Abstract:Abstract BACKGROUND AND AIMS Lithium is an established treatment for bipolar disorder and treatment-resistant depression. Despite awareness of potential kidney damage, there is a lack of research evidence to inform on the existence and magnitude of the risk. Observational studies to date show conflicting findings, possibly explained by inadequate control populations, prevalent user bias and a lack of information on kidney function or serum lithium levels. METHOD We conducted a cohort study to compare kidney outcomes in adults who started lithium or valproate therapy in Stockholm, Sweden, during 2007–18. Within lithium users, we also compared outcomes by average serum lithium concentrations during the first year of therapy. Kidney outcomes were CKD progression (composite of >30% eGFR decline and kidney failure) and AKI (by diagnosis or KDIGO transient creatinine elevations). Propensity score weighted Cox regression was used to estimate hazard ratios with [95% confidence intervals (95% CI)] and balance 46 identified confounders. Complete collection of repeated dispensations at Swedish pharmacies allowed modelling of the time-dependent risk associated with cumulative lithium exposure. Sensitivity analyses included restricting to patients with a diagnosis of bipolar disorder, apply a 1-year lag, use of alternative weighting methods and evaluation of eGFR monitoring rates to ascertain surveillance bias between groups. RESULTS We included 16 645 individuals, of whom 5308 initiated lithium and 5638 valproate therapy. Their median age was 45 years (57% women) and median eGFR was 99 mL/min/1.73 m2. A total of 179 CKD progression events and 234 AKI were identified during a median of follow-up of 4.3 and 4.2 years, respectively. After propensity score weighting, the adjusted hazard ratio for the risk of CKD progression was 1.12 (95% CI 0.86–1.47) and for AKI 0.88 (95% CI 0.7–1.09). There was a weak, non-statistically significant association between cumulative exposure to lithium and the risk of CKD progression that was not observed for cumulative use of valproate. Results were consistent among patients with a bipolar disorder diagnosis and robust to 1-year lag or alternative weighing methods. A total of 3913 lithium users were on therapy for at least one year and underwent routine serum lithium monitoring. Compared with patients with average serum lithium levels <1.0 mEq/L, those with serum lithium ≥1.0 mEq/L (n = 270) were at a higher risk of both CKD progression (HR: 2.18; 1.28–3.71) and AKI (HR: 1.37; 0.81–2.34). CONCLUSION In this analysis of patients from routine clinical practice, the risk of kidney outcomes associated with lithium therapy did not differ from that of valproate. Modest risk magnitudes need to be offset with the effectiveness and anti-suicidal benefits of lithium. However, elevated serum lithium levels strongly predicting kidney risk, emphasizing the need for close monitoring and lithium dose-adjustment.

Dose response relationship between lithium serum levels during pregnancy and birth outcomes

Obstetric Outcomes in Women on Lithium: A Systematic Review and Meta-Analysis

Lithium Use During Pregnancy and the Risk of Miscarriage.

Lithium Use During Pregnancy in 14 Countries

Retrospective Review of Postpartum Lithium Use Including During Lactation

The Dilemma of Lithium Discontinuation in Bipolar Disorder During Pregnancy Planning: A Case Report and Literature Review

A Case of Severe Polyhydramnios During Pregnancy Associated with Long-Term Use of Lithium

Epidemiology studies on effects of lithium salts in pregnancy are confounded by the inability to control for other potentially teratogenic factors

A case of pregnancy with severe polyhydramnios related to long‐term use of lithium

Lithium from breast-milk inhibits thyroid iodine uptake and hormone production, which are remedied by maternal iodine supplementation

MO502: Use of Lithium, Valproate and the Risk of Acute or Chronic Kidney Disease: An Observational Study From Routine Care Data

Lithium in Breast Milk Transiently Affects the Renal Electrolytic Balance of Infants

Lithium and its effects: does dose matter?

Twenty-Three Years of Declining Lithium Use: Analysis of a Pharmacoepidemiological Dataset from German-Speaking Countries

Folic acid supplementation improved cognitive deficits associated with lithium administration during pregnancy in rat offspring

The identification of biomarkers predicting acute and maintenance lithium treatment response in bipolar disorder: A plea for further research attention

Long-term lithium treatment in bipolar disorder: effects on glomerular filtration rate and other metabolic parameters

Retrospective analysis of lithium treatment: examination of blood levels

Exploring the clinical factors affecting lithium dose and plasma level and the effect of brand

International Trends in Lithium Use for Pharmacotherapy and Clinical Correlates in Bipolar Disorder: A Scoping Review

Risk of adverse pregnancy, delivery and neonatal outcomes associated with bipolar disorder and prenatal use of mood stabilizers: A population-based cohort study