Association between cataract and fatty liver diseases from a nationwide cross-sectional study in South Korea

This study examined the link between fatty liver disease (FLD) and cataracts, as previous research has suggested that FLD may contribute to metabolic syndrome, systemic inflammation, and potentially cataracts. We studied a nationwide cross-sectional cohort of the Fifth Korean National Health and Nutrition Examination Survey 2010–2011. FLD was defined as nonalcoholic FLD (NAFLD) and metabolic dysfunction-associated FLD (MAFLD). Multinomial logistic regression was utilized to investigate the relationship between cataracts and FLD after adjustment for potential confounders. Participants with cataracts had higher liver fibrosis scores, including the NAFLD fibrosis score (NFS; P < 0.001), fibrosis-4 index (FIB4; P < 0.001), and fatty liver index (FLI; P = 0.001). NAFLD was not associated with a higher odds ratio (OR) for cataracts in the fully adjusted model (OR = 1.23, P = 0.058). MAFLD was significantly associated with a higher OR (OR = 1.34, P = 0.006). After adjusting for all factors, the severity of FLD was linked to an increased risk of cataracts, with significant linear trends (P values for linear trends of NFS, FIB4, and FLI < 0.05). After adjusting for well-known cataract risk factors, MAFLD was significantly associated with cataracts. Our analysis suggests that FLD may serve as an independent risk factor for cataracts.


Dataset
We used a comprehensive cross-sectional health examination dataset based on the fifth Korean National Health and Nutrition Examination Survey (KNHANES V; available online at https:// knhan es.kdca.go.kr/ knhan es/ eng/ index.do) conducted between 2010 and 2011.The data collection protocol was approved by the Institutional Review Board of the Korean Centers for Disease Control and Prevention (no.2010-02CON-21-C and 2011-02CON-06-C).Additionally, this study was exempted from ethical review by the Institutional Review Board of the Korean National Institute for Bioethics Policy because it used publicly accessible data.All participants signed forms consenting to the use of their health information.The KNHANES is a nationwide, populationbased, cross-sectional survey conducted by the Division of Chronic Disease Surveillance of the Korea Centers for Disease Control and Prevention 15 .All participants were randomly selected using stratified sampling in which the following factors were considered: population, sex, age, regional area, and residential area.KNHANES comprised health records based on health interviews, health examinations, and nutrition surveys.Each participant provided health and socioeconomic information regarding age, household income, alcohol use, smoking status, hypertension, and diabetes 16 .Alcoholics were surveyed through a survey on whether they had experience with counseling for drinking problems.All participants underwent blood tests to evaluate their lipid profiles and liver enzyme levels after overnight fasting.
In this study, Fig. 1 displays the inclusion and exclusion criteria.The KNHANES surveyed comprehensive eye examinations and liver enzyme, including gamma-glutamyl transferase (GGT), during 2010 and 2011.KNHANES datasets from difference time periods were excluded because of the lack of GGT or cataract evaluation.Initially, 17,476 participants were included in the KNHANES 2010-2011 dataset.We excluded 5,348 participants with incomplete ophthalmologic examinations, 7,765 participants aged (< 40 years) with incomplete

Definition of FLD
As the KNHANES did not conduct liver steatosis imaging tests, we adopted indirect indicators commonly utilized in large-scale epidemiological studies.For determining NAFLD and MAFLD, we referred to recent literature 17 .NAFLD was defined using previously validated formulas, such as the NAFLD fibrosis score and fibrosis-4 index (FIB4) 18,19 .Threshold values were applied using those described in a previous study 17 .MAFLD was defined using a previous formula based on the fatty liver index (FLI) and metabolic syndromes 20 .For MAFLD diagnosis, at least one evidence of metabolic syndrome must be present while satisfying an FLI ≥ 30 21 .The detailed formulas used to calculate the FLIs are described in the Supplementary Materials (See Supplementary Table S1).They were calculated based on a previous study using data from the KNHANES 22 .This score-based approach has been widely adopted in large epidemiological studies 23 , and showed very good predictability in a FLD-detection validation study 24 .According to the protocol of the previous study 17 , serum albumin was excluded from the NFS calculation due to insufficient data in KNHANES.NFS, FIB4, and FLI values calculated for analysis were coded into quartiles.

