Sage Journals: Discover world-class research

Abstract

Objective

This study aimed to evaluate the associations between body mass index (BMI), dietary nutrient intake, and the risk of branch retinal vein occlusion (BRVO) in a nationally representative Korean population, with a focus on sex- and BMI-specific differences.

Methods

This population-based cross-sectional study analyzed data from adults aged ≥40 years who participated in the 7th and 8th Korea National Health and Nutrition Examination Surveys (KNHANES) conducted between 2017 and 2021. BRVO was diagnosed using standardized fundus photography and optical coherence tomography interpreted by certified retinal specialists. BMI was categorized according to Asia–Pacific criteria, and dietary intake of macronutrients and micronutrients was assessed using structured nutrition surveys. Comparisons were performed between participants with and without BRVO, stratified by sex and BMI category. Survey-weighted logistic regression models were applied to account for the complex sampling design and to adjust for age, smoking status, alcohol consumption, hypertension, diabetes mellitus, and BMI.

Results

A total of 15,790 participants (6,629 men and 9,161 women) were included, of whom 96 (43 men and 53 women) were diagnosed with BRVO. Participants with BRVO were significantly older and had a higher prevalence of hypertension (all P < 0.001). Overall, individuals with BRVO showed significantly lower intakes of protein, crude fiber, iron, sodium, and potassium (all P < 0.05). Although overall BRVO prevalence did not differ significantly across BMI categories, sex-stratified analyses revealed an increasing prevalence with higher BMI in women. Associations between dietary nutrient intake and BRVO varied across BMI strata.

Conclusions

Lower nutrient intake and metabolic risk factors, including hypertension and BMI, may be associated with BRVO risk. The observed sex- and BMI-specific differences highlight the importance of tailored nutritional and metabolic risk assessment. Longitudinal studies are warranted to clarify causal relationships.

Keywords

branch retinal vein occlusion nutritional intake body mass index diet hypertension

Introduction

Retinal vein occlusion (RVO), encompassing both central (CRVO) and branch (BRVO) subtypes, is the second most common retinal vascular disorder after diabetic retinopathy.^1,2 More than half of patients with RVO regain functional visual acuity sufficient for daily activities when they receive appropriate treatment.^3,4 RVO occurs more frequently in retinal veins than arteries and often carries a relatively favorable prognosis.⁵ Recent global epidemiologic data indicate that RVO is a relatively common retinal vascular disorder, with an overall pooled prevalence of approximately 0.8% in the adult population, and branch retinal vein occlusion accounting for the majority of cases.⁶ This considerable worldwide burden highlights the need to better understand modifiable metabolic and lifestyle-related risk factors for RVO. In the Korean population aged ≥19 years, the prevalence of RVO was reported to be 0.6% (standard error 0.1%) between 2008 and 2012, and the incidence of clinically diagnosed RVO was approximately 48 per 100,000 person-years.⁷

Advancing age is a major risk factor for RVO.⁸ Previous studies have reported that certain cardiovascular and systemic conditions, particularly hypertension, are associated with RVO, whereas the associations with other factors such as smoking or generalized atherosclerotic changes have been inconsistent across populations and analyses.^9–12 Metabolic syndrome, defined by elevated blood pressure, hyperglycemia, dyslipidemia, and central obesity, has been shown to increase the risk of RVO among smokers compared with non-smokers.¹³ Furthermore, recent large population-based cohort studies have reported that patients with RVO have an increased risk of cardiovascular morbidity and all-cause mortality, suggesting that RVO may be a marker of systemic vascular disease.^14,15

Dietary patterns and specific nutrient intake are increasingly recognized as modifiable determinants of cardiovascular health. Diets rich in fish, fruits, vegetables, whole grains, and nuts, as well as nutrients such as omega-3 fatty acids, antioxidant vitamins (A, C, D), and folic acid, may improve vascular health and potentially reduce the risk of RVO.¹⁶ However, no Korean study has investigated the associations of systemic conditions, body mass index (BMI), and nutrient intake with BRVO.

This study aimed to evaluate the associations between BMI, dietary nutrient intake, and the risk of BRVO in a nationally representative Korean population using a recently released dataset from the Korean National Health and Nutrition Examination Survey (KNHANES) conducted between 2017 and 2021.

Methods

This study was based on data obtained from the 7th and 8th Korea National Health and Nutrition Examination Survey (KNHANES), a nationwide, population-based program administered by the Ministry of Health and Welfare of Korea. All KNHANES participants had provided written informed consent, and the survey protocol complied with the Declaration of Helsinki of 1975 as revised in 2024. The present analysis of anonymized public data was additionally approved by the Institutional Review Board of Kangbuk Samsung Hospital, Seoul, Republic of Korea on August 23rd, 2024 (IRB No. 2024-08-022). This study was reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines.¹⁷

Study population

This population-based cross-sectional study analyzed data from adults aged ≥ 40 years who participated in the 7th (2016–2018) and 8th (2019–2021) KNHANES. Each survey cycle consisted of three components: a health interview, a nutrition survey, and a health examination. KNHANES is a nationwide survey conducted by the Korea Disease Control and Prevention Agency (KDCA) using a multistage, stratified cluster sampling design based on the National Census Registry. A rolling sampling method was employed to ensure comparability across survey years and maintain national representativeness.¹⁸ The quality of ophthalmologic data was validated by the Epidemiological Survey Committee of the Korean Ophthalmological Society. Sampling weights were applied to account for the complex survey design, non-response, and post-stratification,¹⁹ enabling estimation of population-level parameters for the Korean adult population.²⁰ The detailed methodology of KNHANES has been described previously.²¹

The number of CRVO cases in the KNHANES dataset was very small (1 case in 2017, 2 in 2018, 1 in 2019, 3 in 2020, and none in 2021). Therefore, CRVO was not included in the analysis to avoid unstable estimates and potential bias.

