Authors: Yaqiong Wang, Kuangyang Chen, Jie Yu, Yifan Tu, Yuxin Zhao, Yikai Zhang, Yepeng Hu, Hong Yang, Han Yan, Chao Zheng
Categories: Research, MASLD, Systemic inflammation, Prognosis, All-cause mortality, CVD mortality
Source: BMC Gastroenterology
Authors: Yaqiong Wang, Kuangyang Chen, Jie Yu, Yifan Tu, Yuxin Zhao, Yikai Zhang, Yepeng Hu, Hong Yang, Han Yan, Chao Zheng
Inflammation plays a key role in the onset of MASLD, progression to MASH, and is associated with poor prognosis of MASLD. The aim of this study was to thoroughly investigate the association of different systemic inflammatory indicators with all-cause mortality and CVD mortality outcome in MASLD.
Data were from the NHANES cohort. Mortality data were from the NDI. Systemic inflammatory indicators analyzed included lymphocytes, monocytes, neutrophils, platelets, CRP, MLR, NLR, and PLR. Associations of inflammatory indicators with mortality were analyzed using Kaplan-Meier curves, Cox regression, and restricted cubic spline analysis. Time-dependent ROC curve was performed to analysis the predictive value of inflammatory indicators for mortality.
10,308 MASLD adults were included. Systemic inflammatory markers were associated with mortality outcome in MASLD. Specifically, lymphocyte and platelet count showed an L-shaped association with all-cause mortality in MASLD patients, monocyte and neutrophil count and CRP level demonstrated a positive nonlinear correlation, NLR and MLR presented a linear positive correlation, and PLR presented a U-shaped association (all P < 0.001). Time-dependent ROC curve analysis showed that NLR and MLR exhibited higher predictive values among all these markers for mortality in MASLD adults.
Systemic inflammatory indicators were associated with mortality outcome in MASLD and may serve as valuable prognostic biomarkers for MASLD.
The online version contains supplementary material available at 10.1186/s12876-025-04018-3.
Metabolic dysfunction-associated fatty liver disease (MASLD) is the most prevalent chronic liver disease, affecting approximately 35% of adults worldwide [1]. The disease spectrum of MASLD encompasses benign steatosis, metabolic dysfunction-associated steatohepatitis (MASH), MASH-related fibrosis, cirrhosis and hepatocellular carcinoma (HCC), all of which significantly contribute to increased morbidity and mortality [2–4].
Historically, the pathogenesis of MASLD was attributed to lipid accumulatio, which triggers hepatocyte mitochondrial dysfunction, the release of inflammatory mediators, and activation of hepatic stellate cells [2]. These processes exacerbate hepatic lipid accumulation, inflammation and fibrosis [2]. However, emerging evidence suggests that inflammation may occur prior to hepatic steatosis and is involved in the entire process of MASLD onset and progression [2, 5, 6]. Chronic low-grade inflammation and hepatic metabolic dysfunction promote the development of hepatic steatosis, progression to MASH, and even the development of fibrosis [6, 7]. The 2023 Delphi consensus statement has highlighted metabolic dysfunction as a key driver in the pathogenesis of MASLD. However, inflammation, a central component in the pathogenesis of MASLD and its progression to MASH, remains underappreciated and inadequately evaluated.
Monitoring peripheral immune cell counts and phenotypes holds potential as a valuable prognostic tool in MASLD. Despite this, the role of peripheral blood components, particularly leukocytes, in assessing the progression and prognosis of MASLD has long been overlooked. To address this gap, this present study aims to explore the relationship between various blood-based inflammatory markers—including blood cell counts, C-reactive protein (CRP), and derived composite markers—and mortality outcomes in MASLD patients. These blood-based inflammatory markers are commonly available in large epidemiological datasets, facilitating external cohort validation and providing a comprehensive and cost-effective reflection of systemic inflammation. Utilizing a representative U.S. population database, this study seeks to investigate how these inflammatory markers influence MASLD-related mortality and prognosis.
As mentioned in previous studies, data were gathered from the National Health and Nutrition Examination Survey (NHANES) datasets in the United States between 1999 and 2018 [8]. All participants signed a written informed consent. Detailed information on the NHANES datasets is accessible online from its official homepage.
