The objective of this laboratory study was to assess the cardiorespiratory consequences related to the use of different back-support exoskeletons during a repetitive lifting task. Fourteen women and thirteen men performed a dynamic stoop lifting task involving full flexion/extension of the trunk in the sagittal plane. This task was repeated for 5 min with a 10 kg load to handle. Four conditions were tested: with a passive exoskeleton (P-EXO), with two active exoskeletons (A-EXO1 and A-EXO2), as well as without exoskeleton (FREE). The oxygen consumption rate and cardiac costs were measured continuously. Results showed a significant reduction in the oxygen consumption rate for all exoskeletons as compared to FREE. This reduction was significantly more important for A-EXO1 (~28%) compared to A-EXO2 (~13%) and P-EXO (~7%). Compared to FREE, the cardiac cost was only reduced for A-EXO1 (~24%). Several factors can explain these differences on the cardiorespiratory parameters observed between exoskeletons: the technology used (passive vs active), the torque provided by the assistive device, the weight of the system, but also the level of anthropomorphism (related to the number of joints used by the exoskeleton). Our results also highlighted the lack of interaction between the exoskeleton and sex. Thereby, the three back-support exoskeletons tested appeared to reduce the overall physical workload associated with a repetitive lifting task both for men and women.