Association of Systemic Immune-Inflammation Index with Non-Alcoholic Fatty Liver Disease: A Population-Based Cross-Sectional Study

Introduction

Non-alcoholic fatty liver disease (NAFLD) is a clinical syndrome characterized by steatosis of the liver cells and excessive accumulation of lipids without any previous history of excessive alcohol consumption.¹ The incidence of NAFLD continues to increase year by year due to changes in lifestyle, diet structure, reduced physical activity, and an accelerated pace of life, which has become a global public health concern.² The exact incidence of NAFLD is unknown. The prevalence of NAFLD in the general population in each country is 10–24%, which is significantly higher in women than in men.³ Among them, the prevalence of NAFLD in the United States (US) is estimated at 16%–23%.⁴ The pathogenesis of NAFLD is very complex and involves intracellular biochemical metabolism.⁵ More and more researchers believe that NAFLD is a manifestation of metabolic syndrome in the liver. It may be related to islet resistance and fat oxidation disorders.⁶ When left untreated, this disease may develop into liver cancer, which is difficult to treat and may even lead to death.⁷ As a result, we should pay great attention to the early detection and treatment of this disease.

NAFLD is associated with an increased burden of inflammation.^8,9 Moreover, it is related to metabolic disturbance.¹⁰ Hu et al first developed the systemic immune-inflammation index (SII) in 2014, which integrated three types of inflammatory cells, including platelets, neutrophils, and lymphocytes, and was calculated by platelet count × neutrophil count lymphocyte count.¹¹ The SII index is dependent on various cell counts in the hemogram, such as neutrophils, lymphocytes, and platelets. The markers based on these cell counts were also reported to be associated with inflammatory conditions such as thyroid conditions,¹² gastrointestinal diseases,¹³ thyroiditis,¹⁴ diabetes mellitus,¹⁵ irritable bowel disease,¹⁶ and COVID-19 infection.¹⁷ On the other hand, the SII index is considered a good and stable index that reflects both local immune responses and systemic inflammation in the body as a whole.^18,19 It has been identified to predict outcomes in patients with multiple cancers, heart failure, acute ischemic stroke, and acute kidney injury.^20–24 However, the effect of the SII index on NAFLD has not been fully elucidated. In addition, epidemiological research has not been able to determine whether the SII index is associated with the ZJU index, BARD score, NAFLD fibrosis score, or the risk of NAFLD in the general US population. Considering the detrimental effects of NAFLD, recognizing risk factors and devising measures to avoid or control the consequences as soon as possible are highly beneficial. As a result, we examined the association between the SII index and the prevalence of NAFLD by analyzing data from the Nutrition and Health Examination Survey (NHANES) for the 2017–2018 years.

Materials and Methods

Study Population

The NHANES database is a population-based cross-sectional survey designed to gather information about the health and nutrition of representative American households. It combines demographics, dietary, examination, laboratory, questionnaire, and limited access data. For data users and researchers throughout the world, survey data are available on the internet.²⁵ The NHANES data for the present study from 2017 to 2018 were used and analyzed. Among the 8897 participants in the total sample, we excluded participants with insufficient NAFLD data (n = 2949) and SII index (n = 257), respectively. Moreover, excluding participants who did not have data on the ZJU index, BARD score, or NAFLD fibrosis score (n = 3069) and participants with missing covariate data (n = 1425). Finally, a total of 1197 individuals were included in this research (Figure 1). The National Center for Health Statistics Ethical Review Board approved all protocols, and each participant provided written informed consent.²⁶ Detailed study design proposals are publicly available online (https://www.cdc.gov/nchs/nhanes/).

Figure 1 Study flow chart. Abbreviations: NHANES, National Health and Nutrition Examination Surveys; NAFLD, non-alcoholic fatty liver disease; SII index, systemic immune-inflammation index.

