Associations between body circumference and testosterone levels and risk of metabolic dysfunction-associated fatty liver disease: A Mendelian randomization study

doi:10.21203/rs.3.rs-2314023/v1

BACKGROUD: Body circumference and testosterone levels has been reported as associated with metabolic dysfunction-associated fatty liver disease (MAFLD) risk. However, whether body circumference and testosterone levels play a role in the development of MAFLD remains inconclusive.

METHODS: Using a large database of genome-wide association studies, genetic loci that are independent of each other and strongly associated with body circumference and testosterone levels were selected as instrumental variables, the causal relationship between body circumference and testosterone and risk of MAFLD was investigated by two-sample Mendelian randomization methods such as inverse variance weighted (IVW), MR-Egger regression, and weighted median estimator (WME), using the odds ratios (ORs) as evaluation indicators.

RESULTS: A total of 377 SNPs were included as instrumental variables in this study, including 3 for neck circumference, 165 for waist circumference, 78 for hip circumference, and 131 for testosterone levels. Using the two-sample Mendelian randomization method described above to derive a causal association between exposure and outcome, the IVW method was used as the primary method of analysis, and the results of the analysis showed a causal association between all four exposure factors and MAFLD, with an OR of neck circumference was 0.87,95%CI:0.78~0.98, P=0.018; and an OR of waist circumference of 2.88, 95%CI: 1.73 to 4.81, P<0.001, OR of hip circumference was 0.48, 95%CI:0.24~0.96, P=0.039, and an OR of testosterone level of 1.75, 95%CI:1.17~2.62, P=0.006. Cochran Q test by IVW and MR-Egger method showed that SNPs did not have intergenic heterogeneity, and the test for pleiotropy indicated a weak possibility of pleiotropy in the causal analysis.

CONCLUSION: The results of the two-sample Mendelian randomization analysis showed that waist circumference and testosterone level were risk factors for MAFLD, and the incidence of MAFLD increased as both increased; neck circumference and hip circumference were protective factors for MAFLD, and the risk of MAFLD decreased as both increased.

Neck circumference1

Waist circumference2

Hip circumference3

Testosterone4

Metabolic dysfunction-associated fatty liver disease5

Two-sample Mendelian randomization analysis6

Causal Inference7

Metabolic dysfunction-associated fatty liver disease (MAFLD) was renamed from Non-alcoholic fatty liver disease (NAFLD) in March 2020 (1). The disease is characterized by intrahepatocellular lipid deposition combined with systemic multisystemic metabolic disorders (2), the prevalence is increasing year by year and has become the number one chronic liver disease in China and the global prevalence is about 25% (3). MAFLD can not only cause steatohepatitis and liver fibrosis, liver cancer, but also cause extra-hepatic complications such as cardiovascular disease and chronic kidney disease due to MAFLD-related metabolic disorders, which are more serious to people's health (4). However, so far, there are no effective preventive and therapeutic drugs for MAFLD, therefore, it is significant to identify the factors affecting the development of MAFLD and to intervene early for the development of MAFLD. Obesity, especially abdominal obesity, is the primary risk factor for MAFLD, and body circumference, especially waist circumference, is a practical indicator of body BMI and obesity (5), previous studies (6–8) have found that the incidence of fatty liver disease increases with the growth of neck circumference and waist circumference, and hip circumference and waist-to-hip ratio are also risk factors affecting the development of MAFLD (9). It has also been shown that sex hormone levels are significantly associated with various specific disorders in the metabolic syndrome, testosterone is an important regulator of glucose and lipid metabolism in the body and is an intrinsic indicator of hepatic lipid metabolism (10–12), and the severity of disease in MAFLD patients is significantly correlated with blood testosterone levels (13). However, the perspective of the above studies is limited to traditional observational epidemiology, which is vulnerable to unknown confounding factors and reverse causation. In recent years, Mendelian randomization (MR) has been developed as an important method for causal inference, and it uses exposure-related genotypes as instrumental variables (IVs) and can overcome the drawbacks of traditional epidemiological studies such as difficult data acquisition and poor extrapolation of results (14). However, traditional MR requires genotypes from the same individual as well as information on exposure and outcome, and data detection costs are high. In recent years, the two-sample Mendelian randomization (2-sample MR) method has been gradually developed, which allows gene and exposure association data and gene and outcome association data from two independent sample populations of the same overall population (15), respectively, compared with traditional MR, greatly improving the efficiency and feasibility of etiological studies, and has been widely used in causal association studies of risk factors and disease outcomes.

