Association of Vitamin D and Diabetic Retinopathy in Patients with Type 2 Diabetes Mellitus

Introduction

With the rapid increase in the incidence of diabetes mellitus (DM) in recent years, the complications related to diabetes have become a larger health problem. Inadequately controlled diabetes precipitates various long-term microvascular (blindness, neuropathy and nephropathy) and macrovascular (cardiovascular and stroke) complications (1, 2) which form the leading cause of morbidity and mortality in diabetic patients worldwide (1). Diabetic retinopathy (DR) being one of the most prevalent complication of diabetes, is considered as a major culprit of blindness globally (1, 3). DR is also the most common cause of non-traumatic visual loss in the working-age population. It has been estimated that the worldwide prevalence of DR was 93 million (35%) and the prevalence of visionthreatening DR was 28 million (10.2%) among diabetic patients in 2010 (4). World Health Organization estimates that throughout the time DR accounts for approximately 5% of the worldwide prevalence of blindness, the prevalence rises sharply to 15-17% in developed countries (3).

There are various risk factors that have been suspected in the pathogenesis of DR, like hypertension and hyperglycemia. Although, hypertension and hyperglycemia have demonstrated strongest association, interventions directed at improving these factors have shown moderate success. Thus, the interactions among neural and retinal vascular dysfunction and the mechanisms leading to retinal pathology including neovascularisation have been questioned recently (3).

In addition to the classical role in skeleton and bone health, vitamin D (VD) has been identified to exert non-classical pleiotropic effects such as anti-inflammatory, antiangiogenic, antiproliferative, and immunomodulatory properties (5). It is also considered to positively regulate hypertension and blood glucose levels (6). Moreover, it has been established that maintaining VD at adequate levels can be a useful technique to prevent type 2 diabetes mellitus (T2DM) (7). T2DM and VD deficiency (VDD) have been considered as pandemic diseases with a number of health consequences. Furthermore, VDD is known to be more common in patients with diabetes (5). VDD has an established role in developing the risk of diabetic complications as VD is considered to affect the risk for retinopathy, due to its immunomodulatory properties (3). However, the effect of low levels of VD on causation of type 1 DM (T1DM) is well established, the association between VDD and T2DM is not consistent. Evidences collected from many epidemiological studies, indicate that most of the T2DM cases are attributable to manageable habits and lifestyle changes. Therefore, the identification of easily modifiable risk factors is crucial for the prevention of diabetes and its complications (1). Since, the evidence regarding the association of vitamin is limited (8), and, the prevalence of VDD is contentious in T2DM patients (9), this systematic review has been conducted to establish or refute the association between VD and DR.

A recent PubMed search (till March 31, 2017), with the use of terms “vitamin D” and “retinopathy,” yielded only 159 publications. Thus, few studies have addressed the question of the potential implication of VD in the pathogenesis of DR. Among these studies there were only 9 publications that met the inclusion criteria. The aim of this study was to collect available evidence on the association between VD and DR, and to summarize the results by performing a systematic review of published studies according to Meta-analysis of Observational Studies in Epidemiology (MOOSE) group recommendations (10). The primary objective of this review is to assess whether there is an association between VD status and DR in T2DM.

Methods

Results

Search Results

The literature search results are summarized in the flowchart below (Fig. 1). After application of the inclusion and exclusion criteria, predefined in the study protocol, a total of 9 randomized controlled trials (RCTs) were included in the final systematic analysis. Amongst the nine studies included in the systematic review, six studies compared the VD levels in controls and in T2DM patients with DR (1, 9, 11-14), and 4 studies demonstrated the difference between controls and T2DM patients without DR (1, 11-13), one study demonstrated the difference in the levels in T2DM patients with or without DR, and further with Grade 1 or Grade 2-4 DR (8), one study compared the VD levels in patients without DR with non-sight threatening DR (NSTDR) and sight threatening DR(STDR) (15), two studies demonstrated the difference between control, T2DM patients with proliferative DR(PDR) (9, 14) and one study compared the VD levels in controls, T2DM with PDR and non-proliferative DR (NPDR) (16).

Close

Fig. 1:

Flow diagram of the literature search strategy. Number of studies identified and rejected at each stage are described.

Export to PPT

Discussion

DM being the most common endocrine disorder with its increasing prevalence every year needs to be addressed urgently (17), as the complications of diabetes may become larger health problem. As evident by many epidemiologic studies, most T2DM cases are attributable to modifiable habits and lifestyle factors. Thus, the recognition of easily malleable risk factors is crucial for the prevention of diabetes and its complications (1).

Inadequately controlled DM precipitates various long-term microvascular (blindness, nephropathy, and neuropathy) and macrovascular (cardiovascular and stroke) complications (1, 2), which forms the leading cause of morbidity and mortality in patients of diabetes worldwide (1). DR being one of the most prevalent microvascular complications is considered as a major culprit of blindness globally (1, 3). Various risk factors have been implicated in the pathogenesis of DR. Although, hypertension and hyperglycemia have demonstrated strongest association with the DR, interventions directed at improving these have shown moderate success. Thus, the interactions among neural and retinal vascular dysfunction and the mechanisms leading to retinal pathology including neovascularisation have been questioned recently (3).

