18F-FDG PET/CT discriminates whether rheumatoid arthritis patients are in ultrasound remission or not.

Objectives: 18F-FDG PET/CT is used for cancer evaluation but can also quantify joint inflammation in RA. We evaluated its ability to identify patients in remission as defined by US. Methods: Sixty RA patients underwent clinical, PET/CT and US evaluations (hands and wrists). For PET/CT, the maximum standardized uptake value (SUVmax) was determined. At the patient level, the highest SUVmax, number of PET-positive joints and cumulative SUVmax were recorded. US evaluation included gray-scale and power Doppler. Results: At the joint level, SUVmax was significantly lower when the joint was in US remission (P < 0.0001). Receiver operating characteristics curves identified a threshold of 0.86 to define joint remission with a high positive predictive value of 95.2%. At the patient level, the highest SUVmax and cumulative SUV were significantly lower in patients in US remission (respectively P = 0.0023 and P = 0.045). The highest SUVmax threshold of 2.36 demonstrated a high negative predictive value of 92.9% for determining US remission. For the cumulative SUVmax threshold, the negative and positive predictive values for determining US remission were both 73%. In contrast, these PET/CT parameters did not discriminate if a patient is in clinical remission or not. Conclusions: PET/CT can identify RA patients in remission as defined by US. We show that the highest SUVmax and cumulative SUV are the two discriminative PET/CT parameters able to determine whether a patient is in US remission. This provides an additional advantage of PET/CT as an objective tool to evaluate RA.