Urine Stable Isotope Ratios Are Associated with Proportional Intakes of Animal Protein and Added Sugars in a 15-Day Controlled Feeding Study.

Background: Carbon and nitrogen isotope ratios (CIR and NIR) of blood and hair have shown promise as candidate dietary biomarkers, particularly for sources of dietary protein and sugars. However, whether urine CIR and NIR can provide similar dietary information has not been assessed in United States populations.

Objectives: This was an exploratory analysis that aimed to evaluate the urine CIR and NIR as biomarkers of dietary animal protein ratio (APR; animal protein/total protein) and added sugars ratio (ASR; added sugars/total sugars). We also explored associations with foods contributing to those patterns.

Methods: One-hundred adults [55% females, median age 39 (interquartile range 30-53)] from the Phoenix metropolitan area participated in a 15-d controlled feeding study, where all foods and drinks consumed by the participants were provided by the metabolic kitchen at Arizona State University, and diets were individually designed to simulate participants' usual intakes. Participants collected 6 nonconsecutive 24-h urines and CIR and NIR were measured by elemental analysis-isotope ratio mass spectrometry. The effects of diet and participant characteristics on urine CIR and NIR were evaluated using linear mixed models.

Results: Urine CIR was correlated with APR (r = 0.79, P < 0.001) and ASR (r = 0.37, P < 0.001). In the final linear mixed model, daily and 15-d mean intakes of APR, ASR, and corn intake, urine NIR and C:N were all significant predictors of urinary CIR (marginal R²= 0.73). Urine NIR was correlated with APR (r = 0.68, P < 0.001) and fish intake (r = 0.43, P < 0.001); however, the final linear mixed model had a low marginal R² of 0.16.

Conclusions: Despite its strong association with APR, urine CIR has limitations as a sole biomarker of animal protein due to its association with other dietary factors, primarily added sugars. Nonetheless, its potential to contribute intake information when combined with other biomarker measures should be further explored.