Reliability of one-point blood sampling method for calculating input function in Na18F PET

• Published in: Nuclear medicine communications Vol. 26; no. 6; p. 519
• Main Authors: Hirata, Tetsurou, Wakita, Kazuo, Fujioka, Mikihiro, Nakamura, Fuminori, Imahori, Yoshio, Ido, Tatsuo, Itani, Kenji, Nishimura, Tsunehiko, Kubo, Toshikazu
• Format: Journal Article
• Language: English
• Published: England 01.06.2005
• Subjects: Adult; Aged; Aged, 80 and over; Algorithms; Bone Diseases - blood; Bone Diseases - diagnostic imaging; Diagnostic Techniques, Radioisotope; Female; Fluorine Radioisotopes - blood; Fluorine Radioisotopes - pharmacokinetics; Humans; Image Enhancement - methods; Image Interpretation, Computer-Assisted - methods; Male; Middle Aged; Positron-Emission Tomography - methods; Radiopharmaceuticals - blood; Radiopharmaceuticals - pharmacokinetics; Reproducibility of Results; Sample Size; Sensitivity and Specificity; Sodium Fluoride - blood; Sodium Fluoride - pharmacokinetics
• ISSN: 0143-3636; 1473-5628
• DOI: 10.1097/00006231-200506000-00007

Conventional methods of quantitative Na18F positron emission tomography require multiple arterial blood sampling in order to obtain the input function, and the procedures are invasive and complicated. This study aims to establish a simplified and reliable technique for obtaining the input function. Multiple arterial blood sampling was performed on 12 persons. The time point for one-point sampling was determined as the time when (1) the plasma radioactivity obtained showed the highest correlation to the real integrated value, which was calculated from the input function, and (2) the coefficient of variation of the real integrated value divided by plasma radioactivity obtained at each time point became the minimum. Scaling factors were obtained in order to estimate the plasma radioactivity at each time point, and a reference table was produced in order to estimate the input function. The optimal timing for one-point sampling was 12 min after intravenous injection of Na18F. The estimated integrated value obtained from arterial blood sampling at 12 min and the reference table was highly correlated with the real integrated value (P<0.001). Levels of plasma radioactivity of arterial blood and venous blood were almost the same at 12 and 40 min after Na18F injection. Percentage errors in the estimation of the integrated value were 2.63% (n=12) for the arterial blood collected at 12 min and 4.14% (n=12) for the venous blood collected at 30 min. This simplified method is clinically applicable and would replace traditional methods that require multiple blood sampling.