Purpose: This study aimed (1) to assess the influence of age, sex, blood glucose, and body mass index on the 18F fluoro-deoxy-glucose (18F-FDG) uptake in normal spinal cord; (2) to quantitatively evaluate contamination of the spinal cord SUVmax by the adjacent vertebral marrow activity; and (3) to investigate the validity of normalizing spinal cord SUVmax against lumbar thecal sac SUVmax.
Methods: Two hundred positron emission tomography–computed tomography examinations of subjects with normal spinal cord were retrospectively reviewed. SUVmax of spinal cord and vertebral body was obtained at C2, C5, T6, T12, and L3 levels. Pearson correlation coefficients (r) were obtained at each level between spinal cord SUVmax and vertebral marrow SUVmax, age, body mass index, and blood glucose. Cord to background ratio (CTB) was calculated as the ratio between SUVmax of spinal cord and SUVmax of L3 thecal sac. The coefficient of variation (CV) of spinal cord SUVmax was compared with the CV of CTB.
Results: Spinal cord SUVmax was highest at C2 (mean, 1.76) and lowest at T6 (mean, 1.37) with SD of 0.32 to 0.36 SUV. Sex (P > 0.45), age (r: −0.25 to −0.06), body mass index (r: 0.19 to 0.27), and blood glucose (r: −0.17 to 0.22) had no impact on the spinal cord SUVmax. A moderate to strong positive correlation (r: 0.66–0.80) was found between spinal cord SUVmax and the corresponding vertebral marrow SUVmax. The CV of CTB was greater (0.28–0.32) than the CV of spinal cord SUVmax (0.19–0.25) across all levels.
Conclusions: Of the variables studied, only contamination from adjacent vertebral marrow activity significantly affected the SUVmax of spinal cord. This contamination should be corrected for when reporting spinal cord FDG uptake. Lumbar thecal sac is not a valid reference for normalizing spinal cord FDG uptake.
From the Department of Radiology, Mayo Clinic, Jacksonville, FL.
Received for publication March 4, 2016; accepted May 12, 2016.
Correspondence to: Vivek Gupta, MD, Department of Radiology, Mayo Clinic, 4500 San Pablo Rd, Jacksonville, FL 32224 (e-mail: firstname.lastname@example.org).
The authors declare no conflict of interest.