Vitamin D and Ocular Diseases: A Systematic Review

The contributory roles of vitamin D in ocular and visual health have long been discussed, with numerous studies pointing to the adverse effects of vitamin D deficiency. In this paper, we provide a systematic review of recent findings on the association between vitamin D and different ocular diseases, including myopia, age-related macular degeneration (AMD), glaucoma, diabetic retinopathy (DR), dry eye syndrome (DES), thyroid eye disease (TED), uveitis, retinoblastoma (RB), cataract, and others, from epidemiological, clinical and basic studies, and briefly discuss vitamin D metabolism in the eye. We searched two research databases for articles examining the association between vitamin D deficiency and different ocular diseases. One hundred and sixty-two studies were found. There is evidence on the association between vitamin D and myopia, AMD, DR, and DES. Overall, 17 out of 27 studies reported an association between vitamin D and AMD, while 48 out of 54 studies reported that vitamin D was associated with DR, and 25 out of 27 studies reported an association between vitamin D and DES. However, the available evidence for the association with other ocular diseases, such as glaucoma, TED, and RB, remains limited.


Introduction
Vitamin D has diverse functions in maintaining human health, including regulating gene expression, immune system, inflammation, cell proliferation and differentiation, apoptosis, and angiogenesis [1,2]. Vitamin D 3 , or cholecalciferol, is produced from its precursor, 7-dehydrocholesterol, in the epidermal layer of skin under exposure to sunlight, or is obtained from the diet. It is metabolized in the liver and kidneys to its biologically active forms, 25-hydroxyvitamin D (25(OH)D 3 ) and 1,25-dihydroxyvitamin D (1,25(OH) 2 D 3 ), respectively. The latter is also known as potent steroid hormone calcitriol. Reduced sun exposure will lead to vitamin D deficiency [3,4]. Low vitamin D levels have been associated with many diseases, including cardiovascular diseases [5,6], hypertension [7], diabetes mellitus [8,9], and cancers [10].
The vitamin D status of an individual is usually determined by serum 25(OH)D 3 instead of 1,25(OH) 2 D 3 because of its longer circulating half-life and higher concentration Moreover, 1,25(OH) 2 D 3 can penetrate the cell membranes, either as a free molecule or in DBP-1,25(OH) 2 D 3 complexes. It then binds to the vitamin D receptor (VDR), facilitating the interaction of VDR with the retinoic X receptor (RXR) [16]. This VDR-RXR heterodimer binds to both positive and negative vitamin D response elements in target genes, influencing gene transcription [17]. Hence, the presence of VDR suggests the local activity of vitamin D [18]. In particular, VDR has been detected in different parts of the eye, including the epithelium and endothelium of the cornea, lens, ciliary body, retinal ganglion cells (RGCs), inner nuclear layer, photoreceptors, and retinal pigment epithelium (RPE) [19,20]. Genetic alternations of the VDR gene could lead to defects in gene function, calcium metabolism, cell proliferation, and immune function. DBP is mainly responsible for the transportation of vitamin D and its metabolites.
Levels of active vitamin D in the body are regulated by the enzymes 25-hydroxylase, 1α-hydroxylase, and 24-hydroxylase [15]. In a recent study, the 25(OH)D 3 and 1,25(OH) 2 D 3 generating enzymes 25-hydroxylase (CYP2R1 and CYP27A1) and 1α-hydroxylase (CYP27B1), as well as the deactivating enzyme 24-hydroxylase (CYP24A1), were found to be strongly localized at the complementary regions of the ciliary body, RPE, neural retina, corneal epithelium and endothelium, and scleral fibroblast, suggesting that vitamin D in the eye is locally produced, activated, and regulated [21,22]. Moreover, vitamin D-dependent calcium binding protein calbindin, a vitamin D metabolizing protein, was shown to be expressed throughout the human retina [23]. Some of the cohort studies reported the correlation of metabolic enzymes in ocular diseases. In diabetic patients, retinal CYP27B1 was found to correlate strongly with VEGF-A in the eyes [24]. In a cohort of patients with Vogt-Koyanagi-Harada disease, a non-synonymous variant of CYP2R1 was found in 17 of 39 patients, suggesting that the variant in CYP2R1 may play a role in VKH pathogenesis [25].
Some of the vitamin D regulating proteins, such as ferredoxin reductase participating in the activation of vitamin D in the kidney, are metalloproteins. Vitamin D is able to interact with the matrix metalloproteinase. Metal deficiency may affect ocular condition. However, only one study found significantly lower serum calcium levels in blepharospasm patients, but no significant difference in magnesium, phosphorus, or vitamin D [26].
Therefore, the potential of vitamin D to regulate various processes of potential relevance to ocular diseases has been acknowledged. Studies investigating the roles of vitamin D in ocular tissues and ocular disease pathogenic pathways have been carried out and will continue to contribute towards our understanding of ocular disease mechanisms and help establish effective intervention.

Vitamin D and Ocular Diseases
The potential effect of vitamin D deficiency on human health is a big concern. Recently, especially over the past few years, since the last published review articles related to vitamin D and ocular diseases, more and more studies investigating the relationship between serum vitamin D level and ocular diseases were published, including some prospective studies examining this relationship and therapeutic effects of vitamin D. Currently, review articles related to vitamin D and ocular disease are available [18,20]. To update this and reach a comprehensive understanding, hence, we performed a systematic review here to summarize the evidence revealing the association between vitamin D and ocular diseases.

