Is There Any Association between Nutrition and Myopia? A Systematic Review

Abstract

SIGNIFICANCE 

This systematic review highlights the possible role of nutrition in myopia based on qualitative analysis of vast and diverse literature that investigated this association.

PURPOSE 

We systematically reviewed the outcomes of the studies that previously investigated the association between nutrition and myopia.

METHODS 

EMBASE, MEDLINE, and PubMed were searched by two independent authors to identify cross-sectional, cohort, retrospective, or interventional studies that assessed the association of nutrition with myopia from inception to the year 2021. Furthermore, the reference list of the included articles was screened. The data from the included studies were extracted, and qualitative analysis was performed. Quality assessment for noninterventional studies and interventional trials was performed using the Newcastle-Ottawa Scale and Cochrane RoB 2, respectively.

RESULTS 

Twenty-seven articles were included in the review. Most of the nutrients and dietary elements investigated in noninterventional studies showed inconsistencies in their association with myopia, with the majority indicating no association. Nine studies showed a significant association of diverse nutrients and dietary elements with either an increase (odds ratio, 1.07) or a decrease (odds ratio, 0.5 to 0.96) in the risk of myopia development. However, a majority of these studies have minimal odds ratios with wider or overlapping confidence intervals, implicating weaker associations. All three nutrients and dietary elements assessed in the interventional trial had implications for myopia control, with two trials indicating a clinically minimal effect.

CONCLUSIONS 

This review implies that there is some evidence to indicate a potential influence of specific nutrients and dietary elements in myopia development, which are supported by several theories. However, given the vast, diverse, and complex nature of nutrition, more systematic investigation is warranted to comprehend the extent to which these specific nutrients and dietary elements are associated with myopia through longitudinal studies by subduing the limitations in the existing literature.

Parents often believe that food habits influence eye health and refractive error.^1,2 One of the frequently asked questions to the eye care practitioner by the parents of children with myopia is regarding dietary recommendations to slow down the progression of myopia. Apart from this general perception, few studies showed that an increase in height and body composition is associated with a healthy dietary intake,^3,4 whereas malnourishment is correlated with smaller organ size.^5,6 Hence, it is reasonable to assume that nutrition may have a potential role in myopia-related ocular elongation.

Experiments conducted on the animals showed that administration of chemical substances such as insulin and glucagon (hormones related to the metabolism of nutrients),⁷ retinoic acid (vitamin A precursor),⁸ 7-methylxanthine (7-MX, caffeine metabolite),⁹ and crocetin (a component of saffron)¹⁰ had an impact on the myopia development as well as in regulating ocular growth. All these aforementioned substances are either derived from or influenced by the diet.

In addition to studies conducted on the animal, there is ample literature related to the association between nutrition and myopia in the human population. There is evidence that dates back to an interventional trial conducted in London in the year 1958, in which myopic children who had a dietary modification to include more animal protein had a slower rate of myopia progression as opposed to the control subjects.¹¹ A prospective study that evaluated the nutrient intake using the dietary record method in Hong Kong Chinese children found the incident myopes to be deficient in energy intake and various nutrients that included protein, fat, vitamin B₁, vitamin B₂, vitamin C, phosphorus, iron, and cholesterol in comparison with those who did not develop myopia.¹² A cross-sectional study in Poland that measured the serum concentration of various trace elements (usually obtained via diet) including zinc, copper, selenium, and manganese reported myopic individuals to have lower serum concentrations of zinc and selenium.¹³ Longer axial length (AL) was associated with higher consumption of saturated fats and cholesterol in another cross-sectional study conducted in Singapore Chinese children using a food frequency questionnaire.¹⁴

Despite these studies that support the association between nutrition and myopia, until now, no nutritional intervention has been recognized as an effective myopia control strategy.¹⁵ One possible reason for the existence of lacunae can be a lack of comprehensive understanding of the previous literature, which included studies of different designs (noninterventional and interventional) that examined diverse nutrients (energy intake, macronutrients, and micronutrients) and dietary elements (breastfeeding, solid food weight, and consumption of various food items and supplements), used dissimilar methods to evaluate nutrient/dietary intake (subjective methods: 24-hour recall or food frequency questionnaire and objective methods: measurement of nutrient in serum/hair), and exhibited differential outcomes for each nutrient, making it difficult to judge if any nutrient or dietary element has any influence on myopia, which forms our rationale for conducting the systematic review. Therefore, we systematically reviewed and qualitatively analyzed the outcome of the prior studies that evaluated the association of nutrition in myopia development and progression.

