https://orcid.org/0000-0003-1169-7467
Figures
Abstract
Background
Some studies found that red blood cell count (RBC) was an unrecognized risk factor for non-alcoholic fatty liver disease (NAFLD). While the epidemiological data underpinning the evidence is very limited. As there are some differences between the latest criteria of metabolic dysfunction-associated fatty liver disease (MAFLD) and NAFLD, itis necessary to evaluate the relationship between RBC and MAFLD.
Methods
We performed a cross-sectional analysis of the National Health and Nutritional Examination Survey (NHANES)2017-2018 cohort, including 4477 participants. Hepatic steatosis was determined when the value of controlled attenuation parameter (CAP) obtained by Fibroscan was ≥274 dB/m. Multivariate logistic regression analysis was used to estimate the association between RBC and MAFLD. We estimated the adjusted odds ratio (OR) of RBC for MAFLD, and the nonlinear relationship between RBC and MAFLD was further described using smooth curve fittings and threshold-effect analysis.
Results
We found that MAFLD risk was significantly higher according to RBC quartiles. The adjusted odds ratio (OR) and 95% confidence intervals (CIs)for the highest RBC quartile were 1.5(1.0, 2.3) for male and 1.1 (0.8, 1.6) for female, respectively. As for male, a non-linear relationship was discovered between RBCs and MAFLD, with a RBC threshold of 4.2. The effect sizes and confidence intervals on the right side of the inflection point were 1.5 (1.0, 2.0) (P for nonlinearity = 0.027). The sensitivity analysis showed a similar result.
Citation: Dai X, Zhou G, Xu L (2022) Associations between red blood cell count and metabolic dysfunction-associated fatty liver disease(MAFLD). PLoS ONE 17(12): e0279274. https://doi.org/10.1371/journal.pone.0279274
Editor: Gulali Aktas, Bolu Abant İzzet Baysal University: Bolu Abant Izzet Baysal Universitesi, TURKEY
Received: October 12, 2022; Accepted: December 3, 2022; Published: December 27, 2022
Copyright: © 2022 Dai et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the paper and its Supporting information files.
Funding: The author received no specific funding for this work.
Competing interests: The authors have declared that no competing interests exist.
Introduction
About a quarter of the global population suffered from fatty liver, which was frequently accompanied by metabolic dysfunction and there was no approved medication therapy for this disease. Experts agreed that metabolic dysfunction-associated fatty liver disease (MAFLD) can be a a better general term as patients with nonalcoholic fatty liver disease (NAFLD) are usually heterogeneous [1]. The MAFLD population shows higher liver enzyme levels as well as more lipid and glucose metabolism related diseases. As an revised definition of MAFLD can particularly identify more at risk populations in clinical practice as the metabolic syndrome dramatically raised the risk of fibrosis, cirrhosis, and hepatocellular carcinoma [2]. Metabolic disorder is key to the diagnosis of MAFLD, however, patients who consumed alcohol or had other chronic liver diseases were not excluded from the diagnosis [3]. MAFLD and NAFLD differ significantly in prevalence, risk factors, and all-cause mortality [4]. As MAFLD will take the role of NAFLD in the nomenclature transition, further research into the disease’s intricate pathophysiology is required.
Erythrocytes not only reversibly transport oxygen and carbon dioxide mainly by hemoglobin(Hb), but also contain a variety of additional proteins, with many functions in modulating oxygen transport, vascular cell communication, and inflammation being discovered [5]. Previous studies showed that RBC is associated with insulin resistance and metabolic syndromes(Mets),suggesting it might be served as a potential predictor of Mets risk [6–8]. Researchers found that individuals with obesity and insulin resistance have increased aggregation of red blood cells in their peripheral blood [9]. Imbalance of energy intake and expenditure in MAFLD population can cause insulin resistance in numerous tissues [10].
However, the relationship between RBC and MAFLD is yet to be clarified. As a result, we performed a retrospective analysis on the data from the National Health and Nutrition Examination Survey (NHANES) 2017–2018. This study aims to explore an appropriate index of hematology to predict the onset of MAFLD.
