形态与功能并重：多节段腰椎管狭窄症责任节段判定的影像学进展

Integrating Morphology and Function: Advances in Imaging Methods for Identifying Responsible Segments in Multilevel Lumbar Spinal Stenosis

摘要: 多节段腰椎管狭窄症的责任节段判定是脊柱外科临床决策的关键难题。常规MRI虽能清晰显示多节段解剖狭窄，但其形态学发现与临床症状之间的不匹配现象长期困扰临床实践。本文系统综述近十年来责任节段判定影像学方法的研究进展，从传统形态学评估、功能成像技术、有创诊断方法及人工智能新范式四个维度展开论述，并对所引文献的研究设计、样本量、证据等级进行批判性分析。在此基础上，本文进一步探讨各项技术在临床实践中的整合应用策略，提出基于阶梯式决策逻辑的责任节段判定流程图：以临床评估为基础，优先采用无创功能成像(DTI)进行初筛，对于结果不明确或高危病例，升级至有创诊断(CTM或SNRB)，并以人工智能辅助分析作为定量化决策支持工具。未来研究应开展前瞻性、多中心、大样本研究，建立标准化的影像采集与分析方法，并开发整合多模态信息的临床决策支持系统，以提高责任节段判定的循证医学证据等级和临床可操作性。

Abstract: Identifying the responsible segments in multilevel lumbar spinal stenosis remains a critical chal-lenge in spinal surgery clinical decision-making. Although conventional MRI can clearly demon-strate multilevel anatomical narrowing, the persistent mismatch between morphological findings and clinical symptoms has long perplexed clinical practice. This article systematically reviews the progress over the past decade in imaging methods for identifying responsible segments, addressing four dimensions: traditional morphological assessment, functional imaging techniques, invasive diagnostic procedures, and the emerging paradigm of artificial intelligence. A critical appraisal of study design, sample size, and level of evidence for the cited literature is provided. Building upon this analysis, this article further explores the integrated application strategies of these techniques in clinical practice, proposing a stepwise decision-making algorithm for responsible segment identi-fication: commencing with clinical assessment, prioritizing non-invasive functional imaging (DTI) for initial screening, escalating to invasive diagnostics (CTM or SNRB) for cases with equivocal find-ings or high-risk profiles, and utilizing artificial intelligence-assisted analysis as a quantitative deci-sion support tool. Future research should prioritize prospective, multicenter, large-sample studies, establish standardized imaging acquisition and analysis protocols, and develop clinical decision support systems integrating multimodal information to enhance both the level of evidence and clinical operability for responsible segment identification.