Method of Literature Search
This systematic review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines [27]. The protocol is described as follows.  228), hypertensive retinopathy (PubMed: 2; WOS: 2), retinal arterial occlusion (PubMed: 0; WOS: 0), retinal venous occlusion (PubMed: 1; WOS:0). The search results from both databases were exported and imported in Covidence, which is a software for literature screening in systematic reviews. Among these 6813 results, the system detects that 3533 results were duplicated. They have been removed prior to the screening of the articles (Figure 2).

Inclusion and Exclusion Criteria
The inclusion criteria for studies were: (1) written in English; (2) evaluating the association between blood vitamin D and different ocular diseases in a randomized controlled trial, prospective study, cross-sectional study, or case-control study. After a review of abstracts, relevant articles were retrieved and reviewed. Bibliographies of these articles provided further references. All retrieved records were reviewed by two independent reviewers (HNC and XL). Uncertainties were resolved via discussion with another reviewer (XJZ).

Risk of Bias Assessment
Included interventional studies (both randomized controlled trials and clinical controlled trials) were assessed for quality according to the RoB tool for randomized trials from the Effective Practice and Organisation of Care (EPOC) Group. The assessment for the clinical controlled trial was assessed according to the suggestion from previous literature that both "random sequence generation" and "allocation concealment" were scored as "high risk", while grading the remaining items as RCT [28]. We further modified the RoB tool by allocating 1 point to "low risk", 0.5 point to "unclear risk" and 0 points to "high risk". There are a total of 9 items to be assessed using the RoB tool, and hence, the total number of points for the RoB tool is 9 points, while those cohort, case-control, cross-sectional studies were assessed for quality according to the LEGEND (Let Evidence Guide Every New Decision) System designed for Cincinnati Children's Hospital [29].

Myopia
Myopia is an important public health problem worldwide [30]. The etiology of myopia is complex, with both genetic and environmental risk factors [31][32][33][34]. Epidemiologic evidence indicates that time spent outdoors is a protective factor against myopia development [34][35][36][37][38], yet the underlying mechanism is unclear. Since the main source of vitamin D is sunlight exposure, vitamin D is linked to myopia, hypothesizing that a vitamin D pathway may mediate the protective effect of time spent outdoors on myopia. Evidence from studies on the relationship of vitamin D and myopia is summarized in Table 1. There is no association between serum 25(OH)D concentration and myopia in the 6-14 years old Chinese children.
4a * NIH quality assessment tool for before-after (Pre-Post) study with no control group; # LEGEND for case-control, cohort, and cross-sectional studies, rating of the studies follow the guidelines from LEGEND.
As demonstrated in Table 1, the association between vitamin D and myopia is controversial in cross-sectional studies. Many studies suggest that the serum 25(OH)D3 level shows an inverse association with myopia and may have a protective effect on myopia [39,[41][42][43][44]47,50,51,53,55]. However, several case-control studies from Australia [52], Denmark [46], and the US [54] found that the risks of myopia are not related to their neonatal vitamin D levels.
Nevertheless, it is important to distinguish the causation between vitamin D and myopia. A large longitudinal cohort study found that 25(OH)D 3 levels correlated with self-reported time spent outdoors; however, no evidence suggested that the participants' serum vitamin D levels were independently associated with myopia [40]. Another study of preterm children also suggested that more time spent outdoors was associated with a lower risk of myopia, despite serum 25(OH)D 3 concentrations not being shown to relate to myopia [48]. However, an Australian perspective study showed that, in young adults, myopia was most strongly associated with recent 25(OH)D3 concentrations, which is a marker of time spent outdoors [49].
Our meta-analysis found that the risk of myopia is inversely associated with blood 25(OH)D 3 concentration after adjusting for sunlight exposure or time spent outdoors. However, this relationship was not significant among individuals under 18 years of age [57]. Polymorphisms in the vitamin D pathway genes may affect the development of myopia. One study reported the association of VDR polymorphisms, rs2853559, with myopia [58]. However, the results of other studies suggested that the true contribution of the vitamin D pathway to myopia could be negligible [42,45,59]. Our meta-analysis suggested that polymorphisms in the VDR gene are not associated with myopia [57]. On the other hand, animal studies proved that violet light (VL, λ = 360-400 nm) can suppress myopia progression, whereas no therapeutic effects were observed with UVB radiation (λ = 290-315 nm) [60], suggesting that UVB exposure and its dependent vitamin D synthetic pathway may not have a protective effect on myopia progression.
In conclusion, from the literature evidence, we know that, although blood 25(OH)D 3 concentration is inversely associated with the risk of myopia, it seems unlikely that vitamin D has a direct protective effect on myopia progression. Instead, vitamin D levels may only serve as a biomarker for outdoor exposure.

