The Lithium Calculator

 weight: lbs kg age: serum Li+: serum Cr: male: female: inpatient: TCA: calculate  |  reset Yukawa: Zetin: mg/day lithium carbonate

Text-only version of this document for saving.  Windows users:  change the .htm extension to .hta to simplify the display.

Run this document as a smart phone app!  Specific instructions on how to do it on Android here

A number of methods of predicting lithium dosage have been published.  One of the simplest is that of Zetin and colleagues:

lithium carbonate, mg/day = 486.8 + (746.83 * level) - (10.08 * age) + (5.95 * weight) +
(147.8 * sex) + (92.01 * inpatient) - (74.73 * TCA)

where level is the target serum lithium concentration in meq/L; age is in years; weight is in kg; sex is 1 for male, 0 for female; inpatient is 1 for yes, 0 for no; and TCA is 1 for concurrent tricyclic antidepressant use, 0 for not.

This method does not account for the serum creatinine level, and was found by Wright and Crismon to often overpredict the lithium dose.  Several other methods do account for serum creatinine; that of Yukawa and colleagues was found by Grasmäder to be highly accurate:

lithium carbonate, meq/day = target serum lithium concentration, meq/L * lithium clearance, L/day

Lithium clearance is calculated as:

31.6 + ((0.225 * weight) - 7.79) / Cr

or, for age > 50:

31.6 + ((0.225 * weight) - 7.79) / (Cr - (0.634 * (age - 50)))

where Cr is serum creatinine in mg/dL.  1meq lithium = 36.95mg lithium carbonate.

The script above calculates a daily lithium carbonate dose using both the Yukawa and Zetin methods.  If serum creatinine is not known, only the Zetin value will be calculated.  Please note that these figures are hypothetical and do not recommend a lithium dose for any individual patient.  Many factors influence treatment decisions and clinical judgment is always required.

References:

Grasmäder K.  Pharmacokinetics of antidepressants and lithium:  variability and clinical implication for individual dose adaptation.
hss.ulb.uni-bonn.de:90/ulb_bonn/diss_online/math_nat_fak/2003/grasmaeder_katja/index.htm

Jermain DM, Crismon ML, Martin ES (1991).  Population pharmacokinetics of lithium.  Clin Pharm 10:376-381

Lesar TS, Tollefson GK, Koch M (1985).  Relationship between patient variables and lithium dosage requirements.  J Clin Psychiatry 46:133-36

Pepin SM, Baker DE, Nance KS et al.  Lithium dosage calculation from age, sex, height, weight and serum creatinine.  15th Annual Midyear Clinical Meeting, American Society of Hospital Pharmacists, San Francisco, 9 December 1980

Wright R, Crismon ML (2000).  Comparison of three a priori methods and one empirical method in predicting lithium dosage requirements.  Am J Health-Syst Pharm 57(18):1698-1702
www.medscape.com/viewarticle/406913_print

Yukawa E, Nomiyama N, Higuchi S, Aoyama T (1993).  Lithium population pharmacokinetics from routine clinical data:  role of patient characteristics for estimating dosing regimens.  Ther Drug Monit 15:75-82

Zetin M, Garber G, De Antonio M et al. (1983).  Prediction of lithium dose:  a mathematical alternative to the test dose method.  J Clin Psychiatry 44:144-145

Script © 2006 by Russell W. Cottrell, M.D.