This study used a two-sample MR method to explore the causal associations between four exposure factors of body circumference (including neck circumference, waist circumference, and hip circumference) and testosterone level and MAFLD.

2.1 Study Design

In this study, body circumference (neck circumference, waist circumference, hip circumference) and testosterone level were used as exposure factors, and single nucleotide polymorphisms (SNPs) loci significantly associated with the above exposure factors were selected as instrumental variables(IVs), and the outcome variable was MAFLD, and the causal association analysis between exposure and outcome was performed using a two-sample MR analysis approach based on a publicly available genome wide association study (GWAS) database of large samples, and Cochran Q test to assess heterogeneity, and finally sensitivity analysis to verify the reliability of the causal association results. MR analysis needs to satisfy the following three core hypotheses: ①there is a strong association between instrumental variable Z and exposure factor X; ②instrumental variable Z is not associated with any confounding factor U of the exposure-outcome association; and ③the instrumental variable Z does not affect the outcome Y, except possibly by association with the exposure X. The two-sample MR study model is shown in Figure 1.

2.2 Data sources

Three important body circumferences (neck circumference, waist circumference, and hip circumference) and serum testosterone levels were used as exposure factors, SNPs significantly associated with the above exposure factors were used as IVs, and the outcome factor was MAFLD. The pooled data used to conduct the two-sample MR study were obtained from the IEU Open GWAS database (https://gwas.mrcieu.ac.uk/ ), neck circumference (GWAS ID: ebi-a-GCST90017134), waist circumference (GWAS ID: ukb-a-382), hip circumference (GWAS ID: ukb-a-388), testosterone (GWAS ID: ebi-a-GCST90012102), and MAFLD (GWAS ID: finn-b-NAFLD), all of the above databases were European populations or mixed populations. All datasets used in this study were from the public domain, and summary information is presented in Table 1.

TABLE 1 Summary of the GWAS included in this two-sample MR study

Variable	ID	Sample size	Number of SNPs	Consortium	Population	Sex	Year
Neck circumference	ebi-a-GCST90017134	6358	7297174	-	Mixed	Males and Females	2021
Waist circumference	ukb-a-382	336639	10894596	Neale Lab	European	Males and Females	2017
Hip circumference	ukb-a-388	336601	10894596	Neale Lab	European	Males and Females	2017
Testosterone levels	ebi-a-GCST90012102	188507	16139906	-	European	Males and Females	2020
MAFLD	finn-b-NAFLD	218792	16380466	-	European	Males and Females	2021

2.3 Selection of instrumental variables

SNPs with significant correlation with body circumference and testosterone level (P < 5. 0×10^-8 ) were screened, and the interference of linkage disequilibrium (LD) was excluded (16), setting parameter r² = 0. 001, kb = 10000, and the echo SNPs were excluded, and the SNPs with significant heterogeneity were excluded by heterogeneity test, If the number of SNPs filtered according to the above criteria is large, each SNP should be queried on the PhenoScanner website (http://www.phenoscanner.medschl.cam.ac.uk/), SNPs affected by confounding factors that violated Mendelian randomization core hypothesis②and③were excluded, and finally valid SNPs significantly associated with exposure factors that met Mendelian core hypothesis were obtained as IVs.F > 10 indicates the absence of weak instrumental variable bias, which is calculated as follows: , where N is the sample size of the exposed database, k is the number of SNPs, and R² is the proportion of variance explained by SNPs in the exposed database. R² is calculated as , where EAF is the effect allele frequency, β is the allele effect value, and SD is the standard deviation.

