HealthDay News — A model comprising 6 clinical variables could be used to guide lithium dosage, according to a study published in the June issue of The Lancet Psychiatry.

Vincent Millischer, MD, from the Karolinska Institutet in Stockholm, and colleagues conducted a population pharmacokinetic study involving 2 clinical Swedish cohorts with 2357 patients who received a lithium dose and had clinically relevant plasma concentrations of lithium. The natural logarithm of total body clearance of lithium (CLLi) associated with clinical variables was examined.

The researchers found that clinical predictors of CLLi included age, sex, estimated glomerular filtration rate (eGFR), comedication with diuretics, and agents acting on the renin-aldosterone-angiotensin system. There was an association seen between CLLi and serum lithium, with lower CLLi associated with higher serum lithium. In a genome-wide association study, one locus (rs583503) was associated with change in CLLi. Enrichment of the associations with genes expressed in the medulla and cortex of the kidney was also seen, as well as associations with polygenic risk score for eGFR, body mass index, and blood urea nitrogen. A model including six clinical predictors explained 61.4 and 49.8% of the variance in CLLi in cohorts 1 and 2, respectively. No major improvement of the models was seen with addition of genetic markers.

“Replacing the current time-consuming empirical titration with dose prediction models based on individual clinical data might decrease the time needed to reach therapeutic lithium concentrations, increasing patient compliance and therefore improving clinical care for patients with bipolar disorder,” the authors write.

Several authors disclosed financial ties to the pharmaceutical industry.

Abstract/Full Text (subscription or payment may be required)