Twelve studies (n=539 participants) were included in the review, comprising six studies of 213 healthy volunteers and six studies of 326 patients with affective disorders. Eight studies used double-blind designs. Nine studies compared cognitive performance of individuals who received lithium with the cognitive performance of the same individuals when not taking lithium.
The meta-analysis of both healthy volunteers and patients with affective disorder showed small, but statistically significant impairments with the use of lithium in immediate verbal learning and memory (ES 0.24, 95% CI 0.05 to 0.43; 10 studies) and creativity (ES 0.33, 95% CI 0.02 to 0.64; three studies). There were no significant differences between participants who received lithium and those not receiving lithium in delayed verbal memory, immediate and delayed visual memory, attention, processing speed, executive function or psychomotor performance. There was evidence of statistical heterogeneity for immediate and delayed memory, processing speed and psychomotor performance.
For the analysis of patients with affective disorders, statistically significant impairments were observed with the use of lithium in immediate verbal learning and memory (ES 0.29, 95% CI 0.07 to 0.51; six studies), creativity (ES 0.34, 95% CI 0.00 to 0.68; two studies) and psychomotor performance (ES 0.62, 95% CI 0.27 to 0.97; two studies). There was no evidence of statistical heterogeneity.
There were no significant differences between healthy volunteers who received lithium and those who received a placebo for any of the cognitive domains examined, although there was some evidence of statistical heterogeneity for processing speed.
The authors stated that visual appraisal of the funnel plots showed no evidence of publication bias.