ORIGINAL RESEARCH article

Front. Med., 09 December 2025

Sec. Hepatobiliary Diseases

Volume 12 - 2025 | https://doi.org/10.3389/fmed.2025.1687078

Association between metabolic dysfunction-associated fatty liver disease and risk of incidence of gallstones, evidence from a 5-year railway safety staff retrospective cohort study

  • 1. Department of Blood Transfusion, Wuhan Wuchang Hospital, Wuhan University of Science and Technology, Wuhan, China

  • 2. Wuhan Children's Hospital, Wuhan Maternal and Child Healthcare Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China

  • 3. Physical Examination Center, Renmin Hospital of Wuhan University, Wuhan, China

  • 4. Physical Examination Center, Wuhan Wuchang Hospital, Wuhan University of Science and Technology, Wuhan, China

  • 5. Department of Orthopedics, Wuhan Wuchang Hospital, Wuhan University of Science and Technology, Wuhan, China

Article metrics

View details

762

Views

38

Downloads

Abstract

Background:

Current evidence from cohort studies on the association between metabolic dysfunction-associated fatty liver disease (MAFLD) and gallstone formation remains limited. This 5-year retrospective cohort study aimed to evaluate the impact of MAFLD on gallstone incidence among railway safety staff.

Methods:

Data were obtained from the physical examination data of railway safety staff at Wuchang Hospital, Wuhan City, Hubei Province, China. The diagnosis of MAFLD was consistent with the international expert consensus. Independent associations between MAFLD and participants' risk of gallstone incidence were assessed by multiple Cox regression. Stratified analyses and interaction tests were performed to investigate whether covariates modified the association between MAFLD and gallstone incidence.

Results:

Among 1,422 participants, the risk of gallstone incidence was 9.27% (37/399) higher in the exposed group with MAFLD than in the control group 5.67% (58/1,023). Multiple cox regression showed that people with MAFLD were associated with a risk of gallstone incidence (HR = 1.71, 95% CI = 1.13–2.58). Stratified analysis by sex showed that men with MAFLD had a significantly increased risk (HR = 1.54, 95% CI: 1.00–2.38, P = 0.050), whereas women did not reach statistical significance (P = 0.605), but the sex interaction was not significant (P-interaction = 0.664). In age-stratified analyses, both the 45–60 years group (HR = 1.58, P = 0.078) and the 30–45 years group (HR = 1.89, P = 0.179) showed a trend toward increased risk, but the interaction was not statistically different (P-interaction = 0.270).

Conclusions:

There is a significant association between MAFLD and the incidence of gallstones. Attention should be paid to the prevention and treatment of MAFLD to promote the health of railway safety staff and reduce the burden ofanalysis gallstone-related diseases.

Background

Gallstone disease (GD) has become a major global health issue, affecting 10%−15% of adults in Western countries, with a rapidly rising incidence in Asia linked to changes in lifestyle and diet (13) As one of the most common gastrointestinal disorders, GD imposes a significant economic burden, especially when complications necessitate surgical interventions such as cholecystectomy (4, 5). Although mortality of gallstone is relatively low, its high morbidity and associated costs place substantial pressure on families and healthcare systems, especially in populous developing countries like China (6, 7). In China, the epidemiology of liver disease has shifted markedly, while viral hepatitis was historically the leading cause of liver-related morbidity, vaccination and antiviral advances have reduced its burden. Meanwhile, metabolic dysfunction-associated fatty liver disease (MAFLD) has become the most prevalent chronic liver condition, affecting over 29.2% of the population (810). MAFLD indicates that this disease is closely related to metabolic dysfunction and metabolic syndrome, thereby better reflecting the etiology of fatty liver and shifting from exclusion diagnosis to diagnosis based on positive criteria (11).

The biological association between MAFLD and GD is reasonable, as they share a common metabolic disorder mechanism. Both conditions are strongly associated with obesity, insulin resistance, type 2 diabetes, and dyslipidemia, which are key phenotypic components of metabolic syndrome (1214). From a mechanistic standpoint, MAFLD may promote gallstone formation through several pathways: hepatic insulin resistance activates sterol regulatory element-binding protein-2 (SREBP-2), enhancing cholesterol synthesis and leading to cholesterol-supersaturated bile (15, 16). Additionally, inflammatory mediators derived from visceral adipose tissue in MAFLD patients can impair gallbladder motility by downregulating contractile protein expression and reducing acetylcholine sensitivity in smooth muscle, thereby predisposing to stone formation (17). Emerging epidemiological evidence suggests a significant association between MAFLD and an increased risk of GD. A meta-analysis of eight studies involving 326,922 participants confirmed this association. MAFLD patients had a higher incidence of GD (OR: 1.71, CI: 1.63–1.79, P < 0.001) and this risk may be increased in females (OR: 4.18; CI: 1.21–14.37) and those with high BMI (OR: 1.80; CI: 1.36–2.56) (18), the results of a cohort study showed that GD and cholecystectomy were independently associated with MAFLD (OR = 1.75, 95% CI: 1.43–2.15 for GD; OR = 2.77, 95% CI: 2.01–3.83 compared with no GD) (19). At the same time, studies have found a bidirectional association between MAFLD and incidence of gallstones. MAFLD and its severity are independently associated with an increased incidence of gallstones, while gallstones and cholecystectomy are also associated with the incidence of MAFLD (20, 21). These bidirectional associations suggest a complex interplay that may involve shared metabolic, inflammatory, and possibly gut-liver axis mechanisms.

Despite growing evidence linking MAFLD with GD, significant knowledge gaps persist. Current evidence, predominantly from cross-sectional studies, limits causal inference regarding the relationship. Furthermore, the generalizability of findings across diverse populations is constrained by genetic, dietary, and lifestyle variations, and the potential confounding by metabolic comorbidities remains a challenge. Railway safety staff present an ideal model for investigation. They frequently work in shifts, which are factors contributing to metabolic abnormalities associated with MAFLD and gallstones. Additionally, Railway safety staff is exposed to distinct occupational stressors that may modulate liver metabolism via neuroendocrine pathways. Crucially, the comprehensive and standardized annual health examination system within the railway sector offers robust, longitudinal data, effectively minimizing socioeconomic confounders due to the population's homogeneity in work models and healthcare access. The 5-year follow-up period is sufficient to observe the complete pathological process of metabolic abnormalities to gallstone formation. This study aims to investigate the association between MAFLD and GD based on the data of retrospective railway safety staff for up to 5 years, providing a scientific basis for the prevention of GD.