2.4 Statistical analysis for Mendelian randomization

We used the TwoSampleMR package (version 0.5.6) in R program (version 4.2.1) to integrate and analyze the data. In this study, inverse variance weighted (IVW) (17) was used as the main analysis method, while MR-Egger regression (18), weighted median estimator (WME) (19), simple mode and weighted mode (20) were used together for MR analysis. The principle of IVW is to weight the inverse of the variance of each IV as the weight while ensuring that all IVs are valid, the regression does not consider the intercept term, and the final result is the weighted average of the effect values of all IVs. The major difference between MR-Egger regression and IVW is that the regression takes into account the presence of the intercept term, and in addition, it also uses the inverse of the ending variance as a weight for the fit. The WME is defined as the median of the weighted empirical density function of the ratio estimates, which allows consistent estimation of causality if at least half of the valid instruments are present in the analysis.

2.5 Heterogeneity and sensitivity test

There may be heterogeneity in the 2-sample MR analysis due to differences in analysis platforms, experimental conditions, including populations and SNPs, which may bias the estimation of causal effects. Therefore, the main IVW and MR-Egger methods were tested for heterogeneity in this study. The heterogeneity test was used to test the differences between individual IVs, and Cochran's Q statistic and P-value were used to determine whether there was heterogeneity, and P < 0.1 represented the presence of heterogeneity; Pleiotropy test mainly tests the presence of horizontal pleiotropy for multiple IVs (21), and the P-value of the pleiotropy test was used in this study to measure whether there was pleiotropy in the analysis, if P > 0.05, it is considered that the possibility of pleiotropy in the causal analysis is weak; Leave-one-out sensitivity test is mainly to calculate the MR results of the remaining IVs after eliminating them one by one (22), if the estimated MR results of other IVs after eliminating one IV are very different from the total results, it means that the MR results are sensitive to that IV. The presence of pleiotropy in the analysis was also determined in this study using the MR-pleiotropy residual sum outlier (MR-PRESSO) (23).

3.1 Instrumental variables

After removing IVs with linkage disequilibrium and excluding IVs that violate MR core hypotheses ②and③as well as outliers, a total of 377 SNPs were included in this study, including 3 for neck circumference, 165 for waist circumference, 78 for hip circumference, and 131 for testosterone level. The basic information of partial SNPs is shown in Table 2, and the part not listed is similar to it. The F-statistics corresponding to a single SNP were all bigger than 10, indicating that the causal association was less likely to be affected by weak instrumental variable bias.

TABLE 2 Basic information of SNPs

Description of exposure	SNPs	CHR	EA	OA	exposure			outcome(MAFLD)
Description of exposure	SNPs	CHR	EA	OA	Beta	SE	P-value	Beta	SE	P-value
Neck circumference	rs117907541	9	A	G	1.2386	0.2241	3.26E-08	-0.201	0.1207	0.0958804
	rs35130773	2	A	G	1.6571	0.2898	1.07E-08	-0.2089	0.166	0.2084
	rs78885261	15	T	G	1.2765	0.2227	9.87E-09	-0.1611	0.1393	0.2475
Waist circumference	rs10172196	2	A	G	0.0140208	0.00234352	2.19563E-09	-0.0415	0.051	0.4155
	rs10236214	7	T	C	0.0145481	0.00225631	1.13682E-10	-0.0286	0.0522	0.584201
	rs10237306	7	T	G	0.0133558	0.00221471	1.63648E-09	0.0323	0.0488	0.508
	rs10269774	7	A	G	0.0132003	0.00230082	9.6323E-09	-0.0257	0.0517	0.6192
	rs10459088	12	A	G	0.0135454	0.00245603	3.48747E-08	-0.0092	0.0564	0.87
	rs1056441	20	C	T	0.0128077	0.00229345	2.34617E-08	0.0072	0.0547	0.8952
	……	……	……	……	……	……	……	……	……	……
Hip circumference	rs10153248	17	G	A	-0.0170889	0.0024225	1.74E-12	0.0112	0.0493	0.8196
	rs10210468	2	C	T	-0.0150017	0.00244268	8.19E-10	-0.0638	0.0481	0.1845
	rs10953513	7	G	A	-0.0188736	0.00244305	1.12E-14	0.0182	0.0479	0.704001
	rs1118151	2	G	T	0.016606	0.00269097	6.79E-10	-0.015	0.049	0.759401
	rs11245480	10	G	C	0.0155066	0.00256762	1.55E-09	-0.0443	0.0541	0.4128
	rs113364497	8	T	C	-0.0157612	0.00282326	2.37E-08	0.0234	0.0553	0.671699
	……	……	……	……	……	……	……	……	……	……
Testosterone levels	rs10048173	17	G	A	-0.0242187	0.00252788	1.7E-22	0.0121	0.0479	0.8009
	rs10136290	14	G	A	0.0162591	0.00257663	7.4E-11	0.1105	0.0487	0.0231798
	rs10504255	8	A	G	-0.0173977	0.00268685	6.1E-10	-0.0532	0.0494	0.2813
	rs1050541	17	G	T	-0.0486949	0.00261308	1E-79	0.0212	0.0484	0.6615
	rs10743398	12	A	G	0.0163223	0.00281544	9.2E-10	-0.0413	0.0538	0.4433
	rs10757893	9	G	A	-0.0137428	0.00260112	9.3E-09	-0.0149	0.0491	0.7619
	……	……	……	……	……	……	……	……	……	……

