Responsiveness of the Serum Carbon Isotope Ratio to Dietary Added Sugar Reduction in a Randomized Controlled Trial among Youth with Steatotic Liver Disease.

BACKGROUND

The carbon isotope ratio (CIR) has been proposed as a biomarker for added sugar intake, but few studies have been conducted in youth, with data from controlled, feeding studies particularly lacking.

OBJECTIVES

This study aims to examine associations of added sugar intake with serum CIR in a randomized, controlled dietary sugar reduction intervention in youth.

METHODS

Data were collected from 40 boys (11-16 y) with histologically diagnosed steatotic liver disease who completed a randomized controlled trial and were provided either a low-free sugar diet (<3% of calories from added sugar or juice) or usual diet for 8 wk. Serum CIR was assessed by isotope ratio mass spectrometry using samples from baseline and week 8. Added sugar intake as a percentage of energy (% kcal/d) was assessed by 24-h recalls. Associations of added sugar intake with serum CIR, both cross-sectionally at baseline and as 8-wk change values, were analyzed by multivariable-adjusted linear regression models.

RESULTS

In the full sample at baseline, higher added sugar intake was associated with higher serum CIR {standardized β [95% confidence interval (CI)]: 0.27 (0.05, 0.48)}. From baseline to week 8, added sugar intake decreased in both intervention arms but this change was significantly greater in the low-sugar diet arm compared with usual diet arm [mean change (95% CI): -9.5% (-11.5%, -7.6%) compared with -2.8% (-4.7%, -0.8%)]. In parallel, serum CIR also decreased in the low-sugar diet arm [-0.16 (-0.29, -0.04)], but not the usual diet arm [-0.12 (-0.24, 0.01)], though this was not a significant between-arm difference. There were no significant associations between 8-wk changes in added sugar intake and changes in serum CIR in either arm.

CONCLUSIONS

Our findings support serum CIR as a biomarker for the percentage of added sugar intake in youth with steatotic liver disease cross-sectionally, but not for within-participant change values during the 8-wk intervention. Larger controlled feeding studies, with more generalizable samples of youth, are needed to confirm these findings.