Hyponatremia (hypoNa) is a potentially serious adverse event of treatment with antidepressants. Previous research suggests that risk of drug-induced hyponatremia differs between antidepressants.

This meta-analysis sought to determine the risk of antidepressant-induced hypoNa, stratified by different compounds and classes.

PubMed and Web of Science were searched for studies reporting on incidence or risk of hypoNa in adults using antidepressants (PROSPERO, CRD42021269801). We modelled random-effects meta-analyses to compute overall incidence and risk of any and clinically relevant hypoNa for each compound and class, and ran head-to-head comparisons based on hypoNa incidences. We conducted subgroup analyses for geriatric populations, study context and sodium cut-off value.

Thirty-nine studies (n = 8,459,033) revealed that exposure to antidepressants was associated with significantly increased odds of hypoNa (OR = 2.82 (1.79 – 4.45)). The highest event rates were found for SNRIs (7.17%), SSRIs (5.20%), and TCAs (2.26%); the lowest for mirtazapine (1.02%) and trazodone (0.89%). The highest odds ratios were found for MAOIs (4.12 (1.92 – 8.86)), SNRIs (3.16 (1.77 – 5.67)), and SSRIs (2.78 (1.57 – 4.91)); the lowest for mirtazapine (2.82 (1.87 – 4.21)) and TCAs (1.85 (1.28 – 2.69)). Compared to SSRIs, SNRIs were significantly more likely (OR = 1.27 (1.13 – 1.42), p < 0.001) and mirtazapine significantly less likely (OR = 0.61 (0.39 – 0.96), p = 0.032) associated with hypoNa.

Our meta-analysis demonstrated that, while no antidepressant can be considered completely risk-free, for hypoNa-prone patients mirtazapine should be considered the treatment of choice and SNRIs should be prescribed more cautiously than SSRIs and TCAs.

None Declared