An increased risk of adverse biological effects of metals in sediments may be accompanied by high labile metal fluxes as measured by the diffusive gradients in thin films (DGT) technique. To improve the usefulness of the DGT technique for sediment quality risk assessments, we used the simpler and more cost-effective piston DGTs rather than planar DGT probes to measure bioavailable metal fluxes in naturally contaminated sediments with widely varying composition (properties, metals and concentrations) and assessed their prediction of toxicity to amphipod reproduction in a flow-through microcosm. DGT pistons were deployed in sediments under different conditions, both in the field (in situ) and in the laboratory in sediment cores (lab-equilibrated) and in homogenized sediments (lab-homogenized). We demonstrated that the metal flux toxic units, DGTTU, measured in situ best predicted the magnitude of toxicity to amphipod reproduction. For sediments that had been highly disturbed before testing, DGTTU were less predictive for observed toxicity, but the copper flux alone (DGTTU–Cu) was effective, indicating copper was the primary cause of toxicity in these highly perturbed sediments. Overall, our study highlighted that the adverse effects induced by excessive bioavailable metals in contaminated sediments can be consistently sensed by the DGT pistons.