CINNAMON: AN EFFECTIVE HERBAL TREATMENT FOR DIABETES?

Ryan Bradley, ND, MPH     July, 2010

The search for effective treatments for diabetes, especially type 2 diabetes, remains a challenge for medical researchers, regardless of their background. As many of you know, many of the oral medications have significant side effects that are not popular with patients, have limited effectiveness, or lack evidence of impacting the course of the disease, including the development of complications. This challenge, combined with an increasing interest in herbal and complementary medicine has led to a search for effective natural therapies that have significant effects on blood sugar levels, blood pressure, etc.

Cinnamon, the brown, mildly bitter, mild spicy cooking herb, has gained popularity as an herbal treatment for diabetes because it is readily available and relatively inexpensive. Since this original publication, several others have studied cinnamon in different groups of people and challenged the original findings, while additional publications suggested the type of cinnamon and/or how the extract was made may be significant factors in the effects cinnamon has on blood sugar. Also, many research publications have focused on laboratory measures, which are important in diabetes, but do not fully account for risk; some publications suggest cinnamon may have other benefits, like antioxidant effects and production against "type 3 diabetes" or insulin resistance in the brain. In this article I will review the recent clinical evidence on cinnamon, and provide some guidance on if you should try it yourself!

What Makes Cinnamon "Work"?

Cinnamon contains rather high levels of compounds called "polyphenols"1. Polyphenols refer to the chemical structure of an entire class of compounds that have rather high antioxidant activity, and are typically very effective in reducing the oxidative processes that change fats and lipids into potent inflammatory compounds (e.g. peroxidation). Cinnamon is not the only food/herbal medicine that has high amounts of these polyphenols; other examples include green tea, darkly colored berries, red grapes, some nuts including peanuts, and dark chocolate. Translational research by Roussel et al. supports the antioxidant effects of a water-soluble cinnamon extract (a preparation method that concentrates polyphenols) 2. Some recent studies suggest, for safety and better efficacy, that the water-soluble compounds in cinnamon are the most effective, however good clinical research has not yet compared this extract (and commercially available product) to the whole herb3.

Does Clinical Research Support Cinnamon as an Effective Treatment?

The first significant clinical study evaluating Cinnamon was published by Khan et al. in 2003.4 In this trial, 60 Pakistani participants took Cinnamon (Cinnamon cassia) orally for 40 days at three different doses (1, 3, or 6 grams per day).After 40 days, significant reductions in fasting blood sugar, cholesterol (including LDL or "bad" cholesterol) and triglycerides were noted at each dose, although the group receiving the 6 grams per day dose did have a larger percent reduction in blood glucose compared to the 1 and 3 gram per day dosing groups (29% vs. ~25%). One limitation in this study is that the results may not be applicable to all people, as the trial was performed in Pakistan, where dietary intake of polyphenol compounds is quite limited, and thus introducing dietary polyphenols in such significant amounts may have different effects than it would in groups of people with more diverse dietary intake of polyphenol compounds, including the United States. The flip side of this criticism is that people in Pakistan are relatively medication naive and take fewer supplements, thus creating a nice "blank slate" for research that avoids potential confounders.

The next trial, published by Mang, et al. in 2006, evaluated a water-soluble extract, corresponding to 3 grams of cinnamon, in 79 German participants with type 2 diabetes for four months.5 Although fasting glucose was lowered at the end of the trial, the reductions were modest compared to the findings by Khan et al.; average reductions were about 10%. Unlike the findings of Khan et al., changes in cholesterol and triglycerides were not significant, nor were changes in hemoglobin A1c.

Additional clinical research on cinnamon was published by Suppapitiporn et al. also in 2006.6 In their clinical trial, 60 Thai participants with type 2 diabetes took 1.5 grams per day of encapsulated whole cinnamon powder or placebo for 3 months. At the end of the trial period, no significant changes were evident in either fasting glucose, hemoglobin A1c or lipids. There was suggestion that a greater percentage of patients in the cinnamon group achieved glucose "control" (i.e. reached a hemoglobin A1c value of < 7%), however this difference (35% vs. 15%) was not statistically significant.

The saga continues with the clinical trial published by Vanschoonbeek et al. in which 25 post-menopausal women with type 2 diabetes, from the Netherlands, were randomized to take either 1.5 grams of ground, whole herb cinnamon or placebo for 6 weeks.

Evaluating the effectiveness of cinnamon at the end of 2006 would have led, I'm afraid, to an assessment of "not effective" based on rather small and highly variable fasting glucose reductions seem across studies. Fear not, we only had to wait a year to get more evidence!