Abbreviations: SNP, single nucleotide polymorphism;Chr, chromosome; EA, effect allele; OA, other allele; Beta, beta coefficient; SE, standard error; MAFLD, metabolic dysfunction-associated fatty liver disease.

3.2 Results of two-sample MR analysis

In this study, the IVW method was used as the main analysis method. In the MR analysis results, with an OR value bigger than 1, exposure was considered a risk factor for the outcome, and vice versa as a protective factor for the outcome, and when the P value was less than 0.05, it was considered statistically significant and the causal association was established. The IVW method of the random effects model showed that all four exposure factors were causally associated with MAFLD, with waist circumference (OR=2.88,95% CI:1.73~4.81, P<0.001) and testosterone level (OR=1.75,95%CI:1.17~2.62, P=0.006) being risk factors, and the prevalence of MAFLD increased with the increase in both; neck circumference (OR=0.87,95%CI:0.78~0.98, P=0.018) and hip circumference (OR=0.48,95%CI:0.24~0.96, P=0.039) were protective factors, and the risk of MAFLD decreased with the increase of both. The results of the specific analysis of the five methods are shown in Table 3. Visualization of MR results is shown in the Supplementary Material (MR results visualization charts) .

TABLE 3 MR estimates of associations between 4 types of exposure and MAFLD risk

Exposure	Number of SNPs	MR methods	outcome	OR(95%CI)	p-value
Neck circumference	3	MR Egger	MAFLD	0.95(0.38,2.38)	0.93
		WME		0.88(0.77,1.01)	0.067
		IVW		0.87(0.78,0.98)	0.018
		Simple mode		0.88(0.75,1.04)	0.273
		Weighted mode		0.88(0.76,1.02)	0.242
Waist circumference	165	MR Egger		3.96(0.46,34.41)	0.214
		WME		2.36(1.12,4.99)	0.024
		IVW		2.88(1.73,4.81)	<0.001
		Simple mode		3.51(0.41,30.21)	0.254
		Weighted mode		4.27(0.67,27.34)	0.127
Hip circumference	78	MR Egger		0.04(0.003,0.52)	0.017
		WME		0.38(0.14,1.03)	0.058
		IVW		0.48(0.24,0.96)	0.039
		Simple mode		0.21(0.02,2.26)	0.2
		Weighted mode		0.27(0.03,2.48)	0.248
Testosterone level	131	MR Egger		1.82(0.83,3.99)	0.139
		WME		1.7(0.87,3.32)	0.122
		IVW		1.75(1.17,2.62)	0.006
		Simple mode		2.14(0.56,8.16)	0.268
		Weighted mode		1.58(0.82,3.03)	0.176

Abbreviations: CI, confidence interval

3.3 Result of heterogeneity and sensitivity test

The Cochran Q test for IVW and MR-Egger method indicated that there was no intergenic heterogeneity in SNPs (P > 0.1); the test for pleiotropy indicated that the possibility of pleiotropy in the causal analysis was weak (P > 0.05), and the above results are detailed in Table 4.