Age-Related Macular Degeneration
As a chronic, progressive, degenerative disease, age-related macular degeneration (AMD) is a major cause of central blindness among people aged 60 years or over worldwide [61,62]. Oxidation, inflammation, and angiogenesis contribute to the pathogenesis of AMD, resulting in the dysfunction of RPE [63], Bruch's membrane, and choriocapillaries [64]. In an aging retina, the complement cascade [65,66] and the tissue resident macrophage (retinal microglia) activation pathway [67] ultimately cause protein damage and aggregation, and degeneration of the RPE [68]. Angiogenesis, often caused by oxidative stress and inflammatory reactions, plays a major role in the development and progression of exudative AMD, potentially leading to severe and permanent visual impairment.
The results of studies on cell lines and animal models have shown that vitamin D can protect cells or reduce oxidative stress [69][70][71]. Vitamin D has an anti-inflammatory role in chronic inflammatory diseases by decreasing the proliferation of T-cells and the production of pro-inflammatory agents [72,73]. On the other hand, vitamin D exerted an inhibitory effect on the angiogenesis signaling pathway [74,75], which may play a protective role in exudative AMD development and/or progression. Morrison et al. studied the variants in the vitamin D catabolizing enzyme, CYP24A1, and reported that variants (rs1570669, rs1570670, rs2274130, rs2296239, and rs4809957) were associated with reduced risk for AMD [76]. Table 2 summarized the studies on vitamin D and AMD. Case-control studies with small sample sizes suggest that AMD patients always have relatively low levels of serum vitamin D [77][78][79][80][81][82][83][84][85][86][87], except in a Iranian study, which did not find any significant correlation between serum vitamin D level and AMD [79]. However, this association seems to change in cross-sectional studies with larger sample sizes. Population-based studies held in France [88], the United States [89,90], and Israel [91] did not support a specific role for vitamin D in AMD, but vitamin D may work in some specific populations. An analysis of a sample of 1313 US participants indicated that high serum 25(OH)D3 concentrations may protect against early AMD in women less than 75 years old [92], while another US study supported the fact that levels of serum vitamin D were inversely associated with early AMD but not advanced AMD [93]. A Korean study had 17,045 participants and found that a high level of vitamin D was inversely associated with late AMD in men but not women [94]. Vitamin D deficiency in the European population was found to be associated with nvAMD, but the adjusted OR was small, and cannot exclude residual confounding [95]. Although evaluation of serum 25(OH)D 3 was higher in unaffected individuals than in their affected siblings, but the finding did not reach statistical significance.  Prospective studies, however, have not found a consistent association between vitamin D and the risk of developing AMD. In a large prospective cohort study of 2146 participants with a mean follow-up time of over 9 years, high dietary intake of vitamin D was significantly associated with a 40% lower risk of progression to advanced AMD [99]. However, recently, a nationwide, placebo-controlled, randomized clinical trial found that supplementing vitamin D had no significant overall effect on AMD incidence or progression in healthy people [98]. For this trial, 25,871 participants with a median age of 67.1 years were divided into four groups, receiving vitamin D supplements (2000 IU/day), ω-3 fatty acids (1 g/day), a combination of both, and placebo, respectively. After a median follow-up period of 5.3 years, no significant differences were found in the incidence or progression of AMD when compared with baseline [98]. This study suffered from a lack of stratification by clinical manifestations of AMD, a relatively short follow-up period for chronic disease, and a reliance on self-reported AMD diagnosis, leading to inconsistencies with the previous two studies [98,104].
In summary, cross-sectional studies suggest that vitamin D may have a protective effect on AMD formation, but this effect is small or may only work in a specific population. Furthermore, evidence from prospective cohort studies showed that continuously supplementing vitamin D may not reduce the risks of AMD over a period of several years.

