Lithium (Li) treatment for bipolar affective disorders is associated with a variety of renal side effects. The metabolic response of the kidney to chronic Li treatment has rarely been studied. We applied a novel method of ¹H-nuclear magnetic resonance (NMR)-based metabonomics to integrate metabolic profiling and to identify the changes in the levels of metabolites in the kidney and urine from rats with Li-induced NDI. Metabolic profiles of urine and kidney homogenate [3 different zones (cortex, outer medulla, and inner medulla) or whole kidney] were investigated using high-resolution NMR spectroscopy coupled with pattern recognition methods. The accurate concentrations of metabolites in kidney homogenates and urine were rapidly measured using the target-profiling procedure, and the difference in the levels of metabolites was compared using multivariate analysis, such as principal component analysis and orthogonal partial least squares-discriminant analysis. Major endogenous metabolites for kidney homogenates contained products of glycolysis (glucose, lactate) and amino acids, as well as organic osmolytes (e.g., betaine, myo-inositol, taurine, and glycerophosphocholine). Many metabolites revealed changes in their levels, including decreased levels of organic osmolytes and amino acids in the inner medulla. A number of urinary metabolites were changed in Li-induced NDI, and in particular, elevated urinary levels of acetate, lactate, allantoin, trimethylamine, and creatine could suggest Li-induced renal cell stress or injury. Taken together, metabonomics of kidney tissue and urine based on ¹H-NMR spectroscopy could provide insight into the effects of Li-induced renal effects and cell injury.