Materials and methods
Study population
The National Health and Nutrition Examination Survey (NHANES) is an ongoing repeated cross-sectional study at the Centers for Disease Control and Prevention (CDC) conducted by the National Center for Health Statistics (NCHS). The NHANES program was approved by the NCHS Ethics Review Board and all participants provided written informed consent. All NHANES data and information are publicly available at https://www.cdc.gov/nchs/nhanes/. Data from NHANES that were collected during 2017–2018 were included in this analysis. The participants provided their written informed consent to participate in this study. Ethical review and approval was not required for the study on human participants in accordance with the local legislation and institutional requirements.
We evaluated individuals over the age of 18 who had complete liver elastography measurements. Participants with incomplete test results or who were ineligible for hepatic elastography were eliminated from the study. Those with missing data on body mass index, RBC and important biochemistry were also excluded, leaving 4477 subjects for the analysis.
RBCs and other laboratory measurements
RBC, Hemoglobin (Hb) and white blood cell count (WBC) was measured on venous blood in the mobile examination centers using the Beckman Coulter DxH 800 instrument. Alanine Aminotransferase (ALT), aspartate aminotransferase (AST), γ-glutamyl transferase (GGT), cholesterol (TC), Triglycerides (TG) and uric acid (SUA) were measured using the Roche Cobas 6000 chemistry analyzer (Cobas 6000). Direct high-density lipoprotein (HDL) was processed, stored, and shipped to the University of Minnesota, Minneapolis, MN for analysis. HbA1c was measured on whole blood using Trinity Biotech Boronate Affinity HPLC. In addition, insulin resistance was calculated using the homeostasis model assessment of IR (HOMA-IR) as described by Matthews et al [11]: Fasting plasma glucose (mmol/L) *Fasting insulin (μU/mL)/22.5. HOMA-IR was divided into four categories to distinguish low, middle, high and missing data.
Measurements of other covariates
Demographic covariates included age, sex, race/ethnicity, poverty-income ratio(PIR) and smoking status which was obtained by self-report during the interview. Smoking status was categorized into current smokers, former smokers, or never smokers, determined by the participants’ answers of two questions in “Smoking-Cigarette Use” questionnaire–a) Do you now smoke cigarettes? b) Have you smoked at least 100 cigarettes in your entire life? If they didn’t smoke cigarettes now but had ever smoked at least 100 cigarettes, they were defined as former smokers. Sociodemographic characteristics were collected using a standardized self-administered questionnaire. Weight, height, and waist circumference (WC) were measured by well-trained health technologists according to the anthropometry procedure manual. Body mass index (BMI) was calculated as weight (kg) divided by height squared (m2). Blood pressure was averaged over three measurements of systolic and diastolic pressure. In addition, we defined hypertension as mean systolic blood pressure ≥140 mmHg or diastolic blood pressure ≥90 mmHg or currently taking antihypertensive medication according to self-report questionnaire. Diabetes were defined as hemoglobin A1c≥6.5%, having either diagnosed diabetes by doctor by self-report questionnaire.
Outcome definitions
In the 2017–2018 cycle, vibration-controlled transient elastography was performed by NHANES technicians after a 2-day training program with an expert technician, using the FibroScan® model 502 V2 Touch (Echosens, Paris, France) equipped with a medium (M) and extra-large (XL) probes. All participants aged 12 years and over were eligible except for those who were currently pregnant, unable to lie down, or had an implanted electronic medical device, had lesions where measurements would be taken. Only patients with complete exams (10 or more complete stiffness (E) measures, fasting time of at least 3 hours, and a liver stiffness interquartile range / median <30%) were included in the present analysis.
