A new administration dose index for optimization of acquisition time at PET imaging using count data using an ionization type survey meter

Objectives In general, acquisition time at PET imaging is determined by administration dose of 18F-FDG per body weight (D/W) and noise equivalent count density (NECD). However, D/W may not accurately reflect patient constitution because there is generally no linear relationship between administration dose and count data. We will propose a new administration dose index, “count data measured by an ionization chamber type survey meter (CDi)”, which is [count data measured by an ionization chamber type survey meter] divided by [abdominal circumference].We investigated the clinical validity of CDi.

Methods Firstly, we retrospectively analyzed data of consecutive 30 patients undergoing PET examinations (14 men, 16 women; median age 61, age range 30-86). Median administration dose of 18F-FDG per body weight is 4.8 MBq/kg (range, 2.2-7.9 MBq/kg). First, eight observers (5 radiologists and 3 radiological technologists) visually evaluated homogeneity of accumulation in the liver by using Sheffé's paired comparison in 435 (30C2) images and we calculated image score for each image. Then, we investigated correlation between image score and D/W or CDi using Spearman's rank correlation coefficient (Rs). Secondly, we investigated correlation between D/W or CDi and NECD by using Spearman's rank correlation coefficient (Rs) in 100 patients (the previous 30 patients plus another consecutive 70 patients undergoing PET examinations).

Results Rs between visual score and D/W or CDi were 0.82 (p<0.01) and 0.92 (p<0.01), respectively. Rs between D/W or CDi and NECD were 0.66 (p<0.01) and 0.85 (p<0.01). Rs between CDi and NECD were statistically significantly higher than that between D/W and NECD (p<0.01).

Conclusions CDi showed better correlation with visual assessment and NECD than D/W. CDi may be a more useful index for determination of acquisition time at PET imaging than D/W.