Here, swept-source optical coherence tomography (SS-OCT) is used to compare retinal and choroidal thickness in adults with and without malnutrition, contributing to a better understanding of the pathogenesis of ocular diseases in malnourished individuals.

Despite improvements in reducing hunger in recent years, undernutrition remains a global public health problem. This study utilizes the swept-source optical coherence tomography (SS-OCT) technique to assess changes in retinal and choroidal thickness in underweight subjects. Ophthalmic examinations were conducted on all adults participating in this cross-sectional research. Depending on their body mass index (BMI), the participants were divided into two groups: the underweight group and the normal group. The study included the right eyes of the underweight adults and an equal number of age- and gender-matched normal-weight subjects. The retinal thickness showed no significant difference between the underweight and normal groups (P > 0.05 for all). In males, the retina of the center and inner ring in the underweight group was significantly thinner than that in the normal group, while no such results were found in females. The choroid in the underweight group was significantly thinner compared to that in the normal group (all P < 0.05). Being underweight may affect choroidal thickness in both males and females. In comparison with underweight females, underweight males may experience more retinal damage. These findings contribute to a better understanding of the pathogenesis underlying specific ocular diseases in malnourished individuals.

Despite the Health Organization's successful efforts to combat hunger in recent years, undernutrition remains a significant global public health concern. Globally, it was estimated that 9.8% of the population was undernourished in 2022¹. The incidence of undernutrition varies across regions, with higher prevalence among individuals with lower socioeconomic status^2,3,4. Additionally, some individuals, especially young people, lose weight excessively in pursuit of a perfect body shape. Malnutrition, in all its various forms, affects every country in the world⁵.

Being underweight is associated with negative clinical outcomes, including infections, immune dysfunction, delayed wound healing, and growth and developmental retardation^6,7,8,9. A malnourished state is one of the leading risk factors for premature death and the loss of disability-adjusted life years^10,11,12. Studies have shown that the lowest body mass index (BMI) is associated with the poorest binocular ability¹³. Furthermore, research has demonstrated that undernutrition is linked to various ocular issues, such as macular degeneration, decreased dark adaptation, optic atrophy, keratitis, dry eye, and retinoblastoma^{14,15,16,17,18}.

The retina, with its multiple layers and cell types, is a complex tissue, while the choroid is a highly vascularized structure that provides nutrients to the outer layer of the retina and removes metabolic waste¹⁹. The retina and choroid, as critical structures of the eyeball, can be affected by systemic pathologies or physiological conditions^20,21. They have been found to play a significant role in the pathogenesis of specific ocular diseases, including macular degeneration, polypoidal choroidal vasculopathy, uveitis, glaucoma, and myopia-related chorioretinal atrophy^{22,23,24,25,26}. Therefore, ocular function depends on both anatomically and functionally normal retinas and choroids.

While undernutrition has various effects on the eye, there is limited information available on the relationships between malnutrition and retinal or choroidal thickness in different genders. This study aims to assess potential changes in retinal or choroidal thickness in malnourished adults using the swept-source optical coherence tomography (SS-OCT) technique, which represents a significant advancement in retinal and choroidal imaging²⁷. This technology is particularly effective in accurately identifying the choroidal scleral interface (CSI) in eyes with thicker choroids, thanks to its high penetration capabilities through the retinal pigment epithelium (RPE).

In this study, participants were categorized into two groups based on their BMI: the underweight group (BMI < 18.50 kg/m²) and the normal group (18.50 ≤ BMI < 25.00 kg/m²). The study included 996 right eyes of 996 underweight adults and an equal number of age- and gender-matched normal-weight subjects. The average BMI was 17.48 ± 0.75 kg/m² in the underweight group and 21.30 ± 1.75 kg/m² in the normal group.

This research was conducted at Huashan Hospital of Fudan University from January 2020 to October 2020. The study was approved by the Institutional Review Board of Huashan Hospital (No. KY2016-274), and all participating adults provided written informed consent.

1. Selection of participants

1. Record all participants' demographic characteristics, such as age, gender, and a history of systemic diseases. Consider the following as the exclusion criteria: (1) age < 18 years or > 70 years old and (2) a history of systemic diseases related to retinal or choroidal thickness, including diabetes mellitus, hypertension, and thyroid disease.
   ​NOTE: The elderly population, particularly those aged over 70 years, frequently experienced severe cataracts that could affect the quality of OCT images.
2. Let all adult participants involved in the research undergo ophthalmic examinations. Consider the following as the exclusion criteria: (1) intraocular pressure (IOP) >21 mmHg; (2) best-corrected visual acuity (BCVA) worse than 0.1 LogMAR; (3) spherical equivalent more than ± 6 diopters; (4) a history of ocular diseases, including retinal disease, choroidal disease, and glaucoma; and (5) any previous ocular surgery.