Data collection and outcomes

Demographic and clinical variables were collected through structured health interviews and physical examinations, which included blood pressure measurements, anthropometric assessments, laboratory tests, and ophthalmologic evaluations. Ophthalmic examinations were conducted as part of the physical assessment, during which 45° non-mydriatic color fundus photographs were obtained using a fundus camera (TRC-NW6S; Topcon, Tokyo, Japan), and optical coherence tomography (OCT) was performed using the CIRRUS HD-OCT 500 device (Carl Zeiss Meditec, Dublin, CA, USA).

Branch retinal vein occlusion (BRVO) was diagnosed based on characteristic sectoral retinal hemorrhages with associated venous dilatation or tortuosity confined to the distribution of a branch retinal vein on fundus photographs acquired with the Visucam 224 (Carl Zeiss Meditec AG, Germany) and interpreted by board-certified retinal specialists designated by the Korean Retina Society. Each image was independently evaluated by two graders, and in cases of disagreement, final adjudication was made by a senior review committee within the Korean Retina Society.²¹

Body mass index (BMI) was calculated as weight in kilograms divided by height in meters squared (kg/m²), and categorized into four groups: underweight (<18.5 kg/m²), normal weight (18.5–22.9 kg/m²), overweight (23.0–24.9 kg/m²), and obese (≥25.0 kg/m²).

Dietary intake was assessed using detailed food frequency questionnaires administered during health interviews. The analyzed nutrients included total food intake, energy, macronutrients (protein, fat, carbohydrate), and crude fiber, as well as a range of micronutrients, such as calcium, phosphorus, iron, sodium, potassium, and vitamins. Specifically, vitamin intake components included vitamin A, β-carotene, retinol, thiamin, riboflavin, niacin, and vitamin C. Nutrient intake levels and BMI categories were compared between participants with and without RVO, stratified by sex.

Hypertension was defined as systolic blood pressure ≥140 mmHg, diastolic blood pressure ≥90 mmHg or current use of antihypertensive medication. Diabetes mellitus was defined as a fasting plasma glucose level ≥126 mg/dL, glycated hemoglobin (HbA1c) level ≥6.5%, self-reported physician diagnosis, or current use of antidiabetic medication.

Statistical analysis

Continuous variables are presented as means with standard deviations (SD), and categorical variables are presented as numbers and percentages. Group comparisons were performed using the svyttest and svychisq functions from the survey package in R for continuous and categorical variables, respectively. The associations between nutritional intake and BRVO were evaluated using survey-weighted logistic regression models implemented through the svyglm function in R. Both univariate and multivariate analyses were performed. The multivariate models were adjusted for potential confounders including age, smoking status (current/past vs never), alcohol consumption (current vs never), BMI, presence of hypertension, and diabetes mellitus. Results are presented as odds ratios (ORs) with 95% CIs. For subgroup analyses stratified by BMI categories, separate logistic regression models were fitted for each BMI group. The BMI categories were defined as underweight (<18.5 kg/m²), normal weight (18.5-22.9 kg/m²), overweight (23.0-24.9 kg/m²), and obese (≥25.0 kg/m²) according to the Asian-Pacific criteria.

The complex sampling design of KNHANES was incorporated into all statistical analyses using the ‘survey’ package in R. Survey weights were applied to obtain nationally representative estimates. All statistical analyses were performed using R software version 4.4.1 (R Foundation for Statistical Computing, Vienna, Austria), and all statistical tests were two-sided with p-values less than 0.05 considered statistically significant.

This study is a secondary analysis of the KNHANES 2017–2021 data, which is a nationwide survey conducted by the Korea Disease Control and Prevention Agency (KDCA). Therefore, the sample size was not determined by a prospective power calculation but was defined by the number of eligible participants available in the dataset. Specifically, out of the total 38,678 participants in the survey period, we included all 15,790 individuals who met our inclusion criteria (age ≥ 40 years with complete ophthalmologic and nutritional data) to maximize statistical power and precision. As this is one of the largest population-based studies on BRVO to date, the sample size provides sufficient statistical power to detect meaningful associations, as reflected in the narrow confidence intervals of our estimates. An item non-responder analysis was performed by comparing baseline characteristics between included and excluded participants using standardized mean differences (SMDs). To improve model stability and reduce the potential impact of sparse data in certain BMI subgroups, additional sensitivity analyses were conducted by collapsing BMI categories into three broader groups (<23, 23–24.9, and ≥25 kg/m²). Continuous nutritional variables were standardized and analyzed per 1–standard deviation (1-SD) increment. Potential non-linear associations between dietary intake and BRVO risk were further evaluated using restricted cubic spline models.

Results

From 2017 to 2021, 38,678 individuals participated in Korea National Health and Nutrition Examination Survey (KNHANES). Among them, 15,790 participants (6,629 men and 9,161 women) were included in the final analysis (Figure 1). Of these, 96 individuals were diagnosed with branch retinal vein occlusion, including 43 men and 53 women.

Figure 1.

Flowchart of participant selection and inclusion in the Korea national health and nutrition examination survey (KNHNES) between 2017 and 2021.

