Background and objective: Lithium, which is used to treat bipolar disorder, has a narrow therapeutic blood concentration range and quickly reaches clinically toxic levels. We performed a population pharmacokinetic analysis with a lithium tubular reabsorption model including urinary pH and investigated the relationship between blood lithium concentration and tremor as a side effect.
Methods: Routine clinical data, including 389 serum concentrations, were collected from 214 patients orally administered an adjusted amount of lithium carbonate. Pharmacokinetics were described using a one-compartment distribution model with first-order absorption and elimination. The fractions of the MID (Li+ + LiCO3-) and ION (2Li+ + CO32-) forms were calculated using the Henderson-Hasselbalch equation, and the influences of these fractions on clearance (CL) were evaluated. The rate of tremor development was analyzed using a logit model.
Results: Oral apparent CL (CL/F) was explained by nonrenal CL and renal CL, and renal CL was varied by the fractions of lithium forms influenced by urinary pH. The contribution of MID to CL was slightly larger than that of ION. The rate of tremor development was estimated to be more than 30% when the trough lithium concentration was greater than 1.26 mEq L-1.
Conclusion: Renal function and urinary pH are important indices in lithium treatment, so the serum concentration of lithium may be predicted based on the renal function and urinary pH.