In the visualization of MR results, the funnel plot showed that the points representing the causal association effect were roughly symmetrically distributed when a single SNP was used as the IV, indicating that the causal association was less likely to be affected by potential bias. The results of the "Leave-one-out" sensitivity analysis showed that after eliminating each SNP in turn, no SNP with a large effect on the causal association estimates was found.

TABLE 4 results of Heterogeneity and sensitivity test

Exposure	outcome	MR methods	p of pleiotropy	p of Cochran Q
Neck circumference	MAFLD		0.882
		MR Egger		0.819
		IVW		0.957
Waist circumference			0.767
		MR Egger		0.328
		IVW		0.346
Hip circumference			0.053
		MR Egger		0.352
		IVW		0.27
Testosterone level			0.921
		MR Egger		0.13
		IVW		0.143

In this study, the causal relationship between body circumference and testosterone level and risk of MAFLD was investigated using a two-sample MR analysis method using publicly available databases and a large-scale GWAS study. The MR results derived from the IVW method support the existence of a causal relationship between the exposure factors studied and MAFLD. Among these, waist circumference and testosterone levels were risk factors, and the prevalence of MAFLD increased with both, this finding is consistent with the results of several previous studies. A five-year study by S Wang et al. including 12,477 observers showed that the cumulative incidence of MAFLD increased with increasing waist circumference and concluded that waist circumference was an independent risk factor for MAFLD (24). A prospective cohort study in Korea involving 5400 people aged 40~69 years showed that waist circumference was the most important risk factor for MAFLD among the physical indicators of middle-aged and elderly people in Korea, and the threshold of waist circumference values were 81cm for men and 78.5cm for women (25). A Korean cohort study involving 613 women of various ages showed that serum testosterone levels in premenopausal women were positively associated with the risk of developing MAFLD (26). The results of the above studies suggest that waist circumference and testosterone levels are risk factors for MAFLD, which is consistent with the results of our study. However, it should be noted that there are clinical studies showing gender differences in testosterone levels and the risk of MAFLD: a meta-analysis of 16 indicators involving 13,721 men and 5,840 women by Veeravich Jaruvongvanich et al. showed that the lower the testosterone level, the higher the risk of MAFLD in men, and in women, conversely, the lower the testosterone level, the lower the risk of developing MAFLD (27). The multicenter clinical study by Monika Sarkar et al. showed that men with lower testosterone levels were more likely to develop MAFLD (28). The above results also suggest that there may be gender differences in the causal association between testosterone levels and MAFLD, and the database of testosterone levels selected for this study includes both men and women, so the results of the analysis may be biased. To make the study results more reliable, the analysis should be performed separately according to gender.

The results of the analysis in this study concluded that hip and neck circumference are protective factors for MAFLD and that the risk of developing MAFLD decreases with an increase in both, which is contradictory to the results of some of the current studies. A study including 635 observers at a Thai academic institution showed that men and women with MAFLD had higher neck circumference values than men and women without MAFLD, suggesting that neck circumference may be a risk factor for MAFLD (29). A study including 2761 subjects at Changhai Hospital in China showed that MAFLD subjects had significantly higher neck circumference values than normal control subjects (30). Chinese A clinical study on risk factors for MAFLD in Urumqi showed that hip circumference was not a risk factor for the development of MAFLD and that waist-to-hip ratio was more meaningful as a risk factor for predicting the development of MAFLD (31). A clinical study of MAFLD in Taiwan, China, which included 1,969 participants, showed that waist-to-hip ratio was of concern in the incidence and severity of MAFLD, and that the correlation between waist-to-hip ratio and MAFLD was more pronounced in women than in men (32). As can be seen, hip circumference is generally not used as an independent risk factor for predicting the development of MAFLD in clinical studies; it is susceptible to confounding factors such as gender and type of obesity, and is usually analyzed together with waist circumference and height,and the results of this study that hip circumference is a protective factor for MAFLD have little significance in guiding the actual situation; while neck circumference is often considered to be a risk factor for MAFLD, the opposite result in this study may have been biased by the inclusion of too few instrumental variables consistent with MR core hypotheses.

