Coffee - facts pt. 2 - DIABETES

In this segment I plan to cover the controversial topic of whether or not coffee prevents diabetes. I was going to talk about this in part I, but it was too long and complicated. I assure you I do my best to keep it concise and to the point these days. Ok ready? Get your brain-boosting substance of choice in/on you, whether it's a thinking cap or a cup of coffee, or reishi spore oil.. Here we go.

 Note: a lot of the talk here is on caffeine. Coffee and caffeine are different. I will address that fosho.

 Coffee and diabetes: the paradox, the confusion

Almost all of the epidemiological studies tracking the effects of caffeine and coffee intake and diabetes have shown that they are inversely related; that people who drink coffee over a period of several years tend to have a lower risk of diabetes in the future. None have shown a positive correlation. The link is clearer in subjects who lose weight. In this prospective cohort study, the association was ONLY found in people who had lost weight. Caffeine and coffee may also help people lose weight. The paradox though, is that coffee causes acute insulin resistance. The consensus is that if one is healthy, coffee will prevent diabetes, but if one has diabetes, it is not a good idea to drink it. Let's see how true this is.

One oft-cited epidemiological study that you may find from googling coffee and diabetes is this one involving post-menopausal women. Women who drank greater than 4 cups a day had higher levels of sex-hormone binding globulin (SHBG) in their blood plasma and 56% less risk for developing diabetes after a 10-year follow up. Make note that high SHBG and low testosterone only protect WOMEN from diabetes. In men low testosterone is implicated in diabetes and obesity. Men do not want high SHBG, it will bind to free testosterone and reduce the bioavailable amount. Despite being another epidemiological study, the type that I don't like, I cannot argue against these statistics. Maybe drinking coffee daily will create an anti-diabetic hormonal profile, at least in women, and at least when it comes to terms with SHBG, but it may not if weight is gained. 

On a smaller scale of 4 weeks, where mechanisms can be better elucidated, Dam et al. found that the equivalent of 13 cups of coffee a day did not just result in 14 participants (out of 40) withdrawing from the study, but also higher levels of fasting insulin after 4 weeks. They performed a second study (including data from 45 people after drop-outs) with one group receiving 870mg of caffeine a day and the other 52g of coffee grounds a day (0.9 L), and found similar rises in fasting insulin but not in fasting glucose.  The authors concluded that it's not a good idea to recommend coffee to lower type II diabetes risk. Wrong conclusion, imho. Blood glucose levels were unchanged despite such high amounts of coffee. Also we don't see postprandial (after a meal), glucose concentrations (which are better to test for diabetes risk as we'll explain below). 

Dam et al. 2004

This is the difference between BAD insulin resistance with impaired glucose metabolism, and not bad insulin resistance, where other mediators such as epinephrine or free fatty acids keep insulin high, but glucose unchanged. In fact that's exactly what happens with caffeine consumption. We see transient insulin resistance because of epinephrine, but also adenosine antagonism, and other mechanisms that people are still trying to figure out.

If you want to learn more about this benign form of insulin resistance (that shows that more fat is being burned and glucose is not staying elevated), read this post on how butter does the same thing. Highly highly recommended. May just make you go "WOW."And make you eat butter to burn more fat and improve heart health. 

 Ok so in healthy subjects, there isn't much of a problem short term in blood glucose homeostasis. But in diabetics what does coffee and caffeine do? Lane et al. found that a 375mg dose of caffeine elevated blood glucose 21% IN DIABETICS. That's pretty significant, but for 375mg, the amount of caffeine in 3-4 cups of coffee, it may not be much? In the fasted state, there was not much of an effect on insulin and glucose concentrations, but the authors cite that this is exactly why we should be concerned; postprandial (after a meal) glucose concentrations may be a better indicator of diabetes risk they say, citing this study (you don't want high hba1c, an index of hyperglycaemia). The authors had strong data and strong reason to conclude that diabetics should not drink coffee, at least in conjunction with a high carbohydrate meal (the one they used contained 70g carbs from a sugary Boost drink). Except they used caffeine, NOT coffee. Coffee would have probably had the same effect though. 

Right, so although the physiological effects of coffee are mediated mainly by caffeine, I always believe in the whole food, and coffee contains buzzwords like antioxidants and polyphenols. In fact, coffee is probably the largest source of antioxidants for an average American. Recently, as in, like last November, Cheng et al. published research examining how two other major constituents in coffee, chlorogenic acid (CGA) (as well as EGCG from green tea) and caffeic acid (CA), can reduce the formation of human islet amyloid polypeptide (hIAPP), which can cause beta cell apoptosis (beta cells in the pancreas are what release insulin and when they are destroyed you become insulin resistant). I talked about in part 1 how amyloid proteins aggregate in Alzheimer's disease. Well they do in diabetes as well.

Caffeine had the lowest effect on preventing hIAPP formation, CGA had the largest effect, but caffeine did have other beneficial effects on pancreatic cell viability and in preventing fibril formation (there are many steps in the process for hIAPP to form amyloids and is beyond the discussion here, and my head for that matter right now). This paper doesn't directly answer my unanswered question from the first segment on advanced glycation endproduct (AGE) formation, but I suspect that if the process of AGE formation, and amyloid formation by hIAPP is similar, then these polyphenol compounds would reduce AGE formation. Now these researchers conducted an in vitro study. But they conclude that physiological levels of CGA and CA would be greater in vivo than in their study. They also talked about the various metabolites of these compounds which are all polyphenols, some of which they studied here and found similar abilities in preventing hIAPP aggregation.