Table 1 presents demographic characteristics in male and female patients with and without BRVO. Both age and hypertension were significantly associated with BRVO in men and women (P = 0.001 and P < 0.001, respectively). In men, smoking and high-density lipoprotein (HDL) cholesterol were additionally identified as significant factors (P < 0.05). Among women, significant associations were observed with body mass index (BMI), waist circumference, and both systolic and diastolic blood pressure (P < 0.05). Notably, diabetes mellitus (DM) was not significantly associated with BRVO in either sex (P > 0.05).

Table 1.

Demographics of study participants.

Variables	Men (n = 6,629)			Women (n = 9,161)
Variables	BRVO (n = 43)	No RVO (n = 6,586)	P value	BRVO (n = 53)	No RVO (n = 9,108)	P value
Age, years	62.58 (13.37)	55.35 (15.38)	0.001	67.36 (10.77)	55.89 (14.53)	< 0.001
Current smoker, %	37 (86.05)	5,051 (79.57)	0.014	6 (11.32)	859 (9.68)	0.832
Drinker, %	41 (95.35)	6,074 (93.40)	0.260	35 (66.04)	7,295 (80.91)	0.166
Body mass index, kg/m²	24.63 (3.36)	24.45 (3.26)	0.724	24.65 (3.77)	23.69 (3.54)	0.045
Waist circumference, cm	88.66 (8.62)	87.58 (9.19)	0.856	83.70 (9.22)	80.71 (9.82)	0.012
Fasting glucose, mg/dL	110.10 (24.93)	106.34 (25.81)	0.274	99.80 (17.79)	100.56 (22.00)	0.761
Total cholesterol, mg/dL	175.24 (42.44)	188.41 (39.36)	0.303	190.69 (37.47)	195.28 (38.88)	0.550
HDL-C, mg/dL	43.72 (9.68)	47.25 (11.42)	0.018	53.19 (12.61)	54.40 (12.65)	0.474
Triglycerides, mg/dL	139.76 (77.81)	157.56 (127.35)	0.281	117.47 (65.53)	116.65 (77.70)	0.576
Diabetes, %	6 (13.95)	835 (12.68)	0.912	7 (13.21)	876 (9.62)	0.947
Hypertension, %	22 (51.16)	1,883 (28.60)	< 0.001	35 (66.04)	2,396 (26.31)	< 0.001
Systolic BP, mmHg	125.73 (16.31)	121.94 (15.10)	0.080	133.89 (20.32)	119.43 (17.52)	< 0.001
Diastolic BP, mmHg	77.04 (10.48)	76.93 (10.48)	0.494	77.99 (9.58)	74.03 (9.66)	0.006

Abbreviations: BP: Blood Pressure; BRVO: branch retinal vein occlusion; HDL-C: high-density lipoprotein cholesterol; RVO: retinal vein occlusion.

Dietary intake was analyzed between patients with and without BRVO after adjustment for age (Table 2). The BRVO group demonstrated significantly lower intake of protein, crude fiber, iron, sodium, and potassium (all P < 0.05).

Table 2.

Dietary intake of study participants with and without retinal vein occlusion (RVO).

Variables	Mean (SE)		P value	Adjusted P value^a
Variables	BRVO (n = 96)	No RVO (n = 15,694)	P value	Adjusted P value^a
Intake of food, g	1,359.38 (8.61)	1,560.63 (3.14)	0.006	0.165
Energy, kcal	1,739.74 (9.34)	1,912.09 (3.13)	0.044	0.475
Protein intake, g	56.76 (1.76)	70.16 (0.67)	< 0.001	0.031
Fat intake, g	34.4 (1.67)	45.43 (0.67)	< 0.001	0.178
Carbohydrate intake, g	261.79 (3.63)	281.21 (1.17)	0.136	0.215
Crude fiber intake, g	22.52 (1.22)	26.46 (0.43)	0.007	0.014
Calcium intake, mg	459.51 (7.13)	506.29 (1.84)	0.355	0.537
Phosphorus intake, mg	919.69 (7.11)	1,063.24 (2.41)	0.004	0.088
Iron intake, mg	8.85 (0.77)	10.67 (0.29)	0.002	0.038
Sodium intake, mg	2,652.29 (13.26)	3,337.25 (4.62)	< 0.001	0.013
Potassium intake, mg	2,432.32 (12.02)	2,825.47 (4.14)	0.006	0.042
Vitamin A intake, μgRE	356.13 (6.35)	393.79 (2.13)	0.346	0.749
β-carotene intake, μg	2,682.65 (17.94)	2,963.04 (5.97)	0.376	0.484
Retinol intake, μg	132.87 (5.29)	146.87 (1.82)	0.617	0.747
Thiamin intake, mg	1.03 (0.25)	1.21 (0.09)	0.003	0.122
Riboflavin intake, mg	1.33 (0.36)	1.58 (0.11)	0.055	0.461
Niacin intake, mg	10.54 (0.82)	12.72 (0.3)	0.001	0.149
Vitamin C intake, mg	51.69 (2.69)	66.91 (1.17)	0.037	0.276

P values resulting from general linear model for continuous variables.

Abbreviations: BRVO: branch retinal vein occlusion; RVO: retinal vein occlusion; SE: standard error.

^aAdjusted for age.

Compared with included participants, excluded individuals were substantially younger and had lower metabolic parameters, including BMI and waist circumference, whereas differences in other variables such as sex, diabetes, and hypertension were minimal (Supplemental Table S1).

