It has been suggested that urinary biomarkers of tubular injury might help predict progression to end-stage renal disease. In this issue, Hsu et al. report that in those with established chronic kidney disease, this information does not add to what we know by quantifying creatinine and albuminuria. Here we discuss the evidence for urinary tubular injury markers in predicting renal outcomes in chronic kidney disease and the areas where measurement of these molecules might be useful in the future.