Buziau · Diabetes care 2022 · cross-sectional cohort study · n=3,981

Fructose Intake From Fruit Juice and Sugar-Sweetened Beverages Is Associated With Higher Intrahepatic Lipid Content: The Maastricht Study.

Level 3 - non-randomized controlled study

Cross-sectional observational analysis within a population-based cohort

PubMed 35158374 · doi:10.2337/dc21-2123 · record verified 2026-08-26

What was done

Researchers evaluated cross-sectional data from 3,981 participants (mean age 60 ± 9 years, 50% women) in The Maastricht Study. Total fructose intake and fructose sources (whole fruit, fruit juice, and sugar-sweetened beverages [SSB]) were estimated using a food-frequency questionnaire. Intrahepatic lipid (IHL) content was measured by 3T Dixon MRI. Multivariable models adjusted for age, sex, type 2 diabetes status, education, smoking, physical activity, and dietary intakes of total energy, alcohol, saturated fat, protein, vitamin E, and fiber.

What was found

Total fructose intake and fructose from whole fruit were not associated with IHL in fully adjusted models (P = 0.647 and P = 0.767, respectively). In contrast, higher energy-adjusted intake of fructose from fruit juice (P = 0.019) and SSB (P = 0.009) significantly associated with higher IHL. Comparing the highest to the lowest intake tertiles, IHL was 1.04-fold higher (95% CI 0.99 to 1.11) for fruit juice fructose and 1.09-fold higher (95% CI 1.03 to 1.16) for SSB fructose. The association between fruit juice fructose and IHL was stronger among individuals with type 2 diabetes (P for interaction = 0.071).

Why it matters

This study differentiates liquid fructose sources from whole fruit, suggesting that beverage-derived fructose, rather than total fructose or solid fruit intake, is linked to higher liver fat accumulation.

Limits

The cross-sectional design cannot establish causality or determine whether beverage intake preceded hepatic fat accumulation. Dietary exposure relied on self-reported food-frequency questionnaires prone to recall and social desirability biases. The highest tertile effect estimate for fruit juice had a 95% confidence interval spanning unity (0.99 to 1.11), and residual confounding cannot be excluded.

Cited by