Association between folate and non-alcoholic fatty liver disease among US adults: a nationwide cross-sectional analysis.

To the Editor: Non-alcoholic fatty liver disease (NAFLD) becomes a worldwide epidemic with a currently estimated prevalence of 24%. The spectrum of NAFLD includes simple steatosis and non-alcoholic steatohepatitis. Folate, also called vitamin B9, is the generic term given to a family of compounds, including unsubstituted tetrahydrofolate (THF), 5-methyltetrahydrofolate (5-MTHF), 5-formyltetrahydrofolate (5-FTHF), 5,10-methylene-THF (5,10-CH2-THF), and unmetabolized folic acid (UMFA). Folic acid is usually added to products as a dietary supplement. Folate plays a critical role in one-carbon metabolism, which involves the activation and transfer of one carbon unit for biosynthetic processes, including purine and thymidine synthesis and homocysteine remethylation to methionine. NAFLD is closely related to metabolic disorders, whereas the association between folate and NAFLD has not yet been well established. We used data from a nationally representative sample of US adults who were enrolled in the US National Health and Nutrition Examination Survey (NHANES) 2011–2016 to explore the association between folate and NAFLD based on serum and dietary aspects. NHANES was approved by the US National Center for Health Statistics Ethics Review Board, and all participants signed written informed consent forms. For analyses, we constructed combined sample weights for three survey cycles of NHANES 2011–2012, 2013–2014, and 2015–2016 based on analytical guidelines. In total, 29,902 participants were enrolled in these three survey cycles. Given differences in the concentrations of some serum folate forms based on fasting status, the analysis focused on samples obtained from fasting participants. As a result, a total of 6910 non-pregnant participants aged ≥20 years who fasted before attending a morning exam session were selected. Among 6910 participants, measurements of serum folate levels and dietary folate intake were available for 6256 participants. We excluded participants with alcohol consumption >2 drinks/day for men or >1 drink/day for women and those infected with hepatitis B virus or hepatitis C virus (n = 2261). Finally, a total of 3995 participants were included in the analysis [Supplementary Figure 1, https://links.lww.com/CM9/B363]. Using liquid chromatography tandem–mass spectrometry, the levels of total folate and five biologically active folate forms were analyzed in serum samples of participants included in NHANES 2011–2016. All NHANES participants completed up to two 24-hour dietary recall interviews. For participants with data available from both interviews, total estimated dietary folate intake was averaged. If data were only available from one timepoint, the value was used directly. The dietary supplement data were collected from personal interviews during the 30-day period prior to the survey date. For participants reporting supplement use, the total intake of folate was calculated as the sum of dietary and supplement intake. Dietary intake was presented as dietary folate equivalents (DFE) in this study. For the definition of NAFLD, we utilized the US fatty liver index (USFLI), which was first introduced by Ruhl and Everhart.[1] A USFLI ≥30 was selected to exclude fatty liver as recommended, and those with excessive alcohol consumption (>2 drinks/day for men or >1 drink/day for women) and those suffering from hepatitis were excluded. SAS 9.4 (SAS Institute Inc., Cary, NC, USA) was employed for all analyses in this study. Complex survey methods in SAS were exploited as needed. Surveymeans and Surveyfreq procedures were used to describe variables in weighted forms. A quartile-based analysis was used by dividing folate levels into quartiles, and the lowest quartile was set as a reference. We also assessed collinearity using Pearson correlation coefficients and multicollinearity diagnostic statistics (tolerance and variance inflation factors [VIF]). VIF >10 indicates multicollinearity. Then, a complex logistic regression model was used to evaluate the risk factors for NAFLD with increasing degree of adjustment. A two-sided P < 0.05 was considered to be statistically significant. In this study, the prevalence of NAFLD was 33.9% (1355/3995) based on the USFLI score. We present the characteristics of participants stratified by NAFLD status in Supplementary Table 1, https://links.lww.com/CM9/B363. NAFLD patients tended to be older, predominantly male, and were more likely to have metabolic comorbidities, such as metabolic syndrome. We also found that NAFLD patients had significantly lower serum total folate and 5-MTHF than healthy controls, whereas levels of the minor component UMFA were similar between NAFLD patients and healthy controls. In addition, we found that NAFLD patients consumed less folate not only from food but also from dietary supplements compared with healthy controls [Supplementary Table 1, https://links.lww.com/CM9/B363]. These findings suggested that NAFLD was associated with lower serum folate levels and less dietary folate intake compared with healthy controls. As shown in Table 1, after adjusting for demographic and socioeconomic factors, the estimated risk of NAFLD decreased with increasing quartiles of total folate (P for trend <0.001). The odds ratio (OR) (95% confidence interval [CI]) was 0.37 (0.28–0.49) in quartile 4 compared with the lowest quartile. Similarly, serum 5-MTHF levels were negatively associated with the risk of NAFLD, and the OR (95% CI) was 0.37 (0.28–0.48) in quartile 4 compared with the lowest quartile. Consistently, these negative associations remained significant after adjusting for factors in Models 2 and 3. However, we did not find any significant association of UMFA with the risk of NAFLD. Subgroup analysis revealed that participants without diabetes were more inclined to exhibit a