Role of metformin accumulation in metformin-associated lactic acidosis.

OBJECTIVE

To investigate the role of metformin accumulation in the pathophysiology of metformin-associated lactic acidosis.

RESEARCH DESIGN AND METHODS

We used high-performance liquid chromatography to measure plasma metformin concentrations in 14 patients who experienced lactic acidosis (pH < 7.35 and lactate concentration 5 > mmol/l) while receiving chronic metformin treatment. Their treatment was generally based on alkalinization and dialysis therapy.

RESULTS

Clinical shock and/or evidence of tissue hypoxia was found in all patients with the exception of one who had a nonsteroidal anti-inflammatory drug-induced anuria. Ten patients had significant metformin accumulation (plasma metformin concentrations 4.1-84.9 mg/l, normal value 0.6 +/- 0.5 mg/l before drug intake), generally because of failure to withdraw metformin despite intercurrent pathological conditions affecting its renal elimination (serum creatinine concentrations ranging from 269 to 1,091 mumol/l). There was no metformin accumulation (plasma metformin 0.03-0.7 mg/l) in the four other patients, who had less severe renal failure (serum creatinine 140-349 mumol/l). The severity of the patient's general condition did not predict early hospital mortality (death before discharge from the intensive care unit) even in patients in shock. Whereas it was high in those without metformin accumulation (only 1 of 4 patients recovered), early hospital mortality was low in the 10 patients with metformin accumulation and was not related to its extent (3 patients died with end-stage hepatic failure or cardiac failure). Correlation studies showed a positive correlation between serum creatinine and plasma metformin and between plasma metformin and arterial lactate but, for the latter correlation, only in patients with metformin accumulation.

CONCLUSION

Metformin-associated lactic acidosis is not necessarily due to metformin accumulation; true type B (aerobic) lactic acidosis, i.e., without an apparent associated hypoxic factor, seems exceptional. Neither the severity of the clinical picture nor the degree of metformin accumulation predicted survival; rather, the prognosis was dependent upon the severity of the associated pathological conditions.