Aims: To study the quality of pharmacovigilance reporting in cases of so-called 'metformin-associated lactic acidosis' and, ultimately, whether or not the criteria for this condition are indeed met.
Methods: We searched for cases meeting the criteria for metformin-associated lactic acidosis [arterial pH < 7.35, blood lactate > 5 mmol/l (45 mg/dl) and detectable plasma metformin concentration] in a 15-year period (1995-2010) in a pharmacovigilance database of the license holder for metformin (Merck Serono, Lyon, France).
Results: We found 869 reports stated as 'metformin-associated lactic acidosis' from 32 countries. The respective criteria for pH, lactate concentration and metformin concentration were met in 51.2, 53.3 and 13.9% of cases. All three criteria were met in just 10.4% of cases. By year, each of the percentages remained roughly stable throughout the study period.
Conclusions: The role of metformin in triggering metformin-associated lactic acidosis was assessed incorrectly in most patients and the quality of reporting did not improve over time.
© 2012 The Authors. Diabetic Medicine © 2012 Diabetes UK.