The association between vitamin D level and diabetic peripheral neuropathy in patients with type 2 diabetes mellitus: An update systematic review and meta-analysis

Introduction

Diabetes mellitus (DM), a world significant health concern, is a metabolic disease caused by the defection of insulin secretion or the obstacle of insulin function. The number of people with diabetes will rise up to 366 million in 2030 which is estimated by the World Health Organization (WHO), and these diabetes epidemic will continue [1]. Peripheral nervous system can be affected by diabetes disease in various ways, and the prevalence of neuropathy in patients with diabetes is approximately 30%, and up to 50% will eventually develop neuropathy during the course of their disease [2]. Diabetic peripheral neuropathy (DPN), a common complication of diabetes mellitus, is found in 50% patients living with diabetes mellitus [3]. About 11% patients with DPN have chronic, painful symptoms, which seriously affect patients with diabetes, and DPN, is one of the main causes of morbidity and increased mortality [2], [3], [4]. Some epidemiological studies have indicated the main risk factors of DPN including the high level of blood glucose and glycated hemoglobin, the duration of diabetes, elevated albumin excretion rates, obesity and hypertension [4], [5]. However, the pathological progress of DPN is still not completely clear. It’s essential to identify potential risk factor in diabetes mellitus for the better understanding of the mechanism of DPN. Vitamin D is a steroid hormone which has multifarious and extensive effect in the human body. Vitamin D has already been proven to be associated with the regulation of bone metabolism in previous studies, while, several epidemiological studies also indicate that vitamin D has a strong association with multiple autoimmune diseases [6]. Appropriate supplementation of vitamin D may play a potential therapeutic role in preventing the severity and progression of type 1 diabetes mellitus and type 2 diabetes mellitus [6]. 25-Hydroxyvitamin D [25(OH) D], a biomarker of vitamin D, is measured in the blood sample to monitor the vitamin D level in patients [7]. The concentration of 25-Hydroxyvitamin D ranges from 20 to 30 ng/ml or 50 to 75 nmol/L is considered as vitamin D insufficiency. Vitamin D deficiency is defined as 25(OH) D less than 20 ng/ml or 50 nmol/L [8], [9]. Increasing researches have suggested that low vitamin D level, in addition to result in rickets in infants and chondrosteoma in adults as we all known, may have a potential role in the generation of DM and its underlying diseases [10], [11]. Vitamin D deficiency, a common symptom in diabetic patients who has distal symmetrical polyneuropathy, has been associated with type 1 or 2 diabetes and its microvascular and macrovascular complications [12].

Numerous studies also showed the potential correlation between vitamin D deficiency and DPN in patients with type 1diabetes or type 2 diabetes [10], [12], [13]. Meanwhile, the association between vitamin D deficiency and incidence of neuropathy in type 2 diabetes was confirmed in studies [14]. Thus, vitamin D deficiency can cause the risk of DM and its complication including neuropathy, but owing to small sample size, the difference of study design as well as the population and regions of studies, there is still no clear conclusion about the association between vitamin D and DPN in diabetic patients. Therefore, we performed this systematic review and meta-analysis to explore the impact of vitamin D level in the development of DPN in T2DM.

Methods

Results

We searched all potential relevant articles on PubMed, web of science, CBM and CNKI databases. A total of 381 publications were identified. After browsed the title and abstract of these articles, 332 articles were excluded. After evaluated the full text of the surplus literature, only ten publications met the study selection criteria (Fig. 1).

The main characteristics of included studies were exhibited in Table 1. Of included ten studies, four studies were cross-sectional study design, six studies were case-control study design. 1368 type 2 diabetic patients were included in this meta analysis. Nine studies [22], [23], [24], [25], [36], [37], [38], [39], [40] were considered as moderate quality with 4–6 scores. Only one study [21] got 7 scores as high quality. Most of the studies were conducted in China (70%). Most measurement of 25(OH) D level of patients was the method of Electrochemiluminescence immunoassay (ELICA). One study (Wang N et al.) [39] did not report the measurement of 25(OH) D level. Diagnostic standards of diabetic peripheral neuropathy were defined in most studies. One study [22] did not note the measurement of DPN in detail.

In order to assess the influence of 25(OH)D level in the development DPN in diabetic patients, we collected mean and standard deviation (SD) of 25(OH) D level of ten included studies, after analyzed all the studies, there was obvious heterogeneity in those ten studies (I² = 94.1%, Cochran Q test P < 0.001) (Fig. 2). To identify the source of heterogeneity, subgroup analysis was performed by the measurement of 25(OH) D level, study design, and race. Heterogeneity was obviously decreased after the analysis of study design and race subgroup. There are low heterogeneity between studies which are cross-sectional study design (SMD = −0.63, I² = 37.9%) and performed in the Caucasian population (SMD = −0.56, I² = 16.9%) (Fig. 3, Fig. 4). Sensitivity analysis showed one study (Cui LJ et al.) [25] had great influence on the summary estimate (Fig. 5), after excluded that one, heterogeneity was not obvious reduced (Q test p < 0.01, I² = 91.7%).