2. Body mass index calculation

1. Measure the participants' height and weight using a height-weight measuring instrument (see Table of Materials).
2. Calculate the BMI using the formula: weight / (height x height) (kg/m²).
3. Classify the subjects into two groups based on the World Health Organization International Classification²⁸: the underweight group (BMI <18.50 kg/m²) and the normal group (18.50 ≤ BMI < 25.00 kg/m²).

3. Swept-source optical coherence tomography scan

1. Turn on the Power switch on the SS-OCT device (see Table of Materials) with a 1050 nm wavelength.
   NOTE: This SS-OCT system, capable of 1,00,000 scans/s, has recently undergone significant improvements, enhancing the visualization of the retina and choroid.
2. Click on the Radial Dia.6.0mm Macula Overlap 4 button to access the scanning interface.
3. Capture high-quality images for each eye of the participants during the scanning process.
   NOTE: The OCT scans were performed by experienced ophthalmologists between 8 and 10 a.m. daily to minimize diurnal variations²⁹.
4. Generate a thickness map following the standard early treatment diabetic retinopathy study (ETDRS) grid.
5. Define retinal thickness (Figure 1A,B) and choroidal thickness (Figure 2A,B) as previously described^27,30.
   NOTE: To ensure accurate measurements, it was imperative to manually review the segmented lines within the OCT scans^27,30.
6. Exclude poor OCT images resulting from media opacity or unstable fixation.

4. Statistical analysis

1. Launch the SPSS software (see Table of Materials). The analysis exclusively considered the right eye of the participants³¹.
   NOTE: Present continuous data as mean ± standard deviation (SD) and categorical data as frequency (percentage).
2. Perform group comparison using a t-test for continuous variables and a Chi-square test for categorical variables. Conduct correlation analyses employing Pearson's correlation.
   NOTE: A significance level of P < 0.05 (two-tailed) was used to determine statistical significance.

A total of 996 right eyes from 996 underweight adults were evaluated in this study, with 1:1 age- and gender-matched normal-weight subjects. The demographic characteristics of both groups are summarized in Table 1. The underweight group had an average BMI of 17.48 ± 0.75 kg/m² (range: 14.60-18.40 kg/m²), while the normal-weight group had an average BMI of 21.30 ± 1.75 kg/m² (range: 18.50-24.90 kg/m²).

Table 2 presents the comparison of retinal thickness in nine sectors of the ETDRS grid between the underweight and normal groups. No significant differences were observed in retinal thickness between the two groups.

In Table 3, the retinal thickness in males between the underweight and normal groups are compared. The underweight group had an average BMI of 17.44 ± 0.79 kg/m², while the normal-weight group had an average BMI of 22.34 ± 1.61 kg/m². The center and inner ring regions showed significantly thinner retinal thickness in the underweight group compared to the normal group, while no significant difference was observed in the outer ring.

Table 4 compares the retinal thickness in females between the underweight and normal groups. The underweight group had an average BMI of 17.49 ± 0.73 kg/m², and the normal group had an average BMI of 21.02 ± 1.68 kg/m². No significant differences in retinal thickness were found between the two groups in any region of the ETDRS grid.

Table 5 provides a comparison of choroidal thickness in nine sectors of the ETDRS grid between the underweight and normal groups. The results indicate that the choroidal thickness was significantly reduced in all sectors of the underweight group compared to the normal group. In Table 6, the analysis focuses on choroidal thickness in males within the underweight and normal groups. It shows that the inner temporal region of the underweight group had notably thinner choroidal thickness in comparison to the normal group. Table 7 presents the analysis of choroidal thickness in females between the underweight and normal groups. This analysis reveals that the inner nasal and outer nasal regions of the underweight group exhibited significantly thinner choroidal thickness compared to the normal group.

Table 8 and Table 9 provide correlation analyses between retinal or choroidal thickness and weight, height, or BMI. Weight and height showed significant positive correlations (using Pearson's correlation analysis) with retinal thickness in most areas of the ETDRS grid, while BMI had significant positive correlations with retinal thickness in four areas. For choroidal thickness, weight and height exhibited significant positive correlations in all areas, whereas BMI had significant positive correlations in three areas.

Figure 1: Retinal thickness. (A) Retinal thickness is the distance from the internal limiting membrane (ILM) to retinal pigment epithelium (RPE). (B) Retinal thickness in nine sectors of the early treatment diabetic retinopathy study (ETDRS) grid. Please click here to view a larger version of this figure.

