Ellis, below is the chemical structure of metformin. As you can see, it looks nothing like a sugar. Sugars don't contain nitrogen, and metformin only has five carbons. Glucose, the simplist sugar, has six, and they are configured in a totally different way in all three sterioisomers.
Metformin hydrochloride is a white to off-white crystalline compound with a molecular formula of C4H11N5 • HCl and a molecular weight of 165.63. Metformin hydrochloride is freely soluble in water and is practically insoluble in acetone, ether, and chloroform. The pKa of metformin is 12.4. The pH of a 1% aqueous solution of metformin hydrochloride is 6.68. Its structural formula is:

Interesting research data on metformin and non-diabetics:
Once Off the Drug Many Participants Immediately Became Diabetic
The conclusion that the use of Metformin alone without changes in diet might prevent diabetes was undercut by a follow-up study which found that within weeks of going off the drug, a quarter of that apparent risk reduction disappeared, when a surprising number of people in the Metformin group developed diabetic blood sugar levels as soon as they came off the drug. This suggested that the drug had no effect on the underlying process leading to diabetes, but had only suppressed peoples' blood sugar levels while the underlying mechanism of damage went on unchecked. Unlike the original finding that Metformin appeared to prevent diabetes, this later finding received no play in the media and many doctors are still unaware of it.
So the fact seems to be that taking Metformin might slow the progress of blood sugar deterioration by a very small amount, but it does not bring blood sugars down to the normal level that might truly prevent diabetes or prevent the damage done by abnormally high blood sugars.
The reason for this is made clear by the findings of the university of Texas researchers: The big problem with Metformin is that it works best when blood sugars are high--and in people who are obese. In people of near-normal weight the UT researchers found that at normal blood sugar levels Metformin actually decreased rather than enhanced glucose uptake. This is in line with the finding that the primary way in which Metformin appears to work to lower fasting blood sugars is by limiting the production of glucose by the liver--a phenomenon more likely to occur in people whose fasting blood sugar is high--much closer to the ADA diabetes diagnostic cutoff than to normal.
That this is true was made clear by the note in the published report on the Diabetes Prevention Program study that "The effect of Metformin was less with a lower body-mass index or a lower fasting glucose concentration than with higher values for those variables."