Cardiovascular disease (CVD) is the leading cause of premature mortality in people with bipolar disorder, making the long-term cardiovascular effects of mood-stabilizing treatment an important clinical question. Lithium remains a first-line treatment with established benefits for mood stabilization and suicide prevention, but its cardiovascular profile is complex: therapeutic exposure has been linked to anti-inflammatory and cardioprotective pathways, whereas toxic levels can cause arrhythmias and cardiomyopathy. This study evaluated whether lithium treatment was associated with incident CVD in adults with bipolar disorder.

Investigators conducted a claims-based longitudinal cohort study using the Beijing Medical Claim Data for Employees database. Adults newly diagnosed with bipolar disorder from 2010 through 2017 were included after a two-year washout period; those with CVD before the index date were excluded. Participants were classified according to whether lithium was part of their initial treatment strategy. Inverse probability weighting was used to balance baseline demographic characteristics, cardiovascular risk factors, comorbidity, health care utilization, and concomitant medications. Weighted Cox proportional hazards models estimated the risk of incident overall CVD, ischemic heart disease (IHD), and stroke.

Key Findings
The final cohort included 2945 patients (mean age, 44.5 years; 56.7% female): 1129 initiated lithium and 1816 did not. Over a mean follow-up of three years, 60 participants developed CVD, including 29 cases of IHD and 26 strokes. After weighting, lithium use was associated with a lower risk of overall CVD compared with nonuse (hazard ratio [HR], 0.38; 95% CI, 0.18-0.78; P = .009). The association was also significant for IHD (HR, 0.20; 95% CI, 0.06-0.67), but not for stroke (HR, 0.69; 95% CI, 0.27-1.75).

Hazard ratios compare lithium initiation with no lithium after inverse probability weighting. Stroke is shown in gray because the association was not statistically significant.

Subgroup analyses suggested that the lower CVD risk was most apparent in patients aged 50 years or older, males, and those with pre-existing cardiovascular risk factors. After correction for multiple comparisons, the association remained significant in participants with cardiovascular risk factors (HR, 0.27; 95% CI, 0.11-0.71). Among patients classified as adherent to lithium (proportion of days covered at least 80%), lithium use was associated with lower CVD risk (HR, 0.31; 95% CI, 0.11-0.86); the estimate in nonadherent patients was not statistically significant. Results were generally consistent after excluding follow-up shorter than one year, extending the washout period, using propensity-score matching, and comparing lithium with atypical antipsychotic treatment. Comparisons with valproate were imprecise because of small groups and few events.

The findings are biologically plausible. Lithium inhibits glycogen synthase kinase-3 beta and may influence mitochondrial survival signaling, oxidative stress, endothelial inflammation, and atherosclerotic pathways. However, the absence of a significant association with stroke suggests that any potential cardiovascular benefit may differ by outcome. The small number of cardiovascular events — particularly arrhythmia and heart failure — also limits conclusions about individual disease subtypes.

Limitations and Caveats
These results should be interpreted as an association rather than evidence that lithium prevents cardiovascular disease. The observational design cannot eliminate residual confounding, and the database lacked information on body mass index, smoking, physical activity, diet, socioeconomic status, and lithium serum concentrations. Exposure was defined by the initial prescription rather than time-varying use, mean follow-up was only three years, and the insured urban Beijing cohort may not represent other populations.

Reference:
Xu N, et al. J Affect Disord. Epub ahead of print. Abstract