VDD has been associated with increased risk of chronic diseases, such as cancer, T2DM, cardiovascular disease, and autoimmune diseases including T1DM and multiple sclerosis, as evident from epidemiologic studies (1, 8, 18). Moreover, a wide prevalence of VDD and DM has been observed across all ages, socioeconomic conditions, races, and geographic regions. Despite, the well-established effect of hypovitaminosis D on causation of T1DM, the relationship between VD insufficiency and T2DM is inconsistent (1).

Nutritional status, particularly micronutrients, might influence the risk for DR by affecting the biochemical mechanisms underlying DR (1, 3). Moreover, studies have suggested that VD might influence the risk for diabetes and its complications (1). VD has shown pleiotropic effects such as, suppression of cell-mediated immunity, regulation of cell proliferation, stimulation of neurotropic factors like nerve growth factor, glial cell line-derived neurotrophic factor, neurotropin, suppression of renin–angiotensin–aldosterone system, reduction of albuminuria, immunomodulatory effects (12), anti-inflammatory effects, antiangiogenic, and anti-fibrotic properties (12, 19) and therefore all these VD effects may provide a potential link between VDD and diabetic complications (3) including DR, as there are rising evidences that, inflammation and angiogenesis are involved in the initiation and propagation of DR (19). Findings from some recent studies show a significant inverse association between 25-hydroxyvitamin D [25 (OH)D] levels and the prevalence of diabetic microangiopathy, thus suggesting that VD and microvascular complications are strictly interrelated. Nonetheless, whether lower levels of VD are the cause or result of the microvascular complications are currently unknown (2). Moreover, the possible role of vitamins in the pathogenesis of diabetic complications is of much interest (8).

In a study conducted on T2DM patients low blood VD concentrations were found to be associated with an increased risk of macro- and microvascular disease events (18). Additionally, a study found an inverse relationship between 25(OH)D and the severity of DR, thus concluding that, lower 25 (OH)D status strongly correlates with a higher prevalence of microvascular complications in T2DM patients (2). Another study including both T1DM and T2DM patients, showed an association between the severity of DR and prevalence of VDD (20). An inverse association of 25 (OH)D levels with DR was demonstrated only in men. However, the sex-related variation in the relationship between 25 (OH)D levels and DR is uncertain (19). Furthermore, a metaanalysis demonstrated that VDD is more prevalent in diabetic patients, and that VD supplementation may delay or prevent diabetic complications (21).

In contrast to most reports, serum 25-OH VD concentrations did not differ between subjects with or without diabetes in a retrospective study (21). Another study conducted retrospectively including both T1DM and T2DM patients, found no association between serum 25(OH)D and the presence and severity of DR (3). Thus, the evidence behind the association between VD and retinopathy is non-confirmatory, hence, this systematic review was performed.

The VD levels were found to be inversely associated with the presence of DR. The controls had the highest levels as compared to T2DM patients with or without DR. Additionally, T2DM patients without DR had higher levels than the patients with DR. Moreover, VDD has been found to be associated not only with the presence of DR but also with severity of DR. The concentration of VD decreased at advanced stages. Patients with Grade 2-4 DR had the lower concentration as compared to Grade 1, whereas, patients without DR had higher levels than Grade 1 patients. Similarly, patients with STDR had lower concentration than patients with NSTDR, while patients without DR had highest concentration. Controls had highest concentration as compared to T2DM patients with simple PDR, NPDR and PDR, with PDR patients having the least concentration. All the studies included in this systematic review showed a significant association between the VD levels and DR.

VDD was highly prevalent in the T2DM patients. However, two studies demonstrated that, the prevalence of VDD was higher in patients without DR than in patients with DR (1, 11), suggesting the prevalence of VDD irrespective of the presence of DR, whereas, one study demonstrated a higher prevalence of VDD in T2DM patients with retinopathy as compared to without retinopathy (8). An increased prevalence of VDD was observed in advanced stages of DR. The prevalence of VDD was higher in patients with PDR as compared to simple DR (9). A higher percentage of patients with Grade 2-4 had VDD that patients with Grade 1 DR. Prevalence of VDD was higher in patients with STDR than in NSTDR (15).

Despite, the scarce evidence behind the association of VD and DR, this systematic review suggests an inverse association between the two. However, the question on the potential role of VD in the pathogenesis of diabetic retinopathy is yet to be answered (8). Given previous research indicating possible antiinflammatory and antiangiogenic properties of VD, the connection between VD and DR warrants further studies (20).

Conclusion

Through this systemic review, it was observed that there is an inverse association between the VD levels and DR. Additionally, the concentration of VD decreases as the stage of DR advances. Moreover, the prevalence of VDD is higher in T2DM patients as compared to controls, which further increases in the advanced stages of DR. Based on the results, it is advisable that VD levels should be routinely monitored in T2DM patients, as VD seems to be an easily modifiable risk factor for DR. VD supplementation should be encouraged in population at higher risk for the diabetic complications.