18F-FDG uptake in noninfected prosthetic vascular grafts: incidence, patterns, and changes over time

(18)F-FDG PET/CT is of value in the diagnosis of prosthetic vascular graft infection, but potential pitfalls related to tracer uptake in noninfected implants have been described. The current study assesses the incidence and patterns of (18)F-FDG uptake over time in noninfected grafts, in relationship to prosthetic material and location.

A 12-y PET/CT database was retrospectively searched for cancer patients with prosthetic vascular grafts. Data retrieved from patient files included graft location, material, and time from surgery. Images were reviewed by 2 nuclear medicine physicians in consensus, with the presence and patterns (focal, diffuse homogeneous, inhomogeneous) of increased (18)F-FDG uptake in grafts recorded. The mean standardized uptake value in grafts (SUV-G) and mediastinum (SUV-M) was measured. The ratio of SUV-G to SUV-M (SUV-G/SUV-M) was calculated for each graft.

One hundred seven prostheses were identified in 102 studies in 43 cancer patients. Sixty-seven prostheses were made of Dacron, 33 of Gore-Tex, and 7 were native veins. No increased (18)F-FDG uptake was found in 9 grafts (native veins, 4; Gore-Tex, 3; Dacron, 2). There was diffuse homogeneous uptake in 68 and inhomogeneous uptake in 30 grafts. The homogeneous pattern was more prevalent in Gore-Tex whereas the inhomogeneous uptake was seen more in Dacron vascular grafts. None of the grafts demonstrated focal uptake. The SUV-G range was 0.4-6.3 (average, 1.9), and SUV-M range was 0.6-2.4 (average, 1.4). The intensity of uptake was significantly higher in Dacron (SUV-G = 2.35 and SUV-G/SUV-M = 1.72) than in Gore-Tex (SUV-G = 1.09, SUV-G/SUV-M = 0.91) and native vein grafts (SUV-G = 1.07, SUV-G/SUV-M = 0.75) (P < 0.005). Native vein grafts showed a significant decrease in (18)F-FDG uptake over time whereas synthetic grafts showed no change in intensity for a follow-up of up to 16 y.

Diffuse (18)F-FDG uptake was found in 92% of noninfected vascular prostheses, more in Dacron grafts than with other materials. The intensity of (18)F-FDG uptake of synthetic grafts did not change over time. With knowledge of the presence, patterns, and persistence of (18)F-FDG uptake in noninfected vascular prostheses, misinterpretation of PET/CT studies in patients referred for suspected prosthetic infection and in those assessed for diseases unrelated to their graft status can be avoided.