Diagnosis of MAFLD requires confirmation of the presence of steatotic liver disease (SLD). However, most NHANES survey cycles do not include direct ultrasonographic or histological data for assessing hepatic steatosis. Therefore, we utilized the U.S. multi-ethnic population-modified Fatty Liver Index (USFLI) to diagnose hepatic steatosis [9]. MASLD was defined as the presence of SLD (USFLI ≥ 30), along with one or more of the five cardiometabolic risk factors, in the absence of other identifiable causes of hepatic steatosis. Participants with other causes of SLD or liver disease (based on questionnaire data for viral hepatitis, autoimmune liver disease, and other types of liver disease), or alcohol intake reaching of ≥ 30 g/day for men and ≥ 20 g/day for women (based on 24-hour dietary recall interviews) were excluded [10]. The formula for calculating the USFLI is as described \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$ {\rm{USFLI}},{\rm{ = }},{\matrix{ e(0.3458 \times {\rm{Mexican ,American}} - 0.8073 \hfill \cr \times {\rm{Non}} - {\rm{Hispanic ,Black}} \hfill \cr + ,0.0093 \times {\rm{Age}} + 0.6151 \times \ln \left( {{\rm{GGT}}} \right) \hfill \cr + 0.0249 \times {\rm{Waist ,Circumference}} + 1.1792 \hfill \cr \times {\rm{Ininsulin}} + 0.8242 \times \ln \left( {{\rm{Glucose}}} \right) - 14.7812) \hfill \cr} \over \matrix{ 1 + ,e(0.3458 \times {\rm{Mexican ,American}} - 0.8073 \hfill \cr \times {\rm{Non}} - {\rm{Hispanic ,Black}} + 0.0093 \times {\rm{Age}} \hfill \cr + 0.6151 \times \ln \left( {{\rm{GGT}}} \right) + 0.0249 \times {\rm{Waist ,Circumference}} \hfill \cr + 1.1792 \times {\rm{Ininsulin}} + 0.8242 \times \ln \left( {{\rm{Glucose}}} \right) - 14.7812) \hfill \cr} }
\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$ \times 100{\rm{}}\left( \matrix{ {\rm{"non}} - {\rm{Hispanic\,black"\,and\,"Mexican\,American"}} \hfill \cr {\rm{have\,a\,value\,of}}\,1,{\rm{if\,the\,participant\,is\,of\,that\,ethnicity}} \hfill \cr {\rm{and}}\,0\,{\rm{if\,not\,of\,that\,ethnicity}} \hfill \cr} \right) $$\end{document} The five cardiometabolic risk factors are Body mass index (BMI) ≥ 25 kg/m² or waist circumference (WC) ≥ 94 cm (males) and ≥ 80 cm (females);Fasting blood glucose (FBG) ≥ 100 mg/dL or 2-h postprandial blood glucose level ≥ 140 mg/dL or HbA1c ≥ 5.7% or receiving antidiabetic treatment;Blood pressure ≥ 130/85 mmHg or undergoing antihypertensive therapy;Plasma TG level of ≥ 150 mg/dL or receiving specific drug treatment;Plasma high-density lipoprotein cholesterol (HDL-C) level < 40 mg/dL (male) and < 50 mg/dL (female) or receiving particular drug therapy [10, 11]. ### Population selection A total of 101,316 individuals were initially surveyed across 10 survey cycles (from 1999 to 2018). After excluding participants under the age of 18, 59,204 participants remained. Subsequently, 39,581 individuals were excluded due to missing USFLI components. Additional exclusions included 60 participants with other liver diseases (27 with viral hepatitis, 6 with autoimmune hepatitis, and 27 with other liver conditions based on questionnaire data), 2,304 participants with excessive alcohol consumption (defined as alcohol intake ≥ 30 g/day for men and ≥ 20 g/day for women), 2,624 individuals with a USFLI < 30, and 813 individuals without any of the five cardiometabolic risk factors necessary for the diagnosis of MASLD. his left a total of 13,822 adults diagnosed with MASLD. Finally, we excluded 3,459 participants with missing CRP level, 22 with missing platelet count data, and 33 with missing data on neutrophils, monocytes, or lymphocytes. Ultimately, 10,308 participants were included in the analysis (Fig. 1). Fig. 1The scheme and flowchart of participants in this present study. USFLI, United States fatty liver index; SLD, steatotic liver disease; MASLD, metabolic dysfunction-associated steatotic liver disease ### Systemic inflammation indicators Blood cell counts for lymphocytes, monocytes, neutrophils, and platelets were measured using a Beckman Coulter automatic hematology analyzer at the Mobile Examination Center, with results expressed as ×10^9 cells/µL. The monocyte-to-lymphocyte ratio (MLR), neutrophil-to-lymphocyte ratio (NLR), and platelet-to-lymphocyte ratio (PLR) were calculated as the ratios of monocyte count, neutrophil count, and platelet count to lymphocyte count, respectively. C-reactive protein (CRP) levels were quantified via latex-enhanced nephelometry. ### Outcome ascertainment Mortality data were obtained from the National Death Index (NDI), linked to the NHANES datasets, up to