Integrated lifestyle management to mitigate MASLD risk in patients with chronic kidney disease: a prospective cohort study

Objectives

This study aimed to investigate whether effective management of multiple lifestyle risk factors could mitigate the elevated risk of metabolic dysfunction-associated steatotic liver disease (MASLD) among individuals with chronic kidney disease (CKD).

Methods

We analyzed data from the UK Biobank, including 9,877 patients with CKD and 421,043 non-CKD controls, with follow-up through October 31, 2022. Eight lifestyle-related MASLD risk factors (body mass index, diet, physical activity, alcohol consumption, smoking, social connection, sedentary behavior, and sleep duration) were assessed and jointly summarized to reflect the degree of lifestyle risk factor management. Cox proportional hazards models were used to evaluate associations between cumulative lifestyle risk factor management and incident MASLD. Effect modification by sex, type 2 diabetes (T2D), hyperlipidemia, and hypertension was examined using multiplicative interaction terms. All models were adjusted for a comprehensive set of covariates. Missing data were handled using multiple imputation with chained equations, and several sensitivity analyses were conducted to test the robustness.

Results

Among individuals with CKD, progressively better joint management of lifestyle risk factors was associated with a stepwise reduction in incident MASLD risk. Each additional lifestyle risk factor managed was associated with a 14% lower risk of MASLD. Optimal lifestyle risk factor control (≥ 7 factors managed) was associated with a 64% reduction in MASLD (HR: 0.36; 95% CI: 0.16–0.79). Notably, among CKD patients achieving optimal lifestyle control MASLD risk was attenuated to a level comparable to that of non-CKD controls (HR: 0.95; 95% CI: 0.49–1.84). The protective association of combined lifestyle risk factor management was less pronounced in individuals with hyperlipidemia or T2D compared to those without these conditions (P for interaction < 0.001).

Conclusions

In individuals with CKD, cumulative adherence to optimal lifestyle risk factor management exhibited an inverse association with MASLD development. These findings suggest that comprehensive control of modifiable lifestyle risks may be associated with a mitigation of the elevated MASLD susceptibility observed in CKD populations.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12967-026-07901-z.

Highlights

This prospective cohort study revealed that a higher degree of multifactorial lifestyle control was significantly associated with a reduced risk of MASLD among CKD patients.

Most importantly, our data suggested that achieving optimal control of lifestyle risk factors may attenuate the excess risk of MASLD associated with CKD.

These findings highlight the clinical and public health importance of holistic lifestyle management for mitigating MASLD risk in the CKD population.

Graphical Abstract

Supplementary Information

The online version contains supplementary material available at 10.1186/s12967-026-07901-z.

Introduction

Metabolic dysfunction-associated steatotic liver disease (MASLD) and chronic kidney disease (CKD) have emerged as major global public health challenges, with an estimated prevalence of approxiately32% and 15% among the adult population, respectively [1, 2]. MASLD, formerly known as non-alcoholic fatty liver disease, represents the most common form of chronic liver disorder worldwide [3]. Accumulating epidemiological evidence indicates a strong interrelationship between MASLD and CKD, individuals with MASLD have a substantially increased risk of developing CKD [4, 5], while the prevalence of MASLD is markedly higher among CKD patients than in the general population, suggesting a complex and potential bidirectional association mechanism between the two conditions [6, 7]. This interconnection is not solely from shared metabolic risk factors, such as obesity, insulin resistance, dyslipidemia, and hypertension. Emerging data suggest that overlapping pathophysiological mechanisms, including chronic low-grade inflammation, oxidative stress, and gut microbiota dysregulation, may contribute to the mutual progression of hepatic and renal dysfunction [4]. Given the heightened metabolic vulnerability and adverse clinical outcomes associated with CKD, preventing MASLD in this populations represents a critical intervention target, with potential benefits extending beyond liver-related morbidity to overall cardiometabolic health and disease-free longevity.

At present, the treatment options for MASLD remain limited, and lifestyle modification constitutes the cornerstone of disease management [8]. Evidence consistently demonstrates that individual lifestyle interventions, such as regular physical activity, reduced sedentary behavior, dietary adjustments, and weight reduction, can significantly improve liver function and metabolic profiles in MASLD population [9–12]. However, MASLD is a multifactorial disease influenced by a broader constellation of health behaviors. Lifestyle factors including smoking [13, 14], alcohol consumption [15], sleep duration [16, 17], and social connections [18, 19] have also been implicated in liver disease progression and may exert synergistic effects when combined with traditional metabolic risk factors. From a translational perspective, comprehensive characterization of modifiable lifestyle risks mafacilitate refined risk stratification, and inform the development of integrated, precision-based preventive strategies [18].