METHODS

Search Strategy

A literature search was conducted in EMBASE, MEDLINE, and PubMed with multiple combinations of keywords related to nutrition and myopia. Filters were applied in EMBASE and MEDLINE to eliminate the studies conducted on animals, non-English text, articles without abstracts, and articles without access to the full text. The last date of the search was February 24, 2022. The matrix that was used for the literature search is provided in the Appendix, available at https://links.lww.com/OPX/A604.

Selection of the Studies

The titles and abstracts of the studies were screened for their relevance to our review. The full text of the articles that were found appropriate for the review was assessed for eligibility for inclusion in the qualitative analysis. In addition, the references of the included publications were also inspected. Two independent authors (SC and SKG) were involved in the title/abstract screening and the evaluation of full-text articles for their eligibility. There were discrepancies among the authors regarding the inclusion of studies in the review by 15% (9 of 59 studies), which were resolved by a third reviewer (PKV).

Eligibility Criteria

The studies were included if they (a) were of cross-sectional/cohort/retrospective/interventional design, (b) included participants 30 years or younger, (c) defined myopia, (d) evaluated objective refraction, (e) either assessed the nutrient/dietary element intake subjectively or measured serum concentration of nutrients/dietary elements objectively, (f) assessed the difference in the intake or serum concentration of nutrients/dietary elements between myopes and nonmyopes and/or the odds ratio (OR) for myopia development in case of noninterventional studies, and (g) assessed the change in spherical equivalent refraction (SER) and/or AL (if available) in case of interventional trials.

The studies conducted on nonhealthy participants (such as pre-term born, malnourished, and diabetic individuals) were excluded. We excluded the studies that exclusively measured the serum concentration of nutrients that have a nondietary source, such as vitamin D₃, which is obtained from sunlight exposure. The studies that assessed the diet qualitatively, for example, balanced diet and westernized diet, among others, were excluded. For the interventional trials, we did not have any exclusion criteria based on the study duration.

Data Extraction and Quality Assessment

The following information was extracted from the included studies: (1) study design, (2) the last name of the first author, (3) year of publication, (4) study location, (5) sample size, (6) sample age, (7) method for measurement of refraction (and AL if available), (8) method of assessment of nutritional status, (9) mean difference in nutrients/dietary elements between myopes and nonmyopes (if available), (10) OR with 95% confidence interval (CI)/standard error (if available), and (11) β coefficient of SER and AL with 95% CI/standard error (if available). In each of the studies, the model that has adjusted for a maximum number of confounding variables was chosen, and the OR of myopia and/or β coefficient value of SER and AL (if available) was obtained from that model.

We examined the quality of noninterventional studies using the Newcastle-Ottawa Scale.¹⁶ For the interventional trials, we used Cochrane Risk of Bias 2.¹⁷ The data extraction and quality assessment for the included studies were performed by two authors independently (SC and SKG). Both authors were in good agreement (100%) during the process of data extraction. The disagreements during the quality assessment with the Newcastle-Ottawa Scale (22% [47 of 216 stars]; 24 studies and 9 points for each study) and Cochrane Risk of Bias 2 (13% [2 of 15 domains]; 3 studies and 5 domains for each study) were resolved by the third reviewer (RD).

RESULTS

Search Results

The bibliographic database search identified 1463 studies, of which 391 were duplicates. After screening titles and abstracts, 1013 studies were excluded because they seemed irrelevant in the context of this review, and the full texts of the remaining 59 studies were assessed for their eligibility. The eligibility criteria were fulfilled by 23 studies that were included in this review, and the reasons for the exclusion of the remaining studies are provided in Appendix Table A1, available at https://links.lww.com/OPX/A605. Reference list screening of the included studies identified four additional records that were found eligible to be included in the review. Thus, a total of 27 studies were included in the systematic review (Fig. 1).

FIGURE 1: Preferred Reporting Items for Systematic Reviews and Meta-Analyses flow diagram presenting the process of study identification, screening, and selection of studies.