We defined MAFLD as the presence of metabolic risk factors in the setting of hepatic steatosis, including individuals with other concomitant liver diseases and significant alcohol consumption based on the diagnostic criteria proposed by an international expert panel. MAFLD was diagnosed as the presence of hepatic steatosis with≥1 of the following: i) overweight or obese (body mass index≥25 kg/m2), ii) diabetes mellitus, iii) at least 2 metabolic risk abnormalities. Metabolic risk abnormalities consisted of i) waist circumference≥102 cm for men and≥88 cm for women, ii) blood pressure ≥130/85mmHg or specific drug treatment, iii) fasting plasma triglycerides ≥150 mg/dl or specific drug treatment, iv) plasma HDL-cholesterol <40 mg/dl for men and <50 mg/dl for women or specific drug treatment, v) prediabetes (fasting glucose 100-125mg/dl or hemoglobin A1c 5.7%-6.4%, vi) homeostasis model assessment of insulin resistance score≥2.5, vii) plasma high-sensitivity C-reactive protein level >2 mg/L. Presence of hepatic steatosis was defined by a median Controlled Attenuation Parameter (CAP) ≥ 274 dB/m, a cut-off that yielded 90% sensitivity in distinguishing S0 from S1-S3 in a recent study by Eddowes et al [12].
Statistic analysis
The variables were presented in two forms. Continuous variables with normal distribution were expressed as mean± standard deviation. Categorical variables are expressed as numbers and weighted proportions. One-way analysis of variance (ANOVA) was used to assess continuous variables, while the chi-squared test was used to compare categorical variables.
Because RBCs levels were significantly greater in males than in females, sex-specific RBC quartiles were generated, and all data were evaluated separately in males and females. The odds ratios (ORs) and 95% confidence intervals (CIs) of the connection between RBCs and the probability of MAFLD were calculated using logistic regression models. RBCs were divided into quartiles and the ORs with 95%CI were calculated across the quartiles. Potential confounding factors were gradually included in three models with increasing degrees of adjustment. We selected these cofounders based on their associations with the outcomes of interest or a change in effect estimate of more than 10%. Model1 was adjusted with age, race and BMI. Model 2 was fully-adjusted with age, race, BMI, diabetes mellitus, hypertension, HbA1c, HDL, SUA, TG, Hb, WBC, HOMA-IR and Smoking status. In addition, the threshold effect of RBC on MAFLD risk was explored using piecewise linear regression according to the smoothing plot.
Additionally, we ran two sensitivity analysis. Firstly, We conducted the modeling by taking the potential cofounder of ALT, SBP into account. To fill in missing data, we used multiple imputations of missing data. Secondly, we divided laboratory data into tertiles and made another logistic regression model.
All analyses were conducted using R version 3.4.3 (http://www.R-project.org) and EmpowerStats software (http://www.empowerstat.com) of NHANES.
Results
Clinical characteristics of study participants
The final analysis comprised 4477 patients in total, 2282 females and 2195males. Mean RBCs was higher in males(5.0± 0.5) *106/μL than females (4.5±0.4) *106/μL. The prevalence of MAFLD was 48.8% in males and 38.9% in females. Participant characteristics are shown in Tables 1 and 2 based on the RBC count quartiles for each gender. Participants with greater RBC quantiles had higher BMI, waist circumstance, TC, TG, ALT, and controlled attenuated parameter values than Q1 participants, and had lower HDL in both genders. All individuals were divided into quartiles based on their RBC levels, which were stratified by sex, to examine the relationship between RBC count and the risk of MAFLD. According to the RBC quartiles of the participants, the percentage of MAFLD gradually increased for both genders: 31.6%, 34.0%, 41.3%, 49.0% for females, and 43.2%, 44.7%, 49.7%, 57.2% for males in Q1, Q2, Q3, and Q4, respectively.
Association between RBCs and MAFLD
Binary logistic regression analysis was applied to further delineate the relationship between RBCs and MAFLD. As shown in Table 3, in the female population, we found that although RBCs were positively associated with MAFLD in non-adjusted, model1 and 2, the result was not significant. When RBCs were assessed in quartiles, although there is a remarkable rising trend in non-adjusted and model1(P for trend<0.001), the trend disappeared in model 2. The result was quite different in males. Although in the multivariate regression models, for every one unit increase in the RBCs, the risk of MAFLD was not significantly associated with RBCs (OR = 1.2,95%CI = 0.9,1.6 P = 0.10). But when RBC was assessed in quantiles, in the fully adjusted model, the presence of MAFLD showed a rising trend as the RBC quantiles increased (P for trend = 0.046). The adjusted ORs(95%CI) for the highest quartile(Q4) was 1.5(1.0, 2.3) in male.