We hypothesized that cumulative and integrated management of modifiable lifestyle factors could substantially reduce the risk of MASLD among individuals with CKD. This hypothesis is partial supported by prior studies demonstrating a dose-dependent inverse association between adherence to multiple healthy lifestyle behaviors and MASLD incidence in general populations [13, 20]. However, no existing research has systematically examined the extent to which varying levels of combined lifestyle risk factor modify MASLD risk specifically within CKD populations. Therefore, using data from the UK Biobank, we prospectively evaluated the association between varying levels of combined lifestyle risk factor management and incident MASLD among individuals with CKD, and further assessed whether the optimal lifestyle control could attenuate the excess MASLD risk associated with CKD compared with non-CKD controls.

Currently, no existing research has systematically examined how varying levels of integrated lifestyle risk factor modification correlate with MASLD risk specifically in CKD populations. Accordingly, we prospectively evaluated how varying levels of combined lifestyle risk factor modification associate with MASLD development among CKD participants in the UK Biobank cohort. To assess whether the excess risk of MASLD associated with CKD can be reduced or eliminated, we conducted comparisons between CKD patients and non-CKD controls.

Methods

Study design

The UK Biobank is a large-scale, population-based prospective cohort study that enrolled over 500,000 individuals aged 40 to 69 from 22 centers across the UK between 2006 and 2010. The detailed information of the UK Biobank’s design and study population have been published previously and is available on the UK Biobank website: (https://www.ukbiobank.ac.uk/) [21]. Ethical approval was obtained from the North West Multi-Centre Research Ethics Committee, and written informed consent was secured from all participants. This research was conducted in compliance with the UK Biobank’s data access protocols and was approved under application number 300,908.

CKD was defined using a composite approach based on the International Classification of Diseases, 10th Revision-Clinical Modification (ICD-10-CM) diagnostic codes (N18.x), self-reported physician diagnosis of CKD, or an estimated glomerular filtration rate (eGFR) below 60 mL/min/1.73 m², whichever occurred first. The participant selection process for this study is illustrated in Fig. 1. Of 502,131 individuals initially recruited, 69,951 were excluded due to missing data on key lifestyle factors (body mass index (BMI), smoking, physical activity, alcohol consumption, diet, social connection, sedentary behavior assessed by television viewing, or sleep duration) or CKD status. In addition, 1,260 participants with prevalent MASLD at baseline were excluded to minimize reverse causation. The final analysis included 430,920 participants, including 9,877 individuals with CKD and 421,043 non-CKD at baseline. Participants were followed prospectively from enrollment until the first occurrence of MASLD, death, loss to follow-up, or the end of follow-up (October 31, 2022), whichever came first.

Fig. 1.

Flowchart of participant selection. MASLD, metabolic dysfunction-associated steatotic liver disease; CKD, chronic kidney disease; BMI, body mass index

Assessment of the lifestyle risk factor control

We evaluated eight key modifiable lifestyle risk factors previously implicated in MASLD development [19], including BMI, physical activity, sedentary behavior, diet, social connection, sleep duration, smoking, and alcohol consumption. Detailed assessment procedures for each lifestyle factor are provided in the Supplementary Material.

Never smoking was categorized as low risk. Regular physical activity was considered to be at least 150 min per week of moderate-intensity exercise, 75 min per week of vigorous-intensity exercise, or an equivalent combination of both [22]. Low-risk alcohol consumption was classified as moderate drinking, with a maximum of one standard drink per day for women and two standard drinks per day for men. In the U.K., one standard drink corresponds to 8 g of ethanol, and this amount should be consumed regularly rather than episodically [23]. For diet, it was assessed using a food-based scoring system, with a low-risk level defined as meeting intake of at least five of ten cardiometabolically beneficial food groups. These included increased intake of vegetables, fruits, dairy, fish, whole grains, and vegetable oils, as well as decreased consumption of refined grains, sugar-sweetened beverages, and both processed and unprocessed meats [24]. Body height and weight were measured at baseline, and BMI was calculated as weight (kg) divided by height squared (m²). According to the World Health Organization classification, a BMI between 18.5 and 24.9 kg/m² was considered low risk.

Sleep duration was categorized as low risk if participants reported 7–8 h of sleep per day, consistent with the established J-shaped association between sleep duration and adverse health outcomes [25]. Sedentary behavior was assessed using television viewing time, with <4 h per day classified as low-risk [26, 27]. Social connection was assessed using a composite measure incorporating household size, frequency of visits from friends/family, and engagement in leisure/social activities; absence of social isolation was defined as low risk [28].

Each lifestyle factor was assigned 1 point for a low-risk status and 0 points otherwise. A composite lifestyle score ranging from0 to8 was calculated by summing all eight factors, where higher scores indicating greater adherence to a healthy lifestyle pattern. To ensure adequate sample size within categories, the lifestyle score was grouped into six levels (≤ 2, 3, 4, 5, 6 and ≥ 7). Additionally, for analyses requiring broader classification, participants were categorized into three lifestyle groups: low or unfavorable lifestyle (≤ 2 low risk factors), moderate or intermediate lifestyle (3–5 low risk factors), and high or favorable lifestyle (≥ 6 low risk factors).