In 2007, Altschuler et al. published the first available clinical trial of cinnamon in participants with type 1 diabetes.8 In this trial, 72 adolescents with type 1 diabetes were randomized to receive either 1 gram/day of cinnamon or placebo for 3 months. At the end of the trial period, no significant changes in hemoglobin A1c, insulin sensitivity, or daily insulin use (units/day). In fact, insulin use actually increased in the cinnamon group. It is important to differentiate this study from the other studies discussed above. Type 1 diabetes is a very different condition than type 2, with different causative factors and pathogenesis. It may be possible that the polyphenols of cinnamon have a sensitizing effect on insulin only in the presence of significant insulin resistance (this may be due to the role lipid peroxides have on inducing insulin resistance, which may be improved with antioxidant polyphenol supplementation).

In hopes of addressing the limitations in the trial by Khan et al., specifically the population in the study, in 2007 Blevins et al. published the results of their clinical trial performed in the United States.9 In their study, 60 participants with type 2 diabetes and very stable medications were randomly assigned to take cinnamon (1 gram per day) or placebo for 3 months. At the end of 3 months, there were no significant changes in fasting glucose, hemoglobin A1c, insulin, insulin sensitivity or lipid values. Notably a considerable percentage of the participants (~75%) in this trial were taking insulin sensitizing medications, which may have over-powered any modest action of the cinnamon extract.

As you have probably noticed, many researchers attempted to replicate the findings of Khan et al in different populations, but using similar doses, i.e. 1-1.5 grams/day. However, the basis for the dosing in the original trial wasn't specific, rather it was a somewhat arbitrary attempt at performing a dose-response study, and therefore, although reasonable to try these doses again, new research exploring even higher doses stopped until the 2007 publication by Solomon et al.10 In their study, 7 healthy adults without diabetes performed three oral glucose tolerance tests (An oral glucose tolerance test (OGTT) is performed by having a patient drink a standardized solution containing 75 grams of glucose, and then measuring insulin and glucose simultaneously for several hours following. The values are then plotted as a curve; the area under the curve provides meaningful reflections of glucose tolerance and can be used to estimate insulin sensitivity.)preceded by 5 grams of cinnamon, 5 grams of placebo, or 5 grams of cinnamon given 12-hours before the OGTT. Their results demonstrated significant reductions in post-OGTT blood glucose corresponding to measured improvements in insulin sensitivity.

Solomon et al. continued their translational research evaluating the effects of high dose cinnamon powder on insulin sensitivity in healthy adults by evaluating the effects of sustained short-term intake (3 grams per day or placebo for 14 days) on glucose and insulin measures also following OGTT.11 In this study, cinnamon reduced blood glucose following the OGTT, reduced insulin response to the OGTT and improved insulin sensitivity compared with the placebo. Howevr, the effects were lost however almost immediately upon stopping the cinnamon extract.

Although these small, translational trials provide "proof of concept" evidence, i.e. cinnamon can improve insulin sensitivity in healthy adults, how this evidence applies to people with diabetes is less clear. First of all, the dose used here, 3 and 5 grams per day, are very large. One concern I have is the differences in liver health in "healthy" adults vs. adults with diabetes. As many patients with diabetes also suffer from elevated liver enzymes due to a condition of increased fat deposition in the liver (a condition called non-alcohol steatohepatitis (NASH) or "fatty liver" disease), the effects of such large doses of cinnamon on liver health should be studied more before doses of cinnamon this large are considered for routine use.

Although the clinical trials that have been published on cinnamon generally show very few, if any, side effects, none of these trials have reported very systematic approaches to measuring side effects and none of the studies provide evidence of safety based on laboratory results.

Just when findings were starting to look really quite bleak for a role of cinnamon in lowering blood sugar in people with diabetes, Dr. Crawford published a clinical trial (apparently performed all by himself) evaluating the impact of whole-herb cinnamon (Cinnamon cassia, 1 gram per day for 3 months) vs. placebo in 109 patients with type 2 diabetes recruited from three primary care clinics on a US military base. Cinnamon was added to normal medications and medication changes were not meaningfully different between groups at the end of the study. In this clinical trial, a ~0.5% greater reduction in hemoglobin A1c was evident in the cinnamon group compared to the placebo group; this finding was highly statistically significant. This trial is notable for several reasons, including its size (the largest randomized trial performed thus far) and its "Western" participant group from a typical clinical practice setting.

For ease of reviewing the findings of all of these trials, a summary table of the results is included here as Table 1.