Determining cataract status
Detailed protocols for determining cataract status in the KNHANES have been reported 11,25 .Eye examination quality was controlled by the Epidemiologic Survey Committee of the Korean Ophthalmologic Society (KOS).Participating ophthalmologists or residents were periodically trained by acting staff members of the National Epidemiologic Survey Committee of the KOS.The data quality and collection protocols were verified by the Korea Disease Control and Prevention Agency (KDCA).
In the KNHANES, a slit-lamp (Haag-Streit model BQ-900; Haag-Streit AG, Koeniz, Switzerland) examination was performed without pupil dilation 26 .The overall characteristics of the crystalline lens were examined to determine cataract subtype and severity level.Each lens layer was examined from the anterior to the posterior capsule using a focused slit lamp.During the examination, the Lens Opacities Classification System III (LOCS III) was used to evaluate cataract status 27 .It should be noted that the absence of pupil dilation made the diagnosis of peripheral lens opacities difficult and influenced the detection of cataracts in the peripheral lens.Investigators compared crystalline lens conditions with standard photographs of LOCS III images.Cataract subtypes were categorized as cortical (LOCS III score ≥ C2 for cortical opacity), nuclear (LOCS III score ≥ 4 for nuclear opalescence or nuclear color [NO4 or NC4]), and posterior subcapsular (LOCS III score ≥ P2 for posterior subcapsular opacity) cataracts (See Supplementary Figure S1) 25 .Pseudophakic eyes or eyes with a history of cataract surgery were also classified into the cataract group.For the subtype analysis, participants who had undergone cataract surgery in one eye were categorized as having pseudophakia.Those with primarily cortical cataracts in one eye and no other cataract types were classified as having pure cortical cataracts.Participants with nuclear cataracts were defined similarly.Participants who had minor subtypes of cataract (e.g., anterior and posterior subcapsular cataracts) or a combination of multiple subtypes simultaneously were classified as having other or mixed cataracts.The KNHANES has binarized cataract grading data into presence or absence of each subtype in the raw data according to the official research protocol established by the KOS and KDCA.These criteria and data have been utilized and validated in previous epidemiological studies 28,29 .Moreover, a similar protocol using the LOCS III system to determine cataracts has been verified in a well-known epidemiological study conducted in a Chinese population 30 .

Statistical analysis
The variables according to the cataract or FLD groups were compared using the Wilcoxon rank-sum test for continuous data and χ 2 test for categorical data.We used multinomial logistic regression to estimate the adjusted odds ratios (ORs) for each cataract type compared with the reference of participants without cataracts.Accordingly, logistic regression models determined the association between FLD (MALFD/NAFLD) and cataracts in the primary analysis while accounting for covariates.The adjusted ORs were calculated as follows: (1) model 1, wherein the ORs were adjusted for a priori covariates, including age, sex, and body mass index; and (2) model 2, wherein ORs were adjusted for the following covariates: age, sex, body mass index, current smoking status, alcohol consumption, hypertension, and diabetes mellitus.The KNHANES did not collect information on steroid use.Statistical significance was set at P < 0.05.Analyses used Statistical Package for the Social Sciences Statistics 25.0 (SPSS Inc., Chicago, IL, USA).P values in logistic regression model were adjusted using Benjamini-Hochberg false discovery rate (FDR) correction for multiple testing correction manner 31 .To address confounder collinearity (age, sex, body mass index, current smoking, alcohol consumption, hypertension, and diabetes mellitus), we assessed the correlation matrix, confirming no multicollinearity (all pairwise Pearson's correlation coefficients were < 0.5).

Informed consent
All participants signed forms that consented to the use of their health information during data collection (KNHANES V; available online at https:// knhan es.kdca.go.kr/ knhan es/ eng/ index.do).

Results
Table 1 displays participant characteristics based on their NAFLD or MAFLD status.According to our definitions, 2007 participants (47.3%) had NAFLD, while 2050 (48.3%) had MAFLD.Participants with FLD exhibited higher body mass index (P < 0.001) and a greater prevalence of metabolic conditions, including hypertension and diabetes (P < 0.001).They also showed higher incidence of cataracts or pseudophakia (P < 0.001).Sex, smoking status, and alcohol intake demonstrated significant differences in relation to the FLD classification.www.nature.com/scientificreports/and alkaline phosphatase, and lower total cholesterol, high-density lipoprotein cholesterol, alanine aminotransferase, and platelet count.Participants with cataracts had higher liver fibrosis scores, including NFS (P < 0.001), FIB4 (P < 0.001), and FLI (P = 0.001).
The ORs for each cataract subtype were calculated to examine the association among subtypes and FLD.Table 4 shows the results of multinomial logistic regression analysis for each cataract subtype.In Model 1, NAFLD showed a statistically significant association with all cataract subtypes.After adjusting for all factors (model 2) with FDR correction, NAFLD was not associated with any cataract subtype (FDR-adjusted P = 0.062 for cortical cataracts, 0.155 for nuclear cataracts, 0.694 for other or mixed cataracts, and 0.151 for pseudophakia).MAFLD was significantly associated with all cataract subtypes in Model 1.In the fully adjusted model (Model 2), MAFLD consistently showed significant associations with cortical (FDR-adjusted P = 0.045), nuclear (FDRadjusted P = 0.043), and other or mixed cataract (FDR-adjusted P = 0.027) subtypes.However, there was no significant associations with pseudophakia (FDR-adjusted P = 0.090).