Covariates

In the study, the covariates were as follows: age, sex (male, and female), race/ethnicity (Mexican American, Other Hispanic, Non-Hispanic White, Non-Hispanic Black, and Other Race), family poverty income ratio (PIR), education level (less than high school, high school, and more than high school), marital status (having a partner, no partner, unmarried), the complication of hypertension, and diabetes mellitus (DM), coronary heart disease (CHD), congestive heart failure (CHF), angina pectoris, heart attack, and stroke, smoker (no, former, now), drinker (never, mild, moderate, heavy), body mass index (BMI), waist circumference, fast glucose (FBG), glycosylated hemoglobin (HbA1c), hemoglobin (Hb), high-sensitivity C-reactive protein (hs CRP), alanine aminotransferase (ALT), aspartate amino transferase (AST), gamma-glutamyl transpeptidase (GGT), blood urea nitrogen (BUN), uric acid (UA), serum creatinine (Scr), estimated glomerular filtration rate (eGFR), high-density lipoprotein-cholesterol (HDL-C), total cholesterol (TC), triglyceride (TG), ZJU index, BARD score, and NAFLD fibrosis score.^27–29 Participants self-reported data regarding their age, sex, race or ethnicity, educational level, marital status, smoking, and drinking habits during the home interview. These questions about the complications of CHD, CHF, angina pectoris, heart attack, and stroke were described in the NHANES dataset as 66 MCQ160b-e. In addition, data on FBG, HbA1c, Hb, hs CRP, ALT, AST, GGT, BUN, UA, Scr, eGFR, HDL-C, TC, and TG were obtained from laboratory examination. You can find more information about the variables in this study here https://www.cdc.gov/nchs/nhanes/.

Calculation of the SII Index, ZJU Index, BARD Score, and NAFLD Fibrosis Score

The blood samples were collected from fasting participants in the study. The automated hematology analyzing devices (Coulter^® DxH 800 analyzer) was used to measure blood count (neutrophil, lymphocyte, and platelet counts). In this study, we calculated SII index for each participant as follows: SII index (×10⁹/L) = neutrophil count (×10⁹/L)/lymphocyte count (×10⁹/L) × platelet count (×10⁹/L).^18,30 In addition, the ZJU index formula is expressed as follows: BMI (kg/m²) + FBG (mmol/L) + TG (mmol/L) + 3×ALT (IU/L)/AST (IU/L) ratio (+2, if female).²⁷ The BARD score is calculated by weighted sum of three variables (BMI >28 = 1 point, AAR of >0.8 = 2 points, DM = 1 point).²⁸ Finally, the NAFLD fibrosis score formula is expressed as follows: -1.675 + 0.037 × age (years) + 0.094 × BMI (kg/m²) + 1.13 × IFG/diabetes (yes=1, no=0) + 0.99 × AST (IU/L)/ALT (IU/L) ratio – 0.013 × platelet count (×10⁹/L) – 0.66 × albumin (g/dl).²⁹

NAFLD Measurement

NAFLD was defined using the US fatty liver index (FLI), a well-validated diagnostic index,³¹ which was employed utilizing NHANES III data and calculated as an equation according to a previous study^32,33 that included information on BMI, GGT, TG, and waist circumference. All the information was collected concurrently with the status of iron metabolism. NAFLD was defined as an FLI score of ≥60. The FLI formula is expressed as follows:³⁴

Statistical Analysis

The weighted NHANES sample was used to calculate all estimates. The sample size caused by the missing covariate was deleted in this study. All statistical analyses were calculated using R version 3.6.4 (R Foundation for Statistical Computing, Vienna, Austria) and SPSS version 24.0 (SPSS Inc., Chicago, IL, USA). A P-value < 0.05 was considered statistically significant. Continuous variables were reported as mean ± standard deviation (SD), while categorical variables were presented as numbers (%). Continuous variables were analyzed by the weighted Student’s t-test or one-way ANOVA, and categorical variables were analyzed by the weighted chi-square tests. The SII index was divided into quartiles (Q1: 50.000–292.000, Q2: 292.00–423.059, Q3: 423.060–602.735, and Q4: 602.736–3250.714), and the lowest quartile (Q1) served as the reference group (Q1). Firstly, we performed multivariable logistic regression models to explore the relationship between the independent variable SII index and the risk of NAFLD. Model 1 was adjusted for age and sex. Model 2 was adjusted for model 1 variables plus race/ethnicity, education level, marital status, family PIR, smoke status, drink status, the complication of hypertension, and DM. Model 3 was adjusted for model 2 variables plus the complication of CHD, CHF, angina pectoris, heart attack, and stroke, BMI, waist circumference, FBG, HbA1c, Hb, hs CRP, ALT, AST, GGT, UA, BUN, Scr, eGFR, TG, and HDL-C. Then, after adjusting all the covariates of Model 3 above, restricted cubic spline models (RCS) were analyzed to assess the association between SII index and risk of NAFLD, and generalized additive models and smooth curve fitting were used to explore SII index, ZJU index, BARD score, and NAFLD fibrosis score. Finally, subgroup analyses were stratified by age, sex, hypertension, DM, and BMI to evaluate the association between the SII index and the risk of NAFLD.