The prevalence of BRVO according to BMI category is presented in Table 3. In the total population, the prevalence of BRVO tended to increase with higher BMI, ranging from 0.07% in individuals with BMI <18.5 to 0.44% in those with BMI ≥25, although this association was not statistically significant (P = 0.234). When stratified by sex, no significant relationship between BMI and BRVO prevalence was observed in men (P = 0.550). In contrast, women demonstrated a significant positive association between higher BMI and BRVO prevalence (P = 0.031), with the prevalence rising from 0.08% in the underweight group to 0.48–0.51% in the overweight and obese groups.

Table 3.

The prevalence of BRVO according to BMI.

	BMI category								P value
	<18.5		18.5‒23		23‒25		≥25
	n (%)	95% CI	n (%)	95% CI	n (%)	95% CI	n (%)	95% CI
Total	4,332 (13.95)	13.57-14.34	11,191 (36.03)	35.50-36.57	6,157 (19.82)	19.38-20.27	9,377 (30.19)	29.68-30.71
BRVO	3 (0.07)	0.01-0.20	22 (0.20)	0.12-0.30	29 (0.47)	0.32-0.68	41 (0.44)	0.31-0.59	0.234
Men	1,903 (13.78)	13.21-14.37	4,070 (29.48)	28.72-30.25	3,028 (21.93)	21.25-22.63	4,804 (34.80)	34.00-35.60
BRVO	1 (0.05)	0.00-0.29	10 (0.25)	0.12-0.45	13 (0.43)	0.23-0.73	19 (0.40)	0.24-0.62	0.550
Women	2,429 (14.08)	13.56-14.61	7,121 (41.28)	40.54-42.02	3,129 (18.14)	17.56-18.72	4,573 (26.51)	25.85-27.17
BRVO	2 (0.08)	0.01-0.30	12 (0.17)	0.09-0.29	16 (0.51)	0.29-0.83	22 (0.48)	0.30-0.73	0.031

P values resulting from general linear model.

Abbreviations: BMI: body mass index; BRVO: branch retinal vein occlusion; CI: confidence interval.

Age-adjusted differences in dietary intake between the BRVO and non-RVO groups across BMI categories in men and women are summarized in Tables 4 and 5. As previously shown in Table 3, BRVO prevalence increased with higher BMI, with statistical significance observed only in women. In the BMI ≥25 category, dietary intake did not differ significantly between BRVO and non-RVO groups in either sex. Across various nutrients, including macronutrients and micronutrients, no statistically significant differences were observed (P > 0.05).

Table 4.

Difference of dietary intake between BRVO and non-RVO groups in men.

Variables	BMI category
	<18.5		18.5‒23		23‒25		≥25
	OR (95% CI)	P value	OR (95% CI)	P value	OR (95% CI)	P value	OR (95% CI)	P value
Intake of food, g	1.37 (1.09-1.74)	0.009	0.37 (0.19-0.73)	0.004	0.55 (0.30-1.01)	0.055	0.88 (0.59-1.32)	0.538
Energy, kcal	0.92 (0.76-1.12)	0.398	0.35 (0.18-0.68)	0.002	0.83 (0.56-1.22)	0.344	0.86 (0.50-1.49)	0.593
Protein intake, g	0.82 (0.69-0.98)	0.027	0.28 (0.10-0.76)	0.012	0.57 (0.33-0.96)	0.034	0.64 (0.34-1.21)	0.173
Fat intake, g	0.30 (0.22-0.41)	<0.001	0.21 (0.04-1.07)	0.061	0.71 (0.45-1.11)	0.135	0.55 (0.30-1.02)	0.057
Carbohydrate intake, g	1.47 (1.20-1.79)	<0.001	0.59 (0.36-0.97)	0.038	0.69 (0.40-1.20)	0.191	0.88 (0.48-1.64)	0.697
Crude fiber intake, g	1.14 (0.92-1.41)	0.232	0.66 (0.38-1.16)	0.148	0.37 (0.17-0.83)	0.016	0.85 (0.40-1.80)	0.669
Calcium intake, mg	0.63 (0.50-0.79)	<0.001	0.14 (0.02-0.86)	0.034	0.72 (0.31-1.68)	0.450	0.86 (0.46-1.61)	0.642
Phosphorus intake, mg	0.64 (0.53-0.78)	<0.001	0.30 (0.13-0.71)	0.006	0.64 (0.40-1.02)	0.063	0.72 (0.41-1.26)	0.249
Iron intake, mg	0.22 (0.15-0.32)	<0.001	0.52 (0.11-2.46)	0.413	0.76 (0.41-1.41)	0.378	0.90 (0.58-1.37)	0.613
Sodium intake, mg	0.92 (0.74-1.13)	0.412	0.35 (0.09-1.36)	0.129	0.30 (0.15-0.61)	0.001	0.76 (0.52-1.10)	0.150
Potassium intake, mg	1.01 (0.84-1.21)	0.894	0.45 (0.26-0.76)	0.003	0.53 (0.29-0.98)	0.042	0.66 (0.32-1.37)	0.269
Vitamin A intake, μgRE	0.25 (0.16-0.39)	<0.001	0.03 (0.00-0.60)	0.023	0.35 (0.09-1.39)	0.136	0.92 (0.49-1.70)	0.787
β-carotene intake, μg	0.78 (0.60-1.01)	0.067	0.13 (0.02-1.00)	0.050	0.32 (0.08-1.39)	0.129	1.08 (0.68-1.74)	0.739
Retinol intake, μg	0.01 (0.01-0.04)	<0.001	0.01 (0.00-6.17)	0.169	0.79 (0.22-2.87)	0.720	0.62 (0.20-1.95)	0.417
Thiamin intake, mg	0.96 (0.81-1.13)	0.590	0.49 (0.32-0.74)	0.001	0.59 (0.35-1.01)	0.055	0.61 (0.21-1.75)	0.355
Riboflavin intake, mg	0.49 (0.38-0.64)	<0.001	0.26 (0.05-1.35)	0.110	0.42 (0.22-0.83)	0.012	0.73 (0.40-1.31)	0.294
Niacin intake, mg	0.52 (0.41-0.66)	<0.001	0.51 (0.12-2.17)	0.360	0.62 (0.34-1.12)	0.116	0.57 (0.29-1.11)	0.100
Vitamin C intake, mg	1.48 (0.96-2.29)	0.081	0.20 (0.01-4.84)	0.324	0.17 (0.02-1.37)	0.096	0.87 (0.25-3.01)	0.826

Table 5.