Outcomes

Prevalent MASLD was determined using hospital inpatient records and self-reported medical history. Information on admission dates and diagnostic codes was obtained through linkage with Health Episode Statistics for England and Wales and the Scottish Morbidity Records for Scotland. MASLD cased during follow-up was ascertained using ICD-10 hospital admission codes, specifically K76.0 (fatty liver disease) and K75.8 (other specified inflammatory liver conditions).

Assessment of covariates

Potential confounder, such as age, sex, ethnicity, education attainment, employment status, and the Townsend deprivation index (TDI), were accounted for in the analysis. The TDI, derived from national census data and postal codes, assesses deprivation based on factors like unemployment, overcrowding, and car and home ownership, with lower scores indicating higher socioeconomic status. Additionally, preexisting conditions including hypertension, hyperlipidemia, T2D, stroke and CHD were also included, alongside with relevant treatments such as lipid-lowering, antihypertensive, and antidiabetic medications. Biomarkers, including HbA1c, serum creatinine, triglycerides, cholesterol, low-density lipoprotein (LDL), C-reactive protein (CRP), Alanine Aminotransferase (ALT), and albumin were also incorporated. Information on preexisting conditions was obtained from self-reports and hospital inpatient records.

Statistical analysis

Baseline characteristics were summarized according to levels of lifestyle risk factor control among individuals with CKD and non-CKD. Continuous variables are presented as means ± standard deviations (SD), and categorical variables as counts and percentages. Group differences were assessed using the Kruskal-Wallis test, t-test, or chi-square test, as appropriate. Incidence rate of MASLD was calculated as the number of events divided by the total person-years of follow-up and expressed per 1,000 person-years. Cox proportional hazard regression models were used to examine the association between graded lifestyle risk factor control and incident MASLD among individuals with CKD, with individuals controlling ≤ 2 lifestyle risk factors serving as the reference group. Multivariable models were adjusted for age (years), sex, employment status (employed or not), education level (college/university degree or other), ethnicity (White British or other), TDI, serum cholesterol, low-density lipoprotein, triglycerides, CRP, HbA1c, ALT, albumin, hypertension, diabetes, stroke, CHD, hyperlipidemia, lipid-lowering therapy, glucose-lowering therapy, and antihypertensive therapy.

To examine whether lifestyle risk factor control modified the association between CKD status and MASLD risk, Cox models additionally constructed to compare MASLD incidence between CKD patients and non-CKD controls across strata of lifestyle risk factor control, adjusting for the same covariates. Effect modification by sex, T2D, hyperlipidemia, and hypertension was formally tested using multiplicative interaction terms, and stratified analyses were conducted accordingly. Cumulative incidence of MASLD across groups was estimated using the Kaplan-Meier method, with differences assessed by the log-rank test. To evaluate the non-linear relationship between lifestyle risk factor and MASLD risk, models incorporating restricted cubic spline terms were compared with linear models. Cox regression was further performed to assess association between lifestyle risk factor control and MASLD incidence in the overall population, as well as among non-CKD and CKD participants. Missing covariate data were addressed using multiple imputation with chained equations, with details on the proportion of missingness provided in Supplementary Table 1. Subgroup analyses were performed by age, sex, ethnicity, education level, employment status, and TDI. Sensitivity analyses included excluding MASLD events occurring within the first two years and within the first five years of follow-up to reduce potential reverse causation, as well as repeating analyses using complete-case data to assess the robustness of imputation assumptions. All statistical analyses were performed using R software (version 4.2.2). A two-sided P-value < 0.05 was considered statisticall significant.

Results

Baseline characteristics of the population

Baseline characteristics of participants with CKD non-CKD controls are shown in Table 1. Among the 9,877 CKD individuals, 6.5%, 15.6%, 27.1%, 27.8%, 16.0%, and 7.0% had ≤ 2, 3, 4, 5, 6, and ≥ 7 lifestyle risk factor controls, respectively. With the CKD group, higher levels of lifestyle risk factor control tended to be younger, female, have higher proportions of White people, social economic status, education levels, and employment rate. However, they have relatively lower usage rates of hypoglycemic drugs, antihypertensive drugs and lipid-lowering drugs. Furthermore, the baseline data of participants were categorized based on whether they developed MASLD, as outlined in Supplementary Table 2.

Table 1.

Baseline characteristics of the study group by the degree of joint lifestyle risk factor control

