The notion that artificial sweeteners (and sweet tastes in general) might produce an insulin response is one of those murky memes that winds itself around the blogs, but it’s never stated one way or the other with any sort of confidence. I briefly mentioned the possibility of non-caloric sweeteners influencing satiety hormones in last week’s diet soda post, and a number of you guys mentioned the same thing. Still, I’ve never seen unequivocal evidence that this is the case.
This whole idea first came to my attention some time ago when my dog Buddha got into a bottle of “alternative sleep assists” which contained, among other things, 5 HTP (version of l-tryptophan) and xylitol (sugar alcohol). Long story short, dogs can’t take xylitol because it causes a spike in insulin, which then severely depletes blood glucose. Buddha got past this with a trip to the vet’s at 10:30 Sunday night (thanks, Dr. Dean). But it occurred to me that the same effect might be seen in humans, which is why I pose the question today…
Do artificial sweeteners induce insulin secretion (perhaps via cephalic phase insulin release, which is sort of the body’s preemptive strike against foods that will require insulin to deal with)?
One of the reasons a definitive answer is rarely given is that the question is improperly framed. Artificial sweeteners is not a monolithic entity. There are multiple types of sweeteners, all of them chemically distinct from each other. A more useful question would be “What effect does [specific artificial sweetener goes here] have on insulin?” So let’s go around the circle and ask.
Does aspartame (aka Equal and Nutrasweet) affect insulin?
Aspartame is pretty gross stuff, what with its awful taste and hordes of people who get terrible reactions from consuming it, but that’s not what we’re interested in today. Luckily, there is a good amount of research explaining what, if any, effect aspartame has on insulin secretion.
Another earlier study (full PDF) examined the effects of aspartame on prolactin, cortisol, growth hormone, insulin, and blood glucose levels and found it had none. The authors used the same amount of aspartame you’d find in a standard artificially-sweetened drink but were unable to record any significant hormonal alterations.
A study of forty-eight healthy volunteers found no evidence that aspartame has an effect on insulin levels.
Overall, the evidence seems to suggest little, if any, effect on insulin secretion as a result of tasting or consuming aspartame.
Does saccharin (aka Sweet’N Low) affect insulin?
Although saccharin has lingered in obscurity and consumer banishment (who ever really picks Sweet ‘n’ Low, anyway?) for most of the last couple decades (until recently when the EPA dubbed it safe for human consumption), there is some research on its effects on insulin.
In one study, fasted human subjects swished around eight different taste solutions for 45 seconds, and then spat them out. No swallowing. Only the sucrose and saccharin solutions activated a cephalic phase insulin release.
On the other hand, another study using humans found the opposite: swishing and spitting sweet solutions (even caloric ones using sucrose) did not elicit CPIR, while another study found that neither saccharin nor aspartame influenced insulin secretion in both fasted diabetics and non-diabetics (although aspartame-fed subjects had slightly higher insulin levels than the control and saccharin groups, this was physiologically irrelevant given the steady blood glucose levels).
The evidence for saccharin’s effect on insulin is mixed, but either way, it doesn’t appear to have too big of an impact in real world terms.
Does acesulfame K (aka Sunett and Sweet One) affect insulin?
In one study, researchers found that direct transfusions of acesulfame K increased insulin secretion in rats in a dose-dependent fashion. The same researchers performed an in vitro study, subjecting isolated rat pancreatic islets to acesulfame K solutions, and found that the artificial sweetener was an independent actor on insulin secretion. Both indicate that there is some effect, but it’s difficult to draw any conclusions from in vitro rat studies using isolated pancreatic cells or in vivo rat studies using direct transfusions of sweeteners (as opposed to oral dosing).
Another study using isolated pancreatic cells found that only those artificial sweeteners with a bitter aftertaste (acesulfame K, saccharin, stevia, and cyclamate) augmented the insulin response in the presence of glucose. Aspartame, which does not have a bitter aftertaste, did not affect insulin. Note, though, that this was an in vitro study using isolated cells and that the presence of glucose was a prerequisite for insulin secretion. Of course, dieters slurping down artificial sweeteners do it during meals, most of which tend to feature large amounts of glucose.
Acesulfame K appears to affect insulin levels, although this effect has only been shown in contrived settings – either in the presence of glucose in isolated cells (in vitro), in isolated cells in without glucose (in vitro), or by direct transfusions without the presence of glucose (in vivo). We haven’t seen people orally taking acesulfame K in a fasted state and having an insulin response. Yet.
Does sucralose (aka Splenda) affect insulin?
Sucralose activates the sweet receptors in taste buds, and some in vitro studies have shown that sucralose can stimulate the release of incretin hormones, which increase the secretion of insulin, via the sweet taste receptors in enteroendocrine cells (located in the gut). An in vivo study of sucralose infusions into the gut, however, showed that it does not stimulate the incretin hormones GLP-1 or GIP, does not release insulin, and does not slow gastric emptying.
Another in vivo study, this time using healthy human subjects, got similar results: oral dosing of sucralose did not induce a cephalic insulin response, nor did it affect GLP-1. Not even appetite was affected.
The commercial version of sucralose, Splenda, is cut with dextrose as a bulking agent. Dextrose is essentially glucose, which certainly elicits an insulin response, so there’s definitely the potential for a slight insulin response to Splenda, but there’s not much if any evidence that sucralose has an independent in vivo effect on insulin.
Recently, a review of in vivo studies concluded that “low-energy sweeteners” do not have any of the effects on insulin, appetite, or blood glucose predicted by “in vitro, in situ, or knockout studies in animals.” As far as the clinical studies go, I think I’d have to agree. Am I going to use the stuff? No; there are other potential negative effects to artificial sweetener usage, including gut flora disturbances, the promotion of psychological dependencies on sweets, and long term safety issues, but I think it’s important to be clear on where the science lies. So far as I can tell, according to the literature there isn’t an appreciable insulin effect from most sweeteners.
Still, some people anecdotally report an effect. As Jimmy Moore says, “The bottom line is to check your own blood sugar response and see how it impacts YOU.” Word. If you need to know (and most people don’t), testing yourself would be the way to do it.
Let me know if I’ve missed something, or come up short in my analysis. There’s a lot of stuff out there and it’s possible that I’ve overlooked something. And as always, I’d love to hear about your personal experiences with artificial sweeteners, especially regarding their effect on your weight loss/gain, insulin, appetite, and dietary success/failure. Let me know in the comment section!
P.S. If you’re new here and aren’t sure what all the fuss over insulin is about start here: The Definitive Guide to Insulin, Blood Sugar and Type 2 Diabetes (and you’ll understand it)