RESUMEN
BACKGROUND: Evidence of metformin-associated lactic acidosis (MALA) in advanced chronic kidney disease (CKD) has been limited due to high mortality rate but rare incidence rate. The mechanism of increased MALA in advanced CKD is mainly based on the hypothesis of decreased drug elimination, which might also be confounded by increased comorbidities as CKD progresses. The goal of the study is to analyze the incidence and associated factors of lactic acidosis between metformin user and non-user with advanced CKD. METHODS: This study used a three million population-based, propensity score-matched cohort from 2008 to 2016. The primary outcome was laboratory-defined lactic acidosis. Relationships between the probability of lactic acidosis and various estimated glomerular filtration rate (eGFR) values in advanced CKD patients were also presented in regression analysis. RESULTS: Adults with type 2 diabetes whose eGFR was <30 mL/min/1.73 m2 were enrolled in this study. After the process of propensity score matching, 7707 patients were divided into metformin and non-metformin groups. In linear regression model, metformin significantly increased the risk of lactic acidosis (p=0.0204) as the eGFR declined in advanced CKD over a mean follow up of over 600 days even after confounder adjustment with age, sex and comorbidities. CONCLUSIONS: Metformin was associated with a significant increased risk of laboratory-defined lactic acidosis (p=0.0204) even after adjusting confounder such as age, sex and underlying comorbidities. This "REMIND" study reminds us that metformin-associated lactic acidosis is mainly caused by decreased drug renal elimination other than underlying comorbidities in advanced CKD patients.