Difference of dietary intake between BRVO and non-RVO groups in women.

Variables	BMI category
	<18.5		18.5‒23		23‒25		≥25
	OR (95% CI)	P value	OR (95% CI)	P value	OR (95% CI)	P value	OR (95% CI)	P value
Intake of food, g	0.45 (0.16-1.24)	0.005	0.73 (0.27-1.95)	0.526	0.45 (0.15-1.30)	0.141	0.97 (0.57-1.63)	0.900
Energy, kcal	0.96 (0.93-1.00)	0.087	0.85 (0.46-1.60)	0.623	0.38 (0.17-0.84)	0.017	0.97 (0.53-1.77)	0.915
Protein intake, g	0.03 (0.00-0.48)	0.015	0.75 (0.37-1.51)	0.417	0.23 (0.09-0.57)	0.002	0.80 (0.45-1.41)	0.439
Fat intake, g	0.94 (0.84-1.06)	0.081	0.65 (0.25-1.66)	0.369	0.32 (0.11-0.89)	0.029	0.98 (0.53-1.82)	0.961
Carbohydrate intake, g	1.02 (0.63-1.65)	0.116	0.87 (0.39-1.95)	0.735	0.37 (0.19-0.73)	0.004	1.02 (0.63-1.65)	0.933
Crude fiber intake, g	1.15 (0.98-1.35)	0.015	0.92 (0.39-2.14)	0.843	0.34 (0.12-1.03)	0.057	0.81 (0.53-1.24)	0.337
Calcium intake, mg	0.70 (0.34-1.44)	0.333	0.63 (0.25-1.60)	0.333	0.74 (0.26-2.05)	0.557	1.43 (0.70-2.94)	0.328
Phosphorus intake, mg	0.95 (0.92-0.98)	<0.001	0.73 (0.33-1.63)	0.438	0.41 (0.17-1.01)	0.052	1.10 (0.59-2.06)	0.766
Iron intake, mg	0.01 (0.00-13.19)	0.205	0.70 (0.34-1.41)	0.316	0.23 (0.11-0.50)	<0.001	0.66 (0.30-1.45)	0.301
Sodium intake, mg	0.63 (0.07-5.39)	0.672	0.95 (0.33-2.71)	0.916	0.37 (0.09-1.58)	0.181	0.62 (0.33-1.17)	0.142
Potassium intake, mg	0.78 (0.69-0.89)	<0.001	0.70 (0.26-1.87)	0.475	0.45 (0.14-1.45)	0.179	1.06 (0.62-1.80)	0.842
Vitamin A intake, μgRE	0.69 (0.05-9.77)	0.783	1.08 (0.80-1.45)	0.618	0.99 (0.58-1.71)	0.983	1.15 (0.98-1.35)	0.099
β-carotene intake, μg	1.02 (0.74-1.42)	0.126	1.26 (0.98-1.63)	0.074	0.60 (0.20-1.81)	0.361	1.02 (0.73-1.43)	0.899
Retinol intake, μg	1.15 (0.90-1.46)	0.264	0.32 (0.04-2.56)	0.284	1.09 (0.99-1.21)	0.073	1.13 (1.00-1.28)	0.045
Thiamin intake, mg	0.24 (0.00-110.83)	0.647	0.87 (0.61-1.23)	0.417	0.20 (0.09-0.45)	<0.001	1.02 (0.74-1.42)	0.903
Riboflavin intake, mg	0.44 (0.03-7.30)	0.566	0.49 (0.18-1.31)	0.154	0.57 (0.20-1.63)	0.297	1.33 (0.66-2.69)	0.432
Niacin intake, mg	1.15 (0.98-1.35)	0.400	0.84 (0.39-1.81)	0.658	0.14 (0.04-0.43)	0.001	1.04 (0.60-1.79)	0.890
Vitamin C intake, mg	0.31 (0.01-6.86)	0.457	0.77 (0.31-1.90)	0.568	0.88 (0.47-1.64)	0.683	0.93 (0.58-1.47)	0.744

In the BMI 23–25 category, the overall differences were minimal in both sexes. Among women, iron intake was significantly lower in the BRVO group (r = –0.25, P < 0.001), whereas most other nutrients did not differ. In men, no significant differences were observed in any dietary variable within this BMI range.

In contrast, among men with BMI 18.5–23, several nutrients showed significant differences. Protein intake was lower in the BRVO group (r = –0.03, P = 0.012), along with fat intake (r = –0.04, P = 0.061) and potassium (r = –0.00, P = 0.003). Additionally, reductions in micronutrients such as vitamin A and β-carotene were observed. In women within the same BMI category, no significant dietary differences were found.

In the underweight group (BMI <18.5), both men and women with BRVO had lower food intake compared to non-RVO groups (men: r = 0.00, P = 0.009; women: r = –0.00, P = 0.002).

