Quantitative assessment of 18F-FDG PET in patients with Hodgkin lymphoma: is it significantly affected by contrast-enhanced computed tomography attenuation correction?

The reliability of visual and quantitative response assessment may be impaired owing to inconsistent scanning protocols and image reconstruction methods of 2-deoxy-2-[F]fluoro-D-glucose (F-FDG) PET. Hence, this study investigates the effect of contrast-enhanced computed tomography (CT) attenuation correction in patients with Hodgkin lymphoma. In 10 consecutive patients undergoing either staging or response assessment, F-FDG PET images were attenuation-corrected once on the basis of unenhanced CT and additionally using contrast-enhanced CT. Reconstruction was performed in both cases with ordered subset expectation maximization (OSEM) and ultra-high definition (UHD) algorithm. While maximum and peak standardized uptake value (SUV) were obtained from tumour tissue (lesionSUVmax and lesionSUVpeak), maximum and mean SUVs were determined within the background regions liver (liverSUVmax and liverSUVmean) and mediastinal blood pool (mbpSUVmax and mbpSUVmean). After switching to contrast-enhanced CT attenuation correction, lesionSUVmax and lesionSUVpeak increased on average by 2.55±3.24 (P=0.018) and 3.64±3.22% (P=0.008), respectively, with OSEM and by 4.59±5.49 (P=0.005) and 3.84±5.65% (P=0.005), respectively, with UHD reconstruction. LiverSUVmax and liverSUVmean showed a mean rise of 7.15±4.27 (P=0.005) and 6.97±2.18% (P=0.005), respectively, in the OSEM data sets and of 7.24±6.59 (P=0.017) and 6.29±2.83% (P=0.005), respectively, in the UHD images. The average increases of mbpSUVmax and mbpSUVmean were 10.82±4.89 (P=0.005) and 12.40±3.73% (P=0.005), respectively, after OSEM, compared with 13.11±14.93 (P=0.005) and 11.50±12.19% (P=0.005), respectively, after UHD reconstruction. As the use of CT contrast fluids results in a stronger SUV increase within the liver and mediastinal blood pool than within lymphoma tissue, this may have clinical consequences regarding visual and quantitative response assessment. Ideally, CT scans for PET attenuation correction should therefore be performed in the absence of a contrast agent.