To the Editor: The APA Practice Guideline for the Treatment of Patients With Bipolar Disorder (R15601CHDCDHGC) mentioned two methods to initiate lithium treatment and achieve therapeutic serum levels in patients with bipolar disorder. Lithium may be started in low divided doses and the dose titrated upward according to the response and side effects, or, after a single dose of 600 mg the 24-hour lithium serum level may be used as an indicator of required daily dose. The latter method, in which a nomogram is used to predict the appropriate daily dose (R15601CHDDACGI), is sometimes referred to as "Cooper’s method."

In our clinic, we started lithium treatment according to Cooper’s method in 10 consecutive patients with bipolar disorder. In four of these patients, a 24-hour serum level of 0.08 mmol/liter was observed after administering the 600-mg priming dose. According to the nomogram, these patients should have received a dose of 2700 mg/day of lithium. To avoid toxic side effects, we decided to start with a lower dose than suggested by the nomogram. Eventually, doses of lithium between 1200 and 2000 mg/day proved to render adequate serum levels in these patients (0.5–0.8 mmol/liter).

This observation led to careful review of the reliability of the laboratory test procedures we used for the determination of serum lithium levels. The results of flame atomic emission spectroscopy (R15601CHDEAHIG), which is our routine procedure, and the dry chemistry method in a Vitros 950 analyzer (formerly known as Kodak 950) were both in agreement with the results of the atomic absorption reference method (R15601CHDCFEAJ). Therefore, we concluded that the observed 0.08 mmol/liter after 24 hours was not due to incorrect measurement.

Another approach to the determination of the proper lithium dose is use of the Bayesian technique (R15601CHDEABAF) to calculate patients’ individual pharmacokinetic parameters. The results of these analyses in our patients showed that their individual serum values deviated less than one standard deviation from the expected population values.

On the basis of these findings, we conclude that the use of Cooper’s method (R15601CHDDACGI) may lead to potentially toxic lithium dose recommendations in a substantial proportion of patients. Especially when very low serum levels are being measured 24 hours after giving the 600-mg priming dose, Cooper’s method is probably not a safe method to predict the appropriate daily dose of lithium in patients with bipolar disorder.