December 31, 2019 [12]. The primary outcome was all-cause mortality among MASLD participants, defined by their survival status. The secondary outcome is cardiovascular disease (CVD) mortality according to the ICD-10 classification, defined as death due to diseases of the heart (UCOD_LEADING = 001) or cerebrovascular diseases (UCOD_LEADING = 5). ### Assessment of covariates Demographic data for MASLD patients were extracted from the NHANES datasets. To minimize potential confounding bias, covariates were selected based on prior research and clinical relevance [11]. These covariates included age, gender (male or female), race (Mexican American, non-Hispanic White, non-Hispanic Black, or other), marital status (married or living with partner, without partner, or missing data), and education level (college or above, high school or equivalent, less than high school, or missing data). Additionally, lifestyle habits and comorbidity histories, such as smoking status (never, former, current, or missing data), the presence or absence of cancer, chronic kidney disease (CKD), and CVD were also collected based on questionnaire responses. CVD was defined as participants have been informed of having myocardial infarction, angina pectoris, coronary heart disease, congestive heart failure, or stroke. CKD is defined as participants have been informed of having weak or failing kidneys. Cancer is defined as participants have been informed of having cancer or malignancy. Additionally, physical examination and laboratory test data were collected, including BMI, WC, energy intake (average daily kilocalories from 24-hour dietary recall interviews), fasting insulin levels, TG, FBG, HbA1c, gamma-glutamyl transferase (GGT), alanine transaminase (ALT), and aspartate aminotransferase (AST) levels. ### Statistical analysis All statistical analyses were performed using R software (version 4.4.0), with complex weighting applied to ensure the national representativeness of the U.S. population with MASLD. The Anderson-Darling normality test was performed to assess whether continuous variable followed a normal distribution. Notably, all continuous variables exhibited non-normal distributions and were presented as weighted medians with interquartile ranges. Categorical variables were presented as unweighted numbers and weighted proportions. The Kruskal-Wallis test was used for continuous variables with non-normal distributions, while the Chi-Squared test was employed to compare categorical variables. Missing data were categorized as “missing” in the baseline table, and no imputation methods were applied to address missing values to preserve data integrity. The independent associations of systemic inflammation indicators with all-cause and CVD mortality were determined utilizing univariate and multivariate-adjusted Cox regression with hazard ratios (HR) and 95% confidence intervals (CI). Results from the univariate analysis are presented in Model 1. In Model 2, we adjusted for demographic factors, including age, gender, race, education, marital status, and smoking status, which are known to influence the relationship between inflammatory markers and MASLD. In Model 3, we further adjusted for metabolic factors and comorbidity history, including energy intake, BMI, WC, cancer, CKD, CVD, ALT, and AST levels, as these factors may confound the relationship between inflammation and MASLD outcomes. Kaplan-Meier (KM) survival curves were generated to visualize the survival patterns of MASLD patients across quartiles of inflammatory markers, with differences assessed by the log-rank test. Restricted cubic spline (RCS) regression was performed with the aforementioned multivariable adjustments to examine the relationship between inflammatory markers and mortality. Four knots (at the 5th, 35th, 65th, and 95th percentiles) were chosen for the spline regression to smooth the curve. Subgroup analyses were conducted considering characteristics based on characteristics such as age (< 65 years or ≥ 65 years), gender (male or female), race, education, smoking status, BMI (< 30 kg/m² or ≥ 30 kg/m²), and the presence of comorbidities. Considering the nearly two-decade span of the NHANES dataset utilized in this study, we conducted a sensitivity analysis to evaluate the association between systemic inflammatory markers and mortality outcomes in adults with MASLD within the earlier NHANES cycles. ## Results ### Baseline characteristics of participants A total of 10,308 subjects with MASLD were included between 1999 and 2018 years (Fig. 1). The median age of MASLD