Serum Lithium After Initiation of Sodium-Glucose Cotransporter 2 Inhibitor Treatment

BACKGROUND: Sodium-glucose cotransporter 2 (SGLT2) inhibitors may increase renal clearance of lithium, but prior results are mixed. We investigated this aspect using data from Danish nationwide registers. MATERIALS AND METHODS: We identified all patients with bipolar disorder or unipolar depression between January 1, 2000, and June 1, 2024, who initiated SGLT2 inhibitor treatment (index date). Concurrent lithium treatment was defined by a serum lithium measurement within 90 days before the index date. Changes in serum lithium levels from the pre-index measurement to the first, second, and third post-index measurements were assessed. Furthermore, an analogue positive control analysis based on data from patients who initiated treatment with a thiazide diuretic was performed, as this class of drugs is known to increase lithium levels. RESULTS: We identified 131 patients with pre-index and post-index serum lithium measurements (ie, before and after initiation of SGLT2 inhibitor treatment). Mean serum lithium decreased from 0.59 mmol/L pre-index to 0.52 mmol/L at the first post-index measurement [difference: -0.07 (95% CI: -0.11, -0.03) mmol/L, P<0.001]. At the time of the first post-index measurement, 94 patients (71.8%) had a serum lithium level below the therapeutic index (0.6 to 0.8 mmol/L), compared with 57 patients (43.5%) at the pre-index measurement, corresponding to an absolute increase of 28.2 percentage points (95% CI: 17.3 to 39.2, P<0.001). Subsequent measurements did not differ from the pre-index measurement, likely due to compensatory increases in lithium doses. For patients initiating a thiazide diuretic, the mean pre-index serum lithium level was 0.61 mmol/L and the first post-index mean serum lithium was 0.70 mmol/L, that is, the expected increase [0.09 (95% CI: 0.05 to 0.13) mmol/L, P<0.001]. CONCLUSIONS: Initiation of SGLT2 inhibitor treatment may reduce serum lithium levels to subtherapeutic levels.