Risk-of-bias Assessment

Among the 24 noninterventional studies that were assessed using the Newcastle-Ottawa Scale (Appendix Table A2, available at https://links.lww.com/OPX/A605), 20 studies (83%) have a score ≥6. According to the Cochrane RoB 2, among the three intervention trials, one had a low risk of bias, one had some concerns, and one had a high risk of bias (Appendix Fig. A1, available at https://links.lww.com/OPX/A606).

Characteristics of the Included Studies

The articles included in this review were published before December 2021. Among 27 studies included in this review, there were 24 noninterventional studies and 3 interventional trials. Further information related to the characteristics of the included articles (study design, study location, sample size, and age of the sample) is provided in Table 1.

TABLE 1 - Characteristics and methodology of the included studies

S.N.	Source	Study design	n	Age (y)	Cycloplegia	Nutritional status
1	Gardiner11	Interventional	437	—	—	Intervention
2	Edwards12	Birth cohort	92	10	×	Dietary record
3	Trier et al.18	Clinical trial	68	8–13	✓	Intervention
4	Lim et al.14	Cross-sectional	851	7–9	✓	FFQ
5	Sham et al.19	Cross-sectional	3009	0.5–6	✓	Questionnaire
6	Mutti and Marks20	Cross-sectional	22	13–25	✓	LC-MS/MS and FFQ
7	You et al.21	Cross-sectional	15,066	7–18	×	Questionnaire
8	Choi et al.22	Retrospective	2038	13–18	×	24-h recall, FFQ, and radioimmunoassay
9	Guggenheim et al.23	Birth cohort	5126	7–15	×	HPLC-MS/MS
10	Tideman et al.24	Birth cohort	2666	6	✓	LC-MS/MS
11	Shirzadeh et al.25	Cross-sectional	400	1–5	×	Questionnaire
12	Tideman et al.26	Birth cohort	5711	6	✓	LC-MS/MS
13	Fedor et al.13	Cross-sectional	121	7–17	✓	AAS
14	Liu et al.27	Cross-sectional	527	6–12	✓	Questionnaire
15	Kearney et al.28	Prospective	54	18–20	×	LC/MS-MS and FFQ
16	Chua et al.29	Birth cohort	317	3	✓	24-h recall/3-d food diaries
17	Mori et al.30	Clinical trial	69	6–12	✓	Intervention
18	Burke et al.31	Retrospective	1095	12–19	×	24-h recall
19	Chiang et al.32	Retrospective	9960	12–19	×	24-h recall, HPLC-MS/MS, and LC-MS/MS
20	Fedor et al.33	Cross-sectional	135	7–17	—	AAS
21	Berticat et al.34	Cross-sectional	180	4–18	✓	FFQ
22	Ng et al.35	Birth cohort	642	20	✓	FFQ
23	Burke et al.36	Retrospective	304	12–19	×	ICP-MS
24	Lingham et al.37	Birth cohort	1260	20	✓	EIA, LC-MS/MS
25	Gao et al.38	Cross-sectional	186	6–12	×	HPLC-MS/MS
26	Harb and Wildsoet39	Retrospective	6855	12–25	×	LC-MS/MS and dietary interviews
27	Li et al.40	Birth cohort	467	9	✓	FFQ

✓ = done; × = not done; — = not mentioned; AAS = atomic absorption spectrophotometry; AL = axial length; EIA = enzyme immunoassay: FFQ = food frequency questionnaire; HPLC-MS/MS = high-performance liquid chromatography–tandem mass spectrometry; ICP-MS = inductively coupled plasma–mass spectrometry; LC-MS/MS = liquid chromatography–tandem mass spectrometry; n = sample size; S.N. = study number.