patients was 47 years, with a slightly higher proportion of females than males (51.9% vs. 48.1%). Nearly half of MASLD patients were non-Hispanic whites (66.6%), 52.5% of whom had a college degree or higher. Most participants of MASLD were overweight (70.4%) and 43.5% were obese. A total of 18.1% of the subjects were current smokers. The prevalence of comorbidities was 9.3% for cancer, 2.6% for CKD and 8.1% for CVD. The median follow-up time was 11.33 years, and a total of 2,609 (13.9%) MASLD patients died, of which 25.4% were died of CVD. Compared to survivors, non-survivors tended to be male, older, non-Hispanic white, less educated, unpartnered, have a larger waist circumference, lower energy intake, higher FBG and HbA1c, fewer lymphocytes and platelets, more monocytes, higher CRP, and higher prevalence of comorbidities (Table 1). Table 1Baseline demographic characteristics of participantsVariablesOverallNon-survivorsSurvivors*P* valueN (%)10,30818188490Age47.0[33.0,61.0]70.00[59.00,78.00]44.00[32.00,56.00]< 0.001Energy1961.0[1458.0,2591.0]1681.00[1308.90,2216.00]2020.14[1492.20,2674.00]< 0.001WC100.74[91.50,111.40]102.30[93.00,113.14]100.50[91.34,111.10]0.005TG116.00[82.00,168.00]128.00[91.00,184.00]114.00[80.01,166.00]< 0.001FBG95.00[88.00,103.00]100.00[91.00,117.00]94.00[88.00,102.00]< 0.001GGT21.00[15.00,31.00]23.00[16.00,35.00]20.00[15.00,30.00]< 0.001ALT21.00[16.00,30.00]20.00[16.00,26.00]22.00[16.00,30.00]< 0.001AST22.00[19.00,27.00]23.00[19.00,28.00]22.00[18.0,27.00]< 0.001HbA1c5.40[5.20,5.80]5.70[5.40,6.10]5.40[5.10,5.70]< 0.001Insulin15.14 (14.25)16.28 (16.28)14.95 (13.88)0.004USFLI92.12 (16.14)92.39 (16.05)92.08 (16.16)0.516Leukocyte6.70[5.60,8.10]6.90[5.90,8.40]6.70[5.60,8.10]< 0.001Lymphocyte1.90[1.60,2.40]1.80[1.40,2.30]2.00[1.60,2.40]< 0.001Neutrophil3.90[3.10,5.00]4.20[3.30,5.40]3.90[3.00,4.90]< 0.001Monocyte0.50[0.40,0.60]0.60[0.50,0.70]0.50[0.40,0.60]< 0.001Platelet252.00[214.00,298.00]238.00[199.25,287.00]254.00[216.00,299.00]< 0.001CRP2.30[1.00,5.20]2.90[1.30,6.50]2.20[1.00,5.00]< 0.001NLR2.00[1.53,2.67]2.33[1.73,3.19]1.95[1.50,2.60]< 0.001MLR0.29 (0.12)0.35 (0.16)0.28 (0.11)< 0.001PLR129.44[103.74,163.06]130.59[101.02,171.33]129.22[104.00,161.54]0.213Survival time (month)139.00[92.00,181.00]98.00[54.00,142.00]146.00[114.00,186.00]< 0.001CVD mortality (%) NO9735.0 (95.8)1245.0 (69.7)8490.0 (100.0)< 0.001 YES573.0 (4.2)573.0 (30.3)0.0 (0.0)Gender (%) Female5492.0 (51.9)804.0 (47.5)4688.0 (52.7)0.001 Male4816.0 (48.1)1014.0 (52.5)3802.0 (47.3)Race (%) Mexican American2342.0 (9.7)284.0 (4.0)2058.0 (10.7)< 0.001 Non-Hispanic Black2019.0 (11.1)299.0 (9.3)1720.0 (11.4) Non-Hispanic White4414.0 (66.6)1106.0 (79.9)3308.0 (64.5) Other1533.0 (12.5)129.0 (6.8)1404.0 (13.5)Education (%) College or above4487.0 (52.5)595.0 (38.8)3892.0 (54.7)< 0.001 High school or equivalent2530.0 (27.0)464.0 (29.2)2066.0 (26.7) Less than High school3276.0 (20.4)754.0 (31.9)2522.0 (18.5) Missing data15.0 (0.1)5.0 (0.1)10.0 (0.1) Marital status (%) Married or living with partner5997.0 (62.6)990.0 (58.3)5007.0 (63.3)0.011 Without partner3906.0 (33.7)775.0 (38.8)3131.0 (32.9) Missing data405.0 (3.7)53.0 (3.0)352.0 (3.8)Smoking (%) Former2621.0 (26.1)737.0 (39.2)1884.0 (24.0)< 0.001 Never5390.0 (53.3)765.0 (40.7)4625.0 (55.4) Now1706.0 (18.1)303.0 (19.7)1403.0 (17.9) Missing data591.0 (2.4)13.0 (0.4)578.0 (2.8)BMI (%) < 252017.0 (19.6)444.0 (24.6)1573.0 (18.6)< 0.001 ≥ 25and < 303795.0 (36.9)655.0 (36.3)3140.0 (37.0) ≥ 304472.0 (43.5)707.0 (39.1)3765.0 (44.4)CVD (%) NO9331.0 (91.9)1350.0 (74.6)7981.0 (94.7)< 0.001 YES977.0 (8.1)468.0 (25.4)7981.0 (5.3)CKD (%) NO9987.0 (97.4)1711.0 (94.9)8276.0 (97.9)< 0.001 YES321.0 (2.6)107.0 (5.1)214.0 (2.1)Cancer (%) NO9361.0 (90.7)1450.0 (79.3)7911.0 (92.5)< 0.001 YES947.0 (9.3)368.0 (20.7)579.0 (7.5)Non-normally distributed variables are displayed as median with 1st and 3rd quartile. Categorical variables are displayed as numbers with percentages (n, %). Bold values indicate statistical significance (*p* < 0.05). BMI, body mass index; WC, Waist circumference; TG, triglyceride; GGT, gamma-glutamyl transferase; ALT, glutamic-pyruvic transaminase; AST, aspartate transaminase; FBG, fasting blood glucose; USFLI, United States fatty liver index; CVD, cardiovascular disease; CKD, chronic kidney disease ### Non-linear trend between systemic immune biomarkers and mortality in MASLD RCS analysis, as depicted in Fig. 2, demonstrated a dose–response relationship between systemic inflammatory markers and mortality in MASLD. For all-cause mortality, lymphocyte and platelet counts exhibited