Objective Right ventricular (RV) function is a major determinant of survival in patients with pulmonary arterial hypertension (PAH). Metabolic alterations may precede haemodynamic and clinical deterioration. Increased RV fluorodeoxyglucose (FDG) uptake in positron emission tomography (PET) was recently associated with progressive RV dysfunction in MRI, but the prognostic value of their combination has not been established.

Methods Twenty-six clinically stable patients with PAH (49.9±15.2 years) and 12 healthy subjects (control group, 44.7±13.5 years) had simultaneous PET/MRI scans. FDG uptake was quantified as mean standardised uptake value (SUV) for both left ventricle (LV) and RV. Mean follow-up time of this study was 14.2±7.3 months and the clinical end point was defined as death or clinical deterioration.

Results Median SUV_RV/SUV_LV ratio was 1.02 (IQR 0.42–1.21) in PAH group and 0.16 (0.13–0.25) in controls, p<0.001. In PAH group, SUV_RV/SUV_LV significantly correlated with RV haemodynamic deterioration. In comparison to the stable ones, 12 patients who experienced clinical end point had significantly higher baseline SUV_RV/SUV_LV ratio (1.21 (IQR 0.87–1.95) vs 0.53 (0.24–1.08), p=0.01) and lower RV ejection fraction (RVEF) (37.9±5.2 vs 46.8±5.7, p=0.03). Cox regression revealed that SUV_RV/SUV_LV ratio was significantly associated with the time to clinical end point. Kaplan-Meier analysis showed that combination of RVEF from MRI and SUV_RV/SUV_LV assessment may help to predict prognosis.

Conclusions Increased RV glucose uptake in PET and decreased RVEF identify patients with PAH with worse prognosis. Combining parameters from PET and MRI may help to identify patients at higher risk who potentially benefit from therapy escalation, but this hypothesis requires prospective validation.