Effect of corrections for blood glucose and body size on [F-18]FDG PET standardised uptake values in lung cancer

Standardised uptake values (SUVs) are commonly used as a semi-quantitative index of 2-[F-18]fluoro-2-deoxy-D-glucose (FDG) tracer uptake in positron e mission tomography (PET). Studies have shown that SUVs may depend on body s ize and blood glucose concentration and corrections for these effects have been proposed in the literature. This retrospective study investigated the effect of the proposed corrections on SUVs from a group of 154 patients wit h lung cancer who had scans on a dedicated PET scanner. A total of 252 SUVs were requested as an aid to staging during consideration for surgical rese ction. SUVs were calculated normalised to body weight (SUVW), lean body mas s (SUVLBM) and body surface area (SUVBSA). The following correlations were examined: SUV with height, weight and body surface area for the different b ody size normalisations; SUVW and SUVW x blood glucose (SUVBG) with blood g lucose; SUVW with scan time post injection; and SUVW with apparent lesion d iameter. Significant correlations were only observed between: SUVLBM and he ight (P=0.007); SUVW and scan time (P=0.007); SUVW and lesion diameter (P=0 .0005); and SUVBG and blood glucose (P <0.00001). The correlation between S UVLBM and height suggests that lean body mass as a function of height alone should not be used to normalise SUVs; however, the lean body mass calculat ed from a height and weight nomogram did not show this effect. The strong c orrelation between SUVBG and blood glucose concentration suggests that for non-diabetic fasted patients, lung tumour SUVs should not be adjusted for b lood glucose.