Non-linear relationship between RBCs and MAFLD in males
To verify the linear trend between RBCs and MAFLD, we established GAM to fit the association between RBCs as a continuous variable and MAFLD by the cubic spline smoothing technique. The linear trend was observed in the female population, but the correlation was not significant as shown in S2 Fig. The smooth curve through the analysis of GAM showed that RBCs had a non-linear relationship with the occurrence of MAFLD (adjusted for age, race, BMI, HbA1c, HDL, SUA, TG, Hb, WBC and Smoking status) in the male population (Fig 1). We compared the linear regression model and two-stage linear regression to analyze the relationship between them. The P value for the log-likelihood ratio was less than 0.05. The occurrence of MAFLD increased with an increase in RBCs when RBCs > 4.2 (OR = 1.5 95%CI = 1.0, 2.0 P = 0.025) while no association was found when RBCs were lower than 4.2 as shown in Table 4.
The solid red line represents the smooth curve fit between variables. Blue bands represent the 95% confidence interval from the fit. Models were adjusted for age, race, BMI, diabetes mellitus, hypertension, HbA1c, HDL, SUA, TG, Hb, WBC, HOMA-IR, Smoking status.
Sensitivity analysis
We made two sensitivity analysis. Firstly, we further included ALT and SBP into consideration based on Model2. Imputed data analysis was conducted and the result was shown in S3 Table. Although the upward trend disappeared (P for trend0.07), the adjusted ORs(95%CI) for the highest quartile(Q4) was 1.5(1.0,2,2) P<0.05. Secondly, multivariant logistic regression was conducted when HbA1c, HDL, SUA, TG, Hb, WBC and HOMA-IR were fitted in tertile categories. The covariate screening method was the same as above and the category boundaries were described in the footnote of S4 Table. The result was similar as above. The adjusted ORs(95%CI) for the highest quartile(Q4) was 1.5(1.0, 2.2) in male and P for trend was 0.043.
Discussion
Multivariate analysis result showed that RBC was associated with a greater risk of MAFLD, an association independent of age, race, BMI, and other relevant factors. When RBC were divided into quartiles, the adjusted odds ratio (OR) and 95% confidence intervals (Cis)for the highest RBC quartile was 1.5(1.0,2.3) for male and 1.1 (0.8, 1.6) for female respectively while the upward trend was not significant for female. As for male, a non-linear relationship was discovered between RBC and MAFLD, with a RBC threshold of 4.2*106/μL.
A meta-analysis found that MAFLD prevalence was 50.7% among overweight or obese adults in general population [13]. Another meta-analysis reported the prevalence of 5.37% (95% CI 4.36% to 6.59%) and 29.78% (95% CI 26.06% to 33.79%) of lean and nonobese individuals, respectively [14]. MAFLD criteria is more superior over NAFLD for identifying patients under high risk of hepatic fibrosis, atherosclerotic cardiovascular disease and chronic kidney disease. In patients with MAFLD, even mild alcohol consumption may cause worsening of hepatic fibrosis measures indicators [15–17]. As a result, MAFLD was supported by over 1000 signatories representative of multiple stakeholders as it more accurately reflects the underlying pathogenesis of the disease than NAFLD [18].
Our results have some similarities with previous studies. According to Wang et al, RBC is independently linked to a greater risk of developing fatty liver [19]. However, other potential factors including hemoglobin and uric acid were not adequately adjusted in their study. Mardi suggested that increased erythropoiesis and subclinical inflammation could be part of the Mets [20]. Using a longitudinal cohort design, Zhong et al. discovered that RBC is not only connected with the incidence of NAFLD but also the severity of incident NAFLD development in Chinese population [21]. Kotani confirmed that red blood cell count is an indicator related to Mets and its components, contrary to Kim’s study on Koreans with no relationship discovered between them [22,23]. In this study, RBC was significantly associated with MAFLD in males, with no such trendin females. This gender-specific result is somehow similar to previous findings that Hb concentration is associated with NAFLD and MS in men [24,25]. The relationship between RBC and MAFLD remains unclear. In our analysis, specific information on women’s health, such as menopause status or the use of hormone replacement therapy, was not assessed. And this may help explain these different findings. It has been demonstrated that low-dose estradiol-based hormone can lower liver enzyme levels and may be protective against NAFLD [26,27].