Variable	Degree of joint lifestyle risk factor control (CKD)							P value
	Non-CKD(n = 421,043)	≤2(n = 638)	3(n = 1,540)	4(n = 2,673)	5(n = 2,745)	6(n = 1,586)	≥7(n = 695)
Age, years	56.24 ± 8.07	62.42 ± 6.06	62.82 ± 5.49	62.57 ± 6.03	62.30 ± 6.35	61.62 ± 6.77	61.74 ± 6.63	<0.001
Male, sex	194,171 (46%)	337 (53%)	748 (49%)	1,287 (48%)	1,325 (48%)	716 (45%)	271 (39%)	<0.001
White, n (%)	400,733 (95%)	596 (93%)	1,445 (94%)	2,511 (94%)	2,583 (94%)	1,515 (96%)	664 (96%)	<0.001
TDI	-1.43 ± 3.01	-0.58 ± 3.66	-0.54 ± 3.38	-1.11 ± 3.11	-1.58 ± 2.91	-1.97 ± 2.67	-1.92 ± 2.69	<0.001
College or university degree	168,169 (40%)	193 (30%)	452 (29%)	827 (31%)	1,004 (37%)	598 (38%)	295 (42%)	<0.001
Currently employed	265,264 (63%)	141 (22%)	423 (27%)	888 (33%)	1,032 (38%)	676 (43%)	301 (43%)	<0.001
BMI, kg/m2	27.28 ± 4.70	31.22 ± 5.75	30.89 ± 5.47	30.07 ± 5.12	28.69 ± 4.70	27.11 ± 4.58	24.82 ± 3.51	<0.001
eGFR, mL/min/1.73 m2	91.73 ± 12.02	51.07 ± 13.18	51.34 ± 11.86	52.37 ± 11.11	52.80 ± 10.80	53.76 ± 11.37	54.89 ± 10.49	<0.001
HbA1C (%)	5.43 ± 0.59	5.99 ± 1.15	5.83 ± 0.86	5.74 ± 0.82	5.66 ± 0.78	5.57 ± 0.65	5.49 ± 0.53	<0.001
Serum creatinine, umol/L	71.22 ± 13.37	129.07 ± 82.73	126.25 ± 84.78	119.56 ± 57.40	118.73 ± 60.56	116.29 ± 52.19	109.51 ± 41.93	<0.001
Serum cholesterol, mmol/L	5.71 ± 1.13	5.15 ± 1.29	5.26 ± 1.28	5.32 ± 1.27	5.36 ± 1.27	5.43 ± 1.20	5.57 ± 1.18	<0.001
LDL, mmol/L	3.57 ± 0.86	3.16 ± 0.97	3.25 ± 0.94	3.29 ± 0.95	3.32 ± 0.95	3.36 ± 0.89	3.40 ± 0.87	<0.001
Triglycerides, mmol/L	1.73 ± 1.02	2.30 ± 1.26	2.20 ± 1.19	2.07 ± 1.10	1.95 ± 1.04	1.80 ± 0.97	1.63 ± 0.88	<0.001
C-reactive protein, mg/L	2.49 ± 4.19	5.66 ± 7.59	4.97 ± 7.64	4.03 ± 5.92	3.50 ± 5.83	3.01 ± 5.72	2.31 ± 3.98	<0.001
ALT, U/L	23.51 ± 14.07	23.24 ± 22.41	23.98 ± 13.72	23.31 ± 14.59	22.47 ± 12.51	21.59 ± 12.95	20.73 ± 10.12	<0.001
Albumin, g/L	45.26 ± 2.62	44.02 ± 2.93	44.13 ± 3.08	44.34 ± 2.95	44.47 ± 2.82	44.59 ± 2.83	44.72 ± 2.79	<0.001
Smoke								<0.001
Never	378,160 (90%)	404 (63%)	1,333 (87%)	2,483 (93%)	2,644 (96%)	1,564 (99%)	689 (99%)
Previous/current	42,883 (10%)	234 (37%)	207 (13%)	190 (7%)	101 (4%)	22 (1%)	6 (1%)
Alcohol consumptiona								<0.001
Low risk	302,106 (72%)	207 (32%)	788 (51%)	1,828 (68%)	2,183 (80%)	1,358 (86%)	654 (94%)
High risk	118,937 (28%)	431 (68%)	752 (49%)	845 (32%)	562 (20%)	228 (14%)	41 (6%)
Healthy diet b	141,827 (34%)	23 (3.6%)	176 (11%)	568 (21%)	1,046 (38%)	925 (58%)	571 (82%)	<0.001
Regular physical activity c	138,317 (33%)	18 (2.8%)	121 (7.9%)	409 (15%)	754 (27%)	797 (50%)	531 (76%)	<0.001
Less television watching timed	304,194 (72%)	73 (11%)	388 (25%)	1,180 (44%)	1,881 (69%)	1,354 (85%)	663 (95%)	<0.001
Adequate sleep duration e	288,333 (68%)	98 (15%)	513 (33%)	1,523 (57%)	2,057 (75%)	1,378 (87%)	666 (96%)	<0.001
Appropriate social connection f	384,257 (91%)	308 (48%)	1,203 (78%)	2,426 (91%)	2,620 (95%)	1,551 (98%)	688 (99%)	<0.001
Previous diseases
Hypertension	113,744 (27%)	477 (75%)	1,053 (68%)	1,769 (66%)	1,647 (60%)	841 (53%)	277 (40%)	<0.001
Type 2 diabetes	22,489 (5.3%)	161 (25%)	307 (20%)	455 (17%)	371 (14%)	155 (9.8%)	44 (6.3%)	<0.001
Hyperlipidemia	54,906 (13%)	262 (41%)	559 (36%)	903 (34%)	828 (30%)	385 (24%)	127 (18%)	<0.001
Stroke	5,572 (1.3%)	57 (8.9%)	105 (6.8%)	138 (5.2%)	106 (3.9%)	46 (2.9%)	14 (2.0%)	<0.001
CHD	21,144 (5.0%)	171 (27%)	322 (21%)	479 (18%)	408 (15%)	182 (11%)	63 (9.1%)	<0.001
Lipid-lowering medication	68,193 (16%)	349 (55%)	755 (49%)	1,235 (46%)	1,130 (41%)	540 (34%)	181 (26%)	<0.001
Antihypertensive medication	81,323 (19%)	420 (66%)	919 (60%)	1,574 (59%)	1,449 (53%)	744 (47%)	226 (33%)	<0.001
Glucose-lowering medication	4,062 (1.0%)	59 (9.2%)	103 (6.7%)	127 (4.8%)	123 (4.5%)	46 (2.9%)	8 (1.2%)	<0.001
Follow-up time (year)	13.66 ± 1.22	13.30 ± 2.10	13.43 ± 1.92	13.45 ± 1.66	13.56 ± 1.56	13.62 ± 1.23	13.55 ± 1.30	<0.001