Among included ten studies, four reported the association between vitamin D deficiency and DPN by statistics OR with 95% CI of Asian [22], [23], [24], [25], and only one reported this relationship of Caucasian [21]. Therefore, we only calculated pooled OR value of studies in Asian, and Pooled OR was 1.22 (95% CI:1.17–1.27, I² = 0%), which indicated that low 25(OH) D level in DM patients was an risk factor of DPN (Fig. 6).

There was no evidence of publication bias in meta analysis as showed in performed Begg test and Egger test (P = 0.128 and P = 0.130, respectively).

Discussion

This systematic review and meta-analysis explored how vitamin D level measured in human blood sample affected the development of DPN in diabetic patients. From the result, there was severe heterogeneity in those ten studies (I² = 94.1%) to assess the correlation between 25(OH) D level and DPN. However, after subgroup analysis by study design and race, the heterogeneity was reduced significantly. The cause of these changes may be caused by the low quality study designs and the different regions of studies. When we calculated the standardized mean difference (SMD) of included articles, we found that there was an obvious association between vitamin D and DPN in Caucasian, but no obvious correlation between vitamin D and DPN was observed in Asian. However, after analyzed all reported OR values of four studies, finally, we find that diabetic patients with vitamin D deficiency are 1.22 times suffering from DPN than those patients without vitamin D deficiency in Asian. Thus, vitamin D may be a high risk factor for the occurrence of DPN in diabetic patients. A previous meta-analysis also has demonstrated that vitamin D is involved in the development of DPN, and vitamin D deficiency is very likely to be associated with increased risk of DPN [26]. Appropriate vitamin D supplements can be an effective measurement to prevent the development of DPN in diabetic patients.

The study of Cui LJ et al. [25] had a great influence on the stability of this meta analysis as showed in sensitivity analysis. The mean and standard deviation of vitamin D level in the control group was higher than the case group compared with other studies (57.31 ± 10.27 nmol/L vs 36.99 ± 7.26 nmol/L). The reason might be due to the small sample size, personal error, and Imprecise measurement method of this study. Thus, these factors caused the great influence on the result of this meta analysis.

DPN, a common complication of diabetes mellitus affects patients with DM, is a major cause of morbidity and increased mortality [3]. It is necessary to clarify the potential risk factor of DPN. As we all known, DPN is increased with age and diabetic duration. Recently, more and more studies imply that vitamin D level makes a tremendous impact on the generation and development of DM and its complications [27], [28]. Vitamin D is an essential fat-soluble vitamin in the human body, which is closely linked to human health. Vitamin D deficiency can cause many health problems such as well-known osteoporosis in adulthood [7]. Multiple articles demonstrated that vitamin D might play its therapeutic role effectively through immune-modulatory properties in T1DM. Meanwhile, vitamin D also might play an essential role in protecting pancreatic β-cell function through various channels and in various ways [11].

In animal studies, Vitamin D deficiency has been shown to associate with low levels of neurotrophins (especially nerve growth factor) and defective neuronal calcium homeostasis. Vitamin D through its receptor modulates neuronal differentiation as well as neuronal growth and functions [12], [29]. In addition, vitamin D deficiency has been linked to a lower pain threshold which increases when vitamin D deficiency is corrected [12]. Vitamin D deficiency is common in diabetic patients, and low concentrations are associated with the presence and the severity of sensory neuropathy in diabetes [21]. Some retrospective studies have indicated vitamin D deficiency is an independent risk factor for DPN [21], [22], [23], [24], [25]. Moreover, increasing prospective intervention studies also have shown that oral vitamin D supplementation can reduce the symptoms and pain of DPN significantly [30], [31]. Clearly, serum vitamin D deficiency has been involved in the generation and development of DPN disease in all patients with diabetes mellitus. Thus, vitamin D deficiency may be a potential cause of the generation of DPN in DM, and this causal relationship has been proposed in multiple studies of various ethnic groups [32]. Vitamin D deficiency is considered as a pandemic and a worldwide problem for human health [33], [34], [35]. 25-Hydroxyvitamin D which is a biological marker of vitamin D can be measured to monitor the vitamin D level in human blood sample [7]. A serum level of 25-hydroxyvitamin D more than 75 nmol/L or 30 ng/ml, is required to maximize vitamin D's beneficial effects for health, and vitamin D deficiency is defined as serum 25(OH) D level less then 20 ng/ml or 50 nmol/L [8], [33]. Multiple cross-sectional studies have strongly suggested an inverse relationship between vitamin D level and DPN in DM [21], [36]. However, because of the study design, sample size of study, some confounding factor and the fact of study result, there is still no clear conclusion on the association between vitamin D and DPN. Therefore, further large, well designed, controlled studies are needed to examine the true relationship between vitamin D and DPN.

Conclusion

Our study can tentatively draw a conclusion briefly. This meta-analysis indicates that type 2 diabetic patients with vitamin D deficiency are 1.22 times to suffer from DPN compared with normal vitamin D level in Asian. Vitamin D deficiency is an important cause factor for the development of DPN in Caucasian with diabetes mellitus. It is urgently needed to supply vitamin D appropriately for preventing the generation of DPN in type 2 diabetes. More large, random, prospective studies are required to confirm this association.

Conflict of interest

The authors declare no conflict of interest.