Search for surrogate markers to predict end stage kidney disease in long term lithium users

Background: A surrogate marker (a substitute indicator of the true outcome) is needed to predict subgroups of long-term lithium users at risk of end-stage kidney disease (ESKD). In this narrative review the aim is to determine the optimal surrogate endpoint for ESKD in long-term lithium users in a scientific context. Main: In a literature search in long-term lithium users, no studies on surrogate measurements on ESKD were identified. Therefore, comparable ESKD populations were sought, based on baseline eGFR, age, somatic comorbidity and sex. Articles were scored on comparability and risk of bias. Seventeen studies were included; ten studies evaluated a percentual decline (between 20 and 50% decline in eGFR) and seven studies focused upon a declining slope (from 1.63 to 6 ml/min/1,73m2 decline per year), using an interval of one to five years. Study populations mostly included patients with cardiovascular disease and chronic kidney disease. Conclusion: Currently, the most appropriate marker for ESKD in long term lithium users appears a 30% decline in eGFR in at least one year. In order to confirm this hypothesis, further research in a cohort of long-term lithium users is needed. Better feasible research on lithium induced nephropathy could result in more knowledge about the risk on kidney function decline in lithium users and guide clinical decision making on lithium use.