Height predict incident non-alcoholic fatty liver disease in a general adult population, independent of body mass index and metabolic syndrome

Background Early-life hormonal and nutritional factors can greatly influence the risk of non-alcoholic fatty liver disease (NAFLD). Adult height is a simple marker for these factors. We aimed to evaluate adult height as a predictor of NAFLD. Methods We performed a prospective cohort study of 35, 994 participants recruited from 2010 to 2016 in Tianjin, China. NAFLD was diagnosed by ultrasound. Adjusted Cox proportional hazards regression models were conducted in 2019 to assess the gender-specific association between the quintiles of height and the incidence of NAFLD. Results Participants were followed up for 5.5 years with a mean follow-up of 2.6 years. During the follow-up period, 6,245 of 35,994 participants developed NAFLD. The multiple-adjusted hazard ratios (95% confidence interval) of NAFLD for increasing quintiles of height were 1.00 (reference), 0.90 (0.81, 0.99), 0.97 (0.87, 1.07), 0.86 (0.78, 0.96), and 0.84 (0.75, 0.94) ( P for trend < 0.01) in males and 1.00 (reference), 0.97 (0.86, 1.09), 0.98 (0.86, 1.11), 0.93 (0.81, 1.06), and 0.84 (0.73, 0.96) ( P for trend = 0.02) in females, respectively. Conclusions The study is the first to demonstrate adult height is a new, simple and inexpensive marker for ultra-early prediction of NAFLD.


Diagnosis of NAFLD
Real-time ultrasonography performed by trained and certified technicians was used to diagnose NAFLD. Participants were considered to have NAFLD if (1) they had a selfreported alcohol intake of < 140 g/week and < 70 g/week for males and females, respectively; (2) and at least two of the following abnormal findings of abdominal ultrasound images: diffusely increased liver near field ultrasound echo; increased liver echotexture, compared to the kidneys; vascular blurring and the gradual attenuation of far field ultrasound echo(18).

Assessment of other variables
Waist circumference (WC) was measured using a nonelastic plastic anthropometric tape at the level of umbilicus with subjects standing and breathing normally.Participants rest for at least 5 minutes in a seated position prior to blood pressure measurements. Blood pressure was measured twice from participants' upper right arms using the TM-2655 device (A&D Company Ltd, Tokyo, Japan), and the blood pressure value was recorded in average. Fasting blood samples for the analysis of biochemical values were collected in siliconized vacuum plastic tubes. Fasting blood glucose, total cholesterol, triglycerides, low-density lipoprotein cholesterol, and high-density lipoprotein cholesterol were measured using appropriate kits on a Cobas 8000 analyzer (Roche, Mannheim, Germany).
We defined metabolic syndrome (MetS) according to the American Heart Association scientific statements of 2009(19). Alanine aminotransferase was measured by IFCC method.
Body weight was measured by training nurses, with participants without heavy clothes.
Body mass index (BMI) was calculated as weight (kg) divided by squared height (m 2 ).
Information on family history of cardiovascular disease, family history of hypertension, family history of hyperlipidemia, and history of diabetes, was assessed at baseline using a structured questionnaire.

Statistical analysis
Baseline characteristics of participants were compared using analysis of variance for continuous variables and logistic regression analysis for categorical variables. Continuous variables were shown as geometric mean (95% confidence interval (CI)), and categorical variables were presented as percentage.
Because the interaction between sex and height was statistically significant (P <0.0001), we analyzed the association between height and NAFLD stratified by sex. We fitted four Cox proportional hazards regression models to evaluate the association between baseline height and incident NAFLD. The initial model was unadjusted model (crude model). Model 2 was adjusted for age and WC. In model 3, we additionally adjusted for smoking status, alcohol drinking status, Mets, family history of cardiovascular disease, family history of family history of hypertension, family history of hyperlipidemia, and family history of diabetes. In model 4, we further adjusted for baseline BMI. All P values for linear trends were calculated using the median value for each quintile.
All statistical analyses were performed using SAS version 9.4 (SAS Institute, Inc.). Twotailed P <0.05 was considered as statistically significant.