Additional sensitivity analyses using collapsed BMI categories and standardized nutritional variables yielded results broadly consistent with the primary analyses (Supplemental Table S2 and S3). Restricted cubic spline analyses suggested non-linear associations for most dietary variables, with a few exceptions, including total energy and phosphorus intake in women.

Discussion

This study demonstrates that lower nutrient intake, age, hypertension, and BMI may contribute to the risk of BRVO in the Korean population, based on data from the 2017–2021 KNHANES. As the most recently released dataset, KNHANES 2017–2021 provides an up-to-date representation of the current ocular health status of Koreans. To our knowledge, this is the first study to investigate the associations between systemic conditions, BMI, dietary nutrient intake, and BRVO risk using this nationally representative dataset.

Previous clinical studies have reported that age is significantly associated with the risk of RVO.^22,23 The meta-analysis by Rogers et al. showed that the prevalence of BRVO increased with age, from 1.57 per 1,000 in individuals aged 40–49 years to 10.32 per 1,000 in those aged ≥80 years, which is approximately a sevenfold rise.⁸ Hypertension is also recognized as a traditional risk factor for RVO,²⁴ and lowering blood pressure has been linked to a reduced risk.²⁵ In the present study, both age and hypertension were identified as factors associated with BRVO, regardless of sex.

Although diabetes has been identified as a potential risk factor for RVO in earlier studies,^24,26 a large meta-analysis found a significant association only with CRVO, not with BRVO.²⁷ Supporting this, a recent Korean study reported that well-controlled diabetes was linked to a 55% lower prevalence of BRVO.²⁸ Consistent with these findings, diabetes was not significantly associated with BRVO in either men or women in our study.

Several previous studies have reported a higher risk of BRVO among individuals with elevated BMI.²⁴ A large population-based cross-sectional study involving 155,466 subjects similarly demonstrated that the presence of RVO was associated with increased BMI.²⁹ Park et al. further suggested that the elevated relative risk observed in overweight and obese populations may be explained by obesity-related comorbidities such as diabetes, hypertension, and hypercholesterolemia, as the significance diminished after adjustment for these covariates.³⁰ In line with these findings, our study also revealed an increasing trend of BRVO prevalence from normal weight to overweight and obese groups. However, this association was statistically significant only in women, while no significant relationship was observed in men. These results indicate that the link between BMI and BRVO may be stronger in women, suggesting that sex-specific factors could play an important role. Future studies on BRVO might therefore consider stratified analyses by sex when evaluating the impact of BMI. Moreover, further research is warranted to examine more detailed BMI categories and to compare the odds ratios of BRVO across different obesity subgroups, which may help clarify the dose–response relationship between adiposity and BRVO risk.

Although lower BMI is often accompanied by reduced total energy intake in population studies, previous intervention and cohort studies have shown that dietary composition and specific nutrient patterns are associated with vascular outcomes independent of total caloric intake or body weight.^31,32 Recent evidence highlights the role of diet in the prevention and management of RVO, with studies showing that higher consumption of fish, nuts, fruits, and vegetables exerts protective effects, while deficiencies in vitamins A, C, D, and folic acid act as independent risk factors due to their anti-inflammatory and vascular-protective properties.¹⁶ A large population-based study further reported that sufficient magnesium intake reduced the odds of RVO by 64%,³³ emphasizing the importance of specific micronutrients. In our study, men with BRVO consistently showed lower intake of a wide range of nutrients, suggesting generally lower overall dietary intake, whereas women demonstrated more selective patterns, particularly in protein, iron, fiber, and phosphorus, with iron intake being lowest in the overweight group. Importantly, as even the intake of key macronutrients such as protein and fat was associated with BRVO, we cautiously suggest that ensuring a balanced and sufficient intake of these fundamental nutrients may benefit certain patient groups.

Given the lower total food and energy intake observed in participants with BRVO, the reduced intake of multiple nutrients may partly reflect an overall reduction in dietary quantity rather than isolated deficiencies of specific nutrients. Nevertheless, several nutrient differences remained significant after adjustment for age, suggesting that the observed associations are not solely explained by reduced caloric intake or chronological aging.

These findings, consistent with prior literature, suggest that lower nutrient intake—whether global or nutrient-specific—may be related to the presence of BRVO, and that sex-specific dietary approaches could be considered in future preventive strategies. Nevertheless, lower intake in older individuals may partly reflect age-related physiological and behavioral changes in diet, and residual confounding related to aging and overall undernutrition cannot be fully excluded. Future studies incorporating validated overall diet quality indices, such as Healthy Eating Index–type or dietary diversity measures, may provide a more comprehensive understanding of how overall dietary patterns relate to BRVO beyond single-nutrient analyses.

Several limitations should be acknowledged when interpreting our findings. First, because this was a cross-sectional study, we were unable to establish causal relationships between risk factors and the occurrence of RVO. Future prospective longitudinal cohort studies are needed to clarify causality. Second, the prevalence of RVO may have been underestimated since only macula-centered fundus photographs were evaluated, which could have led to missed diagnoses of peripheral RVO. In addition, previous RVO events without clinically visible residual retinal changes may not have been detected. Third, only participants with gradable fundus photographs were included in the analysis, which may have introduced selection bias; to address this, we compared baseline characteristics between included and excluded participants (Supplemental Table S1). The excluded participants were generally younger and had lower cardiometabolic risk profiles compared with the included population. In addition, dietary intake was assessed using self-reported nutritional surveys without biochemical measurements, and therefore actual nutrient deficiencies or circulating nutrient levels could not be determined. Furthermore, supplementary analyses suggested that several dietary variables may exhibit non-linear associations with BRVO risk, including potential U-shaped patterns at extreme intake levels. Therefore, the associations estimated using linear regression models should be interpreted with caution. Finally, because the study population consisted of Korean adults, caution is warranted when generalizing these findings to other ethnic groups. Nevertheless, this study is the first to investigate the relationship between body mass index, nutritional intake, and retinal vein occlusion in the Korean population. Importantly, our analysis stratified risk factors by sex, providing novel insights into gender-specific associations. Another notable strength lies in the relatively large sample size compared with previous population-based surveys. Furthermore, the large sample size and the use of standardized, nationally representative KNHANES data enhance the reliability and generalizability of our findings.