Methods

Source of data

The retrospective cohort study using de-identified annual physical-examination records of railway safety staff from Wuchang Hospital, Wuhan, China, covering 85% of the railway safety staff in Wuhan prefecture and its three satellite cities between 1 January 2019 and 31 December 2023. This database records detailed information on population statistics, laboratory tests, imaging data, vital signs, and disease diagnosis. Due to the lack of protected information in the database and the anonymity of patients, a waiver of informed consent was granted (Figure 1).

Figure 1

Flowchart showing the selection process of participants from the Wuhan Wuchang Hospital Railway Safety Staff Physical Examination Health Database. Initially, there were 1801 participants aged 18 to 60 from 2019 to 2023. Exclusions included 16 participants with cholecystectomy, 158 with pre-existing gallstones, 196 who did not attend follow-up or had follow-up of less than three years, and 9 with missing data. Ultimately, 1422 participants were analyzed, with 399 having MAFLD and 1023 as the control group without MAFLD.

Flowchart of the study population.

Identification of MAFLD

According to the criteria in the latest international diagnostic consensus (10) diagnostic criteria for MAFLD in this study were based on imaging and blood biomarker evidence of hepatic fat accumulation (hepatocellular steatosis) in combination with one of three conditions: overweight/obesity, type 2 diabetes mellitus, and metabolic dysfunction, in accordance with the criteria set forth in the most recent international diagnostic consensus. Metabolic dysfunction is defined as the presence of at least two risk factors for metabolic abnormalities. These factors include: waist circumference: ≥90 and 80 cm in Asian men and women, respectively; blood pressure: ≥130/85 mmHg or on blood pressure-lowering medication; blood triglycerides: ≥1.7 mmol/L or on lipid-lowering medication; plasma high-density lipoprotein (Hdl) cholesterol: < 1.0 and 1.3 mmol/L in men and women, respectively, or on lipid-lowering medication; prediabetes fasting glucose: 5.6–6.9 mmol/L or 2 h postprandial glucose 7.8–11.0 mmol/L or glycated hemoglobin 5.7%−6.4%; homeostasis model assessment of insulin resistance index: ≥2.5; high-sensitivity C-reactive protein: >2 mg/L.

Assessment of covariates

Clinical covariates potentially associated with MAFLD and gallstones were retrieved from 2019–2023 Wuhan Wuchang Hospital Railway Safety Staff Physical Examination Health Database. Demographic variables for each patient included age and sex. Body mass index (BMI) was calculated as weight (kg)/height2 (m2), and the presence of comorbidities was defined according to pre-index date ICD-10 codes, defined using dichotomous variables, including diabetes mellitus, hypertension. Laboratory results included lipid indices: TC (total cholesterol), TG (triglycerides), LDL-C (low-density lipoprotein cholesterol), HDL-C (high-density lipoprotein cholesterol), which were converted into categorical indices according to the Chinese lipid management guidelines in order to make the lipid indices more clinically informative (22); fasting blood glucose; renal function indices of creatinine and uric acid; liver function indices of Alanine aminotransferase (ALT), Aspartate Aminotransferase (AST), Direct Bilirubin (DBil), Total Bilirubin (TBil). To improve the clinical significance of the results and to detect potential non-linear relationships, continuous variables were transformed into categorical variables based on their distribution within the cohort. Specifically, they were divided into four quartiles (Q1–Q4), where Q1 represents the lowest 25% of values and Q4 the highest 25%.

Statistical analysis

The baseline characteristics of the research subjects are represented as the mean ± standard deviation of continuous variables, while categorical variables are represented as numbers (percentages). Compare the values of each variable between two groups using independent t-test for continuous variables and chi square test for categorical variables. We used Cox proportional hazards regression analysis to evaluate the incidence of new cases of MAFLD and gallstones, as well as the differences in this association between different subgroups. The results are expressed as hazard ratio (HR) [95% confidence interval (CI)]. Compare the HR (95% CI) of gallstones in the MAFLD group with that in the non fatty liver group as a reference. Two-sided P < 0.05 was considered statistically significant. Perform statistical analysis using R 4.5.1 software.

Results

This study enrolled a total of 1,422 participants (Table 1). The 399 participants (28.06%) diagnosed with MAFLD at baseline were classified as the Exposed group, with the remaining participants forming the Control group (n = 1,023, 71.94%). Median follow-up was 4.12 years, yielding 4 750 person-years. Compared with controls, the MAFLD group differed significantly (P < 0.05) in sex, age, BMI, hypertension, diabetes, TC, LDL-C and HDL-C categories, creatinine, uric acid, TG, FBG, ALT, AST and DBil, whereas total bilirubin levels were comparable (P > 0.05).

