Defining PET standardized uptake value threshold for tumor delineation with metastatic lymph nodes in head and neck cancer.

OBJECTIVE

Hot spots of F-18 fluorodeoxyglucose positron emission tomograms are variable in size according to window settings of standardized uptake values. The purpose of this study was to determine the standardized uptake value threshold that represents the target volume.

METHODS

Sixty-three patients who underwent fluorodeoxyglucose positron emission tomographic computed tomography and were diagnosed as having head and neck cancer with cervical lymphadenopathy were studied. The horizontal and vertical diameters of metastatic lymph nodes (LN-CT) were measured at the center of computed tomographic images. Of the corresponding nodes, the maximal standardized uptake value (SUVmax) and standardized uptake value profiles along the central horizontal and vertical axes were calculated on positron emission tomographic images (LN-PET). On the standardized uptake value profiles, the standardized uptake value levels (SUVeq) where the size of LN-PET was equivalent to the diameters of LN-CT were obtained. The regression formula between SUVeq and SUVmax was obtained. The regression formula of SUVeq was validated in subsequent 30 positron emission tomographic computed tomography studies.

RESULTS

The mean horizontal and vertical diameters of LN-CT were 14.9 and 16.4 mm, respectively. SUVmax ranged from 1.88 to 9.07, and SUVeq was between 1.16 and 6.42. The regression formula between SUVeq and SUVmax was as follows: SUVeq = 1.21 + 0.34 × SUVmax (coefficient of correlation: R = 0.69). The validation study resulted in a good correlation between the volume of lymph nodes on computed tomography and positron emission tomographic computed tomography (R(2) = 0.93).

CONCLUSIONS

The formula with a relatively high coefficient of correlation is considered to indicate that SUVeq is not constant, but is a complex of an absolute standardized uptake value and is proportional to SUVmax.