Glaucoma
A leading cause of irreversible blindness, glaucoma is a group of optic neuropathies involving the death of retinal ganglion cells (RGCs) and the loss of their axons [105,106]. Two cross-sectional studies in South Korea reported that vitamin D deficiency is associated with glaucoma [107,108]. Similarly, a Chinese study found that the vitamin D deficiency, along with the presence of the BsmI 'B' allele and TaqI 't' allele of the VDR gene, are relevant risk factors for glaucoma development [109]. Other studies in France [110], Croatia [111], the United States [112,113], and Turkey [114] have demonstrated that glaucoma patients have lower serum vitamin D levels compared to normal controls. However, another Turkish case-control study found no statistically significant difference in serum vitamin D levels between glaucoma patients and control subjects [115]. Similarly, a recent large-sample study in the United States showed that dietary intake, supplements, and serum levels of vitamin D are not significantly related to the risk of glaucoma [116]. Notably, ethnicity may contribute to the pathogenesis of glaucoma, giving rise to different conclusions among these studies [117]. Most of the literature reported the association between vitamin D and glaucoma and that a lower vitamin D concentration was found in glaucoma patients when compared with the control group [108][109][110][111]114], however, there were no findings on the association between vitamin D and the severity. Increases in vitamin D were associated with lower risks of having glaucoma (fourth quintile versus first quintile, OR 0.713, 95% confidence interval, 0.520 to 0.979) [108]. Only a limited study reported no statistically significant difference between the glaucoma group and the control group [108]; significantly lower vitamin D can only be found in advanced glaucoma patients [112] (Table 3). In the multivariable-adjusted logistic analysis, the odds ratio of glaucoma in the fourth quintile was significantly lower than that of the first quintile in females (odds ratio, 0.713; 95% confidence interval, 0.520 to 0.979). Lower 25(OH)D level was significantly associated with an elevated risk of glaucoma in females compared with higher 25(OH)D level. The IOPg values were higher in cases of vitamin D deficiency. Corneal biomechanical differences in patients with lack of vitamin D were not statistically significant. 3b * The Effective Practice and Organisation of Care (EPOC) RoB Tool for randomized trials; # LEGEND for case-control, cohort, and cross-sectional studies, rating of the studies follow the guidelines from LEGEND.
High intraocular pressure (IOP) is an important risk factor for glaucoma. In an animal study on non-human primates, vitamin D treatment modulated the expression of IOP-regulating genes, with IOP falling in a dose-dependent manner [121]. However, a human study found no association between serum 25(OH)D 3 levels and IOP, nor significant changes in participants' IOP levels after receiving 6 months of oral vitamin D supplements (20,000 IU twice weekly) compared to the placebo group [118]. This contradiction may be due to the oral intake of vitamin D, which may lower the availability of vitamin D in the eye. Patients with glaucoma were found to have lower 25(OH)D concentrations in aqueous humor [119], and the IOP values were higher in cases of vitamin D deficiency [120]. Further studies are required to determine if vitamin D can be a potential intervention for glaucoma, especially through testing different supplement approaches.
Some studies have identified vitamin D as an independent risk factor for glaucoma; however, the role that vitamin D plays in relation to glaucoma remains uncertain. Apart from the elevated IOP pathway, vitamin D may participate in the oxidative stress pathway due to its anti-oxidation and anti-inflammatory abilities. In an in vivo study, 1,25(OH) 2 D 3 ameliorated the effects of oxidative stress from hydrogen peroxide-induced toxicity in human RPE cells through antioxidant signaling pathways, leading to lower levels of reactive oxygen species (ROS), cytokines, and vascular endothelial growth factor (VEGF) [122]. Another study demonstrated that vitamin D significantly altered the inflammatory-related genes in glaucoma, suppressing the expression of the angiotensin I-converting enzyme (ACE), carbonic anhydrase (CA), and Ras homologue gene family member A (RhoA), while significantly increasing the expression of the cytokine A20 precursor (CCL20) in the small intestines of rats [123]. ACE inhibitors are neuroprotective for cultured retinal neurons and can lower IOP in humans [124,125], while CA inhibitors can lower IOP and increase blood flow in the retinal vasculature and optic nerve [126]. The suppression of RhoA through subsequent vitamin D treatment can reduce aqueous outflow resistance and enhance fluid outflow [127,128]. Lastly, CCL2, an intraocular pressure responsive cytokine, possesses a potential role in intraocular pressure regulation [129].
In summary, all reported studies are cross-sectional studies (case-control studies and population surveys) and suggested the protective associations of vitamin D on glaucoma. Future studies should employ randomized clinical trial designs to investigate the causal relationship between glaucoma and low vitamin D levels or calcitriol deficiency.

Diabetic Retinopathy
Because of its ability to inhibit neovascularization, vitamin D has been studied in the development of diabetic retinopathy (DR). Many observational studies have examined the relationship between vitamin D levels and the prevalence or severity of DR, with most identifying an inverse association with both type 1 and 2 diabetes . However, a Chinese study has reported a lack of association between vitamin D deficiency and DR after adjusting for all potential covariates, such as demographics, physical measurements, laboratory measurements, related complications, comorbidities, and medications [166]. Another Indian study suggested a possible association of vitamin D deficiency with type 2 diabetes, but not specifically with DR [167]. As demonstrated in Table 4, in general, some of the studies reported an inverse correlation between the serum vitamin D and severity of retinopathy [130,[132][133][134][135][136][137][138][139]152,157,168,169]; similar findings were also reported, for example, the co-existence of low vitamin D and microvascular complications [131] or the association between the severity of DR and the prevalence of vitamin D deficiency [133,161,165] (Table 4); while some studies reported either no association or no significant difference between DR patients and healthy controls [170][171][172][173]. The agreement of the association between vitamin D deficiency and neuropathy is lower when compared with retinopathy. While some studies report that the risk for having diabetic neuropathy is higher in those with vitamin D deficiency [134,137], there is limited research on contrasting findings [174]. Further investigations are warranted. There is an association between severity of diabetic retinopathy and prevalence of vitamin D deficiency (β = 1.3, p = 0.01, unadjusted; and β = 1.2, p = 0.01, adjusted for age and obesity status given their clinical significance). However, regression analysis of these data did not demonstrate a statistically significant relationship between the 2 variables (β = −0.04, p = 0.07, unadjusted; and β = −0.03, p = 0.13, adjusted). The findings were inconclusive about the existence of a relationship between retinopathy severity and serum 25(OH)D 3 concentration.      Hypovitaminosis D is often in T2DM, especially in those with proliferative DR. 3b Diabetic retinopathy (DR), background diabetic retinopathy (BDR), non-proliferative retinopathy (NPR), preproliferative diabetic retinopathy (pre-PDR), proliferative diabetic retinopathy (PDR), Diabetes mellitus (DM), Type 1 diabetes mellitus (T1DM), type 2 diabetes mellitus (T2DM), diabetic kidney disease (DKD); * The Effective Practice and Organisation of Care (EPOC) RoB Tool for randomized trials; # LEGEND for case-control, cohort, and cross-sectional studies, rating of the studies follow the guidelines from LEGEND.
Besides cross-sectional studies, a population-based prospective study also showed that a high level of vitamin D was associated with a lower risk of DR after 3 years [178]. A doubleblind, placebo-controlled trial found that low blood 25(OH)D 3 levels were associated with an increased risk of macrovascular and microvascular disease events among type 2 diabetics [139].
DR is a serious microvascular complication of diabetes. The characteristics of early DR include the loss of pericytes from retinal capillaries, the appearance of acellular capillaries and microaneurysms, and the breakdown of the blood-retinal barrier. In the proliferative phase of DR, neovascularization in the retina may occur, which significantly increases the probability of vision loss [184,185]. Potential mechanisms that explain how vitamin D can prevent DR include insulin resistance, immune regulation, anti-inflammation, and anti-angiogenesis. Animal studies have shown that vitamin D is important for insulin synthesis and can improve the body's sensitivity to insulin, reducing the risk of insulin resistance [186,187]. Other studies have found that vitamin D treatment decreased the retinal expression of VEGF and the transforming growth factor TGF-β1 in rats [188], which may have protective effects on the retina. VDR has also been implicated in the pathogenesis of DR [189]. A meta-analysis of seven studies evaluating the association of the VDR gene polymorphisms with DR found that the FokI polymorphism of the VDR gene has a significant association with DR susceptibility [190]. Apart from the VDR polymorphism, other studies have proposed different protective mechanisms of vitamin D on DR, including protecting the vasculature [191][192][193], reducing oxidative stress [194,195], modulating inflammation and immune responses [180,[196][197][198], inhibiting the renin-angiotensin aldosterone system [199,200], reducing the effects of advanced glycation end products [201,202], reducing endoplasmic reticulum stress [203,204], regulating endothelial cells apoptosis [205], and regulating diabetic leukostasis [206]. Further studies are needed to determine the exact mechanisms of vitamin D on DR.
In summary, even though there are no consistent associations between vitamin D level and DR in observational studies, more than 30 reports suggested an inverse relationship. The same conclusion is made in perspective studies, although the causal relationship has not been identified. Further studies should investigate whether vitamin D supplementation can reduce the risk of DR.