This study explores the causal relationship between body circumference, testosterone, and risk of MAFLD using two-sample MR method, which avoids the disturbance of confounding factors such as social environment and lifestyle because genetic variation is a long-term and stable exposure and can be measured directly. Compared to randomized controlled trials, MR allows for truly random assignment and does not violate ethics. Two-sample MR has a relatively larger sample size, allows for greater confidence, and increases the specificity of genetic variants using multiple IVs compared to a single SNP.

Innovations of this study: ① The datasets used in this study were of large sample size and publicly available, which ensured the quality of the IVs used in the analysis. ② The IVs in this study fully considered and excluded SNPs that could affect the results. ③ Outlier SNPs were excluded using the MR-PRESSO method. The selection of IVs in this study is more detailed, comprehensive and reliable than in previous studies.

Of course, there are limitations in this study: ① The two-sample MR assumes a linear relationship between exposure and outcome, and the method is not applicable if the relationship is nonlinear. (② No gender-specific subgroup analysis was performed, and a more specific effect relationship could be obtained if a two-sample MR analysis with different sexes was performed. ③ This study has limitations in explaining the biological mechanisms of causal effects.

In conclusion, this study validated the feasibility of applying MR methods to the study of the risk of developing MAFLD, and searching for key risk factors for preventable MAFLD at the genetic level, which has guiding implications for the early prevention of MAFLD in China. However, there are few GWAS data in Asian and Chinese populations, and data from other databases are difficult to obtain and organize, so the results of this study need to be further validated in Chinese populations in combination with clinical and randomized controlled trials.

Ethics approval and consent to participate

The data used in this paper are publicly available, ethically approved, and the subjects have given their informed consent.

Consent for publication

Not applicable

Availability of data and materials

The datasets [Exposure/Outcome] for this study can be found in the [IEU Open GWAS] [https://gwas.mrcieu.ac.uk/].

Competing interests

The authors declare that they have no competing interests

Funding

This work was supported by Shandong Provincial Health Commission and Shandong Provincial Administration of Traditional Chinese Medicine: Qilu School of Traditional Chinese Medicine Academic School Construction Project (2022-93); TCM Science and Technology Development Program of Shandong Province 2019-2020 (2019-0114)

Authors' contributions

Study conception and design: L N, JG S; Analyses: L N; Draft: L N; Supervision: JG S and L N; Interpretation of results, critical editing, and manuscript approval: all authors.

Acknowledgments

We express our gratitude to the IEU Open GWAS database (https://gwas.mrcieu.ac.uk/ ) for providing publicly available summary-level GWAS data for our study.