Results

Baseline Characteristics

The basic clinical characteristics and laboratory examinations are shown in Table 1. The 1197 subjects were divided into four groups (Q1, Q2, Q3, and Q4) based on their SII index levels. We computed that the number of participants in this research may be representative of the total population of 64,268,398 in the United States. The prevalence of NAFLD in this study was 28.1%. There were significant differences in sex, smoker, the complication of CHD, BMI, waist, hs CRP, Scr, ZJU score, BARD score, and NAFLD fibrosis score among the different uric acid groups.

Table 1 Characteristics of the Study Population Based on SII Index Quartiles

Association Between SII Index and ZJU Index, BARD Score, and NAFLD Fibrosis Score

We conducted generalized additive models with smooth functions to assess the association between the SII index and the ZJU index, the BARD score, and the NAFLD fibrosis score. The SII index was linearly positive with the ZJU index but negative with the NAFLD fibrosis score (Figure 2A and C). However, there is first a decrease in correlations between the SII index and BARD score, then an increase, and finally another decrease (Figure 2B).

Figure 2 Associations of SII index with ZJU index, BARD score, and NAFLD fibrosis score. (A) Association between SII index and ZJU index. (B) Association between SII index and BARD score. (C) Association between SII index and NAFLD fibrosis score. Abbreviations: SII index, systemic immune-inflammation index; NAFLD, non-alcoholic fatty liver disease.

Association Between SII Index and NAFLD

In the restricted cubic spline (RCS) plot, we can see the SII index is associated with a U-shaped association with the prevalence of NAFLD (P for nonlinearity <0.05, Figure 3). As the SII index increased, the risk of NAFLD decreased significantly. When the SII index reached 422.40, the risk of NAFLD was at its lowest, and then the curve showed an upward trend. Three multivariate logistic regression models (Model 1, Model 2, and Model 3) were constructed to investigate the relationship between the SII index and the prevalence of NAFLD (Table 2). After adjusting for interfering factors, compared with the lowest quartiles (Q1), the odds ratios (ORs) with 95% confidence intervals (CIs) for NAFLD across the quartiles were 0.923 (0.585, 1.455), 0.563 (0.351, 0.901), and 1.061 (0.669, 1.682).

Table 2 Adjusted ORs for Associations Between SII Index and the Prevalence of NAFLD

Figure 3 Restricted cubic spline curve for the relationship between SII index with the prevalence of NAFLD. Abbreviations: SII index, systemic immune-inflammation index; NAFLD, non-alcoholic fatty liver disease; OR, odd ratio; CI, confidence interval.

Subgroup Analyses

We performed subgroup analyses stratified by age, sex, hypertension, DM, and BMI, to determine the link between SII index and risk of NAFLD are shown in Table 3, and Figure 4. The stratified analysis revealed the U-shaped associations of SII index with NAFLD were found among participants in all age groups, male or female, with or without hypertension, with DM, and with BMI of <30 or BMI of > 30 kg/m² (Figure 4A–E). We also observed that SII index positively associated with risk of NAFLD in participants without DM (Figure 4D). The test for interactions was not statistically significant for age, sex, hypertension, DM, and BMI (all P for interactions >0.05, Table 3).

