Comparison of three lithium dosing methods in 950 "subjects" by computer simulation.

Area of Science:

• Pharmacokinetics
• Clinical Chemistry
• Computational Pharmacology

Background:

• Accurate prediction of lithium steady-state concentrations is crucial for effective therapeutic drug monitoring.
• Existing methods like the one-point method (OPM) and repeated one-point methods (ROPM) have varying degrees of accuracy.
• Simulating assay and timing errors is essential to evaluate the robustness of these prediction methods.

Purpose of the Study:

• To compare the predictive accuracy of the OPM, ROPM with a 12-hour interval (ROPM-12), and ROPM with a 24-hour interval (ROPM-24).
• To assess the impact of simulated assay and timing errors on the accuracy of these lithium prediction methods.
• To identify the most reliable method for predicting lithium steady-state concentrations in clinical practice.

Main Methods:

• Generated pharmacokinetic parameters (k, V, ka) for 950 simulated subjects.
• Applied OPM, ROPM-12, and ROPM-24 to predict lithium concentrations.
• Introduced +/- 5% and +/- 10% errors to simulate assay and timing inaccuracies.

Main Results:

• The OPM demonstrated the smallest prediction errors across simulated error levels (SD5%, SD10%).
• ROPM-24 also showed good accuracy under the SD5% error condition.
• A significant correlation was observed between the elimination rate constant (k) and prediction error for OPM and ROPM-24.
• The OPM tended to underpredict lithium concentrations in subjects with longer half-lives.

Conclusions:

• The OPM is the most accurate method for predicting lithium steady-state concentrations among those evaluated.
• While accurate, the OPM's tendency to underpredict in certain patient populations necessitates careful consideration.
• ROPM-24 offers a viable alternative, particularly when assay errors are minimal.