Methodology of the Included Studies

The method of assessment of refractive error and nutritional status for each study is given in Table 1. Refraction was obtained under cycloplegia in 14 studies and the absence of cycloplegia in 11 studies. Two studies did not mention their method of assessment of refraction and stated that they performed refraction/complete ophthalmic examination.^11,33 Among 27 studies, 1 study each defined myopia as SER <0.00 D³⁴ and SER ≤−0.25 D,¹¹ 3 studies defined myopia as SER <−0.5 D,^25,27,35 14 studies defined myopia as SER ≤−0.50 D,^{12–14,19,22,24,26,28,29,33,36–38,40} 3 studies defined myopia based on SER ≤−0.75 D,^18,20,39 and four studies used SER ≤−1.00 D as a cutoff,^21,23,31,32 whereas one study included SER ≤−1.5 D.³⁰

The intake of nutrients was assessed in 11 studies via the methods of dietary records, food diaries, questionnaires, food frequency questionnaires, 24-hour recall, and dietary interviews. The serum/hair concentration of nutrients was measured in eight studies through the techniques of atomic absorption spectrophotometry, liquid chromatography–tandem mass spectrometry, enzyme immunoassay, high-performance liquid chromatography–tandem mass spectrometry, and inductively coupled plasma–mass spectrometry. Five studies have assessed both the intake and serum/hair concentrations of different nutrients/dietary elements. The remaining three studies were interventional trials where a nutrient or dietary element was given as an intervention.

Outcomes of the Included Studies

Of the 24 noninterventional studies, 10 studies provided outcome in the form of mean/median differences in the intake or serum concentration of nutrients and dietary elements between the myopic and nonmyopic groups (Appendix Table A3, available at https://links.lww.com/OPX/A605), 6 studies calculated the OR of myopia in association with intake or serum concentration of nutrients/dietary elements, 7 studies provided both mean difference and OR value, and the remaining 1 study provided the association between the spherical equivalent of refraction with the intake of nutrients/dietary elements. An adjustment was done for age and sex in 17 studies, ethnicity (n = 19), parental myopia (n = 10), time spent outdoors (n = 10) and time spent on near-work (n = 6), and other factors such as family income and maternal education, among others (n = 16).

All three interventional studies reported the difference in SER among the treatment group and control group, whereas two of them provided the difference in AL between the groups as well. The difference in the characteristics between the treatment and control groups was not mentioned in one trial. The remaining two trials reported no difference in age, baseline SER, baseline AL, and baseline corneal curvature between the control and treatment groups. In addition, one trial stated that their groups did not differ in myopia progression and axial growth at baseline, whereas the other trial had groups that were comparable for parental myopia and time spent on near work.

Because of the presence of heterogeneity in methodology and reported outcomes of the included studies, a meta-analysis could not be performed to gauge the cumulative ORs of myopia development based on the intake or serum concentrations of various nutrients and dietary elements. Therefore, we qualitatively described the outcomes of the included literature.

Noninterventional Studies

Among the 24 noninterventional studies that evaluated the association between nutrition and myopia, there were 11 cross-sectional studies, 8 birth cohort studies, and 5 retrospective studies (Table 2). Through these noninterventional studies, 28 different nutrients and dietary elements were assessed based on the evaluation of nutrient/dietary intake levels, whereas 6 different nutrients were evaluated via their levels in serum for their association with myopia. Ten of 28 nutrients assessed based on intake level and 2 of 6 nutrients assessed via serum level were investigated once and lacked replication studies (Fig. 2). The association of each of the nutrients and dietary elements with myopia is presented in Figs. 3 and 4. The odds for myopia development in association with intake of nutrients and dietary elements^{19,21,22,27,32,34,40} and serum concentration of vitamin D^{22–24,32,37,39} that are assessed in a few noninterventional studies are given in Figs. 5 and 6.

TABLE 2: Association of the nutrients and dietary elements with risk of myopia development in noninterventional studies

FIGURE 2: Number of studies that evaluated the association between myopia and intake of specific nutrients and dietary elements (A) serum concentrations of specific nutrients (B).

FIGURE 3: Type of association of myopia with the intake of energy intake (A), macronutrients (B), micronutrients (C), and other dietary elements (D). Positive association: myopic individual = higher intake of nutrients and dietary elements; negative association: myopic individual = lower intake of nutrients and dietary elements.

FIGURE 4: Type of association of myopia with the serum concentration of nutrients. Positive association: myopic individual = higher intake of nutrients and dietary elements; negative association: myopic individual = lower intake of nutrients and dietary elements.

FIGURE 5: Forest plot showing the OR of myopia in association with the intake of the energy intake (A), macronutrients (B), micronutrients (C), and other dietary elements (D) in the studies that reported the OR of myopia. *P < .05; ^#OR for high myopia. The OR of myopia for each study is obtained from the model that has been adjusted for the maximum confounding variables. OR = odds ratio.

