Background: Idiopathic Pulmonary Fibrosis (IPF) is a progressive and irreversible disease leading to terminal respiratory insufficiency. Fluorodeoxyglucose ([18F]-FDG) PET/CT has been proposed to track the activity of the disease. However, IPF is characterized by regional changes in lung density that affects the FDG uptake, a factor generally not taken into account in previous studies. In this work, we studied the relationship between severity of IPF and lung uptake of [18F]-FDG, evaluated by mean and maximum standardized uptake value (SUV), corrected (SUVmean-corr; SUVmax-corr) and uncorrected for lung density (SUVmean-uncorr; SUVmax-uncorr). Methods: [18F]-FDG PET/CT was performed in 31 IPF patients between 2013 and 2017. Lung density was determined on CT. SUV values were correlated with lung function tests, carbon monoxide diffusion (DLCO) and 6-min walking test (6MWT) at baseline and at 1 year. Correlation with the GAP index, a well-validated prognostic score in IPF, was also determined. Results: At baseline, SUVmean-uncorr was highly correlated with lung density (r = 0. 755; p < 0.001). SUVmean-uncorr and lung density were correlated with lung function tests (vital capacity (VC): p = 0.013 and p = 0.003; forced vital capacity (FVC): p = 0.004 and p = 0.001; total lung capacity (TLC): p = 0.001 and p = 0.001, respectively), while SUVmean-corr was not (VC: p = 0.733; FVC: p = 0.667; TLC: p = 0. 382). Interestingly, SUVmean-corr was significantly higher in patients with a GAP index of 3 (p = 0.005), and negatively correlated with DLCO (r = − 0.398; p = 0.026) and desaturation during the 6MWT (r = − 0.401; p = 0.024). But no correlation was found with changes in lung function tests, walk distance and DLCO at 1 year. Conclusion: To evaluate the role of ([18F]-FDG) PET/CT in IPF, correction for lung density appears necessary. As suggested by the correlation with DLCO, densitycorrected SUV seems related to the intrinsic disease activity and particularly to the integrity of the alveolar-capillary barrier. However, ([18F]-FDG) PET/CT has probably a limited prognostic value as no correlation was found between SUVmean-corr and the clinical evolution at 1 year. Further studies with a longer follow-up are warranted.