Conclusion

We found that nutritional inadequacies and metabolic risk factors, such as hypertension and elevated BMI, may increase the risk of BRVO. By analyzing risk factors separately in men and women, we identified sex-specific associations that provide novel insights. These findings highlight the importance of dietary management and metabolic health monitoring as potential strategies for reducing the burden of BRVO in the Korean population.

Supplemental material

Supplemental material - Associations of nutritional intake and body mass index with branch retinal vein occlusion: A cross-sectional study using data from a Korean national survey

Supplemental material for Associations of nutritional intake and body mass index with branch retinal vein occlusion: A cross-sectional study using data from a Korean national survey by Bokyung Kim, Eunseob Kim, Yun Tae Kim and Mingui Kong in Science Progress.

Footnotes

Acknowledgement

Artificial intelligence–based tools were used to assist with language editing and clarity during manuscript preparation. The authors take full responsibility for the content, analysis, and interpretation of the data.

ORCID iD

Mingui Kong

Ethical considerations

This study was conducted in accordance with the tenets of the Declaration of Helsinki, and was approved by the institutional review board IRB at the Kangbuk Samsung Hospital on August 23rd, 2024 (approval number: 2024-08-022) in Korea.

Consent to participate

Given the retrospective design of this study and the use of anonymized data, requirements for informed consent were waived by the institutional review board IRB at the Samsung Medical Center in Korea.

Author contributions

Conceptualization: Kim B, Kong M. Data curation: Kim B, Kim E, Kim YT, Kong M. Formal analysis: Kim YT, Kong M. Investigation: Kim B, Kong M. Methodology: Kim B, Kim E, Kim YT, Kong M. Validation: Kong M. Writing – original draft: Kim BK, Kim YT, Kong M. Writing – review & editing: Kong M.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

Data used in this study are publicly available from the Korea National Health and Nutrition Examination Survey (KNHANES) and can be accessed at . No additional unpublished data are available from the authors.

Supplemental material

Supplemental material for this article is available online.

References

Rogers

McIntosh

Lim

et al. Natural history of branch retinal vein occlusion: an evidence-based systematic review. Ophthalmology 2010; 117: 1094–1101. e1095. https://doi.org/10.1016/j.ophtha.2010.01.058

Laouri

Chen

Looman

, et al. The burden of disease of retinal vein occlusion: review of the literature. Eye (Lond) 2011; 25: 981–988. https://doi.org/10.1038/eye.2011.92

Snider

MJE

Bheemidi

Muste

, et al. Meta-Analyses of Two Treatment Regimens for Retinal Vein Occlusion. Ophthalmic Surg Lasers Imaging Retina 2023; 54: 244–250. https://doi.org/10.3928/23258160-20230320-01

Garweg

Zandi

. Impact of treatment on long-term visual function in retinal vein occlusion-response to the comment on: retinal vein occlusion and the use of a dexamethasone intravitreal implant (Ozurdex(R)) in its treatment. Graefes Arch Clin Exp Ophthalmol 2016; 254: 2479–2480. https://doi.org/10.1007/s00417-016-3459-y

Scott

Campochiaro

Newman

, et al. Retinal vascular occlusions. Lancet 2020; 396: 1927–1940. https://doi.org/10.1016/S0140-6736(20)31559-2

Song

Zha

, et al. Global epidemiology of retinal vein occlusion: a systematic review and meta-analysis of prevalence, incidence, and risk factors. J Glob Health 2019; 9: 010427. https://doi.org/10.7189/jogh.09.010427

Shin

Cho

Kim

, et al. Prevalence and associated factors of retinal vein occlusion in the Korean National Health and Nutritional Examination Survey, 2008-2012: A cross-sectional observational study. Medicine (Baltimore) 2016; 95: e5185. https://doi.org/10.1097/MD.0000000000005185

Rogers

McIntosh

Cheung

, et al. The prevalence of retinal vein occlusion: pooled data from population studies from the United States, Europe, Asia, and Australia. Ophthalmology 2010; 117: 313–319. e311. https://doi.org/10.1016/j.ophtha.2009.07.017

O’Mahoney

Wong

Ray

. Retinal vein occlusion and traditional risk factors for atherosclerosis. Arch Ophthalmol 2008; 126: 692–699. https://doi.org/10.1001/archopht.126.5.692

10.

Yasuda

Kiyohara

Arakawa

, et al. Prevalence and systemic risk factors for retinal vein occlusion in a general Japanese population: the Hisayama study. Invest Ophthalmol Vis Sci 2010; 51: 3205–3209. https://doi.org/10.1167/iovs.09-4453

11.

Hendrick

. Retinal Vein Occlusion Review. Asia Pac J Ophthalmol (Phila) 2018; 7: 40–45. https://doi.org/10.22608/APO.2017442

12.

