Systematic Review and Meta-analysis: The Role of Diet in the Development of Nonalcoholic Fatty Liver Disease.
Level 3 - non-randomized controlled study
Systematic review and meta-analysis of non-randomized observational studies
PubMed 34838723 · doi:10.1016/j.cgh.2021.11.026
What was done
A systematic review and meta-analysis searched literature from January 1985 to May 2021 to evaluate nutritional factors in the presence and severity of nonalcoholic fatty liver disease (NAFLD). Studies were required to use validated dietary and anthropometric assessment tools administered by trained professionals. Comparisons were conducted between patients with NAFLD and healthy controls, as well as between NAFLD and nonalcoholic steatohepatitis (NASH). Risk of bias was evaluated using the ROBINS-I tool.
What was found
Across 60 eligible studies involving 100,621 patients, total caloric intake was significantly higher in patients with NAFLD compared with controls (mean difference, 78.08; 95% confidence interval, 41.03-115.13). Proportions of total calories from macronutrients (protein, fat, and carbohydrate) and daily intake of fiber, caffeine, and vitamins E, A, and C did not significantly differ between patients with NAFLD and controls. Soft drink consumption showed a trend towards association with NAFLD with an odds ratio of 4.4 and very wide confidence intervals. There was no significant difference in any dietary comparison between patients with NAFLD and NASH. All meta-analyses had significant heterogeneity.
Why it matters
This review indicates that overall excess caloric intake is the primary dietary factor associated with NAFLD rather than specific macronutrient distributions or vitamin intakes.
Limits
All included meta-analyses demonstrated significant heterogeneity. The majority of included studies (41 of 60, or 68%) had a moderate risk of bias based on the ROBINS-I tool. The number of patients in the comparison between NAFLD and NASH was relatively small, and confidence intervals for soft drink intake were very wide. Because the included studies were non-randomized observational designs, residual confounding cannot be ruled out.
Cited by
- context Fructose from high-fructose corn syrup is currently causing chronic liver disease in 100 million Americans.