Table 1

Variables Total (n = 1,422) Control (n = 1,023) Exposed (n = 399) Statistic P-value
Sex, n(%)
Male 1,168 (82.14) 780 (76.25) 388 (97.24) χ2 = 86.25 < 0.001
Female 254 (17.86) 243 (23.75) 11 (2.76)
Age, n(%)
18–30 391 (27.50) 318 (31.09) 73 (18.30) χ2 = 31.93 < 0.001
30–45 477 (33.54) 347 (33.92) 130 (32.58)
45–60 554 (38.96) 358 (35.00) 196 (49.12)
BMI, n(%)
Normal 607 (42.69) 572 (55.91) 35 (8.77) χ2 = 389.64 < 0.001
Underweight 46 (3.23) 46 (4.50) 0 (0.00)
Overweight 574 (40.37) 351 (34.31) 223 (55.89)
Obese 195 (13.71) 54 (5.28) 141 (35.34)
Diabetes, n(%)
No 1,107 (77.85) 877 (85.73) 230 (57.64) χ2 = 131.28 < 0.001
Yes 315 (22.15) 146 (14.27) 169 (42.36)
Hypertension, n(%)
No 1,141 (80.24) 884 (86.41) 257 (64.41) χ2 = 87.63 < 0.001
Yes 281 (19.76) 139 (13.59) 142 (35.59)
TC, n(%)
Normal 1,027 (72.22) 792 (77.42) 235 (58.90) χ2 = 52.61 < 0.001
Borderline high 290 (20.39) 177 (17.30) 113 (28.32)
High 105 (7.38) 54 (5.28) 51 (12.78)
LDL, n(%)
Ideal 647 (45.50) 515 (50.34) 132 (33.08) χ2 = 57.84 < 0.001
Optimal 495 (34.81) 353 (34.51) 142 (35.59)
Borderline high 206 (14.49) 117 (11.44) 89 (22.31)
High 74 (5.20) 38 (3.71) 36 (9.02)
HDL, n(%)
Normal 1,240 (87.20) 930 (90.91) 310 (77.69) χ2 = 44.91 < 0.001
Abnormal 182 (12.80) 93 (9.09) 89 (22.31)
Creatinine. quantile, n(%)
Mean ± SD (μmol/L) 72.87 ± 13.46 72.30 ± 14.32 74.33 ± 10.83 t = −2.89 0.004
Q1 355 (24.96) 288 (28.15) 67 (16.79) χ2 = 23.56 < 0.001
Q2 347 (24.40) 227 (22.19) 120 (30.08)
Q3 363 (25.53) 251 (24.54) 112 (28.07)
Q4 357 (25.11) 257 (25.12) 100 (25.06)
Uric acid. quantile, n(%)
Mean ± SD (μmol/L) 387.50 ± 98.97 368.63 ± 95.85 435.90 ± 90.10 t = −12.09 < 0.001
Q1 356 (25.04) 324 (31.67) 32 (8.02) χ2 = 130.06 < 0.001
Q2 355 (24.96) 268 (26.20) 87 (21.80)
Q3 355 (24.96) 241 (23.56) 114 (28.57)
Q4 356 (25.04) 190 (18.57) 166 (41.60)
TG. quantile, n(%)
Mean ± SD (μmol/L) 1.83 ± 1.67 1.49 ± 1.24 2.70 ± 2.23 t = −10.27 < 0.001
Q1 351 (24.68) 340 (33.24) 11 (2.76) χ2 = 276.74 < 0.001
Q2 359 (25.25) 297 (29.03) 62 (15.54)
Q3 355 (24.96) 229 (22.39) 126 (31.58)
Q4 357 (25.11) 157 (15.35) 200 (50.13)
FBG. quantile, n(%)
Mean ± SD (mmol/L) 5.21 ± 1.48 5.04 ± 1.27 5.64 ± 1.83 t = −6.01 < 0.001
Q1 356 (25.04) 279 (27.27) 77 (19.30) χ2 = 81.30 < 0.001
Q2 351 (24.68) 282 (27.57) 69 (17.29)
Q3 354 (24.89) 268 (26.20) 86 (21.55)
Q4 361 (25.39) 194 (18.96) 167 (41.85)
ALT. quantile, n(%)
Mean ± SD (μmol/L) 30.56 ± 25.28 24.72 ± 18.94 45.52 ± 32.37 t = −12.06 < 0.001
Q1 351 (24.68) 337 (32.94) 14 (3.51) χ2 = 289.68 < 0.001
Q2 358 (25.18) 300 (29.33) 58 (14.54)
Q3 356 (25.04) 236 (23.07) 120 (30.08)
Q4 357 (25.11) 150 (14.66) 207 (51.88)
AST. quantile, n(%)
Mean ± SD (μmol/L) 25.46 ± 12.83 23.15 ± 8.93 31.38 ± 18.28 t = −8.60 < 0.001
Q1 355 (24.96) 320 (31.28) 35 (8.77) χ2 = 146.95 < 0.001
Q2 347 (24.40) 261 (25.51) 86 (21.55)
Q3 361 (25.39) 263 (25.71) 98 (24.56)
Q4 359 (25.25) 179 (17.50) 180 (45.11)
DBil. quantile, n(%)
Mean ± SD (μmol/L) 3.00 ± 1.19 3.08 ± 1.24 2.77 ± 1.02 t = 4.87 < 0.001
Q1 339 (23.84) 225 (21.99) 114 (28.57) χ2 = 17.09 < 0.001
Q2 331 (23.28) 229 (22.39) 102 (25.56)
Q3 360 (25.32) 258 (25.22) 102 (25.56)
Q4 392 (27.57) 311 (30.40) 81 (20.30)
TBil. quantile, n(%)
Mean ± SD (μmol/L) 15.28 ± 6.23 15.41 ± 6.34 14.94 ± 5.93 t = 1.26 0.206
Q1 350 (24.61) 249 (24.34) 101 (25.31) χ2 = 1.27 0.737
Q2 350 (24.61) 246 (24.05) 104 (26.07)
Q3 359 (25.25) 260 (25.42) 99 (24.81)
Q4 363 (25.53) 268 (26.20) 95 (23.81)

Baseline characteristics of participants.

t, t-test, χ2, chi-square test; SD, standard deviation. Bold: Statistically different.

Multivariable cox regression (Table 2) showed MAFLD remained independently associated with incident gallstones after adjustment for sex, age, BMI and metabolic variables (HR 1.71; 95 % CI 1.13–2.58; P = 0.011). Age-stratified analysis showed that those aged 45–60 years had a significantly higher risk than the reference group aged 18–30 years (HR=2.82, 95% CI: 1.64–4.84, P < 0.001). Among metabolic related factors, overweight (HR = 1.77, 95% CI: 1.12–2.79, P = 0.014), diabetes (HR = 1.97, 95% CI: 1.29–3.00, P = 0.002), and hypertension (HR = 2.60, 95% CI: 1.71–3.97, P < 0.001) showed independent predictive effects. Males showed a trend of increased risk compared to females (HR = 1.83,95% CI: 0.97–3.42, P = 0.060), but did not reach statistical significance. Among the lipid indices, total cholesterol (TC) and low-density lipoprotein (LDL) stratification did not show a significant association, while the highest quartile group (Q4) of fasting glucose had a 93% increased risk (HR = 1.93, 95% CI: 1.08–3.42, P = 0.025). No significant associations were observed across quartiles of ALT, AST or bilirubin (all P > 0.05).

