Background: Lithium remains an important treatment for bipolar disorder; however, whether elevated lithium levels lead to long-term renal problems is unknown. Previous consensus opinion was that levels should be kept below 0.6 mmol/L in geriatric patients to minimize renal toxicity.
Objective: We hypothesized that elevated serum lithium levels correlate with decreased renal function [estimated glomerular filtration rate (eGFR)] in geriatric psychiatry outpatients.
Study design: This was a 4-year retrospective cohort study (2007-2011).
Setting: We performed this study in three Canadian university-affiliated tertiary care clinics.
Patients: Data from 42 lithium-using geriatric psychiatry outpatients was used. INTERVENTION/EXPOSURE: Our main exposure of interest was mean serum lithium level between 2007 and 2011.
Main outcome measure: Our primary outcome was change in eGFR between 2007 and 2011 (hypothesis formulated before data collection).
Results: Lithium levels did not correlate significantly with change in eGFR at 2- or 4-year follow-up (r < 0.12, p > 0.57). There were no significant predictors of change in eGFR in a multiple linear regression model including hypertension, diabetes, baseline eGFR, lithium duration, and lithium levels.
Conclusion: Lithium levels do not correlate strongly (ρ > 0.5) with decreased eGFR at 2- and 4-year follow-up in geriatric outpatients. These results are not reliably generalizable when treating patients at mean lithium levels greater than 0.8 mEq/L, especially at 4-year follow-up, and larger studies will be necessary to examine the possibility of a smaller correlation. Nonetheless, these data and the existing literature suggest that lithium levels up to 0.8 mmol/L are safe for use in the long-term treatment of geriatric patients with mood disorders without pre-existing chronic renal failure.