Continuous variables are expressed as the mean [standard deviation], while categorical variables are presented as n (%)

a. Low-risk alcohol consumption is defined as moderate drinking, with a limit of one drink per day for women and two drinks per day for men (in the U.K., one drink contains 8 g of ethanol)

b. Refers to adequate intake of at least one-half of 10 recommended food groups

c. Regular physical activity is defined as engaging in a minimum of 150 minutes of moderate-intensity exercise per week, 75 minutes of vigorous activity, or a combination of both

d. Refers to <4 h per day of television watching

e. Adequate sleep duration is categorized as sleeping 7 to 8 hours per day

f. Appropriate social connection was defined as frequent social connection

TDI, Townsend deprivation index; BMI, body mass index; LDL, low-density lipoprotein; ALT, Alanine Aminotransferase; CHD, coronary heart disease; CKD, chronic kidney disease

CKD status, individual and combined lifestyle risk factor control, and MASLD risk

Compared with non-CKD individuals, those with CKD had a 49% higher risk of developing MASLD (HR: 1.49, 95% CI: 1.28–1.74). After adjusting for potential confounders, the analysis of the association between individual lifestyle risk factor control and MASLD indicated that control of these eight individual lifestyle risk factors could reduce the risk of MASLD to varying degrees. However, BMI was most closely associated with MASLD, meaning that reducing BMI has the greatest impact on lowering the risk of MASLD (HR: 0.66, 95% CI: 0.61–0.70). In addition, after adjusting for potential confounders, individuals in the high and moderate degrees of lifestyle risk factor control groups had HRs of 0.49 (95% CI, 0.45–0.53) and 0.72 (95% CI, 0.67–0.77), respectively, compared with those in the low control group (Supplementary Table 3).

Joint lifestyle risk factor control and MASLD risk among individuals with CKD

During a median follow-up of 13.5 years, 435 incident MASLD cases occurred among CKD patients, compared with 9,955 cases among non-CKD controls. Kaplan-Meier analyses demonstrated a graded increase in cumulative MASLD incidence across unfavorable, intermediate and favorable lifestyle categories in the overall population, with consistent patterns observed in both CKD and non-CKD groups (Fig. 2). Among CKD individuals, progressively higher levels of lifestyle risk factor control were associated with a significantly lower risk of incident MASLD (Table 2). For each additional lifestyle risk factor controlled, there was a 14% reduction in MASLD risk (HR: 0.86; 95% CI: 0.80–0.93). Participants achieving optimal lifestyle control (≥ 7 low risk factors) had the lowest MASLD risk (HR: 0.36; 95% CI: 0.16–0.79). These associations remained robust across multiple sensitivity analysis, including exclusion of MASLD events occurring in the first 2 and 5 years of follow-up and complete-case analyses without covariate imputation (Supplementary Table 4). Consistent inverse associations were also observed across subgroups by age, sex, ethnicity, education level, TDI, and employment status (Supplemental Table 5).

Fig. 2.

Association of degree of control lifestyle risk factors with risk of MASLD during the follow-up time in the (A) all population, (B) non-CKD, (C) CKD. MASLD, metabolic dysfunction-associated steatotic liver disease; CKD, chronic kidney disease; Unfavorable lifestyle: low (0–2), Intermediate lifestyle: moderate (3–5), Favorable lifestyle: high (6–8)

Table 2.

