Respiratory motion in Positron Emission Tomography leads to reduced image quality, influencing this way the quantitative accuracy of PET measurements, as shown in numerous studies. However, only few results have been published on its impact on lesion detection. This study intends to evaluate the impact of motion correction on the detection of small lesions (between 8 and 12 mm diameter) using a Computed-Aided Detection (CAD) system on FDG whole-body simulated PET images. We evaluate two types of motion correction techniques, both using motion fields derived from the reconstruction of gated PET images. The first technique consists in averaging the coregistered gated reconstructed PET images, while the second method integrates the motion fields during the iterative reconstruction process.