文献速递糖尿病

GLP-1受体激动剂的神经精神结局:医保2型糖尿病人群研究

Neuropsychiatric Outcomes Associated With GLP‐1 Receptor Agonist Use in Medicare Beneficiaries With Type 2 Diabetes: Variation by Treatment Pattern and Obesity Status
RSS 订阅 · Diabetes, Obesity and Metabolism · 2026-10-26

本研究基于美国Medicare 2型糖尿病患者的理赔数据,比较GLP-1受体激动剂与其他二线降糖药的神经精神结局。结果显示GLP-1RA使用与总体精神障碍风险轻微升高相关,但临床意义不确定。抑郁症风险升高,物质使用障碍、痴呆和自杀/自伤风险降低,且关联随治疗模式和肥胖状态不同而变化。

要点

  • GLP-1RA总体精神障碍风险轻微升高,临床意义不确定
  • 抑郁症风险升高,物质使用障碍、痴呆和自杀/自伤风险降低
  • 单药治疗和升级治疗模式下的风险升高可能反映基线差异

核心结论

  • GLP-1RA的神经精神影响并非统一的类别效应,而是随治疗模式、肥胖状态和结局类型不同而变化的语境依赖性表现。
  • 单药治疗或升级治疗时精神风险升高可能反映基线临床差异,而非药理作用。
  • 物质使用障碍、痴呆和自杀/自伤风险降低具有临床安抚意义。

研究设计与样本

  • 基于人群的队列研究,使用美国Medicare参保2型糖尿病患者的理赔数据。
  • GLP-1RA与其他二线降糖药物进行比较,评估四种治疗模式:单药治疗、换药、升级和联合加强。
  • 采用1:1倾向评分匹配、协变量校正的分层Cox回归,含时变协变量,共分析八种神经精神结局。

主要发现

  • GLP-1RA使用与总体精神障碍风险轻微升高相关(HR=1.04;95%CI:1.03–1.06;每1000人年多1.7例事件)。
  • 抑郁症风险升高(HR=1.11;每1000人年多3.0例事件)。
  • 物质使用障碍风险降低(HR=0.84;每1000人年少4.1例事件)。
  • 痴呆风险降低(HR=0.87;每1000人年少2.8例事件)。
  • 自杀/自伤风险降低(HR=0.81;每1000人年少0.4例事件)。

关键数值与推荐

  • 单药治疗模式显示精神障碍风险升高(HR=1.43)和情绪症状风险升高(HR=1.32)。
  • 升级治疗模式也表现出精神障碍风险升高。
  • 换药和联合加强治疗模式显示风险降低或无显著关联。

临床怎么用

  • 在启动GLP-1RA治疗时,应结合具体治疗模式和肥胖状态评估神经精神风险,尤其是单药或升级治疗场景。
  • 对于有物质使用障碍、痴呆或自杀风险的患者,GLP-1RA可能并非禁忌。

局限性

  • 基于理赔数据的研究可能存在残余混杂和基线临床差异。
  • 总体精神障碍风险升高幅度较小,在大型队列中的临床意义不确定。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

原文

ABSTRACT Aims Glucagon-like peptide-1 (GLP-1) receptors are expressed in brain regions governing mood, reward and stress, yet the neuropsychiatric effects of GLP-1 receptor agonist (GLP-1 RA) therapy remain poorly characterized. We estimated associations between GLP-1 RA use and neuropsychiatric disorders and variation by obesity status and treatment pattern. Material and Methods This population-based cohort study used claims data from Medicare enrolees with Type 2 diabetes mellitus. GLP-1 RAs were compared with other second-line glucose-lowering medications and evaluated across four treatment patterns: monotherapy, switch, escalation and augmentation. Eight neuropsychiatric outcomes were examined using 1:1 propensity score-matched, covariate-adjusted stratified Cox regression with time-varying covariates. Results GLP-1 RA use was associated with a small increase in aggregate psychiatric disorder risk (Hazard Ratio [HR] = 1.04; 95% Confidence Interval: 1.03–1.06; 1.7 excess events/1000 person-years), of uncertain clinical significance given the large cohort size. Risk was elevated for depression (HR = 1.11; 3.0 excess events/1000 person-years) but reduced for substance use disorders (HR = 0.84; 4.1 fewer events/1000 person-years), dementia (HR = 0.87; 2.8 fewer) and suicidal/self-harm (HR = 0.81; 0.4 fewer events/1000 person-years). Associations varied by treatment pattern: monotherapy showed elevated psychiatric (HR = 1.43) and elevated emotional symptom (HR = 1.32) risks; escalation also exhibited elevated psychiatric risk, whereas switch and augmentation demonstrated reduced or null risks. Conclusions Rather than a uniform class effect, the neuropsychiatric profile of GLP-1 RA appears context-dependent, differing by treatment pattern, obesity status and outcome examined. Elevated risk with monotherapy or escalation may reflect baseline clinical differences rather than pharmacological effect. Reduced risks of substance use disorders, dementia and suicidal/self-harm are reassuring.

来源:Diabetes, Obesity and Metabolism,查看原文。版权归原作者所有。

引用

  1. Diabetes, Obesity and Metabolism:Neuropsychiatric Outcomes Associated With GLP‐1 Receptor Agonist Use in Medicare Beneficiaries With Type 2 Diabetes: Variation by Treatment Pattern and Obesity Status

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。