Adjusted ^a HR of incident MASLD according to the degree of joint lifestyle risk factor ^b control among individuals with CKD

	Degree of joint lifestyle risk factor control (CKD)						Per 1 risk factor control	P for trend
	≤2(n = 638)	3(n = 1,540)	4(n = 2,673)	5(n = 2,745)	6(n = 1,586)	≥7(n = 695)
Number of MASLD events	52	86	126	104	46	21	435	-
Incidence rate per 1,000 person-years	6.13	4.16	3.50	2.79	2.13	2.23	-	-
HR (95%CI)	1.00 (Ref)	0.88 (0.54, 1.42)	0.73 (0.46, 1.15)	0.53 (0.33, 0.86)	0.43 (0.25, 0.75)	0.36 (0.16, 0.79)	0.86 (0.80, 0.93)	<0.001

^a Model adjusted for age, sex, ethnicity, education, Townsend deprivation index, employment, serum cholesterol, low-density lipoprotein, triglycerides, CRP, HbA1c, ALT, albumin, hypertension, diabetes, stroke, CHD, hyperlipidemia, lipid-lowering therapy, glucose-lowering therapy, and antihypertensive therapy

^b Lifestyle risk factors included body mass index, alcohol consumption, smoking, diet, physical activity, sleep duration, social connection, sedentary behavior. MASLD, metabolic dysfunction-associated steatotic liver disease; CKD, chronic kidney disease; CHD, coronary heart disease; CRP, C-reactive protein; ALT, Alanine Aminotransferase; HR, hazard ratio; CI: confidence interval

Joint lifestyle risk factor control and MASLD risk in CKD compared with non-CKD participants

Restricted cubic spline (RCS) analyses revealed a significant inverse association between composite lifestyle risk factor control score and MASLD risk when modeled as a continuous variable (Fig. 3). We further examined whether graded lifestyle risk factor control could diminish the elevated MASLD risk associated with CKD (Table 3). Among CKD participants, MASLD incidence rates were highest in those with ≤ 2 controlled lifestyle factors and decreased progressively with increasing levels of control (6.13 vs. 4.16 vs. 3.50 vs. 2.79 vs. 2.13 vs. 2.03 cases per 1,000 person-years). The incidence rate of MASLD for control subjects was 1.73 per 1,000 person-years. Notably, the excess risk of MASLD associated with CKD among those jointly controlling 5 lifestyle risk factors was even attenuated to a level similar to that of non-CKD control participants (HR: 1.29; 95% CI: 1.02–1.58). Among CKD patients with ≥ 7 controlled lifestyle factors, the MASLD risk was not statistically significantly different from that of non-CKD controls (HR: 0.95; 95% CI: 0.49–1.84). Moreover, the robustness of the results was verified through multiple sensitivity analyses. Compared with non-CKD patients, the degree of control of lifestyle risk factors in CKD patients was consistent with the risk of MASLD as determined in the main analysis (Supplementary Table 6).

Fig. 3.

Restricted cubic splines for the association between degree of control lifestyle risk factor and incident risk of MASLD in (A) all population, (B) non-CKD, (C) CKD. Red lines and shaded areas represent the odds ratios and 95% confidence intervals, respectively. Adjustment for age, sex, ethnicity, education, employment, TDI, serum cholesterol, low-density lipoprotein, triglycerides, CRP, HbA1c, ALT, albumin, hypertension, diabetes, stroke, CHD, hyperlipidemia, lipid-lowering therapy, glucose-lowering therapy, and antihypertensive therapy. TDI, Townsend deprivation index; ALT, Alanine Aminotransferase; CRP, C-reactive protein; CHD, coronary heart disease; CKD, chronic kidney disease. MASLD, metabolic dysfunction-associated steatotic liver disease; T2D, type 2 diabetes; HR, hazard ratio; CI: confidence interval

Table 3.

Adjusted ^a HR of incident MASLD according to the degree of joint lifestyle risk factor ^b control compared with control subjects

	Degree of joint lifestyle risk factor control (CKD)
	Non-CKD(n = 421,043)	≤2(n = 638)	3(n = 1,540)	4(n = 2,673)	5(n = 2,745)	6(n = 1,586)	≥7(n = 695)
Number of MASLD events	9,955	52	86	126	104	46	21
Incidence rate per 1,000 person-years	1.73	6.13	4.16	3.50	2.79	2.13	2.03
HR (95%CI)	1.00 (Ref)	2.09 (1.39, 3.14)	1.86 (1.40, 2.46)	1.65 (1.30, 2.09)	1.29 (1.02, 1.58)	1.07 (0.74, 1.55)	0.95 (0.49, 1.84)

^a Model adjusted for age, sex, ethnicity, education, Townsend deprivation index, employment, serum cholesterol, low-density lipoprotein, triglycerides, CRP, HbA1c, ALT, albumin, hypertension, diabetes, stroke, CHD, hyperlipidemia, lipid-lowering therapy, glucose-lowering therapy, and antihypertensive therapy

^b Lifestyle risk factors included body mass index, alcohol consumption, smoking, diet, physical activity, sleep duration, social connection, sedentary behavior. MASLD, metabolic dysfunction-associated steatotic liver disease; CKD, chronic kidney disease; CHD, coronary heart disease; CRP, C-reactive protein; ALT, Alanine Aminotransferase; HR, hazard ratio; CI: confidence interval

Interaction analyses

We further investigated the potential interactions with sex and disease status (hypertension, hyperlipidemia, and T2D) (Fig. 4A and B). In our study, a significant interaction (P for interaction < 0.001) was observed between lifestyle risk factors and hyperlipidemia and T2D. The protective effect of comprehensive lifestyle management against MASLD was significantly attenuated in participants with hyperlipidemia or T2D compared to those without (interaction p < 0.001) (Fig. 4B). However, neither group achieved statistical significance in the high degree of lifestyle risk factors control group (Fig. 4B).