Cheng et al. 2011

Trust me I'm as confused as you. Continuing with the "whole food" mentality and chlorogenic acid, Johnston et. al. found that CGA may have antagonistic effects to caffeine. Nine healthy volunteers were given either caffeinated or decaffeinated coffee, with the same amount of CGA, and the levels of insulin, glucose, glucose-dependent insulinotropic polypeptite (GIP), and glucagon-like peptide 1 (GLP-1), were assessed over a 3 hour period after ingestion. Results? Decaffeinated coffee resulted in lower insulin and glucose levels over the three hour period. Caffeinated coffee resulted in a higher level of insulin at 30 minutes, but not much higher at the other time values. The authors suspect that this is due to CGA, and the fact that levels of GIP and GLP-1 were higher. Those two hormones, classified as incretins, are anti-diabetic, they sensitize tissues to insulin. If you're a nutrition geek, you have to read that paper. It talks also about how the release of these hormones results in glucose being absorbed in more distal parts of the small intestine, due to CGA mediated Na+ electrochemical gradient dissipation.

In C57BL/6J mice fed a high fat diet (these mice are a model for diabetes), coffee and caffeine were both found to improve insulin sensitivity and reduce inflammation, as assessed by inflammatory markers such as interleukin-6 (IL-6). The diet was revealed to only be 13.6% fat though.. so watch out for those "high-fat" studies implicating fat as the problem. The point is, the mice's glucose tolerance was slowly impaired, and coffee, as well as caffeine, ameliorated this. The expression of inflammatory adipocytokines, chemicals that mediate the inflammation process and are implicated as well in diabesity and heart disease, were reduced.

In humans though, this epidemiological study found that coffee drinkers, both male and female, had higher levels of IL-6, as well as c-reactive protein (one of the main causative agents in heart disease). LOL. What does this mean? Coffee can be good as part of a healthy lifestyle. I'm more convinced by the mRNA analyses than the epidemiological evidence, but the epidemiological stuff tells us what happens to normal people who behave normally. It doesn't tell us what could happen under optimal conditions. If you try to be healthy and consume coffee in moderation, you may be healthier and happier! But while being stressed out and sleep deprived, coffee may make things worse and that's not the subject of this article.

To conclude, Wedick et al. found that 5 cups a day of coffee for 8 weeks in healthy overweight subjects also raised IL-6 concentrations (may cause insulin resistance), but also higher adiponectin levels which they cite are inversely associated with T2DM (type II diabetes mellitus). There was not much of an effect on insulin and glucose concentrations after the 8 weeks. The group receiving decaffeinated coffee showed lower fetuin-A concentrations. Fetuin-a is a glycoprotein secreted by hepatocytes and higher levels increase T2DM risk. Caffeinated coffee had no effect on fetuin-A concentrations, and decaffeinated had no effect on adiponectin levels. The authors conclude that over an 8-week period, the acute disturbances to glucose homeostasis mediated by coffee do not remain. Indices of T2DM risk show that coffee may reduce risk.

FINAL NOTES: 
1. Coffee contains a lot of phytochemicals and polyphenols which are anti-diabetic, through several mechanisms, including reduction of hIAPP polymerization, higher adiponectin levels, for secreting GLP-1 and GIP, and for raising epinephrine levels which may aid in fat loss.

2. Coffee taken with a high carbohydrate meal in diabetics may be a bad idea, since it aggravates hyperglycaemia. But consuming large amount of carbohydrates as a diabetic may not be a good idea anyway. On a lower carbohydrate diet, consuming coffee may not pose much of a threat to insulin/glucose homeostasis, and may aid in the fat loss process. However, if one of coffee's main mechanisms is the prevention of hIAPP polymerization, then it would still improve the function of the pancreas despite elevated serum glucose levels.. The influence of coffee on hIAPP formation in conjunction with a high carbohydrate diet has not been studied yet to my knowledge. If coffee's anti-diabetic action is mostly behind the scenes, then the glucose/insulin story won't tell us much at all. I suspect this to be the case.

3. I don't know anyone who wants to "prevent diabetes." If you are normal weight, you probably don't care about preventing it. So who cares? The studies I overviewed show that metabolically, coffee is overall good for glucose homeostasis and prevention of toxic amyloid proteins aggregating in your pancreas and brain, but that doesn't mean diabetes is caused by not drinking coffee. I think the real question is, should diabetics drink coffee? The answer is, not with a lot of sugar and carbohydrates, but surely in conjunction with a diet and exercise intervention designed to improve insulin sensitivity and weight loss. It may be best in the fasted state, before exercise for instance since Lane et al. who found increased glucose levels when coffee was taken with a meal in diabetics found no effect in the fasted state. I don't have the answer to this yet, but perhaps in the fed state, coffee could cause weight gain in susceptible individuals on a high sugar diet, but in the fasted state it could aid fat loss.

4. Add cinnamon to your coffee if you are diabetic. As in cinnamon powder, not cinnamon flavored coffee.

 I hope you enjoyed this if you made it this far, and I hope it answered some questions and it was not confusing and that it was mostly correct. 


 -AS