Dry Eye Syndrome
Dry eye syndrome (DES), or dry eye disease [207], is a common eye disease affecting about 12% of the world's population; prevalence was lowest in North America (4.6%) and highest in Africa (47.9%) [208]. Many factors are related to DES, including hormonal alterations, environmental changes, and aging [209]. DES is accompanied by the inflammation of the ocular surface, which may cause visual disturbances, tear film instability, and potential damage [210]. Whereas an increasing number of studies have shown that a relationship exists between vitamin D and DES, their findings remained controversial. Some cross-sectional studies suggested an inverse correlation between vitamin D levels and ocular surface disease index (OSDI) scores or DES incidence [211][212][213][214][215][216][217][218][219][220][221][222][223][224][225][226], while three others have not reached a significant conclusion [227][228][229][230], as demonstrated in Table 5.  Decreased serum vitamin D was associated with exaggerated symptoms in dry eye patients with mild dry eye signs. An inverse correlation (r = −0.569; p = 0.0110) was observed between serum vitamin D levels and OSDI score in the patient cohort. Tukey's multiple comparisons test showed a significant difference between the OSDI score of patients with serum vitamin D less than or equal to 10 ng/mL and greater than 20 ng/mL.

7/9 * (RoB)
Seok Hyun Bae [219] 2016 South Korea Retrospective observational study 105 DES refractory to conventional treatment and vitamin D deficiency patients Vitamin D supplementation is effective and useful in the treatment of patients with DES refractory to conventional treatment and with vitamin D deficiency. The TBUT in males was increased after 2 weeks compared to pre-treatment and in female it was increased after 2 and 6 weeks compared to pre-treatment (p = 0.041, <0.001 and <0.001, respectively, paired t-test). OSDI score in men was lower at 6 weeks compared to pre-treatment and in women it was lower at 10 weeks compared to pre-treatment (p = 0.033 and 0.012, respectively, paired t-test).  As a consequence of the presence of VDR and 1α-hydroxylase in different parts of the eye, vitamin D replacement improves tear hyperosmolarity that is considered to be induced by ocular surface inflammation. The change of TFO was negatively correlated with the change of 25(OH)D3 before and after replacement in patients with dry eye disease (r = −0.390, p = 0.049). 2b Chih-Huang Yang [233] 2017 Australia Case-control study 29 DES patients and 29 healthy controls Low vitamin D levels (<50 nmol/l) were associated with dry eye symptoms in older individuals but not those diagnosed with dry eye. Vitamin D supplement increased the vitamin D levels, and improved dry eye symptoms, the tear quality and ocular surface conditions. 3a Goktug Demirci [221] 2018 Turkey Cross-sectional observational study 30 Vitamin D deficiency patients and 30 healthy controls Vitamin D deficiency is associated with tear hyperosmolarity and tear film dysfunction. The Schirmer I test values and TBUT measurements for VDD were significantly lower compared with controls (p < 0.001).
Patients with vitamin D deficiency may be prone to dry eye. The OSDI score was decreased in the IM group (intramascular supplementation of vitamin D for 2 weeks) after cholecalciferol supplementation compared with pretreatment, whereas that in the none group and oral group was not different between after cholecalciferol supplementation and pretreatment. The effect of topical carbomer-based lipid-containing artificial tears and hyaluronate was dependent on serum 25HD levels.
4a Serum 25(OH)D 3 level might be associated with dry eye severity in primary SS. Among dry eye parameters, the corneal staining score, conjunctival staining score, Schirmer I value, and TBUT were statistically worse in the serum 25(OH)D3 deficiency group compared with the normal group (p < 0.05). The conjunctival staining score showed a significant difference between the deficiency group and insufficiency group (p < 0.05). The tear measurements of the pediatric type 1 diabetes mellitus were lower than those in the healthy pediatric control group. The accompanying VDD made this situation more pronounced. The correlations between the vitamin D level and the Schirmer test, OSDI score, CSS, TF-BUT measurements, TMA, and TMH values were examined in T1DM group and control groups. Although there was a significant correlation in all measurements in T1DM group, there was no correlation in control groups. High intake of total fat, SFAs, oleic acid, and vitamin D may be inversely associated with the prevalence of MGD in Japanese individuals. Vitamin D intake in the MGD group was significantly lower than that in the non-MGD group (p = 0.039). Multivariate adjusted odds ratios (95% confidence intervals) between extreme quintiles of intake of vitamin D for MGD prevalence was 0.38 (0.17-0.87).

