Acute Metformin Overdose: Examining Serum pH, Lactate Level, and Metformin Concentrations in Survivors Versus Nonsurvivors: A Systematic Review of the Literature
Received 20 June 2008; received in revised form 11 December 2008, 24 April 2009 and 28 April 2009; accepted 29 April 2009. published online 26 June 2009.
Study objective
Metformin is known to cause potentially fatal metabolic acidosis with an increased lactate level in both overdose and therapeutic use. No association between mortality and serum pH, lactate level, or metformin concentrations, though intuitive, has yet been described. This systematic literature review is designed to evaluate the association between mortality and serum pH, lactate level, and metformin concentrations in acute metformin overdose.
Methods
We reviewed the literature by using the MEDLINE, EMBASE, CINAHL, and TOXNET databases for cases of metformin overdose with documented mortality data and values of serum pH, lactate level, and metformin concentrations. When available, patient age, patient sex, and whether patients received intravenous sodium bicarbonate therapy or hemodialysis were also analyzed. Cases meeting inclusion criteria were analyzed to determine whether a difference in distribution of nadir serum pH, peak serum lactate level, or peak serum metformin concentrations existed between overdose survivors and nonsurvivors.
Results
We identified 10 articles that had 1 or more cases meeting our inclusion criteria. In total, there were 22 cases of metformin overdose (5/22 died) that met inclusion criteria. No intentional overdose patients died whose serum pH nadir was greater than 6.9, maximum lactate concentration less than 25 mol/L, or maximum metformin concentration less than 50 μg/mL (therapeutic range 1 to 2 μg/mL). Intentional overdose patients with a nadir serum pH less than 6.9 had 83% mortality (5/6), those with lactate concentration greater than 25 mmol/L had 83% mortality (5/6), and those with metformin concentration greater than 50 μg/mL had 38% mortality (5/12). Nadir serum pH and peak serum lactate and metformin concentration distributions in survivors and nonsurvivors revealed that survivors had a median nadir pH of 7.30, interquartile range (IQR) 7.22, 7.36; nonsurvivors, a median nadir pH of 6.71, IQR 6.71, 6.73; survivors, a median peak lactate level of 10.8 mmol/L, IQR 4.2, 12.9; nonsurvivors, a median peak lactate level of 35.0 mmol/L, IQR 33.3, 39.0; survivors, a median peak metformin level of 42 μg/mL, IQR 6.6, 67.6; and nonsurvivors, a median peak metformin level of 110 μg/mL, IQR 110, 110.
Conclusion
No cases of acute metformin overdose meeting the study's inclusion criteria were found in which patients with a nadir serum pH greater than 6.9, peak serum lactate concentrations less than 25 mmol/L, or peak serum metformin concentrations less than 50 μg/mL died. Patients with acute metformin overdose who died had much lower serum pH nadirs and much higher peak serum lactate and metformin concentrations than those who survived.
aEmory University School of Medicine Department of Emergency Medicine and the Georgia Poison Center, Atlanta, GA
bEmory University Rollins School of Public Health Department of Epidemiology, Atlanta, GA
Address for correspondence: Damon M. Dell'Aglio, MD, Georgia Poison Center, 80 Jesse Hill Jr. Drive SE, Atlanta, GA 30303
Author contributions: DD and LP conducted the literature review, data collection and database search. JA conducted the statistical analysis and statistical editing and revision. DD wrote the initial draft. All authors conceived and designed the study and contributed to editing and revisions. All authors contributed substantially to revisions. DD takes responsibility for the paper as a whole.
Funding and support: By Annals policy, all authors are required to disclose any and all commercial, financial, and other relationships in any way related to the subject of this article that might create any potential conflict of interest. The authors have stated that no such relationships exist. See the Manuscript Submission Agreement in this issue for examples of specific conflicts covered by this statement.
ABSTRACT