Clinical utility of salivary lithium concentration.

Saliva, plasma and erythrocyte lithium concentrations were studied in 40 long-term lithium users. Intersubject variation in saliva to plasma lithium ratio was too great for clinical utility, when based on a group linear regression equation. By contrast, intrasubject data from 8 patients studied on three or more occasions indicated a higher linear correlation (r = 0.91 to 1.00 for 7 of 8 patients). For individual patients the ratio remained stable over varying concentration ranges, and was not affected by time from last dose or length of use of lithium. Based on these data, use of saliva concentrations of lithium for routine monitoring appears feasible after several blood and saliva concentrations are obtained to establish a relationship for a particular patient. Saliva monitoring may facilitate accurate analysis and convenience for some patients.