Table 2

Variables β S.E Z P-value HR (95% CI)
MAFLD
No 1.00 (Reference)
Yes 0.53 0.21 2.54 0.011 1.71 (1.13–2.58)
Sex
Female 1.00 (Reference)
Male 0.60 0.32 1.88 0.060 1.83 (0.97–3.42)
Age
18–30 1.00 (Reference)
30–45 −0.13 0.33 −0.40 0.688 0.87 (0.45–1.68)
45–60 1.04 0.28 3.76 < 0.001 2.82 (1.64–4.84)
BMI
Normal 1.00 (Reference)
Underweight −0.89 1.02 −0.87 0.382 0.41 (0.06–3.02)
Overweight 0.57 0.23 2.45 0.014 1.77 (1.12–2.79)
Obese 0.48 0.31 1.54 0.124 1.61 (0.88–2.95)
Diabetes
No 1.00 (Reference)
Yes 0.68 0.22 3.13 0.002 1.97 (1.29–3.00)
Hypertension
No 1.00 (Reference)
Yes 0.96 0.21 4.45 < 0.001 2.60 (1.71–3.97)
TC
Normal 1.00 (Reference)
Borderline high −0.06 0.27 −0.24 0.812 0.94 (0.55–1.60)
High 0.45 0.34 1.32 0.188 1.56 (0.80–3.03)
LDL
Ideal 1.00 (Reference)
Optimal 0.22 0.23 0.95 0.342 1.24 (0.80–1.93)
Borderline high 0.09 0.32 0.27 0.785 1.09 (0.58–2.05)
High −0.01 0.53 −0.03 0.977 0.99 (0.35–2.76)
HDL
Normal 1.00 (Reference)
Abnormal 0.39 0.35 1.10 0.269 1.97 (1.29–3.00)
Creatinine, quantile
Q1 1.00 (Reference)
Q2 −0.14 0.30 −0.45 0.649 0.87 (0.48–1.58)
Q3 −0.03 0.29 −0.09 0.930 0.97 (0.55–1.73)
Q4 0.23 0.28 0.83 0.405 1.26 (0.73–2.17)
Uric acid, quantile
Q1 1.00 (Reference)
Q2 0.01 0.30 0.04 0.967 1.01 (0.56–1.82)
Q3 0.12 0.29 0.42 0.673 1.13 (0.64–1.98)
Q4 0.08 0.29 0.26 0.791 1.08 (0.61–1.92)
TG, quantile
Q1 1.00 (Reference)
Q2 0.34 0.30 1.15 0.250 1.41 (0.79–2.52)
Q3 0.27 0.30 0.90 0.366 1.31 (0.73–2.37)
Q4 0.14 0.31 0.46 0.649 1.15 (0.62–2.13)
FBG, quantile
Q1 1.00 (Reference)
Q2 −0.02 0.33 −0.06 0.955 0.98 (0.52–1.86)
Q3 0.51 0.29 1.75 0.079 1.67 (0.94–2.94)
Q4 0.66 0.29 2.23 0.025 1.93 (1.08–3.42)
ALT, quantile
Q1 1.00 (Reference)
Q2 0.16 0.29 0.57 0.569 1.18 (0.67–2.07)
Q3 0.22 0.28 0.77 0.439 1.24 (0.72–2.17)
Q4 −0.29 0.32 −0.89 0.375 0.75 (0.40–1.41)
AST, quantile
Q1 1.00 (Reference)
Q2 0.21 0.28 0.75 0.454 1.23 (0.72–2.11)
Q3 −0.32 0.31 −1.02 0.306 0.73 (0.39–1.34)
Q4 0.04 0.29 0.13 0.899 1.04 (0.59–1.83)
DBil, quantile
Q1 1.00 (Reference)
Q2 −0.10 0.28 −0.35 0.727 0.91 (0.52–1.57)
Q3 −0.27 0.29 −0.94 0.347 0.76 (0.43–1.35)
Q4 −0.27 0.28 −0.94 0.346 0.77 (0.44–1.33)
TBil, quantile
Q1 1.00 (Reference)
Q2 0.32 0.31 1.02 0.309 1.37 (0.75–2.53)
Q3 0.28 0.31 0.92 0.357 1.33 (0.73–2.44)
Q4 0.41 0.30 1.37 0.171 1.51 (0.84–2.73)

Risk of incidence of gallstones according to the MAFLD status.

Subgroup analyses showed that the association between MAFLD and risk of incidence of gallstones was heterogeneous across populations (Table 3). Sex stratification showed a significantly increased risk in men with MAFLD (HR = 1.54, 95% CI: 1.00–2.38, P = 0.050), while women did not reach statistical significance (P = 0.605), but the sex interaction was not significant (P-interaction = 0.664). In the age stratification, both the 45–60 years group (HR = 1.58, P = 0.078) and the 30–45 years group (HR = 1.89, P = 0.179) showed a trend of increased risk, but the interaction was not statistically different (P-interaction = 0.270).