Table 3 Subgroup Analysis for Associations Between SII Index and the Prevalence of NAFLD

Figure 4 Restricted cubic spline curve for the relationship between SII index with the prevalence of NAFLD. (A) The association between SII index and NAFLD stratified by age; (B) The association between SII index and NAFLD stratified by sex; (C) The association between SII index and NAFLD stratified by hypertension; (D) The association between SII index and NAFLD stratified by DM; (E) The association between SII index and NAFLD stratified by BMI. Abbreviations: SII index, systemic immune-inflammation index; NAFLD, non-alcoholic fatty liver disease; DM, diabetes mellitus; BMI, body mass index.

Discussion

NAFLD is the most common cause of elevated liver enzyme levels in U.S. adults and the most common cause of cryptogenic cirrhosis.³⁵ NAFLD has attracted increasing attention and research because it can progress to cirrhosis and even liver cancer.³⁶ Inflammation is a feature of non-alcoholic fatty liver disease progression and plays an important role in hepatic steatosis and fibrosis.³⁷ In addition, the pathogenesis and disease progression of NAFLD are closely related to the activation of innate immunity.³⁸ However, its pathogenesis is still not fully understood.

In the study, firstly, we found that the SII index was linearly positive with the ZJU index but negative with the NAFLD fibrosis score. And the SII index and BARD score showed a trend of first decreasing, then increasing, and then decreasing. A correlation between the SII index and the ZJU index, the BARD score, and the NAFLD fibrosis score has not been studied to date. Secondly, we revealed that the SII index is associated with a U-shaped association with the prevalence of NAFLD in the RCS plot. Wang et al found that genes associated with immune infiltration may serve as potential markers for therapeutic targets for NAFLD.³⁹ Xie et al. Revealed that a high SII index is associated with hepatic steatosis but not with liver fibrosis.⁴⁰ In addition, Song Y and his team also found that U.S. adults with a high SII index had an increased risk of hepatic steatosis.⁴¹ However, the findings of Ioannou GN showed that the presence and severity of hepatic steatosis were associated with increased pan-immune inflammation value levels but not with the SII index in obese children and adolescents.³⁶ Additionally, the acute phase of inflammation plays a significant role in liver graft injury. Hong BJ and his team reveal that hepatic transplant rejection is attenuated by inhibiting the inflammasome activation pathway.⁴² In summary, this is consistent with the conclusion of this study. Reasonable control of inflammation in vivo can effectively reduce the occurrence of NAFLD. Thirdly, the stratified analysis showed that the U-shaped associations of the SII index with NAFLD were found among participants in all age groups, male or female, with or without hypertension, with DM, and with a BMI of <30 or > 30 kg/m². Additionally, we also observed that the SII index was positively associated with the risk of NAFLD in subjects without DM. Type 2 diabetes, obesity, and hyperlipidemia are considered to be important risk factors for NAFLD.⁴³ Between 21% and 45% of patients with NAFLD have type 2 diabetes.⁴⁴ Among them, patients with NAFLD combined with type 2 diabetes are more likely to progress to cirrhosis, have an increased risk of cardiovascular disease and kidney disease, and have a higher mortality rate.^45,46 In addition, the prevalence of NAFLD is 4–6 times higher in obese patients than in those of normal weight.⁴⁷ Shi et al also found that obese children have low-grade chronic inflammation.⁴⁸ In contrast, patients with type 2 diabetes are significantly more likely to develop NAFLD, regardless of their BMI.⁴⁹ Among hyperlipidemic patients, those with hypertriglyceridemia are at greater risk of NAFLD than those with hypercholesterolemia.^50,51

NHANES database provides nationally representative estimates based on standardized protocols for data collection. Consequently, the current findings can be generalized widely. However, it is important to note that our study has several limitations. Firstly, the study only included the general population of the United States from of NHANES 2017–2018 years, due to year limitations. Secondly, self-reported confounders may be biased due to self-reporting questionnaire. Finally, as a cross-sectional study, conclusions were limited to associations rather than causality.

Conclusion

In conclusion, the relationship between the SII index and the risk of NAFLD presented a U-shaped curve in the American population. A turning point for the SII index was observed, and the prevalence of NAFLD was lowest when the SII index was 422.40. The potential mechanisms of the SII index in NAFLD need further exploration.