P08 - Miscellaneous

• Book ID: 102900851
• Publisher: John Wiley and Sons
• Year: 2005
• Tongue: French
• Weight: 755 KB
• Volume: 48
• Category: Article
• ISSN: 0022-2135
• DOI: 10.1002/jlcr.988

✦ Synopsis

Animal models are critical for the study of human disease. The manipulation of the mouse and rat genomes has lead to the development of models of human disease as diverse as cancer, cardiac and neurological conditions. With these models it is possible to study their mechanisms of progression and also to monitor response to therapies. Small animal PET cameras have allowed unprecedented imaging of small structures such as the brain and heart of a mouse. In order to obtain certain quantitative information about disease, kinetic models of radiotracers for PET have been developed. In order to use the kinetic models it is critical to measure the input function, or the distribution of a radiotracer in the blood as a function of time, of the radiotracer used in the PET study. There are various methods that can be used to measure the input function such as blood sampling or implanted probes to measure the blood activity concentration directly. These methods are invasive and technically challenging. A much better method that can be used is to measure the blood activity concentration from the images obtained with the small animal scanner. The present work will show that it is possible to obtain the input functions in rats and mice using microPET (CTI Concorde Microsystems LLC) imaging.

BALB/C mice and Sprague Dawley rats were used for this study. On the day of the study animals were anesthetized and maintained with 1.5-2.0 % isoflurane/oxygen (v/v) throughout the duration of the experiment. Once under anesthesia, animals had jugular venous and arterial catheters placed for FDG injection and blood sampling, respectively. The heart rate was monitored with a modified AccuSync 71 (AccuSync Medical Research Corporation) and this equipment was also used as the input trigger for gating on the microPET scanner. Scanning was initiated and at 5 seconds FDG was infused over a 30 second interval. Blood sampling also commenced upon the start of injection of FDG. Image data was collected for 60 minutes in listmode data format. The listmode data was sorted into both dynamic and dynamic-gated sinograms. Gated data was based upon division of cardiac cycle into 8 gate bins. Sinograms were reconstructed with both FBP and MAP with scatter and attenuation corrections applied. A volume-of-interest (VOI) was drawn on the left ventricle of the heart using late frame data when FDG was concentrated in the myocardial wall. The image derived input function was obtained based on this VOI and compared with blood sampling data.

A clear progression of improvement could be seen in the various reconstruction techniques. Gating the cardiac cycle improved the input function over the non-gated method by eliminating the spill over from the myocardium into the blood pool area at late time points. Utilizing the MAP reconstruction method, while only slightly improving the rat input function, was essential in obtaining a good input function for the mouse. Good correlation between the blood sampling and the image derived input function was also observed.