One of the key mechanisms in the emergence of MAFLD is insulin resistance. Insulin resistance is the common pathophysiological basis of metabolic dysfunctions including obesity, type 2 diabetes mellitus, hypertension, dyslipidemia, and metabolic syndrome [10]. Barbieri et al. provided in vivo evidence of a relationship between insulin resistance and red blood cell count [28]. Actually, human insulin has a synergistic effect with erythropoietin on stimulating the proliferation of the erythroid colonies [29,30]. Several studies also revealed that insulin resistance is associated with an enhanced degree of erythrocyte aggregation [9,31]. Secondly, As a a powerful sphingolipid mediator, sphingosine 1-phosphate (S1P)controls a variety of physiological responses, including cell migration, differentiation, apoptosis, lymphocyte trafficking, and inflammation [32]. Persuasive research showed that erythrocytes are the primary source of plasma S1P,which has been demonstrated as an intracellular biolipid that responds to tissue hypoxia by boosting erythrocyte glycolysis and oxygen delivery [33]. While NAFLD can create hypoxia microenvironment in which lipids accumulation in the blood vessels and the liver leads to a limited blood supply to hepatocytes, indicating that S1P may play a role in it [34]. S1P has also been shown to have a relationship with insulin resistance, hyperlipidemia and inflammation [32]. It was reported that SphK1/S1P/S1P2R pathway activation can inhibit the feedback loop of insulin secretion and sensitivity [35]. Experimental animal research and human investigations both found a positive correlation between plasma S1P and LDL cholesterol, demonstrating the significant function of S1P in the setting of hyperlipidemia [36]. Moreover, inflammation has a vital role in the progression of fatty liver disease. According to several reports, S1P activates S1PR2 to prevent macrophages from migrating to the area of inflammation [33]. Importantly, it has been discovered that NAFLD has a modified lipidomic profile, implying that modifications in the quantity and caliber of lipids in the erythrocyte membrane may have an impact on the pathogenesis and development of steatosis or steatohepatitis [37]. Furthermore, arelationship between membrane fatty acid compositions and insulin resistance was also discovered [38].
Based on the new concept of MAFLD, we explored for the first time the relationship between RBC and MAFLD. However, there still existed some limitations in this study. Firstly, although we adjusted for several important covariates, other potential factors such as physical activities, drug used and incompleteness of insulin resistance which was related to MAFLD may introduce bias. Additionally, w sex hormones were not included as covariates, probably leading to bias between sexes. Secondly, the diagnosis of hepatic steatosis was based on liver ultrasound transient elastography and there were some debates concerning the cut-off value of median Controlled Attenuation Parameter for diagnosing MAFLD, hence the diagnostic gold standard remains liver biopsy. Thirdly, it was challenging to determine a cut-off value for predicting MAFLD due to the relatively minor differences between RBC. Therefore, additional prospective research is still required to elucidate the detailed relationship between RBC and MAFLD.
Conclusions
We demonstrated that elevated RBC is associated with the higher risk of MAFLD in males. This positive relationship was not significant in females after full adjustment. This finding provided novel evidence indicating that red blood cell count might be a potential biomarker for MAFLD in males.
