an overall measure of error caused by technique and did not consistently increase between the first, second, or third BBF estimates in any cohort. Cortical BBF data were found to be significantly more reliable than cancellous data (p < 0.01); error caused by technique was least in the midshaft femoral or midshaft humeral cortical samples. Increasing the dose of spheres administered from 0.5 to 3 million/kg for three serial microsphere injections increased the number of reliable samples and did not lead to increased technical error or shunting. This study demonstrates that there is a significant decrease in BBF over time in the anesthetized dog; therefore, serial estimates of BBF can only be interpreted if the results are normalized or if a control group of animals is included.