FIGURE 6: Forest plot showing OR of myopia in association with the serum concentration of the nutrients in the studies that reported the OR of myopia. *P < .05; ^#OR for high myopia. The OR of myopia for each study is obtained from the model that has been adjusted for the maximum confounding variables. OR = odds ratio.

Cross-sectional Studies

Eleven cross-sectional studies evaluated the association of myopia with the intake of nutrients and dietary elements.^{13,14,19–21,25,27,28,33,34,38} Of the 11 studies, 7 reported that myopia is associated with nutrients and dietary elements. Berticat et al.³⁴ reported increased probability of myopia with a higher intake of refined carbohydrates in girls (OR, 1.07; 95% CIs, 1.02 to 1.13) and a lower intake of refined carbohydrates in boys (OR, 0.94; 95% CI, 0.89 to 0.98). Higher intake of protein,²¹ fiber,²⁰ copper,²⁰ magnesium,²⁰ solid food weight,²⁰ and higher zinc concentration in hair,³³ as well as lower serum concentration of vitamin D,^20,38 zinc,¹³ and selenium¹³ and lack of breastfeeding (OR, 0.50; 95% CI, 0.27 to 0.95),²⁷ were associated with the presence of myopia. Higher intake of saturated fats and cholesterol was found to be associated with longer AL in one study despite the lack of association of these nutrients with myopia development in the study.¹⁴ In another study that noted no association between breastfeeding and myopia, breastfed children were found to have more hyperopic SER compared with nonbreastfed children.¹⁹

Birth Cohort Studies

Eight birth cohort studies investigated the association of myopia with the intake of nutrients and dietary elements.^{12,23,24,26,29,35,37,40} Six studies reported that a reduced risk of myopia is associated with the increase in the intake of energy,¹² protein,¹² total fat (OR, 0.96; 95% CI, 0.92 to 1.00),⁴⁰ cholesterol,¹² vitamin B₁,¹² vitamin B₂,¹² vitamin C,¹² phosphorus,¹² and iron,¹² as well as an increase in the serum concentration of vitamin D/D₂^23,24,26,37 (OR, ranging between 0.65 and 0.91; 95% CI, 0.46 to 1.04). One study showed that a lower intake of calcium is associated with more myopic SER, although it was not found to be associated with myopia development.⁴⁰

Retrospective Studies

Five studies evaluated the nutrients and dietary elements for their association with myopia in a retrospective design.^{22,31,32,36,39} In one study, higher energy intake (OR, 0.72; 95% CI, 0.53 to 0.97) and higher serum concentration of vitamin D (OR, 0.55; 95% CI, 0.34 to 0.90) were found to be associated with decreased odds of development of high myopia.²²

Interventional Trials

Three dietary interventional trials were included in this review, with each of these trials assessing different nutrient or dietary elements for their role in myopia progression: proteins, 7-MX, and crocetin supplements (Table 3).^11,18,30 Gardiner¹¹ reported that children with myopia who consumed 10% of their calorie intake as animal protein had less myopia progression by −0.22 to −0.42 D compared with the control subjects with myopia during the treatment period of 1 year. Trier et al.¹⁸ found no significant difference in myopia progression between the children who were treated with 7-MX and those who were not. They noted a lesser axial elongation in the treated children by 98 μm (P = .048) over a duration of 24 months compared with the control subjects but only in the children who had moderate axial growth at baseline. Mori et al.³⁰ reported that children who were supplemented with crocetin progressed less by 0.08 D (P = .049) and had less ocular growth by 30 μm (P = .046) compared with the control group in 24 weeks.