Supporting information
S2 Fig. Association between RBCs and occurrence of MAFLD in female.
https://doi.org/10.1371/journal.pone.0279274.s002
(TIF)
S1 Table. The collinearity screening of baseline characteristics.
https://doi.org/10.1371/journal.pone.0279274.s003
(DOCX)
S2 Table. The adjusting roles of potential confounders on the estimates of RBCs on MAFLD risk.
https://doi.org/10.1371/journal.pone.0279274.s004
(DOCX)
S3 Table. Association between RBC count and MAFLD status.
https://doi.org/10.1371/journal.pone.0279274.s005
(DOCX)
S4 Table. Association between RBC count and MAFLD status.
https://doi.org/10.1371/journal.pone.0279274.s006
(DOCX)
References
- 1. Eslam M, Sanyal AJ, George J. MAFLD: A Consensus-Driven Proposed Nomenclature for Metabolic Associated Fatty Liver Disease. Gastroenterology. 2020;158(7):1999–2014.e1. Epub 2020/02/12. pmid:32044314.
- 2. Lin S, Huang J, Wang M, Kumar R, Liu Y, Liu S, et al. Comparison of MAFLD and NAFLD diagnostic criteria in real world. Liver international: official journal of the International Association for the Study of the Liver. 2020;40(9):2082–9. Epub 2020/06/02. pmid:32478487.
- 3. Eslam M, Newsome PN, Sarin SK, Anstee QM, Targher G, Romero-Gomez M, et al. A new definition for metabolic dysfunction-associated fatty liver disease: An international expert consensus statement. Journal of hepatology. 2020;73(1):202–9. Epub 2020/04/12. pmid:32278004.
- 4. Lim GEH, Tang A, Ng CH, Chin YH, Lim WH, Tan DJH, et al. An Observational Data Meta-analysis on the Differences in Prevalence and Risk Factors Between MAFLD vs NAFLD. Clinical gastroenterology and hepatology: the official clinical practice journal of the American Gastroenterological Association. 2021. Epub 2021/12/07. pmid:34871813.
- 5. Crawford JH, Chacko BK, Kevil CG, Patel RP. The red blood cell and vascular function in health and disease. Antioxidants & redox signaling. 2004;6(6):992–9. Epub 2004/11/19. pmid:15548896.
- 6. Lin JD, Chiou WK, Chang HY, Liu FH, Weng HF, Liu TH. Association of hematological factors with components of the metabolic syndrome in older and younger adults. Aging clinical and experimental research. 2006;18(6):477–84. Epub 2007/01/27. pmid:17255636.
- 7. Kawamoto R, Tabara Y, Kohara K, Miki T, Kusunoki T, Abe M, et al. Hematological parameters are associated with metabolic syndrome in Japanese community-dwelling persons. Endocrine. 2013;43(2):334–41. Epub 2013/01/12. pmid:23307027.
- 8. Wu S, Lin H, Zhang C, Zhang Q, Zhang D, Zhang Y, et al. Association between erythrocyte parameters and metabolic syndrome in urban Han Chinese: a longitudinal cohort study. BMC public health. 2013;13:989. Epub 2013/10/23. pmid:24144016.
- 9. Justo D, Marilus R, Mardi T, Tolchinsky T, Goldin Y, Rozenblat M, et al. The appearance of aggregated erythrocytes in the peripheral blood of individuals with insulin resistance. Diabetes/Metabolism Research and Reviews. 2003;19(5):386–91. pmid:12951646
- 10. Sakurai Y, Kubota N, Yamauchi T, Kadowaki T. Role of Insulin Resistance in MAFLD. Int J Mol Sci. 2021;22(8). Epub 2021/05/01. pmid:33923817.
- 11. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia. 1985;28(7):412–9. Epub 1985/07/01. pmid:3899825.
- 12. Eddowes PJ, Sasso M, Allison M, Tsochatzis E, Anstee QM, Sheridan D, et al. Accuracy of FibroScan Controlled Attenuation Parameter and Liver Stiffness Measurement in Assessing Steatosis and Fibrosis in Patients With Nonalcoholic Fatty Liver Disease. Gastroenterology. 2019;156(6):1717–30. Epub 2019/01/29. pmid:30689971.
- 13. Liu J, Ayada I, Zhang X, Wang L, Li Y, Wen T, et al. Estimating Global Prevalence of Metabolic Dysfunction-Associated Fatty Liver Disease in Overweight or Obese Adults. Clinical gastroenterology and hepatology: the official clinical practice journal of the American Gastroenterological Association. 2022;20(3):e573–e82. Epub 2021/02/23. pmid:33618024.