an L-shaped association, while monocyte, neutrophil and CRP levels showed a positive nonlinear correlation. NLR and MLR exhibited a linear positive association, and PLR presented a U-shaped association (all P for overall < 0.001). For CVD mortality, PLR showed no significant association (*P* = 0.114), while lymphocyte count displayed an L-shaped relationship, platelet showed a linear negative association, and monocyte, neutrophil, MLR, NLR, and CRP exhibited linear positive correlations ((all P for overall < 0.001). Fig. 2Restricted cubic splines reflect the association of systemic inflammatory indicators with mortality among adults with MASLD. **A**-**H**: the association with indicators with all-cause mortality; **I**-**P**: the association with indicators with CVD mortality; MASLD, metabolic dysfunction-associated steatotic liver disease ### Association between systemic immune biomarkers and mortality in MASLD Kaplan-Meier (KM) curves, shown in Fig. 3, illustrated the survival patterns of MASLD patients across quartiles of systemic immune biomarkers. Both all-cause and CVD mortality were significantly higher in MASLD patients with lower quartiles of lymphocytes and platelets, and higher quartiles of monocytes, neutrophils, MLR, NLR, and CRP, when compared with other quartile subgroups (all *P* < 0.01 by log-rank test). The probability of survival was significantly higher in the middle quartiles of PLR compared to the first and fourth quartiles (log-rank test). No significant difference was observed in CVD mortality across quartiles of PLR (*P* = 0.4134 by log-rank test). Fig. 3Kaplan–Meier curves of the survival pattern between systemic inflammatory indicators with mortality among adults with MASLD. **A**-**H**: the survival pattern between indicators with all-cause mortality; **I**-**P**: the survival pattern between indicators with CVD mortality; MASLD, metabolic dysfunction-associated steatotic liver disease Figure 4 demonstrates the unadjusted and multivariate-adjusted Cox regression analysis of the relationship between immune biomarkers and mortality in MASLD. The results revealed that higher levels of monocytes, neutrophils, MLR, NLR, and CRP, as well as lower levels of lymphocytes and platelets, were associated with an increased risk of all-cause mortality in MASLD, irrespective of adjustments for confounders. In the univariate model, NLR exhibited the highest hazard ratio (HR) among the inflammatory markers (HR 2.45, 95% CI 2.14–2.79, *P* < 0.001, Q4 vs. Q1 of NLR). After adjusting for potential confounders, neutrophil count was associated with the highest HR (HR 1.70, 95% CI 1.47–1.96, *P* < 0.001, Q4 vs. Q1 of neutrophil). After multivariate adjustments, CRP, neutrophil count, and NLR remained significantly associated with CVD mortality in MASLD, with higher levels of these markers linked to an increased risk of CVD-related death. Fig. 4Forest plots show the association of systemic inflammatory indicators with mortality among adults with MASLD. **A**-**H**: the association with indicators with all-cause mortality; **I**-**P**: the association with indicators with CVD mortality; MASLD, metabolic dysfunction-associated steatotic liver disease ### Predictive value of systemic immune biomarkers for mortality in MASLD Time-dependent ROC curve analysis was conducted to assess the predictive value of systemic immune biomarkers for mortality in MASLD patients (Fig. 5). The results indicated that the predictive capacity of lymphocytes, neutrophils, and monocytes was lower than that of their derived composite indices. Specifically, the area under the curve (AUC) for NLR in predicting all-cause mortality was 0.660, 0.632, 0.633, and 0.623 at 1, 3, 5, and 10 years, respectively. For CVD mortality, the AUC for NLR was 0.653, 0.601, 0.646, and 0.632 at the same time points. The AUC for MLR for all-cause mortality was 0.658, 0.655, 0.656, and 0.605 at 1, 3, 5, and 10 years, respectively, while for CVD mortality, MLR yielded AUC values of 0.615, 0.674, 0.662, and 0.596, respectively. Platelet count demonstrated relatively stronger predictive power for CVD mortality, with AUC values of 0.637, 0.630, 0.641, and 0.629 at 1, 3, 5, and 10 years, respectively. In contrast, PLR showed no significant predictive value for either all-cause or CVD mortality in MASLD patients. CRP exhibited modest predictive power for all-cause mortality, with AUC values of 0.600, 0.592, 0.570, and 0.546 at 1, 3, 5, and 10 years, respectively. For CVD mortality, the AUC for CRP was 0.562, 0.578, 0.536, and 0.534 at the same time