TABLE 3 - Change in the SER and AL in the dietary intervention trials included in the review

S.N.	Source	Duration of trial	Intervention	Difference in SER(control subjects − treated)	Change in AL(control subjects − treated)
1	Gardiner11	1 y	10% of calorie intake as animal protein/Casilan tablets or no intervention	Children with myopia:  −0.25 to −0.9 = −0.22 D  −1.0 to −1.9 = −0.31 D  −≥2.0 = −0.42 D	NA
2	Trier et al.18	3 y	400 mg 7-MX or placebo tablets	After 12 mo:  Group 1 = −0.11 D (P = .22)  Group 2 = −0.09 D (P = .58) After 24 mo:  Group 1 = −0.22 D (P = .14)  Group 2 = 0.12 D (P = .66)	After 12 mo:  Group 1 = 55 μm (P = .07)  Group 2 = 31 μm (P = .59) After 24 mo:  Group 1 = 98 μm (P = .048)  Group 2 = 37 μm (P = .66)
3	Mori et al.30	24 wk	7.5 mg crocetin soft capsule or placebo capsule	0.08 D at 24 wk (P = .049)	30 μm at 24 wk (P = .046)

Group 1 = children with moderate baseline axial growth = 0.075–0.190 mm per 6 months; group 2 = children with high baseline axial growth = 0.200–0.390 mm per 6 months. 7-MX = 7-methylxanthine; AL = axial length; NA = not applicable; SER = spherical equivalent refraction; S.N. = study number.

DISCUSSION

This review included 27 studies that investigated if nutrition has any relation to the development and/or progression of myopia. The results of the review can be summarized as follows:

1. Each of the studies included in this review has assessed diverse combinations of nutrients and dietary elements for their association with myopia, with a limited number of nutrients being evaluated multiple times.
2. Overall, within the nutrients and dietary elements that were investigated in multiple noninterventional studies, most of the nutrients and dietary elements exhibited inconsistency in association with myopia; that is, the lower intake of a nutrient or dietary element was associated with an increased risk of myopia in one study, whereas the same nutrient or dietary elements when assessed in another study was either associated with a decreased risk of myopia or not associated with myopia.
3. Nevertheless, the outcomes of the noninterventional studies that showed an association of nutrients and dietary elements with myopia implicated rather a limited association due to the following reasons: (a) the ORs indicated minimal effects because most of the OR values lay closer to the line of unity, and (b) the 95% CIs were wide or overlapped with the unity line.
4. All three nutrients and dietary elements investigated in the intervention trials were implicated to have a role in myopia control. However, in two of the trials (Trier et al.¹⁸ and Mori et al.³⁰), the difference in the change of AL between the control and treated groups for a duration of 1 year is clinically minimal.

Hypotheses That Support the Role of Nutrition in Myopia

Although the role of nutrients and dietary elements in myopia remains unclear, there are several theories about how they influence the development and progression of myopia.

Macronutrients

Cordain et al.⁴¹ proposed that the increased consumption of high glycemic load carbohydrates promotes insulin resistance, which leads to an increased concentration of free circulating insulin-like growth factor 1 that sets in a cascade leading to the proliferation of scleral tissue resulting in ocular elongation. This theory is supported by findings from three studies included in the review. Harb and Wildsoet³⁹ noted that an increased level of insulin is associated with increased odds of myopia. Berticat et al.³⁴ reported that a higher intake of refined carbohydrates increased the probability of myopia in girls. Lim et al.¹⁴ found the higher intake of saturated fats and cholesterol to be associated with longer AL.

Micronutrients

There are several assumptions of how vitamin D might affect the development of myopia, which include its antiproliferative effect on scleral growth,²⁰ its role in retinoic acid metabolism,^20,22,24 its influence on the functioning of ciliary muscle via calcium pathway,^20,22 and its impact on the release of dopamine.²⁴ The majority of the studies in this review that investigated serum concentration of vitamin D found it to be lower in individuals with myopia. However, given the contrary findings of no association in the studies that assessed the dietary intake of this nutrient, it is unclear if there is any causal relationship between vitamin D and myopia or if it is a mere biomarker for the time outdoors as indicated by a previous meta-analysis.⁴²

Vitamin A is the precursor for retinoic acid; therefore, it is postulated that alteration of this element will affect the metabolism of retinoic acid,³⁵ which was found to be associated with the development of myopia in animal models.⁸ However, studies conducted on humans failed to replicate this theory, with all the studies included in the review showing no relationship between vitamin A and myopia.