Fig. 4.

Stratified analysis for adjusted HRs of incident MASLD. (A) Sex and Hypertension. (B) Hyperlipidemia and T2D. Diamonds represent point estimates for HRs compared with non-CKD control subjects. Horizontal lines represent the range for 95% CI. The Cox proportional hazards model was adjusted for age, sex, ethnicity, education, employment, TDI, serum cholesterol, low-density lipoprotein, triglycerides, CRP, HbA1c, ALT, albumin, hypertension, diabetes, stroke, CHD, hyperlipidemia, lipid-lowering therapy, glucose-lowering therapy, and antihypertensive therapy. TDI, Townsend deprivation index; ALT, Alanine Aminotransferase; CRP, C-reactive protein; CHD, coronary heart disease; CKD, chronic kidney disease. MASLD, metabolic dysfunction-associated steatotic liver disease; T2D, type 2 diabetes; HR, hazard ratio; CI: confidence interval; IR, Incidence rate per 1,000 person-years

Discussion

In this large-scale prospective cohort study of UK Biobank participants, we provide novel evidence that CKD is associated with substantially increased risk of MASLD, and that cumulative adherence to healthy lifestyle is strongly associated with a reduction in this excess risk. Over a median follow-up of 13.5 years, each additional lifestyle risk factor adequately controlled was associated with a 14% lower risk of incident MASLD among individuals with CKD, with optimal lifestyle control (≥ 7 factors) corresponding to a 64% risk reduction. Importantly, among CKD participants jointly controlling 5 lifestyle risk factors, the excess MASLD risk relative to non-CKD controls was markedly attenuated, and among those achieving optimal control (≥ 7 factors), MASLD risk was no longer statistically distinguishable from that of individuals without CKD. These findings were robust across multiple subgroup and sensitivity analyses.

MASLD and CKD have overlapping pathophysiological pathways and shared risk factors [29, 30]. MASLD may independently promote the onset and worsening of CKD, beyond conventional risk factors. Meanwhile, CKD can exacerbate cardiometabolic dysregulation associated with obesity and contribute to the progression of MASLD [2, 31]. Whether adherence to a healthy lifestyle can modify these effects remain unknown.

Our study highlights a significantly elevated MASLD risk in CKD patients, mitigated by integrated lifestyle interventions. This association extends beyond shared metabolic disturbances (e.g., insulin resistance, inflammation) to reflect dysregulation of the interconnected “liver-kidney axis” or “metabolic organ network” [32]. Recent insights underscore that MASLD and CKD interact bidirectionally via metabolites, inflammatory mediators, and gut-derived signals, creating a self-perpetuating pathological cycle [33, 34]. Additionally, prior studies have shown that the risk of MASLD can be lowered by managing multiple risk factors simultaneously [35, 36], which aligns with evidence from studies on polyphenols, and flavonoids demonstrating multi-pathway protective effects in complex disease models [37, 38].Our study expands this evidence by focusing on individuals with CKD and by directly evaluating whether integrated lifestyle management can attenuate CKD-associated excess MASLD risk.

Unlike studies examining individual lifestyle risk factors in isolation, our results reveal a clear graded association between the degree of combined lifestyle risk factor control and MASLD incidence. Compared with non-CKD participants, individuals with CKD exhibited progressively lower MASLD risk as lifestyle factor control improved, ranging from a more than twofold excess risk among those with unfavorable lifestyles to a non-significant difference among those achieving optimal control. This pattern mirrors findings from cardiovascular prevention studies in T2D, where comprehensive risk factor control substantially reduces the excess cardiovascular risk in this population [21, 39, 40]. Collectively, these findings underscore the importance of multifactorial prevention strategies in complex, multisystem diseases.

We further observed that the protective association of lifestyle management was attenuated among CKD participants with comorbid T2D or hyperlipidemia. This likely reflects the dominant role of profound insulin resistance and inflammation constitute the dominant pathological drivers of MASLD, potentially diminishing the relative impact of lifestyle modifications [41, 42]. These findings align with prior evidence indicating that T2D promotes both MASLD and CKD progression primarily through hyperglycemia and insulin resistance, while hyperlipidemia exacerbates risk via disordered lipid metabolism and enhanced inflammatory responses [43–47]. Consequently, these findings underscore the necessity for risk-stratified management in CKD. For patients with existing metabolic comorbidities, lifestyle intervention must be reinforced as a fundamental component of care and integrated early with pharmacological strategies to achieve synergistic control over metabolic and hepatic risks [48, 49]. Compared to controlling a single risk factor, the reduction in MASLD risk achieved through joint risk factor control is significantly more substantial. A cross-sectional study found that in CKD patients, those with MASLD treated with Angiotensin-Converting Enzyme Inhibitor (ACEI) or Angiotensin Receptor Blocker (ARB) had lower liver stiffness, but there was no statistical difference in the degree of fibrosis or grading of fatty liver between the two groups. This suggests that drug therapy aimed solely at improving renal function may have limited benefits in improving liver lesions in MASLD [50]. Compared to single-component approaches, multifactorial intervention trials targeting cardiovascular disease or mortality have demonstrated superior risk reduction [19, 51–53].

