#2333 Risk of NDI and CKD in individuals treated with lithium

Abstract

Abstract Background and Aims Nephrogenic diabetes insipidus (NDI) is a known side effect of lithium treatment. Lithium has also been associated with the development of chronic kidney disease (CKD). The aims of this study were to examine the risk factors for NDI among persons on lithium treatment and the association of NDI and other risk factors with the development of chronic kidney disease (CKD). Method This was a retrospective cohort study of all persons in Iceland using lithium in the years 2003-2018. Lithium exposure was defined as at least one lithium prescription or at least one serum lithium measurement with a detectable lithium level. Patients with affective disorders (ICD-10 codes F30-F39) attending the outpatient clinics of Landspitali–The National University Hospital Mental Health Services in 2014-2016, without lithium exposure, served as controls. NDI was defined as concomitant serum sodium (SNa) >144 mmol/L and urine specific gravity (SG) <1.005; individuals with no SNa measurement after 2003 were excluded from the analysis. CKD stages 3-5 were defined according to the KDIGO guidelines for CKD as estimated glomerular filtration rate (eGFR) less than 60 mL/min/1.73 m2. The eGFR was calculated from serum creatinine (SCr) using the CKD-EPI equation. The analysis of incident CKD was limited to the period when standardized SCr measurements had become available, in 2008-2018. Individuals with <2 SCr measurements after 2008 and those with CKD stages 3-5 before 2008 were excluded. In addition, this analysis specifically examined the relationship between the mean serum lithium concentration and the development of CKD, comparing those not exposed to lithium to 3 groups of lithium users with a mean serum lithium concentration of 0.3-0.59, 0.6-0.79 and 0.8-0.99 mmol/L. Acute kidney injury (AKI) was defined according to the SCr component of the KDIGO criteria for AKI, and other comorbid conditions were defined based on ICD-9 and ICD-10 diagnosis codes. Multivariable Cox regression with time-dependent covariables was used for risk assessment. Results In the analysis of the development of NDI, the group exposed to lithium consisted of 2252 individuals, of whom 58 (2.6%) met the criteria for NDI compared with 13 (1.0%) of the 1272 controls. When compared with those without lithium exposure. the risk of NDI was lithium concentration dependent, with a hazard ratio (HR) of 1.22 (95% CI, 0.50–2.98), 5.52 (95% CI, 2.81–10.85) and 10.03 (95% CI, 4.51–22.27) for groups with a mean lithium concentration of 0.3–0.59, 0.6–0.79 and 0.8–0.99 mmol/L, respectively. A previous diagnosis of CKD was also a significant risk factor (HR 3.00 [95% CI, 1.50–6.12]). In the CKD analysis, the group of lithium users comprised 1401 individuals. Of those, 192 (13.7%) developed CKD, compared with 39 (2.8%) of the 1406 controls. Analysis adjusted for initial eGFR showed that NDI was a significant risk factor for CKD (HR 1.90 [95% CI, 1.03-3.54]). Other significant risk factors were mean lithium concentration with HR of 1.25 (95% CI, 0.82–1.91), 2.84 (95% CI, 1.94–4.15) and 5.11 (95% CI, 3.22–8.07) for the groups with a mean lithium concentration of 0.6–0.79 and 0.8–0.99 mmol/L, respectively. Other significant risk factors were age per year (HR 1.03 [95% CI, 1.02–1.05]), diabetes (HR 1.62 (95% CI, 1.08–2.46]) and history of AKI (HR 1.92 [95% CI, 1.33–2.77]). Conclusion Lithium use is associated with NDI in a concentration-dependent manner. Long-term lithium treatment is associated with a risk of CKD, which also is concentration dependent, in patients with bipolar and unipolar mood disorders. NDI may be an important risk factor for CKD.