A further prospective evaluation of an equation to predict daily lithium dose

Recently, one prospective study confirmed the safety and accuracy of a lithium-dose prediction equation created by Zetin et al., but no definitive conclusion on the reliability of the equation has been established as yet.

The authors applied the Zetin et al. equation to 18 chronic male schizophrenic inpatients. Predicted doses to reach the serum lithium concentration of 0.4 mmol/L were calculated and prescribed in the form of lithium capsules. At Weeks 1 and 3 after treatment initiation, morning blood samples were collected about 12 hours after the last lithium dose for the measurement of serum lithium concentrations.

None of the 18 patients achieved the desired concentration (0.4 mmol/L) exactly. The mean +/- SD of serum lithium concentrations at Week 1 was 1.01 +/- 0.29 mmol/L (range, 0.2-1.5) and at Week 3 was 0.94 +/- 0.35 mmol/L (range, 0.2-1.8). Lithium concentrations were lower than 0.4 mmol/L in only 1 patient and were higher than 0.4 mmol/L in the other 17 patients. The deviations from the unexpected value were significantly correlated with the renal function (blood urea nitrogen and serum creatinine levels) but not with the neuroleptic doses administered to the patients. Moreover, our patients were relatively older and weighed relatively less than the patients described in the previous prospective study.

The Zetin et al. equation cannot always accurately predict a required lithium dose. Renal function data, even when they range within normal values, may be useful to improve the accuracy of the equation, particularly in patients who are older or weigh less than the norm.