Metformin kinetics in healthy subjects and in patients with diabetes mellitus.

The kinetics of metformin were studied after i.v. and oral administration in four healthy subjects and after oral administration in twelve maturity onset (Type II) diabetic patients. After i.v. administration most of the dose was rapidly eliminated but with a mean `terminal' of 4 h measured up to 12 h in plasma and of 16 h measured up to 60 h from the urinary excretion rate. On average, 80% of the dose was recovered as unchanged drug in the urine with none detected in the faeces. After single oral doses (0.5 and 1.5 g), maximum plasma concentrations and urinary excretion rates were observed at about 2 h with urinary recoveries of unchanged drug of 35-50% and faecal recoveries of about 30%. Urinary recoveries were significantly lower after the higher dose. Absolute oral bioavailability was 50-60% of the dose. Deconvolution analysis showed that after a short lag-time, the available oral dose was absorbed at an exponential rate over about 6 h. Implications for the design of prolonged release dosage forms are discussed. Plasma metformin concentrations measured throughout the seventh and fourteenth days of continuous 0.5 g twice daily treatment were accurately predicted from single dose data, although a discrepancy between observed and predicted trough levels reflected the existence of a slow elimination phase. Implications of the latter for a gradual accumulation of metformin in peripheral tissues and a possible association with lactic acidosis are discussed. Renal clearance of metformin was highly correlated with creatinine clearance. However, a weaker relationship between total oral clearance of the drug and creatinine clearance suggests that the latter may not always be a reliable indicator of potential metformin accumulation owing to variability in absorption and possibly non-renal clearance of the drug,