The antioxidative property of the trace elements, such as zinc, copper, and selenium, which were found to be associated with myopia in few of the included studies, is theorized to counteract the myopia development which is believed to occur as a result of oxidative stress.⁴³ Zinc had a contrasting relationship with myopia when it was evaluated in two different specimens, that is, lower concentration in the serum and higher concentration in the hair samples in different studies by the same authors, who conjectured that this variation might be a consequence of the lesser availability of this nutrient in the myopic eyes because of overaccumulation in their hair.^13,33

Dietary Elements

Breastfed milk contains polyunsaturated fatty acids such as arachidonic acid and docosahexaenoic acid, which are believed to be beneficial for visual and cognitive development.^44,45 This theory is supported by a recent study that has shown that omega-3 polyunsaturated fatty acid supplementation increases choroidal blood flow, which is speculated to inhibit the scleral hypoxia and subsequent scleral remodeling cascade that leads to the development of myopia.⁴⁶ In this review, breastfeeding showed an association with reduced risk of myopia in a single study.²⁷ Although another study did not find a correlation between myopia and breastfeeding, they noted it to be associated with more hyperopic refractive error, which might be because of the younger sample (6 to 72 months) in whom the likelihood of myopia is low.¹⁹ Both of these studies implicated that breastfed children tend to have a higher spherical equivalent of refraction than those who are not.

7-Methylxanthine is hypothesized to control myopia progression by upregulating the collagen content that subsequently improves the biomechanical properties of the sclera, although the exact site of action is unclear.⁹ 7-Methylxanthine is conjectured either to act on the sclera directly or by modulating the muscarinic receptors of dopamine or acetylcholine at the retinal level, which triggers the emmetropization cascade.^9,18 The action of crocetin is believed to occur through activation of the myopia suppressive gene, early growth response 1 (Egr1).^10,30 Another possible pathway by which crocetin is suspected to function is by improvement in choroidal circulation resulting in the increase of choroidal thickness.^10,30 Both of these elements, 7-MX (n = 1) and crocetin (n = 1), showed a minimal effect on controlling the axial elongation in two diverse clinical trials.^18,30

Limitations in Studies Included in the Review

The strength of our review is that we provided a comprehensive report of studies that investigated association of myopia with diverse nutrients and dietary elements. The included studies were conducted in various regions across the globe. Although the quality of most of the included studies in the review was good according to the risk-of-bias assessment, there were a few limitations in these studies. The method of quantification of the nutrients and dietary elements was not the same in all the studies, with the majority of these studies using subjective approaches of food frequency questionnaires or 24-hour dietary recalls. Although some studies did use objective measures such as gauging the concentration of the nutrient in the serum or the hair sample, most of these studies were cross-sectional or retrospective. Given that the minimum number of confounding factors to be adjusted on the Newcastle-Ottawa Scale to gain a score is two, most of the studies included in the review scored well on the NOS, as it was a low cutoff. However, adjustment for all the potential confounding factors that influence the development and progression of myopia was not done in most of the studies (n = 23). The outcome was reported differently in each of these studies, with mean/median differences reported in some studies, whereas other studies reported OR, creating difficulty in understanding the overall effect of the nutrients and dietary elements on myopia development and progression. Also, there were a few studies (n = 3) in which the assessment of nutritional status and refractive error was done at different periods of the study; that is, nutritional status was assessed at an earlier age, whereas the refractive error was evaluated at a later age.

Limitations of the Review

In addition to the limitations of the included studies, this review also has a few limitations. First, because not enough studies examined each nutrient/dietary element, we were unable to quantify the extent of the association of each nutrient/dietary element with myopia development or progression and instead reported the associations qualitatively. Second, considering that fewer studies reported how SER/AL is influenced by the intake of nutrients/dietary elements in the form of the regression coefficient, we mostly reported the association of nutrition with increased risk of myopia and not in terms of refractive error or AL.

CONCLUSIONS

Most of the studies included in the review were of noninterventional design, which did show an association of specific nutrients and dietary elements with myopia. However, these associations were not replicated in the other studies. Similarly, three different interventional trials reported three diverse nutrients and dietary elements to have a myopia control effect. To conclude, this review implies that there is some evidence to indicate a potential influence of specific nutrients and dietary elements in myopia development, which are supported by several theories. However, given the vast, diverse, and complex nature of nutrition, more systematic investigation is warranted to comprehend the extent to which these specific nutrients and dietary elements are associated with myopia through longitudinal studies by subduing the limitations in the existing literature.