The feasibility of performing longitudinal measurements in mice using small animal PET imaging and a microfluidic blood sampling device

To minimize blood loss in a mouse during quantitative PET imaging, we developed an automated blood sampling device using microfluidic technology. The integrated technology was proven reliable in measuring the cerebral glucose metabolism. In this study, we validated the performance of the microfluidic device using in vivo mouse studies. We examined the stability of the mouse in sequential PET scans to access the feasibility of acquiring multiple measurements from the same mouse over time. Methods: Cardiac output and body temperature were chosen as physiological indicators of mouse stability. Three control and three tumor-bearing mice were studied. Sixty minutes of PET data were acquired for each mouse following a bolus injection of 3’-deoxy-3’-[F-18]fluorothymidine (FLT; ∼250 μCi). 18 serial blood samples,0.25 μℓ each, were taken from the femoral artery using the microfluidic device. Body temperatures were monitored before and after the scan. Immediately following the FLT scan, a 40-minute quantitative FDG (∼300 μCi) PET scan was performed on the tumor-bearing mouse. Using a previously described first-pass angiographic method, we calculated two cardiac outputs for each mouse using the first 15-second data of the FLT and FDG scans respectively.