8/9 * (RoB)
Alireza Eslampoor [237] 2022 Iran RCT 100 dry eye disease patients with concurrent vitamin D deficiency Vitamin D supplementation as an adjuvant to routine dry eye treatment improves ocular surface hemostasis parameters, results in better tear stability and a more improved tear osmolarity in patients with vitamin D deficiency.

4a
Jain Nikita [238] 2022 India Hospital-based cross-sectional 60 VDD patients and 60 normal levels subjects A significant difference in the mean values of Schirmer I and Schirmer I test (with anesthesia) (p < 0.001) was seen between the case and control groups. A significant difference in the mean values of TFBUT (p < 0.001) and OSDI scores (p < 0.01) was also seen between the two groups. Several clinical trials investigated the treatment effects of vitamin D supplementation on DES symptoms, as demonstrated. Some of them only involved DES patients [232,236,237], while four other studies set healthy controls [219,231,233,235]. All these studies concluded that vitamin D can improve the DES symptoms of tear quality and ocular surface conditions. However, these studies may suffer from a small sample size or lack of a placebo-control group. Further well-design clinical trials with more samples are required to better understand the relationship between vitamin D and DES.
The key mechanism of vitamin D on DES may involve its antioxidation, anti-inflammatory, and immune-regulatory effects [239][240][241]. Vitamin D deficiency may cause the inflammation of the ocular surface and ultimately DES [233]. Conversely, vitamin D may relieve DES through inhibiting the interleukin-6 (IL-6) inhibitor [233], the key mediator of localized inflammation [242]. Moreover, vitamin D can suppress the release of inflammatory cytokines and stimulate the release of antioxidant cytokines in tears. Lastly, vitamin D can improve corneal epithelial barrier functions [233], which may improve ocular conditions. Apart from the role of vitamin D in the pathogenesis of DES, a study reported that the expression of VDR and CYP27B1 (vitamin D metabolism enzyme) was significantly decreased in DED patients, suggesting the possible involvement of the vitamin D regulatory enzyme in protecting the human eye from dry eye [243]. A study on SNPs of the VDR gene Apa-1, Bsm-1, Fok-1 and Taq-1, reported the association of Apa-1 and Taq-1 with the risk of DES [244].
In conclusion, the exact mechanisms of vitamin D on DES are unclear, but evidence from cross-sectional studies seems to suggest that vitamin D has a protective effect against DES. Limit evidence from clinical trials also suggests that vitamin D supplementation could help to improve DES symptoms, but a further placebo RCT is needed to verify this treatment effect.

Thyroid Eye Diseases
Thyroid eye disease (TED), also known as Graves' ophthalmopathy (GO), is an autoimmune inflammatory disorder. Few studies have examined the relationship between vitamin D and TED (Table 6). A pilot study in Texas, United States found prevalence rates of 20% and 31% for vitamin D deficiency and insufficiency among TED patients, respectively [245]. Another retrospective case-control study comparing vitamin D levels between Graves' disease patients and TED patients found that low serum vitamin D was associated with TED [246]. Assessing and supplementing vitamin D levels may be an important addition to the early management strategies of GD patients. Since vitamin D can regulate immune responses and reduce inflammation, there is a definite need to further evaluate the role of vitamin D deficiency in TED patients. The current evidence for TED suggests that vitamin D would be associated with TED, however, the evidence is limited. It is worthwhile to further investigate the effect of vitamin D on TED. Thyroid eye disease (TED), Graves' disease (GD); # LEGEND for case-control, cohort, and cross-sectional studies, rating of the studies follow the guidelines from LEGEND.

