Vitamin D can shift the balance of the body’s T-cell response toward downregulation of the T-helper-1 immune response [15]. In particular, vitamin D and T1DM vitamin D receptor (VDR) have been considered to play a role in the pathogenesis of T1DM. In 2015, Cheon et al [16]. conducted a study about the contribution of VDR polymorphisms to T1DM susceptibility in a genetically homogenous population in Korea. It was suggested that T and b TaqI and BsmI alleles might be protective against T1DM in Korean subjects. A meta-analysis performed by Qin et al. [17] revealed that the VDR BsmI B allele, bb genotype was closely connected with T1DM risk in Asians, while the bb genotype was linked to T1DM risk in overall populations. There has been skepticism about the effect of vitamin D supplementation as a preventative measure of T1DM. A Norwegian study reported that higher serum 25-hydroxyvitamin D during late pregnancy was related to lower risk of T1DM in the offspring [18], but a Finnish study revealed that there was no association between serum concentrations of 25-hydroxyvitamin D in the first trimester of pregnancy and the risk of T1DM in babies [19]. A retrospective meta-analysis research revealed that infants given vitamin D had lower risks than those without supplementation (odds ratio, 0.71) [20]. The Diabetes Autoimmunity Study in the Young investigated 25-hydroxyvitamin D concentrations in plasma during infancy and throughout childhood and identified no link to islet autoimmunity or advancement to T1DM [21]. However, there is limited supporting evidence from prospective birth cohort studies despite interest in vitamin D supplementation to prevent islet autoimmunity and T1DM.

Most prospective birth cohort studies have not proven association between early exposure to cows’ milk and either islet autoimmunity or T1DM [22-24]. In a double-blind, randomized trial, among 230 infants at risk of T1DM, those given a casein hydrolysate formula showed relatively lower risk of islet autoimmunity than those receiving cow's milk-based formula during the first 6 to 8 months [25]. However, this effect on islet autoimmunity was not confirmed in the larger phase 3 study [26]. Cow's milk ingestion in childhood has been related to both increased risk of islet autoimmunity [27,28] and T1DM [29,30]. Higher intake of cow's milk might boost advancement to T1DM among children with islet autoimmunity [13], which could be mediated by a portion of certain fatty acids in cow's milk and meats [31]. This result could promote a new modality of dietary interventions to stave off T1DM when confirmed by further studies.

Some environmental factors of T1DM are interconnected with the human microbiome. Gut microbes affect glucose and lipid metabolism, immunity, and systemic inflammation outside of the intestines [32-34]. T1DM risk might be modulated by commensal microbiota [32,35], but research has been limited. Some have reported lower microbial diversity in children with islet autoimmunity prior to diabetes progression, compared with a healthy control group [35,36]. Larger studies with carefully selected controls are needed using next-generation sequencing of the microbiome at multiple time-points.

Recent findings have shown that enteroviral VP1 protein immunoreactivity in the β cells of children with T1DM was more frequently detected than in controls [37,38]. In an in vitro study, it was proposed that continuous enteroviral infection of human pancreatic ductal cells might lessen their ability to transdifferentiate into β cells, thus causing β-cell mass reduction owing to autoimmunity [39].

Currently, CGM is approved by the FDA in the outpatient setting as an ancillary instrument, which allows for an increase in information obtained from standard home blood glucose monitoring devices and aids in the recognition of hypoglycemic and hyperglycemic events [52]. The initial approval was given to only one model, the Dexcom G5 sensor [48]. Many CGM users have already found that the CGM provides sufficient information, accuracy, and reliability for insulin dosage adjustments based on the CGM alone [48,51,53].

Most of the early studies with CGM were conducted on people with T1DM who were also using a CSII. There were questions as to whether similar benefits would be seen in people using basal-bolus therapy with MDI. There is now considerable evidence that improvements in the quality of glycemic control are essentially identical in MDI and CSII users [49-51]. The effectiveness of CGM for people with T1DM using MDI as well as CSII was clearly described in recent randomized clinical trials showing that changes in mean glucose were equivalent for users of MDI and CGM [49,50,54].

A recent clinical trial assessed the performance of CGM for patients at high risk to experience hypoglycemia. Van Beers et al. [51] conducted a crossover randomized study and demonstrated a dramatic increase in percentage of time in the target range, which was identical for MDI or CSII users, with an accompanying marked reduction in the hypoglycemia frequency.