points. These findings highlight the prognostic significance of inflammatory markers in predicting mortality in MASLD patients, with NLR and MLR showing superior predictive value for both all-cause and CVD mortality. Fig. 5Time-dependent ROC curves display the predictive value of systemic immune biomarkers for mortality of MASLD. **A**-**H**: the predictive value for all-cause mortality; **I**-**P**: the predictive value for CVD mortality; MASLD, metabolic dysfunction-associated steatotic liver disease ### Subgroup and sensitive analyses Subgroup analyses were conducted to explore the associations between individual inflammatory markers and all-cause mortality in MASLD (Supplementary Figures). Interaction tests revealed that monocytes and CRP consistently maintained their associations with the risk of all-cause mortality across various subgroups defined by age, sex, race, BMI, and comorbidities. Notably, age significantly interacted with the associations between lymphocytes and neutrophils and all-cause mortality in MASLD (P for interaction < 0.05). Furthermore, cancer significantly interacted with the associations between lymphocytes, neutrophils, and platelets and all-cause mortality (P for interaction < 0.05). The association between lymphocyte count and all-cause mortality was more pronounced in MASLD patients aged > 65 years, without CKD, and without cancer. In contrast, neutrophil count demonstrated enhanced predictive significance in males, individuals with BMI < 30 kg/m², and those without CKD. Notably, among adults from the earlier survey periods, inflammatory markers remained significantly associated with mortality outcomes in individuals with MASLD, demonstrating the robustness of the primary study findings (Figure S6). ## Discussion Currently, the relationship between systemic inflammatory biomarkers and mortality in individuals with MASLD has been relatively underexplored. This study provides a comprehensive evaluation of how systemic inflammatory markers correlate with both all-cause and CVD mortality in adults with MASLD, utilizing data from a large prospective cohort. Our findings demonstrate that lymphocyte and platelet counts exhibited an L-shaped relationship with all-cause mortality in MASLD patients. Conversely, monocyte and neutrophil counts, along with CRP levels displayed a positive nonlinear correlation, while NLR and MLR presented a linear positive correlation, and PLR presented a U-shaped association. Time-dependent ROC curve analysis further revealed that both NLR and MLR had superior predictive capacity for all-cause and CVD mortality compared to the other biomarkers examined. Inflammation plays a pivotal role in the initiation and progression of MASLD. Emerging evidence suggests that chronic low-grade inflammation may precede the onset of hepatic steatosis, potentially serving as an early biomarker of the disease [6, 7]. Chronic liver inflammation may arise from extrinsic factors, such as pro-inflammatory diets, adipose tissue, or gastrointestinal disturbances, as well as intrinsic mechanisms, including lipotoxicity, mitochondrial dysfunction, endoplasmic reticulum stress, and cellular apoptosis, etc [13]. Longitudinal cohort studies have identified pro-inflammatory diets as an independent risk factor for hepatic steatosis [14]. Notably, a six-month anti-inflammatory diet in MASLD patients resulted in a significant reduction in hepatic fat content [15]. In experimental models, Estadella et al. demonstrated that hepatic inflammation could precede the onset of steatosis, mediated by the accumulation of visceral fat [16]. Moreover, excessive caloric intake fosters the expansion of adipose tissue, resulting in inflammatory cell infiltration and the release of pro-inflammatory cytokines such as TNF-α and IL-6. These cytokines disrupt insulin signaling pathways, thus promoting extrahepatic lipolysis and intrahepatic lipid deposition [17]. Chronic gastrointestinal inflammation, driven by immune dysregulation and dysbiosis of the gut microbiota, has also been causally linked to the development of hepatic steatosis and inflammation [18–20]. Epidemiological studies have revealed that individuals with inflammatory bowel disease are at an increased risk for developing MASLD [18]. Additionally, fecal microbiota transplantation from MASLD patients to germ-free mice has been shown to replicate key features of liver steatosis and inflammation [19]. Therefore, inflammation may precede the onset of hepatic steatosis and persist throughout the pathogenesis and progression of MASLD, playing a critical