Mohammed Eslam PN, Newsome SK, Sarin QM, Anstee G, Targher, et al. A New Definition for Metabolic Dysfunction-Associated Fatty Liver Disease: An International Expert Consensus Statement. J Hepatol. 2020;73(1):202–9. doi:10.1016/j.jhep.2020.03.039.
Gregory A, Michelotti, Mariana V, Machado. Anna Mae Diehl. NAFLD, NASH and Liver Cancer. Nat Rev Gastroenterol Hepatol. 2013;10(11):656–65. doi:10.1038/nrgastro.2013.183.
Zobair M, Younossi AB, Koenig D, Abdelatif Y, Fazel L, Henry, et al. Global Epidemiology of Nonalcoholic Fatty Liver Disease-Meta-Analytic Assessment of Prevalence, Incidence, and Outcomes. Hepatology (Baltimore, Md.) (2016) 64(1): 73–84. doi: 10.1002/hep.28431.
Sun D-Q, Jin Y, Wang T-Y, Kenneth I, Zheng RS, Rios, et al. MAFLD and Risk of CKD. Metab Clin Exp. 2021;115:154433. doi:10.1016/j.metabol.2020.154433.
Robert Ross IJ, Neeland S, Yamashita I, Shai J, Seidell, et al. Waist Circumference as a Vital Sign in Clinical Practice: A Consensus Statement from the IAS and ICCR Working Group on Visceral Obesity. Nat Rev Endocrinol. 2020;16(3):177–89. doi:10.1038/s41574-019-0310-7.
SoJung Lee JL, Kuk C, Boesch SA. Waist Circumference Is Associated with Liver Fat in Black and White Adolescents. Applied Physiology, Nutrition, and Metabolism = Physiologie Appliquee. Nutr Et Metab. 2017;42(8):829–33. doi:10.1139/apnm-2016-0410.
Nima Motamed M, Sohrabi H, Ajdarkosh G, Hemmasi M, Maadi, et al. Fatty Liver Index vs Waist Circumference for Predicting Non-Alcoholic Fatty Liver Disease. World Journal of Gastroenterology (2106) 22(10): 3023–3030. doi: 10.3748/wjg.v22.i10.3023.
Jian C, Xu Y, Ma X, Shen Y, Wang Y, et al. Neck Circumference Is an Effective Supplement for Nonalcoholic Fatty Liver Disease Screening in a Community-Based Population. Int J Endocrinol. 2020;2020:7982107. doi:10.1155/2020/7982107.
Marta Borges-Canha,João Sérgio. Neves MM, Silva F, Mendonça T, Moreno, et al. Waist-to-Hip Ratio and Inflammatory Parameters Are Associated with Risk of Non-Alcoholic Fatty Liver Disease in Patients with Morbid Obesity. Biomedicines (2022) 10(10): 2416. doi:10.3390/biomedicines10102416.
Daniel M, Kelly S, Akhtar DJ, Sellers V, Muraleedharan KS, Channer, et al. Testosterone Differentially Regulates Targets of Lipid and Glucose Metabolism in Liver, Muscle and Adipose Tissues of the Testicular Feminised Mouse. Endocrine. 2016;54(2):504–15. doi:10.1007/s12020-016-1019-1.
Caldwell ASL, Middleton LJ, Jimenez M, Desai R, McMahon AC. Characterization of Reproductive, Metabolic, and Endocrine Features of Polycystic Ovary Syndrome in Female Hyperandrogenic Mouse Models. Endocrinology. 2014;155(8):3146–59. doi:10.1210/en.2014-1196.
Cao W, Xu Y, Shen Y, Wang Y, Ma X, et al. Associations between Sex Hormones and Metabolic-Associated Fatty Liver Disease in a Middle-Aged and Elderly Community. Endocr J. 2022;69(8):1007–14. doi:10.1507/endocrj.EJ21-0559.
Yim JY, Kim J, Kim D, Ahmed A. Serum Testosterone and Non-Alcoholic Fatty Liver Disease in Men and Women in the US. Liver International: Official Journal of the International Association for the Study of the Liver. 2018;38(11):2051–9. doi:10.1111/liv.13735.
Connor A, Emdin AV, Khera. Sekar Kathiresan. Mendelian Randomization JAMA. 2017;318(19):1925–6. doi:10.1001/jama.2017.17219.
Brandon L, Pierce, Stephen Burgess. Efficient Design for Mendelian Randomization Studies: Subsample and 2-Sample Instrumental Variable Estimators. Am J Epidemiol. 2013;178(7):1177–84. doi:10.1093/aje/kwt084.
Hemani G, Zheng J, Elsworth B, Wade Kh, Haberland V, et al The MR-Base Platform Supports Systematic Causal Inference across the Human Phenome. ELife (2018) 7: 34408. doi: 10.7554/eLife.34408.
Burgess S, Bowden J, Fall T, Ingelsson E. Simon G. Thompson. Sensitivity Analyses for Robust Causal Inference from Mendelian Randomization Analyses with Multiple Genetic Variants. Epidemiology. 2017;28(1):30–42. doi:10.1097/EDE.0000000000000559.
