Lifetime risk of severe kidney disease in lithium-treated patients: a retrospective study

Abstract

Abstract Background Lithium is an essential psychopharmaceutical, yet side effects and concerns about severe renal function impairment limit its usage. Aims Our objectives were to quantify the occurrence of chronic kidney disease stage 4 or higher (CKD4 +) within a lithium-treated population, using age- and time-specific cumulative incidence and age-specific lifetime risk as measures of disease occurrence. Additionally, we aimed to investigate the association between the duration of lithium treatment and the risk of CKD4 + . Methods We identified patients from the Sahlgrenska University Hospital’s laboratory database. We conducted a retrospective cohort study employing cumulative incidence functions that account for competing deaths to estimate cumulative and lifetime risk of CKD4 + . A subdistribution hazards model was employed to explore baseline covariates. For measuring the association between the duration of lithium treatment and CKD4 + occurrence, we used a matched 1:4 case–control study design and logistic regression. Results Considering a 90-year lifetime horizon, the lifetime risk of CKD4 + for patients initiating lithium treatment between ages 55 and 74 ranged from 13.9% to 18.6%. In contrast, the oldest patient group, those starting lithium at 75 years or older, had a lower lifetime risk of 5.4%. The 10-year cumulative risk for patients starting lithium between ages 18 and 54 was minimal, ranging from 0% to 0.7%. Pre-treatment creatinine level was a predictive factor, with a hazard ratio of 4.6 (95% CI 2.75–7.68) for values within the upper third of the reference range compared to the lower third. Moreover, twenty or more years of lithium exposure showed a strong association with an increased risk of CKD4 + compared to 1–5 years of lithium use, with an odds ratio of 6.14 (95% CI 2.65–14.26). Conclusions The risk of CKD4 + among lithium-treated patients exhibited significant age-related differences. Patients under 55 years old had negligible 10-year risk, while the lifetime risk for those aged 75 and older was limited. Duration of lithium treatment, especially exceeding 20 years, emerged as a significant risk factor. For individual risk assessment and prediction, consideration of age, pre-treatment creatinine levels, and the chosen time horizon for prediction is essential.