操作者与超声 modality 依赖的视神经鞘直径超声检测创伤性脑损伤颅内压升高的诊断准确性局限:一项 Meta 回归分析
该系统评价与 Meta 分析纳入 15 篇文献,获得 17 个汇总数据集(14 个成人、3 个儿科)和 41 份诊断报告。双变量随机效应模型显示,成人诊断准确性中等至良好:灵敏度 0.88、特异度 0.76、AUC>0.85(中等确定性证据);儿科数据集准确性较低:灵敏度 0.82、特异度 0.57、AUC 0.77(低确定性证据)。Meta 回归显示超声 modality 和操作者背景是诊断异质性的显著来源:B 型超声轻度提高特异度,非专科操作者诊断准确性降低;常用 ONSD 截断值(如 5.5 mm 二分法或连续分析)与诊断性能无显著关联(p=0.61)。作者认为 ONSD 超声是成人 TBI 颅内压升高的有价值的快速无创辅助手段,但可靠性高度依赖操作者且在儿科有限,需标准化测量方案、结构化操作者培训和儿科特异性验证研究后才能更广泛临床采用。
要点
- 成人 TBI 中 ONSD 超声诊断颅内压升高灵敏度 0.88、特异度 0.76、AUC>0.85(中等确定性证据)。
- 儿科数据集诊断准确性较低:灵敏度 0.82、特异度 0.57、AUC 0.77(低确定性证据)。
- 非专科操作者进行检查与诊断准确性降低相关;B 型超声轻度提高特异度。
- 常用 ONSD 截断值(5.5 mm 二分法或连续分析)与诊断性能无显著关联(p=0.61)。
- 作者认为 ONSD 超声是成人 TBI 颅内压升高的快速无创辅助手段,但可靠性高度依赖操作者。
适用人群与场景
- 成人创伤性脑损伤患者(14 个成人数据集);儿科创伤性脑损伤患者(3 个儿科数据集)。
研究设计与关键结果
- 研究类型:系统评价与 Meta 分析(PROSPERO 注册号 CRD420251052612),纳入 15 篇文献,获得 17 个汇总数据集(14 个成人、3 个儿科)和 41 份诊断报告。
- 成人诊断准确性:双变量随机效应模型汇总灵敏度 0.88、特异度 0.76、AUC>0.85(中等确定性证据)。
- 儿科诊断准确性:灵敏度 0.82、特异度 0.57、AUC 0.77(低确定性证据)。
- Meta 回归:超声 modality 和操作者背景是诊断异质性的显著来源;B 型超声轻度提高特异度,非专科操作者检查与诊断准确性降低相关。
- 常用 ONSD 截断值(5.5 mm 二分法或连续分析)与诊断性能无显著关联(p=0.61)。
证据类型、等级与局限
- 证据类型:系统评价与 Meta 分析;成人中等确定性证据,儿科低确定性证据。
- 局限:原文未报告局限性。
对床旁实践的改变
- 作者认为 ONSD 超声是成人 TBI 颅内压升高的有价值的、快速的无创辅助手段,但可靠性高度依赖操作者。
- 作者认为在更广泛临床采用前,尤其整合入多模态颅内压监测策略前,需要标准化测量方案、结构化操作者培训和儿科特异性验证研究。
- 非专科操作者进行检查与诊断准确性降低相关;B 型超声轻度提高特异度。
提炼说明
- 本条依据论文摘要整理;摘要未涉及「诊断标准与关键阈值」「处置流程与药物剂量」「红旗征与不能漏的情况」「指南或共识出处」,故未列出这些栏目;全文请点「阅读原文」。
英文原文(摘要)· 点击展开
BACKGROUND: Optic nerve sheath diameter (ONSD) sonography has been investigated as a noninvasive tool for detecting elevated intracranial pressure (ICP) in traumatic brain injury (TBI), yet its diagnostic performance remains debated owing to variability in operator expertise and ultrasound modality.
METHODS: This systematic review and meta-analysis, registered in The International Prospective Register of Systematic Reviews (PROSPERO; CRD420251052612), included 15 eligible publications. Two studies reported both adult and pediatric cohorts, and several reported multiple modality or cutoff comparisons, yielding 17 pooled datasets (14 adult and 3 pediatric) and a total of 41 diagnostic reports for quantitative synthesis.
RESULTS: Pooled estimates using a bivariate random-effects model demonstrated moderate-to-good diagnostic accuracy in adults, with a sensitivity of 0.88, specificity of 0.76, and an area under the curve (AUC) > 0.85 (moderate-certainty evidence). In contrast, pediatric datasets showed lower diagnostic performance, with a sensitivity of 0.82, specificity of 0.57, and AUC of 0.77 (low-certainty evidence). Meta-regression analyses identified ultrasound modality and operator background as significant contributors to diagnostic heterogeneity. B-mode ultrasonography modestly improved specificity, whereas examinations performed by nonspecialist operators were associated with reduced diagnostic accuracy. In contrast, commonly applied ONSD cutoff thresholds, including dichotomization at 5.5 mm or continuous analysis, were not significantly associated with diagnostic performance (p = 0.61).
CONCLUSIONS: These findings suggest that ONSD sonography is a valuable, rapid, and noninvasive adjunct for detecting raised ICP in adults with TBI, though its reliability remains highly operator-dependent and limited in pediatric populations. Standardized measurement protocols, structured operator training, and pediatric-specific validation studies are needed before broader clinical adoption, particularly for integration into multimodal ICP monitoring strategies.
参考来源
- Kannady J, Siahaan AMP, Harahap MRA, et al. Operator and Modality-Dependent Limitations of Optic Nerve Sheath Diameter Sonography for Raised ICP in TBI: A Meta-Regression of Diagnostic Accuracy. Neurocrit Care. 2026. PMID: 41882462. doi:10.1007/s12028-026-02493-4
供执业医师抢救与诊疗参考,不替代临床判断;剂量请结合体重、肝肾功能和本院药品规格复核,以指南原文为准。