Uveitis
Uveitis is the inflammation of the uvea driven by the T-cells [248]. It can be described as a failure of the ocular immune system, with the disease resulting from inflammation and tissue destruction. Since vitamin D can inhibit inflammation, influence T-cell responses, and regulate the immune system, it is necessary to examine its involvement in the development of uveitis. In an experimental autoimmune uveitis model, the oral administration of calcitriol was found to prevent and reverse the progression of uveitis by reducing the immunological response [249].
Although in a population-based study, none of the 25 uveitis patients were found to have vitamin D deficiency [250], two large retrospective case-control studies in the United States found an association of lower vitamin D levels with uveitis and scleritis, respectively [251,252]. Other case-control studies have been conducted on patients with specific types of uveitis, such as anterior uveitis (AAU) or non-infectious anterior uveitis [253][254][255], Vogt-Koyanagi-Harada (VKH) disease [256], sarcoidosis-associated uveitis [257], and juvenile idiopathic arthritis (JIA)-associated uveitis [258] (Table 7). All these studies suggested that the vitamin D deficiency is associated with uveitis development. A recent prospective case-control study consistently reported the association of vitamin D levels with active and inactive non-infectious uveitis patients. Vitamin D levels are related to uveitis severity [259].
There are reported associations between VDR polymorphisms and uveitis. Singlenucleotide polymorphisms (SNPs) of the CYP2R1, CYP27B1, CYP24A1 and DHCR7 genes are linked with lower circulating vitamin D levels [260][261][262][263]. A study in a Chinese cohort found that the frequencies of the genotype TT and T allele of DHCR7 rs12785878 were both significantly higher among ocular Behçet disease (BD) patients compared with healthy controls; however, similar associations were not found for the VKH, AAU with ankylosing spondylitis (AS), or pediatric uveitis [263]. Another cross-sectional study found that gene variants involved in vitamin D anabolism and catabolism may be important for VKH pathology [25].
In summary, much evidence has shown that the onset and activity of uveitis are associated with vitamin D level, but most of the above studies were observational and did not provide evidence for causality. In the future, randomized-controlled studies are required to evaluate whether vitamin D can be an option for treating uveitis. Vogt-Koyanagi-Harada (VKH), ankylosing spondylitis (AS), acute anterior uveitis (AAU); # LEGEND for case-control, cohort, and cross-sectional studies, rating of the studies follow the guidelines from LEGEND.

Retinoblastoma
Vitamin D has antineoplastic functions against many types of cancers through influencing cell differentiation, apoptosis regulation, anti-angiogenesis, and cell cycle arrest in various tumors [20]. Animal studies suggested that vitamin D analogues inhibited retinoblastoma (RB) tumor growth in athymic mice by increasing apoptosis, which is associated with the upregulation of both the p53 and p21 proteins [264,265]. However, due to the extremely low incidence rate of RB, few clinical studies have been conducted regarding the effect of vitamin D on RB. Two recent studies in Mexico found that sun exposure in early childhood protects against RB and may decrease the degree of intraocular spread in children with bilateral RB [266,267] (Table 8). However, RB is a childhood cancer, which usually presents at the age of 1 to 2 for bilateral and unilateral RB. Further studies are needed to confirm and understand the association between vitamin D levels and RB.

Cataract
For the past 20 years, the prevalence of cataracts has declined due to the advancement of surgical technology. However, in middle-income and low-income countries, cataract is still the most common cause of visual loss, accounting for 50% blindness [269]. Cataract is caused by losing lens transparency when the lens becomes opaque [269]. Epidemiology studies have shown that ultraviolet radiation is an important factor in increasing the risk of cataract [270][271][272]. Since the natural source of vitamin D is sunlight exposure, vitamin D may be involved in the pathophysiology of cataract. The opacity of the lens is a result of oxidative stress [273]. Since vitamin D can protect cells or reduce oxidative stress [69][70][71], it can be protective against cataract and play a role in lens metabolism.
There are studies attempting to reveal the association between vitamin D and cataract (Table 9). A large cross-sectional study in South Korea, with 16,086 participants aged 40 years or older, revealed that serum 25(OH)D levels were inversely associated with the risk of nuclear cataract [274]. Similarly, another South Korean study on 18,804 subjects also found that cataract risk decreased in men with higher serum 25-hydroxyvitamin D levels compared with those with lower serum 25(OH)D levels, but this association is not significant in women [275]. Other studies in Egypt [276], Turkey [277,278], and the UK [279] also showed that cataract patients often have a low level of serum vitamin D. However, one study [280] reported that serum 25(OH)D levels were not related to nuclear opacities. In summary, all the current studies are cross-sectional or retrospective in design, so the causal relationship between vitamin D and cataract needs to be proved by further large-prospective cohort or RCT. Cataract is essentially a treatable disease. Future study design should aim to find out whether vitamin D supplementation could protect elders against cataract development.