- 14. Chan KE, Koh TJL, Tang ASP, Quek J, Yong JN, Tay P, et al. Global Prevalence and Clinical Characteristics of Metabolic-associated Fatty Liver Disease: A Meta-Analysis and Systematic Review of 10 739 607 Individuals. The Journal of clinical endocrinology and metabolism. 2022;107(9):2691–700. Epub 2022/05/20. pmid:35587339.
- 15. Yamamura S, Eslam M, Kawaguchi T, Tsutsumi T, Nakano D, Yoshinaga S, et al. MAFLD identifies patients with significant hepatic fibrosis better than NAFLD. Liver international: official journal of the International Association for the Study of the Liver. 2020;40(12):3018–30. Epub 2020/10/01. pmid:32997882.
- 16. Sun DQ, Jin Y, Wang TY, Zheng KI, Rios RS, Zhang HY, et al. MAFLD and risk of CKD. Metabolism. 2021;115:154433. Epub 2020/11/20. pmid:33212070.
- 17. Tsutsumi T, Eslam M, Kawaguchi T, Yamamura S, Kawaguchi A, Nakano D, et al. MAFLD better predicts the progression of atherosclerotic cardiovascular risk than NAFLD: Generalized estimating equation approach. Hepatol Res. 2021;51(11):1115–28. Epub 2021/06/16. pmid:34129272.
- 18. Méndez-Sánchez N, Bugianesi E, Gish RG, Lammert F, Tilg H, Nguyen MH, et al. Global multi-stakeholder endorsement of the MAFLD definition. The lancet Gastroenterology & hepatology. 2022;7(5):388–90. Epub 2022/03/07. pmid:35248211.
- 19. Wang HL, Zhang H, Wu SL, Liao GC, Fang AP, Zhu MF, et al. Red blood cell count has an independent contribution to the prediction of ultrasonography-diagnosed fatty liver disease. PLoS One. 2017;12(2):e0172027. Epub 2017/02/12. pmid:28187211.
- 20. Mardi T, Toker S, Melamed S, Shirom A, Zeltser D, Shapira I, et al. Increased erythropoiesis and subclinical inflammation as part of the metabolic syndrome. Diabetes research and clinical practice. 2005;69(3):249–55. Epub 2005/08/16. pmid:16098921.
- 21. Zhong F, Guan L, Lin H, Zhao M, Qin Y, Li Q, et al. Red Blood Cell Count: An Unrecognized Risk Factor for Nonalcoholic Fatty Liver Disease. Front Endocrinol (Lausanne). 2021;12:760981. Epub 2021/12/25. pmid:34950103.
- 22. Kotani K, Sakane N, Kurozawa Y. Increased red blood cells in patients with metabolic syndrome. Endocrine journal. 2006;53(5):711–2. Epub 2006/09/02. pmid:16946564.
- 23. Kim JA, Choi YS, Hong JI, Kim SH, Jung HH, Kim SM. Association of metabolic syndrome with white blood cell subtype and red blood cells. Endocrine journal. 2006;53(1):133–9. Epub 2006/03/18. pmid:16543683.
- 24. Hashimoto Y, Tanaka M, Kimura T, Kitagawa N, Hamaguchi M, Asano M, et al. Hemoglobin concentration and incident metabolic syndrome: a population-based large-scale cohort study. Endocrine. 2015;50(2):390–6. Epub 2015/04/13. pmid:25863486.
- 25. Chung GE, Yim JY, Kim D, Kwak MS, Yang JI, Chung SJ, et al. Associations between hemoglobin concentrations and the development of incidental metabolic syndrome or nonalcoholic fatty liver disease. Digestive and liver disease: official journal of the Italian Society of Gastroenterology and the Italian Association for the Study of the Liver. 2017;49(1):57–62. Epub 2016/11/05. pmid:27810399.