Table 3

Variables n (%) Control (n = 1,023) Exposed (n = 399) HR (95% CI) P-value P for interaction
All patients 1,422 (100.00) 58/1,023 37/399 1.71 (1.13–2.58) 0.011
Sex
Female 254 (17.86) 10/243 1/11 1.77 (0.20–15.36) 0.605 0.664
Male 1,168 (82.14) 48/780 36/388 1.54 (1.00–2.38) 0.050
Age
18–30 391 (27.50) 15/318 2/73 0.58 (0.13–2.55) 0.474 0.270
30–45 477 (33.54) 11/347 8/130 1.89 (0.75–4.76) 0.179
45–60 554 (38.96) 32/358 27/196 1.58 (0.95–2.65) 0.078
BMI
Normal 607 (42.69) 28/572 2/35 1.48 (0.35–6.21) 0.593 0.450
Underweight 46 (3.23) 1/46 0/0
Overweight 574 (40.37) 27/351 21/223 1.21 (0.68–2.13) 0.521
Obese 195 (13.71) 2/54 14/141 2.99 (0.68–13.27) 0.149
Diabetes
No 1,107 (77.85) 42/877 20/230 1.94 (1.14–3.30) 0.015 0.073
Yes 315 (22.15) 16/146 17/169 0.90 (0.45–1.78) 0.764
Hypertension
No 1,141 (80.24) 44/884 17/257 1.37 (0.78–2.40) 0.272 0.993
Yes 281 (19.76) 14/139 20/142 1.35 (0.68–2.67) 0.390
TC
Normal 1,027 (72.22) 45/792 23/235 1.85 (1.12–3.05) 0.017 0.645
Borderline high 290 (20.39) 8/177 9/113 1.75 (0.68–4.55) 0.247
High 105 (7.38) 5/54 5/51 0.83 (0.22–3.08) 0.775
LDL
Ideal 647 (45.50) 28/515 11/132 1.74 (0.86–3.49) 0.121 0.648
Optimal 495 (34.81) 21/353 18/142 2.19 (1.16–4.10) 0.015
Borderline high 206 (14.49) 7/117 6/89 0.89 (0.28–2.80) 0.842
High 74 (5.20) 2/38 2/36 1.21 (0.17–8.57) 0.851
HDL
Normal 1,240 (87.20) 54/930 32/310 1.86 (1.20–2.88) 0.005 0.708
Abnormal 182 (12.80) 4/93 5/89 1.49 (0.40–5.56) 0.552
Creatinine, quantile
Q1 355 (24.96) 16/288 8/67 2.35 (1.00–5.50) 0.049 0.711
Q2 347 (24.40) 10/227 10/120 1.96 (0.81–4.70) 0.133
Q3 363 (25.53) 13/251 10/112 1.79 (0.79–4.09) 0.165
Q4 357 (25.11) 19/257 9/100 1.22 (0.55–2.71) 0.617
Uric acid, quantile
Q1 356 (25.04) 15/324 8/32 6.50 (2.75–15.34) < 0.001 0.023
Q2 355 (24.96) 15/268 7/87 1.50 (0.61–3.69) 0.373
Q3 355 (24.96) 16/241 10/114 1.33 (0.61–2.94) 0.474
Q4 356 (25.04) 12/190 12/166 1.14 (0.51–2.54) 0.749
TG, quantile
Q1 351 (24.68) 18/340 1/11 2.10 (0.28–15.72) 0.471 0.395
Q2 359 (25.25) 19/297 9/62 2.32 (1.04–5.15) 0.039
Q3 355 (24.96) 11/229 15/126 2.49 (1.14–5.42) 0.022
Q4 357 (25.11) 10/157 12/200 1.02 (0.44–2.36) 0.966
FBG, quantile
Q1 356 (25.04) 12/279 8/77 2.62 (1.00–6.49) 0.051 0.127
Q2 351 (24.68) 15/282 3/69 0.78 (0.23–2.70) 0.696
Q3 354 (24.89) 16/268 13/86 2.50 (1.20–5.20) 0.014
Q4 361 (25.39) 15/194 13/167 0.97 (0.46–2.05) 0.945
ALT, quantile
Q1 351 (24.68) 20/337 2/14 4.38 (1.02–18.78) 0.047 0.632
Q2 358 (25.18) 18/300 9/58 2.93 (1.31–6.53) 0.009
Q3 356 (25.04) 15/236 14/120 1.78 (0.86–3.70) 0.120
Q4 357 (25.11) 5/150 12/207 1.75 (0.62–4.98) 0.292
AST, quantile
Q1 355 (24.96) 22/320 2/35 0.97 (0.23–4.11) 0.963 0.183
Q2 347 (24.40) 18/261 11/86 2.02 (0.95–4.28) 0.066
Q3 361 (25.39) 7/263 11/98 3.80 (1.46–9.92) 0.006
Q4 359 (25.25) 11/179 13/180 1.18 (0.53–2.64) 0.685
DBil, quantile
Q1 339 (23.84) 13/225 14/114 2.16 (1.01–4.61) 0.047 0.260
Q2 331 (23.28) 12/229 12/102 2.34 (1.05–5.22) 0.037
Q3 360 (25.32) 13/258 8/102 1.53 (0.64–3.70) 0.342
Q4 392 (27.57) 20/311 3/81 0.66 (0.20–2.22) 0.502
TBil. quantile
Q1 350 (24.61) 10/249 8/101 2.00 (0.76–5.25) 0.160 0.090
Q2 350 (24.61) 10/246 14/104 3.37 (1.50–7.58) 0.003
Q3 359 (25.25) 16/260 9/99 1.43 (0.63–3.23) 0.395
Q4 363 (25.53) 22/268 6/95 0.81 (0.33–2.00) 0.647

Subgroup analysis.

HR, hazard ratio; CI, confidence interval. Bold: Statistically different.

Among the metabolic profile subgroups, uric acid quartile analysis showed a significant interaction (P-interaction = 0.023): patients in the first quartile (lowest uric acid level) had a 6.5-fold higher risk of fatty liver disease (HR = 6.50, 95% CI: 2.75–15.34, P < 0.001). Diabetes stratification showed a stronger fatty liver association in non-diabetic patients (HR = 1.94, P = 0.015; diabetic patients HR = 0.90, P = 0.764), with a critically significant interaction (P-interaction = 0.073). In the lipid subgroups, the risk was significantly increased in the LDL ideal level group (HR = 2.19, P = 0.015) vs. the normal TC group (HR = 1.85, P = 0.017).

Liver function-related indicators showed differential associations: patients in the first tertile of ALT (lowest level) had a 4.38-fold higher risk of fatty liver (P = 0.047), while the third tertile of AST had the highest risk (HR = 3.80, P = 0.006). Bilirubin analysis showed a significant increase in risk in the first two tertiles of direct bilirubin (HR = 2.16, P = 0.047; HR = 2.34, P = 0.037), but total bilirubin was significant only in the second tertile (HR = 3.37, P = 0.003).

