Imaging cardiac sarcoidosis with FLT-PET with comparison to FDG-PET: a prospective pilot study

Purpose: FDG-PET is frequently used for the diagnosis and evaluation of therapy response in cardiac sarcoidosis (CS). The normal biodistribution of FDG requires special dietary preparation in order to suppress myocardial physiological uptake. Unfortunately, a significant proportion of patients will present physiological myocardial uptake despite optimal preparation, limiting the diagnostic value of FDG in these patients. Some non-FDG tracers, such as FLT, have shown uptake in sarcoidosis lesions but do not demonstrate uptake within normal myocardium. This prospective study was designed to compare FLT-PET and FDG-PET for the evaluation of cardiac sarcoidosis (CS).

Methods: 14 subjects with extra-cardiac sarcoidosis (ECS) (11 with CS) were prospectively recruited and imaged with FDG-PET, FLT-PET and rest perfusion PET. Subjects undergoing FDG-PET imaging were prepared with a low-carbohydrate, high fat and protein-permitted diet followed by fasting, as well as IV heparin, while no special preparation was performed for FLT-PET. Two blinded, experienced readers independently reviewed all studies. Heart Rhythm Society criteria served as gold standard for the diagnosis of CS.

Results: The combination of perfusion/FLT-PET and perfusion/FDG-PET performed comparably in this cohort, both having an accuracy for CS of 96% (95% CI: 0.7 - 1.0), with a sensitivity of 91% (95% CI: 0.59 - 1.0) and specificity of 100% (95% CI: 1.0 - 1.0). 7 subjects showed myocardial FDG uptake consistent with active CS, with 6 of these subjects also demonstrating increased FLT uptake. One subject with isolated lateral wall uptake on FDG, a normal variant, showed no corresponding uptake on FLT. Inter-observer agreement between FLT and FDG interpretation was very good for both CS (κ = 0.85 (95% CI: 0.57 - 1.0) vs κ = 0.72 (95% CI: 0.38 - 1.0) and ECS (κ = 0.81 (95% CI: 0.46 - 1.0) vs κ = 1.0 (95% CI: 1.0 - 1.0) The sum rest score was strongly correlated with SUV_total for FLT (r = 0.90, 95% CI: 0.33-0.99, p = 0.014) but not for FDG (p = 0.75).

Conclusions: FLT PET has excellent accuracy for CS and requires no special patient preparation. The distribution of FDG differed from that of FDG, suggesting that the tracers provide different pathophysiological information and that myocardial FLT uptake may serve as a distinct biomarker for CS.

• Download figure
• Open in new tab
• Download powerpoint