Ratio of hepatic arterial-to-portal venous blood flow--validation of radionuclide techniques in an animal model.

The ratio of hepatic arterial-to-portal venous blood flow can be determined from the analysis of a first-pass bolus through the liver by a number of techniques. This study examines the validity of four radiotracer techniques in an animal model. Thirty-four flow studies (3 mCi 99mTc-DTPA/study) were performed in seven anesthetized pigs. Images were acquired for 200 sec and time-activity curves were generated from lung, liver and kidney ROIs. These curves were analyzed using a slope-based (HPI), a height-based (mHAR) and two deconvolution-based methods employing exponential or gamma variate fits. There was an excellent correlation (r greater than 0.9) between results obtained with flow probes and the radiotracer techniques, with the exception of the HPI technique (r = 0.75). The mHAR and deconvolution techniques were inaccurate at very low and high arterial flows, due respectively to noise limitations and hemodynamic instability in the animal. Nevertheless, these techniques appear to be the most promising for routine clinical use.