动脉瘤性蛛网膜下腔出血影像学预测因素:基于入院平扫CT的动脉瘤病因预测模型
回顾性队列研究纳入2016至2021年连续因自发性蛛网膜下腔出血疑似动脉瘤病因入院的259例患者,比较动脉瘤性蛛网膜下腔出血(aSAH)与血管造影阴性非中脑周围蛛网膜下腔出血(an-NPSAH)的入院平扫CT影像特征,构建预测动脉瘤病因的模型,曲线下面积0.71(95%CI 0.62-0.79)。
要点
- aSAH与an-NPSAH的改良Graeb评分中位数相似(6 vs 3.5,p=0.33)。
- aSAH组外侧裂及半球间池血液厚度更大,桥前池血液厚度更小。
- 预测模型区分动脉瘤病因的AUC为0.71(95%CI 0.62-0.79),作者认为可考虑进一步开发应用。
核心结论
- an-NPSAH与aSAH患者入院平扫CT影像特征存在差异。
- 作者提出的风险分层模型可考虑在未来进一步开发并用于临床实践。
研究设计与样本
- 回顾性队列研究,数据来自2016至2021年在一家学术中心连续入院的疑似动脉瘤病因的自发性蛛网膜下腔出血患者的前瞻性收集资料。
- 共纳入259例患者,平均年龄56岁(SD 12.7岁),55%为女性;其中209例为动脉瘤性蛛网膜下腔出血(aSAH),50例为血管造影阴性非中脑周围蛛网膜下腔出血(an-NPSAH)。
- 比较两组入院头颅平扫CT上基底池和外侧裂的血液厚度及改良Graeb评分,并据此构建预测动脉瘤病因的模型。
主要发现
- aSAH与an-NPSAH的改良Graeb评分中位数相似(6 [IQR 2-10] vs 3.5 [IQR 0-8.5],p=0.33)。
- aSAH组外侧裂平均血液厚度大于an-NPSAH组(p=0.010),半球间池平均血液厚度大于an-NPSAH组(p=0.002)。
- aSAH组外侧裂血液扩展程度中位数大于an-NPSAH组(p=0.001),但桥前池平均血液厚度小于an-NPSAH组(p=0.014)。
- 受试者工作特征曲线分析显示模型预测动脉瘤病因具有可接受的准确性(曲线下面积0.71,95%CI 0.62-0.79)。
关键数值与推荐
- 基于影像特征差异构建的评分模型,受试者工作特征曲线下面积0.71(95%CI 0.62-0.79)。
临床怎么用
- 作者认为其提出的风险分层模型可考虑在未来进一步开发并用于临床实践。
局限性
- 原文未报告局限性。
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
OBJECTIVE: Spontaneous angiogram-negative nonperimesencephalic subarachnoid hemorrhage (an-NPSAH) can represent a diagnostic and management dilemma. The authors sought to determine radiographic predictors of aneurysmal etiology based on admission noncontrast head CT scans. METHODS: The authors performed a retrospective cohort study of prospectively collected data from consecutive patients who were admitted for spontaneous subarachnoid hemorrhage (SAH) with suspected aneurysmal etiology to an academic center from 2016 to 2021. They compared blood thickness in the basal cisterns and sylvian fissures and modified Graeb scores on admission head CT scans between the two groups and subsequently developed a predictive model to identify aneurysmal etiology. RESULTS: Of 259 included patients (mean age 56 years [SD 12.7 years]; 55% female), 209 had aneurysmal SAH (aSAH) and 50 had an-NPSAH. The median modified Graeb scores were similar for aSAH and an-NPSAH (6 [IQR 2-10] vs 3.5 [IQR 0-8.5], p = 0.33). The mean blood thickness was greater in the sylvian fissure (p = 0.010) and interhemispheric cisterns (p = 0.002), and there was a greater median degree of extension of blood in the sylvian fissures (p = 0.001) in aSAH than in an-NPSAH patients, but the mean blood thickness was less in the prepontine cistern (p = 0.014). The authors' scoring model was constructed based on differences in radiographic features. Receiver operating characteristic curve analysis showed acceptable accuracy in predicting aneurysmal etiology (area under the curve 0.71, 95% CI 0.62-0.79). CONCLUSIONS: There are differences in radiographic features on admission head CT between an-NPSAH and aSAH patients. The authors' proposed risk stratification model may be considered for further development and use in clinical practice in the future.
引用
- Baumgartner K, Meyer A, Mandel D, et al. Radiographic predictors of aneurysmal etiology in patients with aneurysmal pattern subarachnoid hemorrhage. J Neurosurg. 2022. PMID: 36727566. doi:10.3171/2022.11.JNS222192
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。