成人惊厥性癫痫持续状态持续时间与临床结局之间的关联：一项系统评价。
完整摘要 · 中文译文
目的：癫痫持续状态（SE）的持续时间被广泛认为是决定结局严重程度的关键因素，但其预后价值难以解读，因为各研究对发作起始、终止和复发的定义以及分析处理方式存在显著差异。我们综述了其与死亡率、功能结局（运动和/或认知障碍）及难治性的关联。 方法：我们检索了MEDLINE（通过PubMed）、Cochrane CENTRAL和Scopus，纳入探讨成人SE持续时间与临床结局关系的研究。我们提取了研究设计、SE表型、持续时间定义、分析策略和时间阈值。使用QUIPS工具评估偏倚风险。结局特异性关联采用叙述性综合。 结果：共纳入43项研究（1994-2025年）。SE持续时间的定义和测量方法以及分析方式（连续变量 vs. 分类变量，阈值异质）差异很大。31项研究评估了死亡率，26项评估了功能结局，7项评估了难治性。较长的持续时间与功能结局不良的关联比与死亡率的关联更为一致且独立；死亡率信号较弱可能反映了病因、病前脆弱性和全身严重程度等竞争性决定因素。关于难治性的证据有限，但与长时间发作和治疗抵抗之间的关联相一致。 结论：SE持续时间应被解释为一种依赖于背景的预后标志物，而非绝对阈值。其意义 似乎受到特定结局、电临床表型、病因和诊断方法的影响。标准化的发作起始与终止标准、明确的复发规则以及统一建模将提高跨研究可比性并加强预后评估。
完整摘要 · 英文原文
OBJECTIVE: Duration of status epilepticus (SE) is widely considered a key determinant of outcome severity, but its prognostic value is hard to interpret because definitions of seizure onset, termination, and recurrence, as well as analytical handling, differ substantially across studies. We reviewed its association with mortality, functional outcome (motor and/or cognitive impairment), and refractoriness in adults. METHODS: We searched MEDLINE (via PubMed), Cochrane CENTRAL, and Scopus for studies examining SE duration in relation to clinical outcomes in adults. We extracted study design, SE phenotype, duration definitions, analytical strategy, and temporal thresholds. Risk of bias was assessed with the QUIPS tool. Outcome-specific associations were synthesized narratively. RESULTS: Forty-three studies (1994-2025) were included. The definitions and measurement of SE duration, as well as analytical approaches (continuous vs. categorical, with heterogeneous thresholds), varied widely. Mortality was assessed in 31 studies, functional outcome in 26, and refractoriness in 7. Longer duration was more consistently and independently associated with worse functional outcome than with mortality; the weaker mortality signal likely reflects competing determinants such as etiology, premorbid vulnerability, and systemic severity. The evidence on refractoriness was limited but consistent with a link between prolonged seizures and treatment resistance. CONCLUSIONS: SE duration should be interpreted as a context-dependent prognostic marker rather than as an absolute threshold. Its meaning appears to be shaped by the specific outcome, electroclinical phenotype, etiology, and diagnostic approach. Standardized onset and termination criteria, explicit recurrence rules, and harmonized modeling would improve cross-study comparability and strengthen prognostic assessment.