Medical Myths 001: Metformin mythology
Metformin is a good drug. Has been 1st line oral diabetic agent for years and is effective for other problems – PCOS etc. However, it has for a long time been shrouded by the perception that it causes lactic acidosis and is therefore a bit dangerous… but is this true? Looking at the literature there are a lot of case reports from ICUs and EDs of patients who present with profound metabolic acidosis on metformin. They have a high mortality and sound quite impressive – one might start to think – is Metformin as benign as we all think?
Recent large retrospective studies have looked at the rate of lactic acidosis anmongst diabetics on metformin and found that it is really a very rare complication, and is always associated with another more acute cause of lactic acidosis – renal failure, sepsis, acute heart failure, hypoxia etc – all common in diabetics.
This 2003 review in Arch of Int Med shows that there is no difference in rates of lactic acidosis between diabetics on metformin vs those on other oral agents. Another review in 2004 Diabetes Care showed the same, essentially that metformin in overdosage (suicidal doses) may cause lactic acidosis, but at therapeutic doses it is not seen. Metformin is the innocent scapegoat where there is often a more acute cause.
The Fremantle Diabetes study here in WA looked at this issue and came to the same conclusions in 2008. The incidence of lactic acidosis was more related to the incidence of diabetic co-morbidity (renal, heart, aging) than to the dose or duration of metformin treatment.
So, what does all this mean for our day-day practice?:
- Metformin is safe at usual doses (3g/day)
- Patients with acute conditions whom are unstable / at risk of end-organ failure / shock should have their metformin stopped (unfortunately it is usually already on board). Basically if your patient is sick enough to come in to hospital you should think – “should I hold the metformin?”
- Those at higher risk – sepsis, hypoxia, heart failure / ACS – possible cardiogenic shock , those with impaired renal function who are at risk of acute deterioration.
- Consider patients who need for IV contrast imaging – should hold it off until the study then afterwards for 3 days.
- In outpatients – watch renal function: esp in the older pt, alcoholic or malnourished.
Anyone out there got any points to add, cases to show?
I’m a little confused – you build a good case for relaxing the accepted contraindications citing several relevant papers and then list practice points (the second and third at any rate) that are consistent with the fear of metformin causing lactic acidosis. If they’ve got renal impairment, sure, reduce the dose – it is renally excreted after all – but it doesn’t cause renal impairment. The Agincourt award has to go to the contrast / cease metformin meme – I must revisit the evidence for that theoretical (squared) risk.
One last point – we rarely use 3g metformin here in the kimberley – Chronic Disease Guidleines don’t advocate >2g.
Great work though, keep blogging Cheers