Results
In this study, 41.3% of participants were males and 58.7% were females. Mean age (standard deviation) years was 42.2 (12.9) in males and 39. cholesterol, triglycerides, fasting blood glucose, systolic blood pressure, diastolic blood pressure, and alanine aminotransferase (all P < 0.0001). Participants with NAFLD tended to be male, smoker, ex-smoker, everyday drinker, sometimes drinker, while those without NAFLD tended to be non-smoker and non-drinker (all P < 0.0001). In addition, a higher proportion of participants with NAFLD had MetS and a family history of cardiovascular disease, hypertension and diabetes (all P < 0.0001). Table 2 and Table 3 show the crude and adjusted associations between quintiles of height and NAFLD in male and female participants, respectively. In the third multivariate model which didn't adjust for BMI, the adjusted HRs (95% CI) for NAFLD across height quintiles

Discussion
In this large-scale prospective cohort study, we found that higher level of adult height was inversely associated with the risk of NAFLD among males and females in China. The inverse association remained even after controlling for potential confounding factors. To our knowledge, this is the first study to investigate the association between adult height and NAFLD.
We adjusted for multiple potentially confounding factors in our analysis. This study indicated that numerous factors (such as age, BMI, WC, smoking status, metabolic syndrome and family history of some diseases) correlated positively with NAFLD. We used crude model (model 1) first and results showed negative association between height and NAFLD in both males and females. It is well-recognized that NAFLD and height are related to age and WC(20, 21), so we adjusted for these two variables in model 2. Adjustment for age and WC made the associations in males more obvious compared with model 1; however, in females, this adjustment didn't significantly influence the associations in model 1, leading us to conclude that age and WC are major confounding factors in males but not in females. Since NAFLD was associated with WC, smoking status, drinking status, metabolic syndrome, family history of cardiovascular disease, hypertension, hyperlipidemia and diabetes(22, 23), we subsequently adjusted for these variables in model 3. After adjustments for these factors, the associations didn't change significantly in both males and females, implying these factors may not confound the association between height and NAFLD. In model 4, the present study adjusted BMI and variables in model 3 to confirm the role of BMI in association between height and NAFLD. This adjustment made the association less obvious in both males and females, suggesting that BMI play an important role in association between height and NAFLD.
To date, no studies have investigated the association between height and NAFLD. Several studies investigated the association between height and T2DM. Reports about the risk of T2DM and height have produced conflicting results(24, 25). A meta-analysis showed negative association between height and risk of T2DM in woman only(26), whereas a cohort study of Finnish men showed adult height is associated with decreased risk of T2DM (1). Since T2DM and NAFLD both result from metabolic dysregulation, these results are, to some extent, consistent with our novel findings that shorter people were associated with higher incidence of NAFLD. Compared to previous cohort studies using Chinese adults with overall incidence of NAFLD ranging from 15.2% to 24.8%(27),the overall incidence of NAFLD in our study is 17.4%, consistent with previous studies.
The mechanisms of association between adult height and NAFLD may involve epigenetic changes induced by early life adversity. Interacting with epigenetic mechanisms, hormonal and nutritional conditions in early life influence both attained height and later susceptibility to NAFLD. GH and IGF-1 play essential roles in linear growth as well as in the liver metabolism(28). Therefore, epigenetic modification of IGF-1 gene might play a critical role in the development of NAFLD and the growth in height. NAFLD-specific expression and methylation differences between healthy controls and morbidly obese patients with all stages of NAFLD were seen for the IGF-1 gene, one of the key drivers of the liver's phenotype(29). Furthermore, the methylation of the IGF-1 P2 promoter showed a strong negative association with serum IGF-1 levels and height as a child(30). In addition, data from animal models have indicated that nutritional perturbation of epigenetic regulation is a likely link between prenatal and early postnatal nutrition and health status in later life(31). A study in mice showed that exposure to prenatal and postweaning western-style diet predisposed male mouse offspring to the development of NAFLD in adulthood and induced alterations in DNA methylation in key metabolic genes(32). On the other hand, a recent study has demonstrated that height plays an important partial role in determining several aspects of a person's socioeconomic status(33). Since lower socioeconomic status is closely associated with increased levels of health impairment(34), socioeconomic status maybe an explanation for the association between height and NAFLD. Further research is warranted to validate the hypotheses.
The present study has several limitations. Firstly, NAFLD was diagnosed by abdominal ultrasound rather than liver biopsy, which is the gold standard for diagnosis of NAFLD.
However, abdominal ultrasound is not invasive and widely used in large-scale populationbased studies. Moreover, this noninvasive method has a sensitivity of 89% and a specificity of 93%(35). Second, although we adjusted for a considerable number of potential confounding factors in the present study, residual confounding cannot be excluded.

Conclusions
Adult height was negatively associated with incident NAFLD in males and female, independent of BMI and MetS. The present results indicate that adult height may be a useful predictor for NAFLD to identify high-risk populations and prevent NAFLD at an early age. Furthermore, this study provides clues to the mechanism of the link between early life experiences and NAFLD. Future studies are needed to elucidate the mechanism of association between adult height and the risk of NAFLD.

List Of Abbreviations
NAFLDNon-alcoholic fatty liver disease