Discussion

In the 5-year retrospective cohort study of railway safety staff, we identified that MAFLD was significantly associated with an increased risk of incidence of gallstones compared to the control group (HR = 1.71, 95% CI: 1.13–2.58). To our knowledge, this is the first study to verify the association between MAFLD and incidence of gallstones using longitudinal follow-up data in a population with special occupation of railway safety staff. Our findings extend beyond previous cross-sectional evidence and provide new insights into the relationship between MAFLD and gallstones, with implications for targeted prevention strategies.

The observed association between MAFLD and incidence of gallstones in railway safety staff cohort provides further evidence supporting the metabolic interplay between hepatic steatosis and biliary disease. While previous studies have predominantly focused on general populations or clinical cohorts, our findings highlight unique risk patterns in an occupational group characterized by high-stress environments and irregular lifestyles (23). In a propensity-matched analysis of 7,922 adults from the 2017–2020 National Health and Nutrition Examination Survey (NHANES), the presence of MAFLD increased the odds of GD by 49% (adjusted OR = 1.49; 95 % CI: 1.27–1.75) after adjustment for BMI (24). Similarly, a prospective cohort study of 58,862 participants in China revealed that metabolically healthy obese people had a 1.95 times higher risk of developing gallstones than metabolically healthy normal-weight people (25).

Previous experimental studies previously have shown that a high-fat diet induces hepatic lipid accumulation and alters bile composition. Six weeks of a high-fat diet increased total cholesterol and triglycerides by 211 and 231%, respectively, and significantly activated sterol regulatory element-binding proteins (SREBPs), suggesting activation of the lipid synthesis pathway and an increase in bile cholesterol, which is directly related to gallstone formation (26) -fat (LD) diet increased biliary cholesterol secretion two-fold in mice and significantly elevated the cholesterol saturation index (CSI), which directly led to increased gallbladder cholesterol stone formation (27) findings suggest that dysregulated lipid metabolism in fatty liver creates a biliary environment that favors gallstone formation. Lipid disorders commonly associated with insulin resistance in patients with fatty liver may further promote gallstone formation by altering bile acid metabolism (28, 29). In addition, elevated leptin levels in patients with fatty liver disease may impair gallbladder contractility, leading to cholestasis and stone formation (30).

The risk of gallstones in the participants of the present study was predominantly male, which contrasts with epidemiologic data showing female susceptibility to gallstones, and may reflect the impact of gender differences in occupational stress on metabolic dysregulation in railway safety staff (31). We hypothesize that this may reflect the negative impact of occupational factors (such as shift work, circadian rhythm disruption, and high stress) on metabolic homeostasis, thereby amplifying the role of MAFLD in the risk of gallstones among male railway safety staff. The smaller sample size of women in this study may also be one possible reason for the discrepancy with other research findings. However, this interpretation requires validation based on additional occupational exposure data and prospective studies. A particularly novel finding is that MAFLD patients in the lowest uric acid quartile have a dramatically elevated 6.5-fold risk, suggesting that hypouricemia may paradoxically promote gallstone formation in the setting of hepatic steatosis. This view is consistent with emerging evidence that uric acid, in addition to its role in gout, regulates oxidative stress and lipid peroxidation-a key process in cholesterol crystallization (21). Mechanistically, low uric acid levels may impair antioxidant defenses, allowing reactive oxygen species to promote cholesterol nucleation in bile, but this hypothesis needs to be tested in experimental models (32). Furthermore, stratified analyses revealed differential effects of MAFLD on gallstone risk across varying levels of TC, LDL-C, HDL-C, and triglycerides, underscoring the central role of lipoprotein metabolism in gallstone pathogenesis. In particular, the overproduction of cholesterol-rich LDL by the steatotic liver may directly raise biliary cholesterol saturation, thereby accelerating gallstone formation (33).

In our study, there was no significant association between BMI and gallstones, although there was also a strong association in the general population, which may reflect occupational confounders such as high levels of physical activity counteracting the effects of obesity (34, 35). In the setting of MAFLD, mild to moderate hepatic injury can indicate GDrisk by elevated ALT and AST. A meta-analysis showed that the OR of gallstones increased by 7% for every 10 U/L increase in ALT, and the pooled OR of elevated AST was 1.19. The mechanism is as follows: increased permeability of hepatocyte membrane to ALT/AST leakage, accompanied by bile acid synthesis obstruction and bile cholesterol supersaturation; ALT/AST is also positively correlated with insulin resistance and inflammation, and can amplify the nucleation environment. Therefore, ALT/AST is not only a marker of liver injury, but also a simple predictor of GDsusceptibility in patients with fatty liver (36) association of direct bilirubin (DBil) and indirect bilirubin (IBil) with MAFLD and the development of gallstones may be due to the fact that these two markers are not only signaling for cholestasis and hepatocellular damage, but may also influence cholesterol crystallization through two pathways: oxidative-antioxidant homeostasis and lipid metabolism (37, 38).

Our research has several limitations. Firstly, as the study population was occupationally exposed, the generalizability of the findings is somewhat limited. Future studies should examine these associations among different occupational groups to assess the broader applicability of our research results. Secondly, the retrospective cohort of this study was established based on physical examination data for up to 5 years. Although the completeness and accuracy of the data was relatively high, confounding factors such as smoking and alcohol intake can affect the veracity of responses due to the ban on alcohol and tobacco by railroad safety staff to the extent that they were not collected through questionnaires or interviews. Although the study population was highly homogeneous, these potential confounding factors should be further considered in clinical interpretation and future research. Finally, due to baseline inclusion reasons, the baseline GD population was insufficient in our research design, making it impossible to determine the causal relationship between gallstones and the incidence of MAFLD. The bidirectional association between the two was not confirmed. Early inclusion in cohort studies for observation and more prospective studies on different populations are needed to confirm the causal relationship between MAFLD and incidence of gallstones relationships.

In conclusion, our study advances the understanding of MAFLD and incidence of gallstones relationships by identifying occupational and metabolic modifiers of risk. In this longitudinal cohort analysis, lipid, uric acid, and bilirubin levels all affected the risk of gallstone development with MAFLD. These findings support the consideration of integrative interventions with other metabolic markers within the existing MAFLD management framework. Follow-up research directions should focus on elucidating mechanistic studies of the pathophysiologic determinants of MAFLD combined with multiple metabolic factors affecting gallstone occurrence, and expanding the observed population to more specifically exposed populations of interest.

