Is (18)F-fluorodeoxyglucose positron emission tomography scanning a reliable way to assess disease activity in Takayasu arteritis?

Abstract

Objective: (18)F-fluorodeoxyglucose-positron emission tomography (FDG-PET) scanning has been proposed as a new way of assessing disease activity in Takayasu arteritis (TA), but previous studies have used the nonvalidated National Institutes of Health (NIH) global activity criteria, and thus might be biased. This study was undertaken to determine the value of PET scanning for assessment of disease activity in TA, by comparing PET scan data with clinical, biologic, and magnetic resonance imaging (MRI) data assessed separately.

Methods: Twenty-eight patients with TA (according to the American College of Rheumatology criteria) underwent a total of 40 PET scans. Images were reviewed by 2 pairs of independent nuclear medicine physicians and assessed for pattern and intensity of vascular uptake. TA activity data were obtained within 15 days of the PET scans.

Results: PET scanning revealed abnormal vascular uptake in 47% of the 40 examinations. The uptake intensity grade was 0 in 7 scans, grade 1 in 7 scans, grade 2 in 13 scans, and grade 3 in 13 scans. Morphologic analysis was conducted by grading the pattern of the vascular uptake as diffuse (73%), segmental (20%), or focal (13%). There was a trend toward an association between clinically active disease and the semiquantitative assessment of FDG uptake (P = 0.08). We found no statistical association between levels of acute-phase reactants and intensity of uptake. There was no significant association between the semiquantitative assessment of FDG uptake and the presence of vascular wall thickening (P = 0.23), gadolinium uptake (P = 0.73), or the presence of vascular wall edema (P = 0.56).

Conclusion: Our findings indicate that there is no association between FDG vascular uptake intensity and clinical, biologic, or MRI assessment of disease activity. Previous studies using the nonvalidated NIH global activity criteria are likely biased.