Diabetes mellitus is a group of metabolic disorders described by elevated blood glucose. The criteria for diag- nosing diabetes include a glycated hemoglobin (A1C) > 6.5%, fasting plasma glucose (FPG) > 126 mg/dL, or a 2-hour plasma glucose test > 200 mg/dL [1] . Currently, 25.8 million children and adults in the United States (8.3% of the population) have diabetes. Considering the 7.0 million undiagnosed individuals along with noncompliance in those with diabetes, the economic burden for the United States is vastly increasing every year, with a current estimate of $176 billion from medical expenses [2] .

Several approaches can be taken to reduce the economic burden and improve patients’ quality of life. Lower- ing a patient’s A1C to less than 7% can help prevent macrovascular and microvascular complications of diabetes [3] . Treatment of diabetes involves not only pharmacotherapy but also an emphasis on diet and exercise. Adults with diabetes are advised to perform at least 150 minutes per week of moderate-intensity aerobic physical activ- ity, spreading over three days per week with no more than two consecutive days without exercise [3] . Despite pharmacologic treatments and healthy lifestyle choices, optimum diabetes control is not always maintained. Therefore, patients may seek other alternatives, such as natural products to help control their diabetes.

The use of natural products has increased in the past decade. A natural product is defined as a “vitamin, a mineral, an herb or other botanical, an amino acid, a dietary substance for use by man to supplement the diet by increasing the total daily intake, or a concentrate, metabolite, constituent, extract, or combination of these ingre- dients [4] ”. According to a 2007 government survey, Americans spend $33.9 billion on herbal products or her- bal-related physician visits [5] . Furthermore, a national survey from 1997-1998 reports 57% of patients with di- abetes used some form of complementary or alternative medicine. Of these patients, 16% utilized herbal reme- dies, commercial diets, or folk remedies specifically intended for diabetes [6] . Even though patients may be us- ing these herbal products for type 2 diabetes mellitus (T2DM), very few products have evidence showing their benefits for the treatment of T2DM. The purpose of this article is to evaluate the efficacy of guar gum, magne- sium, oat bran, blond psyllium, and soy in patients with diabetes. This article focuses on these natural products and their effects on blood glucose and/or A1C levels in T2DM. The goal is to provide healthcare practitioners with information that can be incorporated into their clinical assessment and management of patients with dia- betes.

Figure 1 displays the literature search and selection process used to identify clinical trials for this review. Natu- ral Medicines Comprehensive Database was initially searched to identify natural products advocated for the treatment of T2DM. Natural products categorized as both “possibly effective” and “likely safe” were chosen, which included guar gum, magnesium, oat bran, blond psyllium, and soy. A literature review was performed in MEDLINE (1950-March 2013) using the keywords diabetes mellitus type 2, guar gum, magnesium, oat bran, psyllium, and soy. Additional references related to the topic were identified through primary literature, review articles, and textbooks. The references identified from the literature review were evaluated for the treatment of T2DM. All MEDLINE searches published within the last ten years (January 2003-March 2013) and pertinent ar- ticles published prior to 2003 were included in this review. Trials were required to contain measurable doses, be written in English, involve human subjects, and evaluate the natural medicine’s effect on blood glucose and/or A1C. Diabetes prevention studies were not selected for this review.

Of the clinical trials reviewed, the most promising natural products are the fiber products such as psyllium and oat bran. Although there is no strong evidence from large, randomized, controlled clinical trials to support its use, fiber can safely be recommended in doses of 25 to 30 grams each day. High fiber foods include oats, barley, whole grain cereals, brown rice, beans, peas, lentils, nuts, fruits, and vegetables. Most patients do not get ade- quate fiber in their diet from such foods; therefore recommending them provides additional benefits including lowering blood glucose levels [52] .

Psyllium has been shown to have the most promising preliminary evidence. All studies found significant

F = Female; M = Male; NS = Not Significant; T2DM = Type 2 Diabetes Mellitus Patients.

A1C = hemoglobin A1C; BGL= blood glucose level, not specified as pre-prandial or post-prandial; F = Female; FPG = Fasting Plasma Glucose; M = Male; PPG= Post-prandial Plasma Glucose; T2DM = Type 2 Diabetes Mellitus Patients.

improvements in glycemic control. However the majority of trials included a sample size of less than 50, which makes it difficult to assess whether the results could be extrapolated to the T2DM population. Typically with herbal products, it is difficult to evaluate which dosage form to recommend based on the variety of products used in studies. For psyllium, glycemic control was achieved even with various forms. None of the trials eva- luated the efficacy of psyllium compared to conventional therapies, therefore more head-to-head trials would need to be conducted before treatment of psyllium in T2DM can be recommended.

Oat bran evaluation included two clinical trials which resulted in lower blood glucose levels [30] [31] . The trials had several limitations such as small study samples (n < 12) and durations (<6 months). A longer duration is necessary to evaluate the treatment of T2DM. Although the trials did not justify the use of oat bran in patients with T2DM, increasing oat bran can be safely recommended to patients with T2DM. Oat bran does appear to be useful in other co-morbid conditions such as hypercholesterolemia. Current FDA regulations and guidelines al- low food products containing whole oat to be labeled with a health claim stating that the products may reduce the risk of heart disease if they contain at least 0.75 g of soluble fiber per serving [37] . Even though the clinical data does not support the use of oat bran in T2DM, incorporation of oat in the daily diet may be beneficial for other conditions.