Discussion
In this population-based cross-sectional study of South Korean adults, we found that MAFLD was significantly associated with cataracts after adjusting for general cataract risk factors, including age, diabetes, and hypertension.After adjusting for all the factors, NAFLD was found to not be associated with any cataract subtype.This

Table 4.
Results of multinomial logistic regression for analysis of the odds ratios for cataract subtype risk.CI: confidence interval, FLD: fatty liver disease, MAFLD: metabolic dysfunction-associated fatty liver disease, NAFLD: nonalcoholic fatty liver disease, OR: odds ratio.*Each OR analysis was adjusted for covariates including age, sex, and body mass index.† Each analysis for the ORs was adjusted for the covariates including age, sex, body mass index, current smoking, alcohol consumption, hypertension, and diabetes mellitus.‡ P values were adjusted by Benjamini-Hochberg false discovery rate (FDR) correction for the multiple testing correction manner.www.nature.com/scientificreports/indicates that MAFLD may be an independent cataract risk factor.Significant relationships were observed with all cataract subtypes; however, no significant relationships were reported for pseudophakia.Our analysis confirmed that overall cataract risk increased in participants with FLD.Furthermore, FLD severity (NFS, FIB4, and FLI) was associated with an increased cataract risk, with significant linear trends.To the best of our knowledge, this is the first study to evaluate the relationship between FLD and cataracts.FLD is a common chronic liver disease and a global health concern 32 , with a prevalence of 25-40% worldwide 33 .As we excluded participants aged < 40 years, our study showed a similar prevalence, highlighting FLD as a significant issue in Korean society.Recently, FLD was recognized as an important pathological condition associated with liver inflammation and metabolic degeneration, and various health problems such as stroke, obstructive sleep apnea, atherosclerosis, heart failure, peripheral artery disease, and chronic kidney disease 34 .Because the eye is also an organ rich in blood vessels, it is also impacted FLD progression.
MAFLD exhibited a stronger relationship with cataracts compared with NAFLD, primarily because of its comprehensive inclusion of liver cirrhosis resulting from alcohol consumption and various metabolic diseases.Considering that cataract can be caused by various metabolic and inflammatory conditions, we hypothesized a link between FLD and cataract development.Figure 3 illustrates the potential pathogenesis of cataracts in patients with FLD, where both metabolic and inflammatory mediators of FLD may contribute to cataract development.NAFLD is a known risk factor for metabolic and cardiovascular diseases 23 , while MALFD, recently renamed to highlight the metabolic factors of FLD, demonstrated a similar association with other chronic diseases 21 .FLD is now recognized as a chronic inflammatory condition associated with type 2 diabetes and dyslipidemia 32 .Previous studies have focused on the relationship between diabetes and cataracts 1 , where the intracellular sorbitol accumulation in the crystalline lens leads to osmotic stress and liquefaction of lens fibers 35 .In patients with diabetes, hyperglycemia accelerates sorbitol production.Because the cytokines in FLD directly affect insulin resistance in the liver 32 , FLD could exacerbate diabetic cataract development.FLD is also closely associated with dyslipidemia 36 , and disruption in lipid metabolism caused by FLD may contribute to cataract development 37 .Furthermore, hepatokines, signaling proteins secreted by the liver, can reach the eyes 38 , while inflammation and oxidative stress resulting from FLD could further contribute to development of cataract.
Although the impact of diabetes on the eye has been extensively studied 39 , the association between liver diseases and ocular conditions remain unexplored.A recent deep learning-based study demonstrated the potential for detecting hepatobiliary diseases through slit-lamp and fundus photography images 40 .Although the prediction accuracy was not high, the study indicated a connection between eye and liver disorders, including liver cirrhosis, hepatitis, and NAFLD.A literature review highlighted the role of circulating hepatokines, such as fibroblast growth factor-21, hepatocyte growth factor, and angiopoietin-like proteins, in FLD, which can contribute to the development of pterygium, age-related macular degeneration, and diabetic retinopathy 38 .Disturbances in glucose, lipids, and amino acids may affect the eyes through various mechanisms 41 , although epidemiological evidence is currently lacking.This pathogenesis may also influence cataract development, as demonstrated in our large cross-sectional cohort study.Further investigations are needed to validate these potential mechanisms linking FLD and ophthalmic diseases.
One strength of our study is that we analyzed data from the KNHANES, a nationwide survey conducted on a large scare.Therefore, we produced representative results regarding the relationship between cataracts and FLD, with minimal selection bias.Another advantage is that we analyzed both NAFLD and MAFLD to investigate the influence of the broad spectrum of FLD, rather than specific conditions alone.We also studied the dose-response association between well-defined fatty liver indices, including NFS, FIB4, and FLI.Additionally, we adjusted for a wide range of cataract-related covariates to determine the independent association between FLD and cataract risk.
This study has several limitations.First, the cross-sectional nature of our study hinders us from establishing causality between FLD and cataracts.To confirm the associated, a longitudinal controlled research design is required.Second, this study was conducted in a single Asian country, raising uncertainty about the generatability of our findings to other countries or ethnic groups.Third, measurements of glucose levels, cholesterol profile, liver enzyme profile, and body mass index may vary depending on the timing of sample collection.Fourth, the Figure 3. Schematic diagram of possible cataract pathogenesis in patients with fatty liver disease.FLD is a chronic inflammatory condition associated with type 2 diabetes and dyslipidemia 32,36 .In patients with diabetes, hyperglycemia accelerates sorbitol production in the crystalline lens 35 .Hepatokines can reach the eyes, while inflammation and oxidative stress resulting from FLD could further contribute to development of cataract 38  www.nature.com/scientificreports/lack of information on steroid use in the KNHANES dataset is another limitation.Because prolonged glucocorticoid use is a major risk factor for posterior subcapsular cataracts 42 , the absence of this information might have confounded our results.However, in this study, most patients had nuclear and cortical cataracts; therefore, the confounding effects of steroid use were estimated to be small.Fifth, the KNHANES did not record the detailed grades according to LOCS III.The cataract grading based on LOCS III might be subjective and inconsistent without pupil dilation.To overcome this problem, the KOS established clear criteria for each cataract subtype and collected data with binary outcomes.The KOS has appropriately trained investigators to effectively collect and analyze data for large epidemiological studies.

