Commentary on a recent review of lithium toxicity: what are its implications for clinical practice?

A recent paper by McKnight et al. in The Lancet has provided the first formal meta-analysis of the more common adverse reactions to lithium. The authors analyzed 385 studies and focused mainly on the harmful effects of lithium on the kidney, the thyroid and parathyroid glands, body weight, skin and congenital malformations. Their contribution is important and welcome, but as a guide for practice, it needs to be complemented by other relevant observations and individual patient-focused perspectives. The findings from that meta-analysis somewhat underestimate the renal side-effects, and distort to some degree or exclude other adverse effects. The glomerular filtration rate is reduced but not more than 0 to 5 ml/min/year of observation; this may not fully reflect the present state of knowledge. A quarter of patients in the study had abnormalities of the thyroid and/or parathyroid gland, and lithium was found to increase body weight significantly less than did olanzapine. Unfortunately, the authors did not consider the observations from spontaneous reporting systems, which may have changed the picture. We feel that some specific limitations of the study were related to the inclusion of patients regardless of adequacy of treatment, quality of monitoring, drug combinations, age and sex, and stabilization response.