Metformin, in contrast, does not appear to be as closely associated with lactic acidosis. Most of the reported cases are in patients who would appear to have other reasons for lactic acidosis (although this case series including patients on dialysis that appears to be reasonably convincing and there are multiple smaller case series in patients on HD). A large meta-analysis of clinical trials if metformin showed no increase in the rate of lactic acidosis in patients on the drug. However, it should be pointed out that these trials did not include patients with advanced CKD. Metformin is not protein-bound and is largely cleared by the kidneys so there is clearly accumulation of the drug in patients with CKD. However, it is unknown what the real “safe” level is and no studies can ethically be performed to establish this.
As a result, in general, guidelines regarding the use of metformin have been conservative. The insert for the drug suggests that it is contraindicated in individuals with a serum creatinine >1.4mg/dl (women) or 1.5mg/dl (men). Personally, I find these guidelines to be insufficient and frustrating because of course, a creatinine of 1.5mg/dl can mean entirely different things depending on the context (age, weight, race etc.) So it is gratifying to see that the diabetes associations have moved forward with newer guidelines.
The ADA published a paper with new guidelines for the use of metformin in 2011 incorporating eGFR. This paper also summarizes very nicely all of the evidence for and against the use of metformin in CKD. In summary, they suggest that for individuals with eGFR>45, metformin can be safely initiated. For those with an eGFR between 30-45, it can be continued with caution if it is already in use. Finally, it is contraindicated in patients with eGFR<
30. These guidelines make more physiologic sense and although they may still be too conservative, in this era of automatic reporting of eGFR, to me, it is the better approach to take.