Statements

Data availability statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Ethics statement

The studies involving humans were approved by Staff Ethics Committee, Wuhan Wuchang Hospital, Wuhan, China. The studies were conducted in accordance with the local legislation and institutional requirements. The Ethics Committee/institutional review board waived the requirement of written informed consent for participation from the participants or the participants' legal guardians/next of kin because Retrospective study involving information that does not contain sensitive information.

Author contributions

LW: Writing – review & editing, Writing – original draft, Resources, Funding acquisition. HW: Writing – review & editing, Writing – original draft. YL: Project administration, Methodology, Data curation, Writing – original draft. JC: Writing – review & editing, Validation. WS: Supervision, Conceptualization, Investigation, Funding acquisition, Writing – review & editing. HL: Funding acquisition, Supervision, Writing – review & editing, Resources.

Funding

The author(s) declare that financial support was received for the research and/or publication of this article. This study was supported by the Natural Science Foundation of Hubei Province (2025AFB830) and Hubei Key Laboratory for Identification and Control of Occupational Hazards Joint Fund (No. JF2024-G17).

Acknowledgments

The authors would like to thank all the researchers for their joint efforts and the staff of the hospital's medical examination department for their support of the work of this project.

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declare that no Gen AI was used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

References

  • 1.

    Patel H Jepsen J . Gallstone disease: common questions and answers. Am Fam Physician. (2024) 109:51824.

  • 2.

    Graf S Cecchini M . Identifying patterns of unhealthy diet and physical activity in four countries of the Americas: a latent class analysis. Rev Panam Salud Publica. (2018) 42:e56. doi: 10.26633/RPSP.2018.56

  • 3.

    Obesity epidemiology worldwide . Gastroenterol Clin North Am. (2016) 45:5719. doi: 10.1016/j.gtc.2016.07.012

  • 4.

    Burden Burden of digestive diseases in the United States part III: liver biliary tract and pancreas . Gastroenterology. (2009) 136:113444. doi: 10.1053/j.gastro.2009.02.038

  • 5.

    Lammert F Wittenburg H . Gallstones: prevention, diagnosis, and treatment. Semin Liver Dis. (2024) 44:394404. doi: 10.1055/a-2378-9025

  • 6.

    Global epidemiology of gallstones in the 21st century: a systematic review and meta-analysis . Clin Gastroenterol Hepatol. (2024) 22:158695. doi: 10.1016/j.cgh.2024.01.051

  • 7.

    Unalp-Arida A Ruhl CE . Burden of gallstone disease in the United States population: prepandemic rates and trends. World J Gastrointest Surg. (2024) 16:113048. doi: 10.4240/wjgs.v16.i4.1130

  • 8.

    Zhou F Zhou J Wang W Zhang XJ Ji YX Zhang P et al . Unexpected rapid increase in the burden of NAFLD in China from 2008 to 2018: a systematic review and meta-analysis. Hepatology. (2019) 70:111933. doi: 10.1002/hep.30702

  • 9.

    Man S Deng Y Ma Y Fu J Bao H Yu C et al . Prevalence of liver steatosis and fibrosis in the general population and various high-risk populations: a nationwide study with 5.7 million adults in China. Gastroenterology. (2023) 165:102540. doi: 10.1053/j.gastro.2023.05.053

  • 10.

    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. J Hepatol. (2020) 73:2029. doi: 10.1016/j.jhep.2020.07.045

  • 11.

    Rinella ME Lazarus JV Ratziu V Francque SM Sanyal AJ Kanwal F et al . A multisociety Delphi consensus statement on new fatty liver disease nomenclature. Hepatology. (2023) 78:196686. doi: 10.1097/HEP.0000000000000520

  • 12.

    Cortés VA Barrera F Nervi F . Pathophysiological connections between gallstone disease, insulin resistance, and obesity. Obes Rev. (2020) 21:e12983. doi: 10.1111/obr.12983

  • 13.

    Liu J Lin H Zhang C Wang L Wu S Zhang D et al . Non-alcoholic fatty liver disease associated with gallstones in females rather than males: a longitudinal cohort study in Chinese urban population. BMC Gastroenterol. (2014) 14:213. doi: 10.1186/s12876-014-0213-y

  • 14.

    Koller T Kollerova J Hlavaty T Huorka M Payer J . Cholelithiasis and markers of nonalcoholic fatty liver disease in patients with metabolic risk factors. Scand J Gastroenterol. (2012) 47:197203. doi: 10.3109/00365521.2011.643481

  • 15.

    Biddinger SB Haas JT Yu BB Bezy O Jing E Zhang W et al . Hepatic insulin resistance directly promotes formation of cholesterol gallstones. Nat Med. (2008) 14:77882. doi: 10.1038/nm1785

  • 16.

    Caballero F Fernández A De Lacy AM Fernández-Checa JC Caballería J García-Ruiz C . Enhanced free cholesterol, SREBP-2 and StAR expression in human NASH. J Hepatol. (2009) 50:78996. doi: 10.1016/j.jhep.2008.12.016

  • 17.

    Colak Y Bozbey G Erim T Caklili OT Ulasoglu C Senates E et al . Impaired gallbladder motility and increased gallbladder wall thickness in patients with nonalcoholic fatty liver disease. J Neurogastroenterol Motil. (2016) 22:4706. doi: 10.5056/jnm15159

  • 18.

    Kichloo A Solanki S Haq KF Dahiya D Bailey B Solanki D et al . Association of non-alcoholic fatty liver disease with gallstone disease in the United States hospitalized patient population. World J Gastrointes Pathophysiol. (2021) 12:1424. doi: 10.4291/wjgp.v12.i2.14

  • 19.

    Gu S Hu S Wang S Qi C Shi C Fan G . Bidirectional association between NAFLD and gallstone disease: a systematic review and meta-analysis of observational studies. Expert Rev Gastroenterol Hepatol. (2023) 17:28393. doi: 10.1080/17474124.2023.2175671

  • 20.