Conclusion
This study revealed that FLD was significantly associated with a higher cataract risk in Korean adults.After adjustment for well-known cataract risk factors, MAFLD exhibited a strong association with cataracts.Our comprehensive analysis suggests that FLD may serve as an independent risk factor for cataracts.These findings underscore the importance of addressing modifiable risk factors, such as diabetes, hypertension, and FLD in the prevention of cataracts.Further larger-scale, longitudinal, and controlled studies are required to validate the impact of FLD on cataracts.

Figure 1 .
Figure 1.Study dataset flowchart in the Korean National Health and Nutrition Examination Survey (KNHANES).
data collection protocol was approved by the Institutional Review Board of the Korean Center for Disease Control and Prevention (No. 2010-02CON-21-C and 2011-02CON-06-C).Additionally, this study was exempt from ethical review by the Institutional Review Board of the Korean National Institute for Bioethics Policy because it used publicly accessible data.https://doi.org/10.1038/s41598-023-50582-7

Figure 2 .
Figure 2. Plots for adjusted odds ratios (ORs) for cataract risk according to quartile of fatty liver indexes.Model 1 was adjusted for age, sex, and body mass index.Model 2 was adjusted for age, sex, body mass index, current smoker, alcohol consumption, hypertension, and diabetes mellitus.CI, confidence interval; FLI, fatty liver index; FIB4, fibrosis-4 index; NFS, nonalcoholic fatty liver disease fibrosis score; OR, odds ratio.
Figure 3. Schematic diagram of possible cataract pathogenesis in patients with fatty liver disease.FLD is a chronic inflammatory condition associated with type 2 diabetes and dyslipidemia32,36 .In patients with diabetes, hyperglycemia accelerates sorbitol production in the crystalline lens35 .Hepatokines can reach the eyes, while inflammation and oxidative stress resulting from FLD could further contribute to development of cataract38 .

Table 2
presents clinical measurements related to FLD (variables for FLD scores) based on cataract presence.Participants with cataracts demonstrated higher waist circumference, fasting glucose, aspartate aminotransferase,

Table 1 .
Comparison of general and clinical characteristics among participants with and without fatty liver disease.MAFLD: metabolic dysfunction-associated fatty liver disease, NAFLD: nonalcoholic fatty liver disease.

Table 2 .
Clinical measurements and fatty liver indices of participants with and without cataracts.MAFLD: metabolic dysfunction-associated fatty liver disease, NAFLD: nonalcoholic fatty liver disease.Cataract (N