The purpose of this study was to (1) establish a technique for ultrasound-guided trans-splenic portal scintigraphy (TSPS) using 99mTcO4−, (2) evaluate portal vein morphology, (3) compare the radiation exposures for TSPS vs. per-rectal portal scintigraphy (PRPS), and (4) compare the quality of numerical data from the TSPS vs. PRPS. Eight juvenile dogs underwent PRPS and TSPS (minimum of 48 h between studies) after initial screening tests. PRPS was done according to established protocol using 425±36 MBq (mean±SD) of 99mTcO4−. TSPS was done with the dogs in right lateral recumbency over the gamma camera. 99mTcO4− (57±13.9 MBq) was injected into the spleen 1–2 s following initiation of the dynamic acquisition. The frame rate was 4 frames/s for 5 min. There was significantly lower radioactivity of 99mTcO4− given and significantly higher total counts recorded in the liver and heart during the TSPS compared with PRPS. The total counts for the TSPS and PRPS were 7120±4386 and 830±523, respectively. Percent absorption from the spleen was 52.5±19.1% compared with 9.2±5.7% for the colon. Calculated transit time for the TSPS studies was 7±2.3 s. In TSPS studies, the splenic and portal veins were clearly identified. Radiation exposure levels of the dogs were significantly lower following TSPS than after PRPS. TSPS appears superior to PRPS as a method to image the portal venous system representing a valid alternative diagnostic test for animals with suspected portosystemic shunts.