Retinopathy is the most common ocular complication of diabetes mellitus (1), characterized by retinal microvascular dysfunction (2). It is thought to be the main reason for blindness in adults over 20 years in the United States (3). Previous research found that patients with retinopathy were more likely to suffer from heart failure, stroke, kidney disease (4), and neurological disorders. Retinopathy may be a marker for the risk of reactive vascular disease (5). This shows that retinopathy affects not only vision but also other vascular complications. Therefore, identifying the risk factors that influence retinopathy is crucial for systemic microvascular health.

Obesity and diabetes are well-established risk factors for retinal diseases (6, 7). Enhanced oxidative stress induced by diabetes plays a central role in the pathogenesis of retinopathy diseases by contributing to retinal cell dysfunction and apoptosis (8). It also triggers the release of pro-inflammatory cytokines, which can lead to basement membrane thickening, disruption of the blood-retinal barrier (BRB), microvascular occlusion, and ultimately, the development of retinopathy (9, 10).

Obesity is a significant public health problem worldwide (11), and the incidence of overweight and obesity has continued to increase in recent years (12). Obesity is also closely linked to retinal diseases, potentially contributing through mechanisms such as insulin resistance, systemic inflammation, oxidative stress, and microvascular dysfunction (13, 14). Although body mass index (BMI) is widely used to assess obesity, it does not accurately reflect fat distribution. Abnormal fat accumulation can occur even in individuals with a normal BMI (15). While imaging techniques such as computerized tomography (CT) and magnetic resonance imaging (MRI) can directly quantify visceral adiposity, their highly cost limits large-scale clinical use (16). Therefore, several alternative anthropometric indices have been developed to provide low-cost, practical assessments of body fat distribution: body roundness index (BRI), conicity index (CI), waist-to-height ratio (WHtR), tri-ponderal mass index (TMI), weight-adjusted-waist index (WWI), and a body shape index (ABSI).

BRI based on eccentricity theory and incorporating waist circumference and height, offers a more nuanced evaluation of visceral fat and abdominal obesity (17, 18). The WWI, introduced by Park et al., is calculated as waist circumference divided by the square root of body weight, enabling the assessment of central obesity independent of total body weight (19). The WHtR is another simple yet effective index that reflects central fat distribution (20). Additionally, the ABSI and the CI are commonly used to evaluate fat distribution, with ABSI more indicative of central obesity and CI focusing specifically on abdominal adiposity (21).

Previous studies have demonstrated that: compared to general obesity, central obesity is more strongly associated with insulin resistance and diabetes mellitus—both of which are key contributors to retinal diseases (22). While much research has been conducted on the association between these anthropometric indices and conditions such as diabetes, hypertension, and kidney stones (23–25), the relationship between these obesity-related measures and retinal diseases remains unclear. Additionally, it is not known which indicator has the greatest diagnostic value for retinal diseases. Therefore, we used data from the National Health and Nutrition Examination Survey (NHANES) from 2005 to 2008 to examine the relationship between various measures and the prevalence of retinal diseases in the U. S. population. We also compared the diagnostic value of novel and traditional indicators for retinal diseases using the receiver operating characteristic (ROC) method.

This study is the first to examine the relationship between obesity indicators and retinopathy using data from 4,867 participants in the NHANES database. After adjustment for relevant covariates, BRI, CI, WHtR, TMI, and WWI were positively correlated with retinopathy. Subgroup analyses showed that this association was stable across populations. BRI, CI, WHtR, and WWI were more predictive of retinopathy than BMI, a conventional measure of obesity.

Obesity refutation has been shown to exist in many chronic diseases, and previous studies have shown that overweight and obese individuals have reduced mortality in patients with coronary heart disease, heart failure, and some types of breast cancer (26, 27). In the present study, we found an ‘obesity refutation’ of the relationship between BMI and retinopathy, and the risk of retinopathy decreases with an increase in BMI above 34 kg/m². This may be related to lipocalin, secreted by white fat cells in obese people. Studies have shown that lipocalin can have anti-inflammatory and antioxidant effects under certain conditions (28, 29). Secondly, obese people are more likely to consume vitamins and minerals, nutrients that can reduce oxidative stress throughout the body and reduce retinal damage (30, 31).

Weight-adjusted-waist index is a new indicator of obesity, as proposed by Park et al. (32) A retrospective study found that the likelihood of having diabetic retinopathy decreased by 31% for every additional unit of WWI (33). Another study found a negative correlation between WWI and diabetic nephropathy when WWI < 11.48 (34). Similarly, we found that people with diabetes with WWI between 11.619 and 12.065 were less likely to develop retinopathy compared to people with diabetes with WWI (9.381–11.136). This may be related to the high levels of C-peptide in obese people with diabetes (35), which has been shown to increase retinal vascular blood flow and improve retinal endothelial cell function (36, 37). It may also counteract the production of ROS and inhibit the production of VEGF (38).

Sun et al. (39) found that WWI was more predictive of diabetes mellitus and hypertension than BMI (40). Several retrospective studies have also shown that newer indices like WWI and visceral adiposity index (VAI) are more predictive of diabetic retinopathy (41, 42). This suggests that BMI, which reflects generalized obesity, has limitations and that indicators reflecting fat distribution have more excellent clinical predictive value. Lee et al. (43) suggested that excess abdominal and visceral fat may be an essential risk factor for diabetic retinopathy. However, the above studies did not focus on fat distribution and other retinopathies. Therefore, we examined the association of these obesity markers with retinopathy in the American population. They were helpful as indicators for assessing the risk of developing retinopathy.

Obesity as a risk factor for retinopathy may be related to several aspects. Obesity is often associated with insulin resistance (44, 45), producing hyperglycemia. Prolonged hyperglycemia damages retinal endothelial cells, triggering retinal vascular hyperglycemia and leakage (46, 47), resulting in retinal damage. Second, visceral fat accumulation can disrupt the function of pancreatic β-cells, leading to persistent hyperglycemia. On the other hand, visceral fat can inhibit the secretion of adipokines (48, 49), leading to increased expression of inflammatory factors, including TNF-1A and IL-6, which damages the retina (50).

Our study is the first to comprehensively examine the association between six anthropometric measures and the prevalence of retinal disease, while also comparing their predictive value. The findings suggest that controlling obesity and improving body composition may help reduce the prevalence of retinal disorders. Among the indicators evaluated, the WWI demonstrated the strongest predictive potential. Therefore, individuals with higher WWI values may benefit from further retinal examinations. Such an approach could facilitate earlier detection and treatment of retinal disease, reduce the risk of vision loss or blindness, and ultimately conserve medical resources and alleviate societal burden. Moreover, given the close relationship between retinal health and chronic conditions such as diabetes and hypertension, paying closer attention to the retinal status of obese individuals may also contribute to improving their overall health outcomes.

We note that the study still has limitations: the population was from the United States, and the universality of these findings needs further exploration. Due to the limitations of a cross-sectional study, it was impossible to explain the specific mechanisms linking the measurements to retinopathy. In addition, it was impossible to fully adjust for relevant covariates and exclude the influence of confounding factors.

In this study, using NHANES data, we found that obesity indicators (BRI, CI, WHtR, TMI, WWI, and BMI) and retinopathy were significantly and positively correlated and had some predictive value for retinopathy. It is recommended that the risk of retinopathy be reduced by keeping these indicators within safe limits through a healthy lifestyle. Combining these indicators by clinicians is helpful in the prevention and early diagnosis of retinopathy.