BACKGROUND Phosphorus-31 (P) magnetic resonance spectroscopy (MRS) has been evaluated in several studies for pretransplant assessment of renal viability. As an indicator of graft quality, the ratio of phosphomonoesters (PME) to an organ's inorganic phosphate (Pio) is the parameter of choice. However, exact calculation of the PME/Pio ratio is disturbed by interference from the signal of phosphorus contained in the preservation solution (Pip). In this article, the authors present an improvement attained by using volume selective P-MRS using chemical shift imaging (CSI) enabling reduction in the disturbing influence of Pip. METHODS Sixteen renal grafts were investigated using a 2.0-T whole-body scanner and a cross-coil setting with an active decoupled receiver coil. P-CSI was used to measure two-dimensional spectra of a 4-cm slice of the graft in a 12 x 12 matrix. Peaks of each spectrum were fitted with magnetic resonance user interface-MatLab software using VarPro and the mean PME/Pio ratio was calculated. RESULTS Significant correlation between the PME/ Pio ratio and clinical kidney function was found. In comparison to former trials using PME/Pio ratios calculated from non-volume-selective spectra, the correlation outcome improved significantly. Furthermore, overlay of CSI-spectra matrix and corresponding slice image of the kidney illustrates the origin of different signals in the spectra. CONCLUSIONS The authors' work demonstrates that the PME/Pio ratio calculated from CSI spectra is a reliable indicator of viability of renal grafts. Early knowledge of graft quality may allow therapy to be adapted to the conditions of the organ, for example, by initial withholding of nephrotoxic calcineurin-inhibitors in kidneys with high risk of delayed graft function.