negative association between serum total folate and NAFLD [Supplementary Table 2, https://links.lww.com/CM9/B363]. These results suggested that serum total folate, especially its major component 5-MTHF, was negatively associated with the risk of NAFLD. Table 1 - Serum folate levels and risk of NAFLD (n = 3995). Serum total folate 5-MTHF UMFA Models OR (95% CI) P trend OR (95% CI) P trend OR (95% CI) P trend Model 1 <0.001 <0.001 0.217 Q1 1 (reference) 1 (reference) 1 (reference) Q2 0.70 (0.57–0.87) 0.66 (0.52–0.84) 0.94 (0.72–1.24) Q3 0.67 (0.53–0.84) 0.63 (0.50–0.79) 0.92 (0.72–1.18) Q4 0.37 (0.28–0.49) 0.37 (0.28–0.48) 0.84 (0.63–1.12) Model 2 <0.001 <0.001 0.127 Q1 1 (reference) 1 (reference) 1 (reference) Q2 0.64 (0.49–0.85) 0.62 (0.47–0.83) 0.98 (0.73–1.31) Q3 0.77 (0.56–1.07) 0.75 (0.54–1.02) 0.86 (0.65–1.15) Q4 0.48 (0.35–0.67) 0.49 (0.36–0.67) 0.77 (0.55–1.08) Model 3 <0.001 <0.001 0.034 Q1 1 (reference) 1 (reference) 1 (reference) Q2 0.68 (0.49–0.94) 0.63 (0.46–0.85) 1.08 (0.77–1.51) Q3 0.75 (0.51–1.10) 0.71 (0.50–1.01) 0.85 (0.60–1.22) Q4 0.40 (0.28–0.57) 0.39 (0.28–0.55) 0.70 (0.48–1.03) Model 1 was adjusted for demographic and socioeconomic factors (age, gender, race/ethnicity, family income to poverty ratio, education levels, and marital status).Model 2 was adjusted for Model 1 plus BMI.Model 3 was further adjusted for Model 2 plus biochemistry factors (alanine aminotransferase, fasting triglyceride, total cholesterol, and uric acid).Quantiles of serum total folate levels: Q1, ≤26.5 nmol/L; Q2, 26.6–38.7 nmol/L; Q3, 38.8–56.9 nmol/L; and Q4, >56.9 nmol/L.Quantiles of serum 5-MTHF levels: Q1, ≤24.6 nmol/L; Q2, 24.7–36.4 nmol/L; Q3, 36.5–54.4 nmol/L; and Q4, >54.4 nmol/L.Quantiles of serum UMFA levels: Q1, ≤0.50 nmol/L; Q2, 0.51–0.68 nmol/L; Q3, 0.69–0.94 nmol/L; and Q4, >0.94 nmol/L. 5-MTHF: 5-methyltetrahydrofolate; BMI: Body mass index; CI: Confidence interval; NAFLD: Non-alcoholic fatty liver disease; OR: Odds ratio; UMFA: Unmetabolized folic acid. In addition to serum analysis, we next explored the association of dietary folate intake with the risk of NAFLD. Dietary folate intake was measured by DFE. We found that participants with the highest quartiles of folate intake from food plus dietary supplements had a significantly lower risk of NAFLD than those with the lowest quartile (OR: 0.69, 95% CI: 0.49–0.98; Supplementary Table 3, https://links.lww.com/CM9/B363). In the subgroup analysis, the association was more pronounced among female participants. Interestingly, hypertension and diabetes were detected as significant modifiers for the association (P < 0.001 for interaction). Participants without these disorders were more likely to have inverse associations with total folate intake and NAFLD [Supplementary Table 4, https://links.lww.com/CM9/B363]. These findings suggested that individuals may benefit from dietary folate intake to reduce the risk of NAFLD. In this study, we aimed to examine the association between folate and NAFLD using a nationally representative sample in the United States. We found that NAFLD patients consumed less folate and had significantly lower levels of serum total folate and its major component 5-MTHF than healthy controls. We also found that serum folate levels were negatively associated with the risk of NAFLD, and inadequate folate intake was associated with an increased risk of NAFLD. These findings suggest a close association between folate and NAFLD. The potential mechanisms by which folate may be related to NAFLD remain unclear. Several hypotheses in view of its core role in one-carbon metabolism are proposed. First, folate can reduce circulation concentrations of homocysteine through transmethylation and improve autophagy.[2] Second, folate protects against lead acetate-induced hepatotoxicity by decreasing nuclear factor kappa B (NF-κB) and interleukin-1β (IL-1β) production and lipid peroxidation-mediated cell injury.[3] Third, recent results suggested that folate deficiency increased lipid accumulation and leptin production of adipocytes; thus, inadequate folate status might be one of the risk factors for adiposity.[4] To date, therapeutics that directly exploit one-carbon metabolism have been less investigated in the treatment of NAFLD. However, there are many drugs in the pipeline that are good candidates to cure NAFLD.[5] One-carbon donors exhibit dose-dependent ameliorating effects on NAFLD because these donors reduce body weight, serum lipids, and liver enzymes, decrease hepatic fat accumulation, and increase hepatic global DNA methylation. Therefore, folic acid supplementation may be beneficial for NAFLD patients. The advantage of this study is that our analysis is based on the large sample size of a nationally representative sample of the United States population. Besides, NHANES provides more detailed serum folate components such as 5-MTHF and UMFA, not just total folate. In addition, we also explored the relationship between dietary folate (from foods and supplements, https://links.lww.com/CM9/B363) and NAFLD. However, study limitations should also be considered. First, although NHANES is a nationwide, updated, and well quality-controlled database, it is a cross-sectional study per se that does not provide any causality inferences. Second, the classification of NAFLD and NAFLD-associated advanced fibrosis was ascertained using non-invasive biomarkers, which is less optimal than the use of invasive biopsy in diagnosis, and misclassification bias may be inevitable. Third, dietary data in our analysis were collected based on self-reported records, which may result in greater measurement errors. Funding This work was supported by the National Natural Science Foundation of China (No. 82070585), the Natural Science Foundation of Zhejiang Province (No. LY21H030001), and the Health Science and Technology Plan Project of Zhejiang Province (Nos. 2021KY1129 and 2022KY1274). Conflicts of interest None.