Figure 2: Choroidal thickness. (A) Choroidal thickness was defined as the distance from retinal pigment epithelium (RPE) to the chorioscleral interface (CSI). (B) Choroidal thickness in nine sectors of the early treatment diabetic retinopathy study (ETDRS) grid. Please click here to view a larger version of this figure.

Table 1: Demographic characteristics of the participants.

Table 2: Retinal thickness. This table compares the retinal thickness in nine sectors of the early treatment diabetic retinopathy study (ETDRS) grid between the underweight and normal groups.

Table 3: Comparison of the retinal thickness in males.

Table 4: Comparison of the retinal thickness in females.

Table 5: Choroidal thickness. This table compares the choroidal thickness in nine sectors of the early treatment diabetic retinopathy study (ETDRS) grid between the underweight and normal groups.

Table 6: Comparison of the choroidal thickness in males.

Table 7: Comparison of the choroidal thickness in females.

Table 8: Correlation analyses between retinal thickness and weight, height, or body mass index (BMI).

Table 9: Correlation analyses between choroidal thickness and weight, height, or body mass index (BMI).

In this study, SS-OCT was employed to compare retinal and choroidal thickness in adults with and without malnutrition. The outcomes of the study showed that, among males, individuals in the underweight group had significantly thinner retinas in the central and inner ring regions compared to those in the normal group. However, no such differences were observed among females. Additionally, the choroid was found to be significantly thinner in the underweight group compared to the normal group in both males and females. These findings indicate that being underweight may have an impact on choroidal thickness in both sexes and on retinal thickness in males specifically.

To our knowledge, there has been limited research on the relationship between undernutrition and retinal thickness in different genders. Wong et al.³² evaluated the association between BMI and central retinal thickness using OCT and found that a higher BMI was correlated with a thicker retina. This study also showed that BMI was positively correlated with retinal thickness in some areas of the ETDRS grid. Interestingly, only the male retina in the center and inner ring exhibited significant differences between the underweight and normal groups, suggesting that male retinas in these areas are more susceptible to undernutrition. This may be attributed to the larger difference in BMI between underweight and normal-weight males (4.90 kg/m²) compared to females (3.53 kg/m²). Another possible factor contributing to these varying results is the presence of sex differences in factors related to undernutrition, such as serum lipids and homocysteine³³. It's important to note that there can be substantial variability in the percentage of body fat for individuals with the same BMI, which is partly influenced by gender².

The association between underweight and choroidal thickness remains inconsistent. In a study conducted in Turkey, malnutrition was found to decrease choroidal thickness in children without clinically reported ocular symptoms³⁴, consistent with our findings. However, Yilmaz et al.³⁵ compared choroidal thickness in underweight and normal-weight subjects using enhanced depth imaging (EDI) OCT instead of SS-OCT and found no significant difference. The use of SS-OCT in this study allowed for precise visualization of the chorioscleral interface (CSI) in all eyes, which was achieved in only 73.6% of eyes using EDI-OCT³⁶. Previous studies often measured choroidal thickness manually at selected points, which could be affected by focal choroidal thinning or thickening due to the irregular shape of the CSI in some cases^37,38. Additionally, manual measurements are susceptible to human error. The use of SS-OCT in this research allowed for automated and averaged measurements of retinal and choroidal thickness, enhancing accuracy³⁹.

This study found that the choroid was significantly thinner in all regions of the underweight group compared to the normal group, while retinal thickness was significantly different in males only, specifically in the center and inner ring. This discrepancy may be due to the choroid's predominant role in ocular blood flow, representing 85% of it⁴⁰. Moreover, the blood-retina barrier serves to protect the eye from harmful substances. Variations in blood flow and barrier function could explain why the retina is less affected than the choroid.

Being underweight can impact the structure and function of various body organs^10,11, including the eyeball^13,32,34. This study demonstrates that underweight individuals exhibit changes in retinal and choroidal thickness. Furthermore, alterations in retinal and choroidal thickness have been associated with specific ocular pathologies. Therefore, changes in retinal and choroidal thickness could serve as early indicators of ocular disorders in undernourished individuals, even before the onset of symptoms.

However, the present study has its limitations. It is a cross-sectional study, and lacks data on changes in retinal and choroidal thickness during fluctuations in body fat accumulation or loss. Future research should consider studying these variations longitudinally. Additionally, subclinical diseases might impact the results, despite our rigorous participant selection.

In summary, being underweight may affect choroidal thickness in both sexes and retinal thickness in males. It is recommended discouraging excessive weight loss, particularly among males. The current findings contribute to a better understanding of the pathogenesis of specific ocular diseases in undernourished individuals.

None of the authors has a financial or proprietary interest in any material or method mentioned.

This study was funded by grants from the National Natural Science Foundation of China (No. 81900879) and the Science and Technology Commission of Shanghai Municipality (No. 20Y11910800).