Prognostic role of serial quantitative evaluation of 18F-fluorodeoxyglucose uptake by PET/CT in patients with cardiac sarcoidosis presenting with ventricular tachycardia

Daniele Muser, Pasquale Santangeli, Simon A. Castro, Jackson J. Liang, Andres Enriquez, Thomas J. Werner, Gaetano Nucifora, Silvia Magnani, Tatsuya Hayashi, Erica S. Zado, Fermin C. Garcia, David J. Callans, Sanjay Dixit, Benoit Desjardins, Francis E. Marchlinski, Abass Alavi

Positron emission tomography (PET) with18F-fluorodeoxyglucose (FDG) has shown to be useful in diagnosis, staging and monitoring of cardiac sarcoidosis (CS) but its interpretation is not standardized. We sought to investigate the clinical impact of serial quantitative FDG uptake analysis in patients with CS presenting with ventricular tachycardia (VT) treated by catheter ablation (CA). We followed 20 patients (51 ± 9 years, 70% males) with CS and VT who underwent CA, with 92 serial FDG-PET scans (3-10 per patient). Myocardial FDG-avid lesions were quantified using three parameters: maximum standardized uptake value (SUVmax), partial-volume corrected mean standardized uptake value (SUVmean) and partial-volume corrected volume-intensity product [lesion metabolic activity (LMA)]. The volume-intensity product of the entire heart [global cardiac metabolic activity (gCMA)] and the background cardiac metabolic activity (bCMA: difference between gCMA and LMA) were also calculated. The primary end-point was the occurrence of major adverse cardiac events (MACE), including death, heart transplant, hospitalization for heart failure and implantable cardioverter defibrillator (ICD) appropriate interventions. Evolution of echocardiographic parameters over follow-up was also assessed. During a median follow-up of 35 (20-66) months, 18 MACE (1 death, 2 heart transplants, 12 ICD appropriate interventions, 3 hospitalizations) occurred in 12 (60%) patients. At univariable analysis, lack of PET improvement (defined by decrease in LMA of at least 25%) was the only variable associated with cardiac events during follow-up. In particular, non-responders had a 20-fold higher risk of MACE at follow-up (HR 18.96, 95% CI 2.26-159.27; p = 0.007). Moreover, a significant linear inverse relationship was observed between changes in LMA and changes in left ventricular ejection fraction over follow-up (β = -20.11; p = 0.003). In patients with CS and VT, temporal change in FDG uptake evaluated by a quantitative approach is associated with parallel change in systolic function. Moreover, reduction in FDG uptake is strongly associated with fewer MACE at long-term follow-up.