The precise biological pathways underlying the cumulative benefits of joint lifestyle risk factor control in reducing MASLD risk were not fully understood yet. However, multiple plausible interpretations exist. These lifestyle risk factors are known to increase MASLD risk via multiple pathways, implying that integrated control could produce synergistic protective effects. These risk factors each instigate damage through unique biological pathways. For instance, nicotine promotes the progression of MASLD to MASH (metabolic dysfunction-associated steatohepatitis) by activating the AMPKα-SMPD3 axis and increasing the production of intestinal ceramides [54]. Pro-inflammatory diets significantly increases the risk of severe MASLD through multiple mechanisms, including activating inflammatory responses, increasing oxidative stress, exacerbating metabolic disorders, causing gut microbiota dysbiosis, and altering gene expression [55, 56]. Weight management not only mitigates cardiovascular and diabetes risk factors but can also lead to liver disease remission [57]. These findings mirror the multi-target protective effects observed with natural products such as polyphenols, naringin, and targeted compounds modulating ATR-CHK1 and other upstream pathways [58]. Therefore, the joint control of these risk factors may help to mitigate the concentrated harmful effects and interrupt the compounded impacts on liver injury. In the context of CKD, systemic and local inflammation, as well as oxidative stress [59], are key factors driving liver injury, especially in the progression of MASLD [60, 61]. In the context of CKD, systemic and local inflammation as well as oxidative stress is critical drivers of liver injury, particularly during MASLD progression. Combined management of risk factors may synergistically mitigate the heightened MASLD risk associated with CKD, highlighting the necessity of addressing multiple factors rather than focusing on a single target. Nevertheless, further in-depth mechanistic research is required to fully elucidate the protective pathways conferred by this integrated approach in preventing MASLD among CKD patients.

Strengths and limitations

Key strengths of this study include its large sample size, prospective design, long-term follow-up, and comprehensive assessment of lifestyle. Several limitations need to be considered. First, lifestyle factors were assessed at baseline only, precluding evaluation of changes over time. The use of baseline assessments and potential behavioral changes over time may lead to non-differential misclassification, which could bias the effect estimates toward the null (i.e., result in conservative estimates). Second, the selected lifestyle factors, while evidence-based and clinically relevant, may not represent the optimal or exhaustive set of modifiable factors. Third, the generalizability of the findings was somewhat restricted, given that over 90% of the UK Biobank cohort consisted predominantly of White individuals who exhibited healthier behaviors than the general UK population. Fourth, as with any observational studies, there was a possibility of residual confounding despite the thorough adjustment for covariates. Fifth, the exclusive use of inpatient records to diagnose incident MASLD could have caused under-detection, as this approach would miss asymptomatic cases not requiring hospital care. In summary, this method offers high specificity but may have limited sensitivity.

Conclusions

This large prospective cohort study demonstrates that cumulative adherence to healthy lifestyle behaviors is strongly and inversely associated with MASLD incidence among individuals with CKD. Our findings suggest that optimal multifactorial lifestyle management may substantially mitigate, and potentially neutralize, the excess MASLD risk associated with CKD. These findings highlight the potential importance of holistic lifestyle management strategies for mitigating MASLD risk in this study population.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Chao Yan contributed to the conception and design of the work. Xinxin Shen contributed to the acquisition, analysis, and interpretation of data for the work. Chao Yan and Xinxin Shen drafted the manuscript. Shaojie Han, Jinzheng He and Tongxu Wang critically revised the manuscript and optimized the graph visualization. Jindong Ding Petersen and Xiaobing Gong contributed to funding acquisition and project concept and management, and they also investigated and supervised the original data as well as the manuscript. All coauthors gave final approval and agreed to be accountable for all aspects of work ensuring integrity and accuracy.

Funding

This study was supported by the Key Science and Technology Project of Hainan Province (ZDYF2024SHFZ064), Hainan Natural Science Foundation Innovation Research Team Project (825CXTD610), and Hainan Medical University “Leading Talent” Scientific Research Initiation Project Fund (XRC2022005).

Data availability

All data analysed during the current study are publicly available upon request at https://biobank.ndph.ox.ac.uk/ukb/. Data sets used for the analysis will be made available under reasonable requests.

Declarations

Ethics approval and consent to participate

UK Biobank received ethical approval from the North West Multi-centre Research Ethics Committee. All participants gave written informed consent before enrolment in the study, which was conducted in accord with the principles of the Declaration of Helsinki.

Consent for publication

Not applicable.

Conflict of interest

All authors have no conflicts of interest to declare.