Prevalence of impaired kidney function in patients with long‐term lithium treatment: A systematic review and meta‐analysis

OBJECTIVES: Although lithium renal effects have been extensively investigated, prevalence rates of chronic kidney disease (CKD) in lithium-treated patients vary. Our aim was to provide prevalence estimates and related moderators. METHODS: We performed a systematic review in PubMed/Embase until November 01, 2021, conducting a random effects meta-analysis of studies evaluating CKD prevalence rates in lithium-treated patients calculating overall prevalence ±95% confidence intervals (CIs). Meta-regression analyses included sex, age, body mass index, smoking, hypertension, diabetes, cardiovascular disease, lithium-treatment dose, duration, and blood levels. Subgroup analyses included sample size, diagnoses, and study design. Pooled odds ratios (OR) were estimated for studies including patients receiving nonlithium treatment. Study quality was assessed using the Newcastle-Ottawa scale. RESULTS: Five, nine, and six trials were rated as high, fair, and low quality, respectively. In 20 studies (n = 25,907 patients), we estimated an overall prevalence of 25.5% (95% CI = 19.8-32.2) of impaired kidney function; despite lack of differences (p = 0.18), prevalence rates were higher in elderly samples than mixed samples of elderly and nonelderly (35.6%, 95% CI = 21.4-52.9, k = 2, n = 3,161 vs. 25.1%, 95% CI = 19.1-31.3, k = 18, n = 22,746). Prevalence rates were associated with longer lithium treatment duration (p = 0.04). Cross-sectional studies provided lower rates than retrospective studies (14.5%, 95% CI = 13.5-15.5, k = 6, n = 4,758 vs. 29.5%, 95% CI = 22.1-38.0, k = 12, n = 17,988, p < 0.001). Compared with 722,529 patients receiving nonlithium treatment, the OR of impaired kidney function in 14,187 lithium-treated patients was 2.09 (95% CI = 1.24-3.51, k = 8, p = 0.005). CONCLUSIONS: One-fourth of patients receiving long-term lithium may develop impaired kidney function, although research suffers from substantial heterogeneity between studies. This risk may be twofold higher compared with nonlithium treatment and may increase for a longer lithium treatment duration.