role in the progression from MASLD to MASH and liver fibrosis. This study further highlights that the MLR serves as a robust prognostic marker for survival outcomes in MASLD patients. The frequency of the peripheral monocytes was significantly elevated in MASLD patients compared to healthy controls [21]. Monocyte-derived macrophages have pro-inflammatory and pro-fibrotic properties in liver [21]. Hepatic Kupffer cells, originating from embryonic precursors, progressively undergo apoptosis due to lipotoxic stress and are unable to effectively self-renew during MASH. Bone marrow-derived monocytes are recruited to sites of hepatic inflammation, where they differentiate into macrophages to replenish Kupffer cells [21]. Inhibition of monocyte recruitment has been shown to reduce hepatic steatohepatitis and fibrosis [22, 23]. In contrast, chronic stress and disruptions in lymphocyte distribution lead to impaired lymphocyte proliferation and enhanced apoptosis, resulting in decreased lymphocyte counts. This reduction reflects a compromised immune response and diminished host defense capacity. Previous cross-sectional studies have reported a negative association between peripheral lymphocyte counts, the lymphocyte-to-monocyte ratio (LMR), and MASLD risk [24, 25]. However, the prognostic significance of these markers for mortality outcomes in MASLD patients had not been extensively investigated. Our findings suggest that an increase in monocyte count coupled with a decrease in lymphocyte count independently contributes to an increased risk of mortality in MASLD patients, marking them as indicators of poor prognosis. The combined “increase-decrease” effect drives the elevation of MLR, rendering it a straight forward yet effective marker of systemic inflammation and immune function. Notably, MLR showed a positive linear correlation with mortality in MASLD, offering superior predictive value for both all-cause and CVD mortality when compared to individual monocyte or lymphocyte counts. These findings underscore the potential of MLR as a practical and reliable biomarker for identifying MASLD patients at increased risk of mortality. Neutrophils, the most abundant granulocyte subpopulation in peripheral blood, are the primary inflammatory cells infiltrating human liver tissue in cases of MASH as evidenced by biopsy specimens [26]. These cells are thought to play a pivotal role in driving tissue damage and inflammation in MASH. Neutrophils exacerbate hepatocellular injury, inflammation, and fibrosis through the production of reactive oxygen species (ROS), release of myeloperoxidase, formation of neutrophil extracellular traps (NETs), and secretion of pro-inflammatory cytokines [26, 27]. Previous studies of small cohorts have reported a strong correlation between elevated NLR and the severity of hepatic steatosis, inflammatory activity, and progression of fibrosis in MASLD [25, 28, 29]. Individuals with moderate to severe steatosis exhibit significantly higher NLR levels than those with mild steatosis [30]. Furthermore, the mean NLR value in MASH patients is significantly higher than in those with MASLD (3.44 ± 1.29 vs. 1.9 ± 0.7; *P* < 0.001) [31]. NLR has demonstrated substantial predictive value for MASH diagnosis among MASLD patients, with reported sensitivity of 82.62% and specificity of 81.22% [32]. While these smaller studies, utilizing elastography or biopsy for MASLD diagnosis, offer enhanced diagnostic precision. Their limitations lie in the inability to expand the sample size. In contrast, the current study diagnoses MASLD using the USFLI score derived from the NHANES large-scale cohort, further emphasizing the clinical utility of NLR. We found that increased peripheral neutrophil count, decreased lymphocyte count, and elevated NLR were significantly associated with both all-cause and CVD mortality in MASLD patients. Our findings regarding NLR are consistent with these of recent study by Dong et al. [33]. This recent cohort study based on NHANES data, encompassing 3,970 individuals, also showed a positive linear correlation between NLR and mortality outcomes in MASLD patients [33]. Although the association between NLR and MASLD-related mortality was slightly weaker than that of neutrophil count alone after adjusting for multiple confounders, time-dependent ROC curve analysis revealed that NLR demonstrated superior prognostic value for survival outcomes in MASLD patients. Platelets are involved not only in coagulation and hemostasis but also in inflammatory processes [34]. In our study, platelet counts were