Daou Badih J, Khalsa Siri Sahib S, Anand Sharath Kumar, Williamson Craig A, Cutler Noah S, Aaron Bryan L, Srinivasan Sudharsan, Rajajee Venkatakrishna, Sheehan Kyle, Pandey Aditya S
1Department of Neurosurgery, University of Michigan, Ann Arbor; and.
2School of Medicine, Wayne State University, Detroit, Michigan.
J Neurosurg. 2021 Feb 5;135(4):1155-1163. doi: 10.3171/2020.8.JNS201273. Print 2021 Oct 1.
Hydrocephalus and seizures greatly impact outcomes of patients with aneurysmal subarachnoid hemorrhage (aSAH); however, reliable tools to predict these outcomes are lacking. The authors used a volumetric quantitative analysis tool to evaluate the association of total aSAH volume with the outcomes of shunt-dependent hydrocephalus and seizures.
Total hemorrhage volume following aneurysm rupture was retrospectively analyzed on presentation CT imaging using a custom semiautomated computer program developed in MATLAB that employs intensity-based k-means clustering to automatically separate blood voxels from other tissues. Volume data were added to a prospectively maintained aSAH database. The association of hemorrhage volume with shunted hydrocephalus and seizures was evaluated through logistic regression analysis and the diagnostic accuracy through analysis of the area under the receiver operating characteristic curve (AUC).
The study population comprised 288 consecutive patients with aSAH. The mean total hemorrhage volume was 74.9 ml. Thirty-eight patients (13.2%) developed seizures. The mean hemorrhage volume in patients who developed seizures was significantly higher than that in patients with no seizures (mean difference 17.3 ml, p = 0.01). In multivariate analysis, larger hemorrhage volume on initial CT scan and hemorrhage volume > 50 ml (OR 2.81, p = 0.047, 95% CI 1.03-7.80) were predictive of seizures. Forty-eight patients (17%) developed shunt-dependent hydrocephalus. The mean hemorrhage volume in patients who developed shunt-dependent hydrocephalus was significantly higher than that in patients who did not (mean difference 17.2 ml, p = 0.006). Larger hemorrhage volume and hemorrhage volume > 50 ml (OR 2.45, p = 0.03, 95% CI 1.08-5.54) were predictive of shunt-dependent hydrocephalus. Hemorrhage volume had adequate discrimination for the development of seizures (AUC 0.635) and shunted hydrocephalus (AUC 0.629).
Hemorrhage volume is an independent predictor of seizures and shunt-dependent hydrocephalus in patients with aSAH. Further evaluation of aSAH quantitative volumetric analysis may complement existing scales used in clinical practice and assist in patient prognostication and management.
脑积水和癫痫对动脉瘤性蛛网膜下腔出血（aSAH）患者的预后有很大影响；然而，缺乏可靠的工具来预测这些预后。作者使用容积定量分析工具来评估aSAH总体积与分流依赖性脑积水和癫痫预后之间的关联。
使用在MATLAB中开发的自定义半自动计算机程序，对动脉瘤破裂后的总出血量进行回顾性分析，该程序采用基于强度的k均值聚类自动将血液体素与其他组织分离。将体积数据添加到前瞻性维护的aSAH数据库中。通过逻辑回归分析评估出血量与分流性脑积水和癫痫之间的关联，并通过分析受试者工作特征曲线（AUC）下的面积来评估诊断准确性。
研究人群包括288例连续的aSAH患者。平均总出血量为74.9 ml。38例患者（13.2%）发生癫痫。发生癫痫的患者的平均出血量显著高于未发生癫痫的患者（平均差异17.3 ml，p = 0.01）。在多变量分析中，初始CT扫描时出血量较大以及出血量>50 ml（OR 2.81，p = 0.047，95%CI 1.03 - 7.80）可预测癫痫。48例患者（17%）发生分流依赖性脑积水。发生分流依赖性脑积水的患者的平均出血量显著高于未发生的患者（平均差异17.2 ml，p = 0.006）。出血量较大以及出血量>50 ml（OR 2.45，p = 0.03，95%CI 1.08 - 5.54）可预测分流依赖性脑积水。出血量对癫痫的发生（AUC 0.635）和分流性脑积水（AUC 0.629）有足够的辨别能力。
出血量是aSAH患者癫痫和分流依赖性脑积水的独立预测因素。对aSAH定量容积分析的进一步评估可能补充临床实践中使用的现有量表，并有助于患者的预后评估和管理。