Other Ocular Diseases
Apart from the diseases described earlier, some studies reported on the association of vitamin D with curable or rare ocular diseases (Table 10). Vernal keratoconjunctivitis patients were found to be at a significantly lower level of serum vitamin D in case-control studies held in Turkey [284,285], Italy [286], and Iran [287]. Deficiency of vitamin D is also more frequently found in patients with keratoconus [288][289][290], retinal venous occlusions [291,292], and optic neuritis [293][294][295]. However, in children with allergic conjunctivitis, the results were contradictory: two studies [296,297] found significantly higher vitamin D levels in patients, while the other studies reported complete opposite conclusion [298,299].
It is notable that vitamin D is associated with most ocular diseases, from anterior segment to retina. One possible explanation is that vitamin D or its metabolite plays a role in maintaining the stability of ocular metabolism and structure. Higher 25(OH)D levels in aqueous humor may have an influence on ocular disease [119,283]. The change of ocular structure and function, including spatial contrast [300], contrast sensitivity [301], choroidal thickness [302][303][304], corneal endothelial [305], and macular thickness [306], may be affected by serum levels of vitamin D. On the other hand, this association is not a causality. Vitamin D may work as a marker of health status. People with poor health or low vision will have little outdoor activities and consequentially less exposure to sunlight. Some studies found a positive association or no association between vitamin D levels and pterygium [275,[307][308][309][310]. Outdoor occupation is a major risk factor for the development of pterygium [311,312].  No difference in retinal nerve fibre layer (RNFL) between three groups. VDD has no effect on the RNFL. However, a positive correlation existed between the macular choroidal thickness (CT) and the vitamin D levels in DM patients with VDD. 4a

Perspective
In this systematic review, we summarize the evidence of vitamin D's effect on different ocular diseases. However, there are significant limitations in many studies, and the interpretation of results should be within these limitations.
First, the definition of vitamin D deficiency is not consistent in different studies, which affects the objective comparison of results from different studies. Until 1998, vitamin D deficiency was defined as a blood level of 25(OH)D, which represents a total concentration of both 25-hydroxyvitamin D2 and 25-hydroxyvitamin D3 of less than 10 ng/mL (25 nmol/L). However, this definition was redefined in 1998 as a blood level of 25(OH)D < 20 ng/mL (50 nmol/L). One reason for the change in the involvement of PTH: adults may require a serum 25(OH)D of at least 50 nmol/L to achieve optimum PTH levels [319]. Moreover, some studies use quintiles or quartiles to perform data analysis. We suggest that future studies should use the common classification to analyze serum vitamin D levels.
Second, vitamin D levels are related to different socioeconomic, lifestyle, and dietary factors. These factors are very important when considering the onset or development of ocular diseases. Future studies should determine these factors and consider them as confounders in statistical analysis.
Finally, few studies provided causal evidence on whether supplementing vitamin D can reduce the prevalence of ocular diseases. Currently, available evidence is insufficient to confirm how vitamin D works in ocular diseases. Moreover, investigations of those metabolic enzymes are also important to understand why significantly low vitamin D is found in patients with ocular diseases. In summary, among all ocular diseases, the association between vitamin D and AMD was unclear, while clinical trials in DES or DR required bigger sample sizes in different ethnic populations. Other ocular diseases such as cataract, myopia, and uveitis, no RCTs or perspective cohorts have been reported, so their associations with vitamin D are still unclear.

Vitamin D and Eye Care
Vitamin D can be a potential intervention for different ocular diseases. Although findings for certain ocular diseases are inconsistent, they can still serve as references for further studies that examine the therapeutic effects of vitamin D on ocular diseases, considering that vitamin D deficiency is a common health issue worldwide, and vitamin D has wide safety doses and rare side effects. Even though more evidence from the randomized controlled trial is needed to confirm the effect of vitamin D on various ocular diseases, it is recommended to maintain blood 25(OH)D 3 at a desirable level (25-50 nmol/L) by spending a short period of time outdoors, baring skin to the sun, and boosting vitamin D intake by a daily supplement of 400-800 international units (10 to 20 µg).
The human body can also obtain vitamin D naturally from sunlight exposure, however, both extended exposure to unprotected sunshine, which also increases the risk for cataracts, and AMD and completely avoiding sunlight by applying UV B sunscreen should be avoided. Several factors affect vitamin D production, and a more efficient production of vitamin D can be achieved when someone is closer to the equator, has a lighter skin color, and/or exposes larger surfaces of skin during summer midday without sunscreen. Generally, 5-30 min of sun exposure on the unprotected face, arms, legs, or back between 10 a.m. and 3 p.m. twice to three times a week is enough for sufficient vitamin D. Even though vitamin D can be beneficial to our ocular health, long term prolonged sun exposure is also associated with corneal sunburn, tissue growths on sclera, cataracts, and macular degeneration, and wearing hats, and sunglasses are recommended to protect the eyes from UV damage.

Conclusions
Various epidemiological and clinical studies have demonstrated a connection between vitamin D deficiency and ocular diseases, such as myopia, age-related macular degeneration, glaucoma, diabetic retinopathy, dry eye syndrome, thyroid eye disease, uveitis, retinoblastoma, and cataract, among others. While vitamin D is associated with potential pathways related to these respective diseases, their pathogeneses are complicated, and current understandings of the underlying mechanisms remain limited. Vitamin D not only affects mineral metabolism homeostasis, but also possesses antioxidation and antiinflammatory properties. It also plays a role in anti-angiogenesis, modulating cell cycle including cell proliferation, differentiation, and apoptosis. VDR and vitamin D regulatory enzymes are present in ocular tissues, and studies have demonstrated that ocular tissues can activate and regulate vitamin D, suggesting the importance of vitamin D in maintaining ocular health.