- 26. Nemoto Y, Toda K, Ono M, Fujikawa-Adachi K, Saibara T, Onishi S, et al. Altered expression of fatty acid-metabolizing enzymes in aromatase-deficient mice. The Journal of clinical investigation. 2000;105(12):1819–25. Epub 2000/06/23. pmid:10862797.
- 27. McKenzie J, Fisher BM, Jaap AJ, Stanley A, Paterson K, Sattar N. Effects of HRT on liver enzyme levels in women with type 2 diabetes: a randomized placebo-controlled trial. Clinical endocrinology. 2006;65(1):40–4. Epub 2006/07/05. pmid:16817817.
- 28. Barbieri M, Ragno E, Benvenuti E, Zito GA, Corsi A, Ferrucci L, et al. New aspects of the insulin resistance syndrome: impact on haematological parameters. Diabetologia. 2001;44(10):1232–7. Epub 2001/11/03. pmid:11692171.
- 29. Aoki I, Taniyama M, Toyama K, Homori M, Ishikawa K. Stimulatory effect of human insulin on erythroid progenitors (CFU-E and BFU-E) in human CD34+ separated bone marrow cells and the relationship between insulin and erythropoietin. Stem cells (Dayton, Ohio). 1994;12(3):329–38. Epub 1994/05/01. pmid:7521243.
- 30. Miyagawa S, Kobayashi M, Konishi N, Sato T, Ueda K. Insulin and insulin-like growth factor I support the proliferation of erythroid progenitor cells in bone marrow through the sharing of receptors. British journal of haematology. 2000;109(3):555–62. Epub 2000/07/11. pmid:10886204.
- 31. Gyawali P, Richards RS, Hughes DL, Tinley P. Erythrocyte aggregation and metabolic syndrome. Clinical hemorheology and microcirculation. 2014;57(1):73–83. Epub 2013/11/07. pmid:24192695.
- 32. Chen W, Lu H, Yang J, Xiang H, Peng H. Sphingosine 1-phosphate in metabolic syndrome (Review). International journal of molecular medicine. 2016;38(4):1030–8. Epub 2016/09/08. pmid:27600830.
- 33. Tsai HC, Han MH. Sphingosine-1-Phosphate (S1P) and S1P Signaling Pathway: Therapeutic Targets in Autoimmunity and Inflammation. Drugs. 2016;76(11):1067–79. Epub 2016/06/20. pmid:27318702.
- 34. Chen J, Chen J, Fu H, Li Y, Wang L, Luo S, et al. Hypoxia exacerbates nonalcoholic fatty liver disease via the HIF-2α/PPARα pathway. American journal of physiology Endocrinology and metabolism. 2019;317(4):E710–e22. Epub 2019/08/21. pmid:31430204.
- 35. Fayyaz S, Henkel J, Japtok L, Krämer S, Damm G, Seehofer D, et al. Involvement of sphingosine 1-phosphate in palmitate-induced insulin resistance of hepatocytes via the S1P2 receptor subtype. Diabetologia. 2014;57(2):373–82. Epub 2013/12/03. pmid:24292566.
- 36. Tong X, Peng H, Liu D, Ji L, Niu C, Ren J, et al. High-density lipoprotein of patients with type 2 diabetes mellitus upregulates cyclooxgenase-2 expression and prostacyclin I-2 release in endothelial cells: relationship with HDL-associated sphingosine-1-phosphate. Cardiovascular diabetology. 2013;12:27. Epub 2013/01/31. pmid:23360427.
- 37. Notarnicola M, Caruso MG, Tutino V, Bonfiglio C, Cozzolongo R, Giannuzzi V, et al. Significant decrease of saturation index in erythrocytes membrane from subjects with non-alcoholic fatty liver disease (NAFLD). Lipids in health and disease. 2017;16(1):160. Epub 2017/08/24. pmid:28830448.
- 38. Bigornia SJ, Lichtenstein AH, Harris WS, Tucker KL. Associations of erythrocyte fatty acid patterns with insulin resistance. The American journal of clinical nutrition. 2016;103(3):902–9. Epub 2016/02/13. pmid:26864364.