    Konyn P Alshuwaykh O Dennis BB Cholankeril G Ahmed A Kim D . Gallstone disease and its association with nonalcoholic fatty liver disease, all-cause and cause-specific mortality. Clin Gastroenterol Hepatol. (2023) 21:9408.e2. doi: 10.1016/j.cgh.2022.04.043

  • 21.

    Chang Y Noh YH Suh BS Kim Y Sung E Jung HS et al . Bidirectional association between nonalcoholic fatty liver disease and gallstone disease: a cohort study. J Clin Med. (2018) 7:458. doi: 10.3390/jcm7110458

  • 22.

    Jiang GH Li S Li HY Xie LJ Li SY Yan ZT et al . Bidirectional associations among gallstone disease, non-alcoholic fatty liver disease, kidney stone disease. W J Gastroenterol. (2024) 30:491428. doi: 10.3748/wjg.v30.i46.4914

  • 23.

    Wang Z Liu J Li J Wu N Lu G Chen Z et al . Guidelines for lipid management in China (2023). Chin Circ Mag. (2023) 38:23771.

  • 24.

    R A Mary Vaishali P Parasuraman G . Beyond the badge: a scoping review among police personnel in South India exploring the impact of yoga and health education on cardiovascular Health. Cureus. (2024) 16:e68370. doi: 10.7759/cureus.68370

  • 25.

    Zhang Y Zhang J Yu D Tu F Liu J Han B et al . Association between metabolic dysfunction associated steatotic liver disease and gallstones in the US population using propensity score matching. Sci Rep. (2025) 15:910. doi: 10.1038/s41598-025-85218-5

  • 26.

    Man S Gao Y Lv J Tong M Yin J Wang B et al . Metabolically healthy obesity was significantly associated with increased risk of gallstones. Eur J Endocrinol. (2022) 186:27583. doi: 10.1530/EJE-21-0802

  • 27.

    Tuo QH Liang L Zhu BY Cao X Liao DF . Effect of Daxx on cholesterol accumulation in hepatic cells. World J Gastroenterol. (2008) 14:43540. doi: 10.3748/wjg.14.435

  • 28.

    Zommara MA Swelam S Raya-Álvarez E Imaizumi K Elmahdy A Alkhudhayri DA et al . Nutritional and potential health benefits of chufa oil, olive oil, and anhydrous milk fat against gallstone disease in a C57BL/6N mouse model. Front Nutr. (2024) 11:1445484. doi: 10.3389/fnut.2024.1445484

  • 29.

    Liew P-L Lee W-J Wang W Lee Y-C Chen W-Y Fang C-L et al . Fatty liver disease: predictors of nonalcoholic steatohepatitis and gallbladder disease in morbid obesity. Obes Surg. (2008) 18:84753. doi: 10.1007/s11695-007-9355-0

  • 30.

    Sun H Warren J Yip J Ji Y Hao S Han W et al . Factors influencing gallstone formation: a review of the literature. Biomolecules. (2022) 12:550. doi: 10.3390/biom12040550

  • 31.

    Sekine K Nagata N Sakamoto K Arai T Shimbo T Shinozaki M et al . Abdominal visceral fat accumulation measured by computed tomography associated with an increased risk of gallstone disease. J Gastroenterol Hepatol. (2015) 30:132531. doi: 10.1111/jgh.12965

  • 32.

    Yang T Zhong J Zhang R Xiao F Wang Y Tao H et al . Different types and numbers metabolic abnormalities and risk of gallbladder stone disease in adults. Front Nutr. (2024) 11:1443575. doi: 10.3389/fnut.2024.1443575

  • 33.

    Soomro U Abbas G Aslam M Kirshan Kumar S Hyder A Tareen K et al . Predictors of gallstones in patients with metabolic dysfunction-associated fatty liver disease (MAFLD). Cureus. (2025) 17:e85073. doi: 10.7759/cureus.85073

  • 34.

    Zhang YP Zhao YL Sun YL Zhu RT Wang WJ Li J . Physical activity and the risk of gallstone disease: a systematic review and meta-analysis. J Clin Gastroenterol. (2017) 51:85768. doi: 10.1097/MCG.0000000000000571

  • 35.

    Wang H So H Ko SW Jung SW Bang SJ Park EJ . Gallstone is associated with metabolic factors and exercise in Korea. Healthcare. (2022) 10:1372. doi: 10.3390/healthcare10081372

  • 36.

    Li S Zhu P Chen F Yu W Xie L Xia J et al . Liver function-related indicators and risk of gallstone diseases-a multicenter study and a systematic review and meta-analysis. Gastroenterol Res Pract. (2024) 2024:9097892. doi: 10.1155/2024/9097892

  • 37.

    Jansen T Daiber A . Direct antioxidant properties of bilirubin and biliverdin. is there a role for biliverdin reductase?Front Pharmacol. (2012) 3:30. doi: 10.3389/fphar.2012.00030

  • 38.

    Stocker R Glazer AN Ames BN . Antioxidant activity of albumin-bound bilirubin. Proc Natl Acad Sci U S A. (1987) 84:591822. doi: 10.1073/pnas.84.16.5918

Summary

Keywords

fatty liver, gallstones, cohort, railway safety staff, survival

Citation

Wei L, Wang H, Li Y, Cai J, Shang W and Liu H (2025) Association between metabolic dysfunction-associated fatty liver disease and risk of incidence of gallstones, evidence from a 5-year railway safety staff retrospective cohort study. Front. Med. 12:1687078. doi: 10.3389/fmed.2025.1687078

Received

16 August 2025

Accepted

03 November 2025

Published

09 December 2025

Volume

12 - 2025

Edited by

Jian Wu, Suzhou Municipal Hospital, China

Reviewed by

Yuzhu Dai, The 903rd Hospital of the PLA, China

Ze Xiang, Zhejiang University, China

Updates

Copyright

*Correspondence: Wen Shang, ; Hao Liu,

†These authors have contributed equally to this work and share first authorship

Disclaimer

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.

Outline

Figures

Cite article

Copy to clipboard


Export citation file


Share article

Article metrics