Soy demonstrated significant improvements in glycemic control with the clinical trials included; therefore it may be beneficial to include soy as part of the diet for T2DM patients. Soy-based products are readily available and can be incorporated into the diet, however adherence to soy-based diets have previously been reported to be poor [46] . The trials reviewed here reported good adherence, which could potentially account for the impro- vement in fasting and postprandial glucose levels. Many of the trials provided education to the subjects at enroll

A1C = hemoglobin A1C; F = Female; FPG = Fasting Plasma Glucose; M = Male; T2DM = Type 2 Diabetes Mellitus Patients.

ment and close monitoring by dieticians and physicians throughout the treatment periods. This must be taken into consideration when extrapolating these findings to the general population since soy supplementation with- out proper counseling on diet adherence may not have similar results. Another concern regarding study design is the use of various soy products throughout different trials. Unless large clinical trials compare the various soy products on the market, it would be difficult to recommend a specific type of treatment. It is important to note that most studies with beneficial effects typically used soy protein products. Therefore, if a patient chooses to supplement their diet with soy for T2DM, it may be beneficial to recommend a protein-based product. Some tri- al limitations included small sample sizes, typically less than 50 subjects, and a primary focus on cardiovascular endpoints such as lipid levels. Since T2DM and glycemic parameters were not common primary endpoints, it may be beneficial to have further studies evaluating larger patient populations and longer durations of soy ther- apy with emphasis on FPG and A1C to evaluate its effects on T2DM.

Based on limited available data, there appears to be some potential benefit of magnesium supplementation for the reduction of FPG and A1C. The majority of studies demonstrated no effect on fasting plasma glucose or A1C. Many of these studies, however, were of short duration (four to six weeks) and small population size (~60 subjects or less). The short duration of magnesium use makes it difficult to assess its impact on chronic man- agement of diabetes through parameters such as A1C. In addition, variances in dosing and product selection make it challenging to determine an optimal magnesium salt form and dose for adequate supplementation and improvement in diabetic markers. For those longer term studies of 12 to 16 weeks, only one evaluation of 63 subjects taking glibenclamide with either 2.5 g MgCl₂ or placebo showed improvements in FPG and A1C at 4 months when compared to baseline and placebo. Subjects in each group were poorly controlled at baseline with an average A1C of at least 11.5% and demonstrated a 30.4% and 14.4% decrease in A1C in the magnesium and placebo groups, respectively, with significant improvement in the magnesium group compared to placebo [26] . Although this study showed positive results, its distinct study population of poorly controlled patients with di- abetes, small study size, limited duration of therapy, and supplementation dose may limit its use in general prac- tice. Since patients with poorly controlled diabetes are at higher risk of hypomagnesemia, magnesium supple- mentation may be beneficial in those with deficiencies to correct magnesium levels. However based on current data there appears to be potential for use but not a clear benefit in improving A1C and FPG in the general popu- lation of patients with diabetes.

Guar gum use in T2DM remains controversial. A few trials show it may lower blood glucose levels, while others show no benefit. Many of the trials had appropriate treatment times (3 - 10 months) to evaluate guar gum’s effect on T2DM. However all the trials had small samples sizes (n < 41). The type of guar varied how- ever the doses used were similar (15 g per day). Based on the evidence at this time, guar gum should not be recommended for the treatment of T2DM.

Overall, all five herbal products have limited data to support their use over conventional therapy. Large, rando- mized, controlled clinical trials are necessary to determine efficacy. Many of the trials lacked adequate sample sizes, control groups, and duration. In addition, the clinical trials available lack standardization of the type of product being investigated.

As the number of people with diabetes in the United States increases and the goals of therapy are not met, pa- tients may seek non-conventional therapies such as natural products. Since the FDA prohibits the use of health claims for items sold as food supplements, products will not have indications. Therefore, it is especially impor- tant to educate patients and emphasize that they should discuss the use of natural medicines with their providers. Although patients may view natural products as safer routes for treatment of T2DM with fewer side effects, many natural products have similar pharmacologic effects on conventional medications, which can result in ad- ditional toxicities. For this reason, monitoring patients for hypoglycemia with concomitant use is vital. In addi- tion, patients should be informed not to replace their conventional medications with natural products.

A few agents, such as psyllium or soy, may play adjunctive roles in achieving the therapeutic goal for a pa- tient with T2DM and should be discussed with a healthcare provider before using them. As more supplements become available, the need for healthcare professionals to familiarize themselves on the use, efficacy, and safety of these products is essential. Until additional data are collected from well-designed trials, natural products in T2DM are not recommended over the use of conventional drug therapies.

We acknowledge Nasreen Mahmood, Anita Siu, and Sneha Srivastava for editorial review.