BMC Psychiatry. 2025 May 21;25(1):514. doi: 10.1186/s12888-025-06925-x.
ABSTRACT
BACKGROUND: Suicidal ideation (SI) is a common symptom of bipolar disorder (BD). Patients with BD and suicidal ideation (BDSI) have been shown to exhibit abnormal spontaneous brain activity and homocysteine (Hcy) levels. Additionally, cognitive deficits are also considered to be a critical symptom in BD. However, the relationship among spontaneous brain activity, Hcy levels, and cognitive deficits in patients with BDSI remains unclear.
METHODS: A total of 74 participants were enrolled, comprising individuals with BDSI (n = 20), BD patients without suicidal ideation (BDNSI) (n = 24), and age-/sex-matched healthy controls (HC) (n = 30). Each participant underwent cognitive performance assessments, and blood samples were collected to measure Hcy levels. We then calculated the amplitude of low-frequency fluctuation (ALFF) from resting-state functional magnetic resonance imaging data. Mediated-effects analysis was conducted to explore the association among these three variables.
RESULTS: Hcy levels were significantly higher in the BDNSI group than in the BDSI group (t = 2.33, P = 0.024). Specifically, a significant positive correlation was observed between Hcy levels and the fractional amplitude of low-frequency fluctuation (fALFF) signals in the left posterior cingulate gyrus in the BDSI group (r = 0.644, P = 0.005). Mediation analyses revealed that the left posterior cingulate gyrus significantly mediated the negative relationship between Hcy levels and both visual learning /verbal learning performance (95% confidence intervals for the indirect effects ranging from [Formula: see text]0.592 to [Formula: see text]0.069 and [Formula: see text]0.465 to [Formula: see text]0.042, respectively) in the BDSI group.
CONCLUSIONS: Our data suggest that patients with BDSI and BDNSI may exhibit distinct Hcy-neurocognitive-brain function profiles, which could be further verified by investigating the underlying pathophysiological mechanism of BDSI.
PMID:40399851 | DOI:10.1186/s12888-025-06925-x
AI-Assisted Evidence Search
Share Evidence Blueprint
Search Google Scholar