Semi-quantitative analysis of F fluorodeoxyglucose uptake in the assessment of disease activity and therapeutic response in rheumatoid arthritis: An institutional experience.

F fluorodeoxyglucose (F-FDG) positron emission tomography (PET) can be used to image synovial inflammation in patients with rheumatoid arthritis (RA). Recently, clinical application of novel therapies for RA, such as tumor necrosis factor-α (TNF-α) inhibitor and anti-interleukin-6 receptor antibody, has been introduced. The radiological assessment of disease activity changes of patients who underwent these therapies will help the clinicians to obtain more information about the patients and to decide drug withdrawal or change of medication. It is considered that F-FDG PET scan is generally very expensive; however, the information from F-FDG PET about patients during biological treatments helps discontinuation of these treatments with incomplete response despite its high costs and with possible side effects such as malignant lymphoma. In this study, we evaluated if the 18F-FDG uptake of the affected joints represented by standardized uptake value (SUV) correlated with the clinical assessment of patients with RA. In addition, we would like to evaluate if there was a correlation between the difference of SUV and improvement of clinical findings in RA patients undergoing anti TNF therapies. RA patients who underwent anti-TNFα therapies in a tertiary care hospital were assessed using whole-body F-FDG PET/computed tomography (CT). PET assessments were performed on hip joints, knees, shoulders, wrists, ankles, MCP, and PIP for a total of 28 joints in each patient. The F-FDG uptake was then quantified using the maximum SUV (SUVmax) prior to, and 6 months after the initiation of treatment with anti-TNF-α drugs. Disease activity score (DAS28 and DAS28-C-reactive protein [CRP]) were recorded and white blood cell, matrix metalloproteinases (MMP-3) and rheumatoid factor (RF) were examined in all patients. The average of SUV among measured joints, or the sum of these joints (total SUV), correlated with DAS28 ( = 0.671, < 0.001), DAS28-CRP ( = 0.623, < 0.001), ESR ( = 0.542, < 0.001), CRP ( = 0.411, = 0.002), MMP-3 ( = 0.399, = 0.006), and RF ( = 0.447, = 0.002). There were correlations between ΔSUV and ΔDAS28 ( = 0.651, < 0.001), ΔSUV and ΔDAS28-CRP ( = 0.682, < 0.001), ΔSUV and ΔESR ( = 0.449, = 0.023), and ΔSUV and ΔMMP-3 ( = 0.457, = 0.027), respectively. The number of PET-positive joints and the cumulative SUV significantly correlated with the DAS, which is a composite disease activity score (DAS) that combines the swollen and tender joint counts, the erythrocyte sedimentation rate (DAS-ESR) or CRP serum levels (DAS - CRP) or RF (DAS - RF) or metalloproteinases-3 (DASMMP-3). At baseline and at 6 months' post-treatment with anti-TNFα drugs, there was a significant correlation between the PET results, either visual, the cumulative SUVs or the composite SUV index, and the comprehensive clinical assessment (DAS), the CRP levels and the number of joints positive for RA, and cumulative synovial thickness. By reflecting inflammatory activity, F-FDG PET may enhance the diagnostic performance and expectation of disease prognosis in RA, especially with early synovial inflammation. The intensity of uptake varied from mild to intense (SUV values from 3.10 to 6.0). Overall, these values correlated well with the clinical evaluation of involved joints. F-FDG PET imaging data provided a distribution of joint involvement with varying degrees of severity and phase of disease activity (moderate, low, and remission) in the same patient. PET/CT imaging with F-FDG shows better image quality, provides more confirmative diagnostic information, and will be promising imaging modality in diagnosis and management of RA.