Objective: Lithium, due to its low therapeutic index, should be closely monitored during clinical practice. Moreover, its clinical effects need 7-14 days and it doesn't have effects at subtherapeutic serum level. These problems necessitated development of several prediction formulas of lithium dosage requirements. Authors compared 4 prediction formulas (Lobeck, Zetin, Terao, and Yugawa method) to identify the most suitable method for Korean bipolar patients.
Methods: Medical records of 30 bipolar patients were reviewed. Steady-state(at least 6 days of usage) dosages and serum levels of lithium were identified. We calculated predicted dosages through each prediction formulas using variables such as age, body weight, height, BUN, creatinine. We compared correlations between real doses and predicted doses calculated through each prediction formulas. Absolute mean error and mean error were calculated to assess the precision and bias of each method.
Results: Pearson's correlation coefficients were Zetin method 0.544 (p=0.002), Lobeck method 0.442 (p=0.014), Terao method 0.459 (p=0.011), Yugawa method 0.396 (p=0.030), each. Zetin method was most closely related with real dose of lithium. Absolute mean dosage prediction error showed no significant difference among 4 methods.
Conclusions: Zetin method was the most suitable formula for predicting lithium dosage requirement in our bipolar patients. But clinical significance of accuracy of methods needs to be further investigated.
|