CGM usage in the inpatient setting and ICU remains a work in progress [55,56]. There has been considerable interest in using CGM in the hospitalized patients for glycemic control, in particular for patients continuing use of outpatient therapies, for control of insulin infusions, and for use in the intensive care unit [48]. Continuous improvements in the accuracy, robustness, and usability of CGM sensors offer a promising outlook for their role in the inpatient setting [48]. Thabit et al. [57] have recently reported substantial improvements in glucose control in a randomized parallel-arm study of 40 inpatients with type 2 DM using a closed-loop control without premeal boluses. Specifically, the proportion of time in the target range increased from 38.1% in the control group to 59.8%, a 21.8% change. In practice, there are safety concerns regarding data analysis. Occasionally, inexperienced experts might make hasty or inappropriate decisions about doses because of the enormity of data output. Further studies are recommended to provide sufficient information about the possible benefits of CGM for hospitalized patients (Table 2). Since CGMs are often used along with insulin pumps, guidelines should be established for both technologies in hospitalized patients.

Metformin lowers hepatic glucose production, reduces glucose absorption in the intestines, and fosters glucose uptake and insulin sensitivity in peripheral tissues, resulting in a reduction of fasting and postprandial glucose [61]. Several clinical studies have been conducted for metformin therapy in T1DM pediatric patients (Table 3). Hamilton et al. [62] examined pediatric patients (n=27) with T1DM who required insulin doses >1 unit/kg/day. HbA1c was lower by up to 0.6% in the metformin group than in the placebo group. Fasting plasma glucose (FPG) was significantly reduced with metformin (-21.6±36 mg/dL vs. 1.8±45 mg/dL, P=0.004). Insulin doses were notably lower in the metformin group (-0.14±0.1 units/kg/day vs. 0.02±0.2 units/kg/day, P<0.01). It was observed that body mass index (BMI) was lower in the metformin group, but it was not significant. In another clinical trial, 30 adolescent patients were randomly given metformin or placebo for three months. HbA_1c level significantly decreased (9.6% to 8.7%) in the metformin group, but it did not change in the placebo group. Individuals having lower insulin sensitivity showed greatest benefit from metformin therapy [63]. A randomized, double-blinded placebo-controlled study examined adolescent patients (n=74) with T1DM for 6 months [64]. They were randomized to either metformin 500 mg twice daily or placebo. Overweight or obese patients with metformin showed a trend toward lower HbA_1c. Daily insulin dose (DID) in units and units/kg was significantly reduced by metformin. Further, metformin dramatically decreased BMI z scores. To assess metformin added to insulin therapy in overweight or obese patients with T1DM, 2 RCTs were conducted. In the first trial of enrolled patients (n=140, aged 12–20 years) [65], HbA_1c dropped significantly around 3 months, but it was not maintained around 6 months. Mean total DID per kg was lower in the metformin group than the placebo group (-0.1 [-0.2 to -0.0], P<0.001). The metformin and placebo groups showed BMI reductions of at least 10% from baseline (24% vs. 7%, P=0.01). The second trial included patients (n=28, aged 10–20 years) with T1DM [66], and there was no significant difference in HbA1c and FPG levels between the metformin and placebo group for 9 months. The change in total DID was 0.31 units/kg between the metformin and placebo group.

Dipeptidyl peptidase-4 (DPP-4) inhibitors, including sitagliptin and vildagliptin, are known to increase GLP-1 and decrease glucagon levels after a meal without interrupting counterregulatory response during hypoglycemia [67-69]. In small study, it was observed that sitagliptin therapy for 4 weeks significantly decreased HbA_1c levels by 0.2%–0.3% [70]; however, longer study demonstrated no significant reductions [71]. Changes in insulin requirements also range from no change [67,68] to minimal reductions of 0.05–0.13 units/kg/day; however, larger reductions were seen in postprandial insulin use [70,71]. According to a meta-analysis, DPP-4 inhibitors were not beneficial for patients with T1DM because of the limited glucose-lowering effect and the risk of severe hypoglycemia [72]. Considering the inadequate evidence for decreases in HbA_1c levels and reduction of insulin dose, the application of DPP-4 inhibitors in the clinic setting is restricted at present.

SGLT2 inhibitors may contribute to weight reduction in overweight/obese patients with T1DM and decreased glucose levels by blocking SGLT2 in the renal proximal tubule, interrupting glucose reabsorption [73,74]. However, evidence on the beneficial effect of SGLT2 inhibitors as adjunct to insulin treatment in T1DM remains insufficient. A randomized, double-blind study investigated patients (n=62) with T1DM concerning safety [75]. The results of efficacy revealed no significant difference between dapagliflozin and placebo in terms of mean glucose level, except reductions in FPG and TDID. Canagliflozin was examined in patients (n=351) with T1DM by a randomized, double-blind trial (phase 2 study) [76]. After 18 weeks, the HbA_1c level was decreased by ≥0.4%; however, body weight change was not observed in patients. Further, an average rate of hypoglycemia was not different among groups.