文献速递神经创伤神经重症

Infrascanner 2000™在巴基斯坦颅脑创伤患者中检测颅内出血的诊断效能

Diagnostic Performance of the Infrascanner 2000™ for Detecting Intracranial Hemorrhage in Traumatic Brain Injury Patients in Pakistan.
PubMed 文献 · Neurosurgery · 2026-10-08

该前瞻性观察研究在巴基斯坦白沙瓦两家三级医院纳入141例轻中重度颅脑创伤患者,以非增强CT为参考标准评估Infrascanner 2000™检测颅内出血的准确性。CT证实125例(88.6%)存在出血,设备灵敏度91.2%、特异度50.0%、总体准确率86.5%,平衡准确率0.706(95%CI: 0.626-0.786)。结论为灵敏度良好但特异度中等,可作为CT延迟时的快速分诊辅助,不能替代CT。

要点

  • Infrascanner对颅内出血灵敏度91.2%,阴性似然比0.18,适合快速筛查。
  • 特异度仅50.0%,阳性预测值93.4%,阴性预测值42.1%。
  • 假阴性可能反映小或深部病灶,假阳性可能与颅外软组织损伤有关。

核心结论

  • Infrascanner 2000™检测颅内出血灵敏度良好但特异度中等,与已发表系列一致。
  • 该设备可作为CT获取延迟时的快速分诊辅助工具,但不应替代CT成像。

研究设计与样本

  • 前瞻性观察性诊断准确性研究,遵循STARD指南报告。
  • 在巴基斯坦白沙瓦Northwest General Hospital和Lady Reading Hospital两家三级医院神经外科急诊纳入141例轻、中、重度颅脑创伤患者。
  • 所有患者先行Infrascanner检查(评估者间kappa 0.82),再以非增强CT为参考标准。

主要发现

  • CT证实141例中125例(88.6%)存在颅内出血。
  • Infrascanner灵敏度91.2%、特异度50.0%、阳性预测值93.4%、阴性预测值42.1%、总体准确率86.5%。
  • 阳性似然比1.82,阴性似然比0.18。
  • 平衡准确率0.706(95%CI: 0.626-0.786),为单一操作点而非ROC曲线下面积。
  • 真阳性以浅表血肿为主;假阴性可能反映小或深部病灶;假阳性可能与颅外软组织损伤有关,但未正式分级。

临床怎么用

  • 在CT不可及或延迟的低资源环境中,可用Infrascanner快速筛查颅内出血以辅助分诊。
  • 设备阴性结果不能排除颅内出血,尤其需警惕小或深部病灶。
  • 阳性结果需结合临床表现,并尽快以CT确认。

局限性

  • 假阳性可能源于颅外软组织损伤,但该因素未正式分级。
  • 研究为两家中心设计,作者建议进一步开展包括院前环境的多中心研究。

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

BACKGROUND AND OBJECTIVES: Intracranial hemorrhage (ICH) is a major cause of morbidity and mortality after traumatic brain injury (TBI). Computed tomography (CT) is often unavailable in low-resource settings, delaying diagnosis. The Infrascanner 2000™, a handheld near-infrared device, has been proposed as a rapid, noninvasive ICH screening tool. The objective of this study was to determine the diagnostic accuracy of the Infrascanner 2000™ for detecting ICH in TBI patients, compared with noncontrast CT, at 2 tertiary care hospitals in Peshawar, Pakistan. METHODS: In this prospective observational study, 141 patients with mild, moderate, or severe TBI presenting to the emergency department of Neurosurgery at Northwest General Hospital and Lady Reading Hospital, Peshawar, Pakistan, underwent Infrascanner examination (inter-rater kappa 0.82), followed by noncontrast CT as the reference standard. Sensitivity, specificity, predictive values, likelihood ratios, balanced accuracy, and overall accuracy were calculated. Balanced accuracy (mean of sensitivity and specificity) replaced area under the curve, as the device yields a single binary operating point. Results followed Standard for Reporting Diagnostic Accuracy Studies guidelines. RESULTS: CT confirmed hemorrhage in 125 of 141 patients (88.6%). The Infrascanner showed sensitivity of 91.2%, specificity of 50.0%, positive predictive value of 93.4%, negative predictive value of 42.1%, and overall accuracy of 86.5% (positive likelihood ratio 1.82; negative likelihood ratio 0.18). Balanced accuracy was 0.706 (95% CI: 0.626-0.786), a single operating point rather than a conventional receiver operating characteristic-derived area under the curve. True positives were predominantly superficial hematomas; false negatives may reflect small or deep lesions. False positives could reflect extracranial soft-tissue injury, though this was not formally graded. CONCLUSION: The Infrascanner showed good sensitivity but modest specificity for ICH detection in this two-center Pakistani cohort, consistent with published series. It may serve as a rapid triage adjunct where CT access is delayed but should not replace CT imaging. Further multicenter studies, including prehospital settings, are warranted.

来源:Neurosurgery,查看原文。版权归原作者所有。

引用

  1. Khan Z, Khan MN, Khan WD, et al. Diagnostic Performance of the Infrascanner 2000™ for Detecting Intracranial Hemorrhage in Traumatic Brain Injury Patients in Pakistan. Neurosurgery. 2026. PMID: 42847715. doi:10.1227/neu.0000000000004241

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。