Auxilia Anna Maria, Olié Emilie, Dubois Jonathan, Capuzzi Enrico, Aubin Valérie, Aouizerate Bruno, Bellivier Frank, Belzeaux Raoul, Dubertret Caroline, Januel Dominique, Haffen Emmanuel, Lefrere Antoine, Pelletier Agnès, Polosan Mircea, Rey Roman, Roux Paul, Samalin Ludovic, Schwan Raymund, Walter Michel, Yrondi Antoine, Llorca Pierre Michel, Leboyer Marion, Etain Bruno, Courtet Philippe
Department of Emergency Psychiatry and Acute Care, CHU Montpellier, IGF, Univ. Montpellier, CNRS, INSERM, Montpellier, France.
Fondation Fondamental, Créteil, France.
Eur Psychiatry. 2025 Jul 22;68(1):e117. doi: 10.1192/j.eurpsy.2025.10068.
Suicidal behaviors (SB) in bipolar disorder (BD) are major adverse outcomes that may influence disease progression. While staging models exist for psychiatric disorders, none include suicide. This study aims to stratify suicidal risk in BD, propose a staging model for SB, and assess its clinical utility.
Participants from the FondaMental Advanced Centers of Expertise for Bipolar Disorder (FACE-BD) cohort were categorized into five stages (St) based on SB: St0 (no suicidal ideation [SI]), St1 (SI but no suicide attempt [SA]), St2a (non-severe/violent SA), St2b (severe /violent SA), and St3 (multiple SAs). Stages were analyzed based on demographic, clinical, cognitive, and biological characteristics using logistic regression.
Key differences emerged between stages. St1 showed longer untreated illness and higher lability and lower functioning than St0. St2a was linked to anxiety, substance use disorders, and longer disorder duration, while male gender and lithium bitherapy were protective. St2b exhibited shorter untreated illness and higher childhood trauma (CTQ) scores, with male gender and alcohol use as risk factors. St3 was associated with BD-II, alcohol use, longer disorder duration, and more depressive episodes, but less anxiety. No biochemical or cognitive differences were found across stages. The model was significantly associated with SA occurrence (LRT = 28.74, < 0.0001).
This staging model for suicidality in BD provides a multifaceted approach to risk stratification and predictive insights, although further refinement is needed.
双相情感障碍（BD）中的自杀行为（SB）是可能影响疾病进展的主要不良后果。虽然存在针对精神疾病的分期模型，但均未纳入自杀因素。本研究旨在对双相情感障碍中的自杀风险进行分层，提出自杀行为的分期模型，并评估其临床实用性。
来自双相情感障碍基础高级专业中心（FACE-BD）队列的参与者根据自杀行为被分为五个阶段（St）：St0（无自杀观念[SI]），St1（有自杀观念但无自杀未遂[SA]），St2a（非严重/暴力自杀未遂），St2b（严重/暴力自杀未遂），以及St3（多次自杀未遂）。使用逻辑回归基于人口统计学、临床、认知和生物学特征对各阶段进行分析。
各阶段之间出现了关键差异。与St0相比，St1显示出未治疗疾病时间更长、情绪更不稳定且功能更低。St2a与焦虑、物质使用障碍以及疾病持续时间更长有关，而男性性别和锂盐联合治疗具有保护作用。St2b表现出未治疗疾病时间较短且儿童创伤（CTQ）评分较高，男性性别和饮酒是危险因素。St3与双相II型障碍、饮酒、疾病持续时间更长以及更多抑郁发作相关，但焦虑较少。各阶段未发现生化或认知差异。该模型与自杀未遂的发生显著相关（似然比检验=LRT = 28.74，P < 0.0001）。
双相情感障碍自杀行为的这一分期模型提供了一种多方面的风险分层方法和预测见解，尽管还需要进一步完善。