Kolar

. Risk factors for central and branch retinal vein occlusion: a meta-analysis of published clinical data. J Ophthalmol 2014; 2014: 724780. https://doi.org/10.1155/2014/724780

13.

Lim

Shin

Han

, et al. Differential Effect of the Metabolic Syndrome on the Incidence of Retinal Vein Occlusion in the Korean Population: A Nationwide Cohort Study. Transl Vis Sci Technol 2020; 9: 15. https://doi.org/10.1167/tvst.9.13.15

14.

Frederiksen

Stokholm

Frederiksen

, et al. Cardiovascular morbidity and all-cause mortality in patients with retinal vein occlusion: a Danish nationwide cohort study. Br J Ophthalmol 2023; 107: 1324–1330. https://doi.org/10.1136/bjophthalmol-2022-321225

15.

Voigt

Elbaz

Bohm

, et al. Incidence of Retinal Vein Occlusion and Its Association with Mortality: Results from the Gutenberg Health Study. Ophthalmology 2025; 132: 869–877. https://doi.org/10.1016/j.ophtha.2025.02.007

16.

Lendzioszek

Mrugacz

Bryl

, et al. Prevention and Treatment of Retinal Vein Occlusion: The Role of Diet-A Review. Nutrients 2023; 15: 3237. https://doi.org/10.3390/nu15143237

17.

von

Altman

Egger

, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. J Clin Epidemiol 2008; 61: 344–349. https://doi.org/10.1016/j.jclinepi.2007.11.008

18.

Yoon

Choi

Lee

, et al. An Overview of Ophthalmologic Survey Methodology in the 2008-2015 Korean National Health and Nutrition Examination Surveys. Korean J Ophthalmol 2015; 29: 359–367. https://doi.org/10.3341/kjo.2015.29.6.359

19.

Kweon

Kim

Jang

, et al. Data resource profile: the Korea National Health and Nutrition Examination Survey (KNHANES). Int J Epidemiol 2014; 43: 69–77. https://doi.org/10.1093/ije/dyt228

20.

Kim

. The Korea National Health and Nutrition Examination Survey (KNHANES): current status and challenges. Epidemiol Health 2014; 36: e2014002. https://doi.org/10.4178/epih/e2014002

21.

Song

Choi

Han

, et al. Methodology and Rationale for Ophthalmic Examinations in the Seventh and Eighth Korea National Health and Nutrition Examination Surveys (2017-2021). Korean J Ophthalmol 2021; 35: 295–303. https://doi.org/10.3341/kjo.2021.0016

22.

Cheung

Klein

Wang

, et al. Traditional and novel cardiovascular risk factors for retinal vein occlusion: the multiethnic study of atherosclerosis. Invest Ophthalmol Vis Sci 2008; 49: 4297–4302. https://doi.org/10.1167/iovs.08-1826

23.

Lim

Cheung

Wang

, et al. Prevalence and risk factors of retinal vein occlusion in an Asian population. Br J Ophthalmol 2008; 92: 1316–1319. https://doi.org/10.1136/bjo.2008.140640

24.

The Eye Disease Case-control Study Group . Risk factors for branch retinal vein occlusion. The Eye Disease Case-control Study Group. Am J Ophthalmol 1993; 116: 286–296.

25.

Kim

Lee

, et al. Retinal Vascular Occlusion Risks in High Blood Pressure and the Benefits of Blood Pressure Control. Am J Ophthalmol 2023; 250: 111–119. https://doi.org/10.1016/j.ajo.2023.01.023

26.

Klein

Moss

, et al. The epidemiology of retinal vein occlusion: the Beaver Dam Eye Study. Trans Am Ophthalmol Soc 2000; 98: 133–141, discussion 141-133.

27.

Wang

Wen

, et al. Diabetes mellitus as a risk factor for retinal vein occlusion: A meta-analysis. Medicine (Baltimore) 2020; 99: e19319. https://doi.org/10.1097/MD.0000000000019319

28.

Lee

Kim

. Influence of Smoking and Controlled Medical Conditions on the Risks of Branch Retinal Vein Occlusion in South Korea: A Population-Based Study. Ophthalmic Epidemiol 2025; 32: 44–51. https://doi.org/10.1080/09286586.2024.2321902

29.

Wong

Larsen

Klein

, et al. Cardiovascular risk factors for retinal vein occlusion and arteriolar emboli: the Atherosclerosis Risk in Communities & Cardiovascular Health studies. Ophthalmology 2005; 112: 540–547. https://doi.org/10.1016/j.ophtha.2004.10.039

30.

Park

Lee

Han

. Underweight: another risk factor for cardiovascular disease? A cross-sectional 2013 Behavioral Risk Factor Surveillance System (BRFSS) study of 491,773 individuals in the USA. Medicine (Baltimore) 2017; 96: e8769. https://doi.org/10.1097/MD.0000000000008769

31.

Estruch

Ros

Salas-Salvado

et al. Primary prevention of cardiovascular disease with a Mediterranean diet. N Engl J Med 2013; 368: 1279–1290. https://doi.org/10.1056/NEJMoa1200303

32.

Appel

Moore

Obarzanek

et al. A clinical trial of the effects of dietary patterns on blood pressure. DASH Collaborative Research Group. N Engl J Med 1997; 336: 1117–1124. https://doi.org/10.1056/NEJM199704173361601

33.

Kim

Lee

, et al. Sufficient Magnesium Intake Reduces Retinal Vein Occlusion Risk: National Health and Nutrition Examination Survey Analysis. Nutrients 2025; 17: 1285. https://doi.org/10.3390/nu17071285

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.26 MB