Metformin is the only hypoglycemic agent proven to reduce mortality in Type 2 DM. As such, we really should be making every effort to prescribe it, and keep patients taking it. Unfortunately, mostly due to it's relationship to parent compound phenformin, metformin has developed a reputation for causing lactic acidosis (LA). Prescribing guidelines instruct witholding the drug if there is 'renal impairment'. This is a controversial issue, reviewed here. Here are some points to consider:
- A Cochrane review found no difference in LA rates between those taking and not taking metformin, around 6 cases per 100,000
- The prognosis (or lactate level) does not correlate with metformin levels in cases of LA where metformin has been implicated, but does correlate with the severity of the underlying hypoxic condition.
- A post-hoc analysis of the GOAL-A1C study, showed that 18% of patients taking metformin in the community already had CKD III
- Many patients have received metformin in the presence of an absolute contraindication and the rate of LA is not seen to be increased. One such study includes patients with CI's such as a creatinine > 2.5mg/dL and NYHA IV heart failure.
Personally, I have seen a few severe cases of LA in patients taking metformin that clearly had another precipitant. I do believe that the lactate levels in those cases were higher than I would have otherwise expected. Maybe this is the metformin effect - to aggravate a LA in evolution. Critical illness notwithstanding, I continue metformin down to a CrCl of 30ml/min, but would be interested in hearing what others out there are doing.
I discontinue when GFR is lower than 45. I have seen a couple cases of High anion gap acidosis more so in obese patients in CKD -3 patients prompting me to discontinue metformin
ReplyDeleteDespite the benefits of metformin base on so many studies, I tend to worry about the issue of lactic acidosis in worsening renal function patients...therefore my vote is a an earlier threshold to stop metformin; GFR <50 or sudden worsening of renal function. Better not to cause harm, especially when we know it has happened before...
ReplyDeleteHere in Canada, where litigation concerns are not the same, consensus is stopping when eGFR is < 25 to 30(by CG calculation not by MDRD), in my center (20 nephrologists) at least... I have personally seen a few cases where Metformin played a role in LA, yet since it's the only drug that saves lives in DM2, discontinue only at GRF <25
ReplyDelete