significantly lower in MASLD patients who succumbed to the disease compared to those who survived. Thrombocytopenia may reflect disease progression in MASLD, potentially due to excessive platelet consumption driven by chronic inflammation or hypersplenism associated with cirrhosis [35]. Some studies suggest that adenine nucleotides and hepatocyte growth factors released from platelet granules possess antifibrotic properties [36, 37], whereas others reported that platelet activation and adhesion contribute to hepatic fibrosis [38–40]. Notably, antiplatelet therapy with aspirin has been shown to mitigate hepatic steatosis, slow fibrosis progression in MASLD, and reduce the risk of HCC [38–40]. These findings collectively imply that both excessively low and high platelet levels may be linked to an increased risk of MASLD-related mortality. However, in our large-scale cohort analysis, lower platelet counts were associated with an elevated risk of mortality, while higher platelet counts did not demonstrate a significant correlation. Moreover, platelet counts exhibited a strong predictive capability for CVD mortality in MASLD patients. The relationship between the PLR and MASLD incidence and progression remains contentious. Some studies have reported a U-shaped association between PLR and both the development of MASLD and its progression to cirrhosis [41, 42], while others found no significant relationship between PLR and MASLD incidence [43]. Our findings show that while PLR has a U-shaped correlation with all-cause mortality, it does not exhibit a significant association with CVD mortality. Furthermore, PLR lacks predictive utility for both all-cause and CVD mortality in MASLD patients. CRP produced by hepatocytes in response to acute or chronic inflammation, is a well-established marker of systemic inflammation [44]. CRP may contribute to the development of MASLD/MASH through various mechanisms, including inhibition of the leptin and insulin signaling, mitochondrial damage, activation of NF-κB signaling [44]. Clinical studies have demonstrated a nonlinear positive correlation between CRP level and the risk of MASLD [24].CRP not only differentiates MASH from non-progressive simple steatosis but also correlates with the severity of hepatic fibrosis in MASH [45]. However, some studies have failed to establish a significant link between CRP and the degree of hepatic steatosis, necroinflammation, or fibrosis [46]. After adjusting for all confounders, CRP was positively and nonlinearly associated with all-cause mortality and linearly associated with CVD mortality in MASLD. These findings suggest that CRP could serve as a valuable early marker for prognostic outcomes in MASLD patients. The primary significance of this work is that it comprehensively investigates the prognostic impact of peripheral blood inflammatory markers on mortality outcomes in MASLD adults in the United States. By utilizing a sizable cohort size and adequate follow-up period, we carefully evaluated the association between multiple blood inflammatory markers and mortality outcomes in MASLD adults, providing valuable insights to guide clinical management strategies. Additionally, to ensure the robustness of our findings, we conducted a sensitivity analysis and observed that, despite the extended timespan of the NHANES dataset, the primary results remained stable in the earlier NHANES cycles (1999–2006). However, there are several limitations of this study that merit consideration. First, the diagnosis of SLD relied on USFLI criteria rather than more precise histologic or ultrasound assessments. Second, resulting from the cross-sectional design of NHANES, systemic inflammatory markers were calculated from baseline data. We were unable to determine a causal relationship between systemic inflammatory biomarkers and MASLD. Finally, given that this study was based on a US population, the generalizability of the results warrants further validation. ## Conclusion In conclusion, our study establishes a significant association between systemic inflammatory markers and the risk of all-cause and CVD mortality in MASLD adults. With this understanding, clinicians can more accurately identify individuals at risk for MASLD and provide early intervention to improve the prognosis of MASLD patients. Further prospective investigations are necessary to validate these findings and to elucidate the underlying mechanisms driving this association. ## Electronic supplementary material Below is the link to the electronic supplementary material. Supplementary Material 1