多内分泌代谢性卵巢综合征的心理代谢内表型提议：心理困扰、进食功能障碍、睡眠紊乱、慢性应激与代谢恶化之间的双向相互作用

文献信息

中文摘要

背景： 多囊卵巢综合征于2026年更名为多内分泌代谢性卵巢综合征（PMOS），以更好地体现其多系统生物学特征。心理困扰、进食失调、睡眠障碍和代谢功能障碍常同时存在，但临床诊疗和研究通常将它们视为平行的共病。
目的： 对提议的心理代谢构念进行操作化定义；将其相对于传统表型和数据驱动聚类进行定位；并对心理、行为、睡眠-应激、神经内分泌和代谢通路之间的关联、时间预测、因果关系和治疗反应证据进行分级。
方法： 我们对PubMed/MEDLINE自建库至2026年9月15日的文献进行了有针对性的叙述性综述。七个结构化主题模块返回1,278条记录，经基于PubMed标识符（PMID）的去重后得到1,009个唯一标识符，以及1,008条可获取的文章记录用于标题/摘要筛选。优先纳入指南、系统综述、纵向或因果研究、试验以及有信息价值的人体机制研究；一般人群和临床前证据仅在其间接性被明确说明时使用。
证据综合： 抑郁和焦虑、体像担忧、暴食和贪食谱症状、睡眠不佳以及阻塞性睡眠呼吸暂停在PMOS中的报告频率高于对照人群，但大多数估计为横断面且异质性大。一项三级队列研究发现，基线抑郁和/或焦虑史可预测中位7年内的新发代谢综合征（校正风险比1.45，95%置信区间1.02-2.06）；这支持预后富集，而非因果关系。一般人群的纵向和遗传学研究支持抑郁-代谢的双向关系，而PMOS特异性因果证据不一致。可能的联系包括胰岛素-雄激素-脂肪相互作用、神经活性类固醇信号、卵巢交感输入、睡眠碎片化、应激与自主神经调节、奖赏相关进食、治疗暴露、污名和社会背景。尚无证据验证一个离散的内表型。
结论： 我们将候选心理代谢内表型定义为一种跨领域的、机制上相互耦合的模式，可能出现在任何传统PMOS表型中。临床共现仅是入口特征；内表型地位需要可重复的纵向耦合、增量预后价值、可解释的转变，最好还有治疗效应异质性。四个重叠的原型是工作性表述，而非亚型。一个六步临床路径以共享表述为中心，并在治疗选择中嵌套照护强度层级。

英文摘要

BACKGROUND: Polycystic ovary syndrome was renamed polyendocrine metabolic ovarian syndrome (PMOS) in 2026 to better represent its multisystem biology. Psychological distress, eating dysregulation, sleep disturbance, and metabolic dysfunction frequently coexist, yet clinical care and research commonly treat them as parallel comorbidities.
OBJECTIVE: To define the proposed psychometabolic construct operationally; position it relative to conventional phenotypes and data-driven clusters; and grade evidence for association, temporal prediction, causation, and treatment response across psychological, behavioral, sleep-stress, neuroendocrine, and metabolic pathways.
METHODS: We conducted a targeted narrative review of PubMed/MEDLINE from inception through 15 September 2026. Seven structured topic blocks returned 1,278 records, 1,009 unique identifiers after PubMed identifier (PMID)-based deduplication, and 1,008 retrievable article records for title/abstract screening. Guidelines, systematic reviews, longitudinal or causal studies, trials, and informative human mechanistic studies were prioritized; general-population and preclinical evidence was used only when its indirectness was explicit.
EVIDENCE SYNTHESIS: Depression and anxiety, body-image concerns, binge- and bulimia-spectrum symptoms, poor sleep, and obstructive sleep apnea are reported more often in PMOS than in comparison populations, but most estimates are cross-sectional and heterogeneous. One tertiary cohort found that baseline depression and/or anxiety history predicted incident metabolic syndrome over a median seven years (adjusted hazard ratio 1.45, 95% confidence interval 1.02-2.06); this supports prognostic enrichment, not causality. General-population longitudinal and genetic studies support reciprocal depression-metabolic relationships, whereas PMOS-specific causal evidence is inconsistent. Plausible links include insulin-androgen-adipose interactions, neuroactive steroid signaling, ovarian sympathetic input, sleep fragmentation, stress and autonomic regulation, reward-related eating, treatment exposure, stigma, and social context. None validates a discrete endotype.
CONCLUSIONS: We define a candidate psychometabolic endotype as a cross-cutting, mechanistically coupled pattern that may occur within any conventional PMOS phenotype. Clinical co-occurrence is only an entry profile; endotype status requires replicated longitudinal coupling, incremental prognostic value, interpretable transitions, and preferably treatment-effect heterogeneity. Four overlapping prototypes are working formulations rather than subtypes. A six-step clinical pathway centers shared formulation and nests care-intensity tiers within treatment selection.