Salivary urea is studied as a non-invasive alternative for screening and monitoring of renal diseases. Its high variability prevents a wider clinical use. Animal experiments are needed to identify factors affecting this marker. The aim of this study was to describe the inter-individual variability of salivary urea in healthy mice, establish reference intervals, and analyse the effects of sex, age and body weight. Plasma and saliva samples were obtained from 37 male and 41 female healthy adult CD1 mice aged 13–69 weeks (body weight 22–51 g). The reference interval for salivary urea in heathy mice based on our results is 2.7–8.4 mmol/l (CV = 23 %). Multivariate analysis did not show any significant effect of age, sex, or body weight. In addition, salivary urea did not correlate with its plasma concentrations. The high variability of the promising salivary marker of kidney function in healthy mice requires further research before its use to diagnose or monitor renal failure in animal models of kidney diseases. Other potential confounders should be analysed, including intra-individual and pre-analytical variability. In addition, a normalization factor such as total salivary proteins or salivation rate is likely needed.