Slob Eaw G, Pjf THurik, Ar R, Ca. A Note on the Use of Egger Regression in Mendelian Randomization Studies. Int J Epidemiol. 2017;46(6):2094–7. doi:10.1093/ije/dyx191.
Jack Bowden GD, Smith PC, Haycock, Stephen Burgess. Consistent Estimation in Mendelian Randomization with Some Invalid Instruments Using a Weighted Median Estimator. Genet Epidemiol. 2016;40(4):301–14. doi:10.1002/gepi.21965.
Hartwig FP, Smith GD, Bowden J. Robust Inference in Summary Data Mendelian Randomization via the Zero Modal Pleiotropy Assumption. Int J Epidemiol (2017) 46(6):1985–98. doi:10.1093/ije/dyx102.
Stephen Burgess SG, Thompson. Interpreting Findings from Mendelian Randomization Using the MR-Egger Method. Eur J Epidemiol. 2017;32(5):377–89. doi:10.1007/s10654-017-0255-x.
Quentin F, Gronau E-J, Wagenmakers. Limitations of Bayesian Leave-One-Out Cross-Validation for Model Selection. Comput Brain Behav (2019) 2(1): 1–11. doi:10.1007/s42113-018-0011-7.
Marie Verbanck C-Y, Chen B, Neale R. Do. Detection of Widespread Horizontal Pleiotropy in Causal Relationships Inferred from Mendelian Randomization between Complex Traits and Diseases. Nat Genet. 2018;50(5):693–98. doi:10.1038/s41588-018-0099-7.
Wang S, Zhang J, Jiang XZ, Tong B, Wang Q, et al. Relationship between waist circumference trajectory and new-onset non alcoholic fatty liver disease in the non-obese population. Zhonghua Liu Xing Bing Xue Za Zhi = Zhonghua Liuxingbingxue Zazhi. 2020;41(6):824–8. doi:10.3760/cma.j.cn112338-20190630-00479.
Lee J-H, Jeon S, Lee HS, Kwon Y-J. Cutoff Points of Waist Circumference for Predicting Incident Non-Alcoholic Fatty Liver Disease in Middle-Aged and Older Korean Adults. Nutrients. 2022;14(14):2994. doi:10.3390/nu14142994.
Park J-M, Lee HS, Oh J, Lee Y-J. Serum Testosterone Level Within Normal Range Is Positively Associated with Nonalcoholic Fatty Liver Disease in Premenopausal but Not Postmenopausal Women. Journal of Women’s Health (2002) (2019) 28(8): 1077–1082. doi: 10.1089/jwh.2018.7263.
Veeravich Jaruvongvanich A, Sanguankeo T, Riangwiwat SU. Testosterone, Sex Hormone-Binding Globulin and Nonalcoholic Fatty Liver Disease: A Systematic Review and Meta-Analysis. Ann Hepatol. 2017;16(3):382–94. doi:10.5604/16652681.1235481.
Monika Sarkar K, Yates A, Suzuki J, Lavine R, Gill, et al. Low Testosterone Is Associated With Nonalcoholic Steatohepatitis and Fibrosis Severity in Men. Clinical Gastroenterology and Hepatology: The Official Clinical Practice. J Am Gastroenterological Association. 2021;19(2):400–2. doi:10.1016/j.cgh.2019.11.053.
Sapwarobol Suwimol T, Pisit A, Anchalee K, Narisorn J, Jureeporn, et al. Neck Circumference as a Screening Measure for Identifying NAFLD among a Group of Academic Employees in Bangkok, Thailand. PLoS ONE. 2022;17(2):e0263826. doi:10.1371/journal.pone.0263826.
Yanyan Hu, Chen J, Yang Lu, Chen P, Li J, et al. The Value of Neck Circumference (NC) as a Predictor of Non-Alcoholic Fatty Liver Disease (NAFLD). J Clin Translational Endocrinol. 2014;1(4):133–39. doi:10.1016/j.jcte.2014.07.001.
Sulan Lin Y, Xian Y, Liu W, Cai J, Song, et al. Risk Factors and Community Intervention for Nonalcoholic Fatty Liver Disease in Community Residents of Urumqi, China. Medicine. 2018;97(9):e0021. doi:10.1097/MD.0000000000010021.
Lin I-Ting, Lee M-Y, Wang C-W, Wu D-W, Chen S-C. Gender Differences in the Relationships among Metabolic Syndrome and Various Obesity-Related Indices with Nonalcoholic Fatty Liver Disease in a Taiwanese Population. Int J Environ Res Public Health. 2021;18(3):857. doi:10.3390/ijerph18030857.

No competing interests reported.

supplymaterial.pdf

Associations between body circumference and testosterone levels and risk of metabolic dysfunction-associated fatty liver disease: A Mendelian randomization study

Status:

Journal Publication

Version 1

Abstract

Figures

Introduction

Materials And Methods

Results

3.1 Instrumental variables

3.3 Result of heterogeneity and sensitivity test

Discussion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1