Cho Soo Buem, Baek Hye Jin, Ryu Kyeong Hwa, Moon Jin Il, Choi Bo Hwa, Park Sung Eun, Bae Kyungsoo, Jeon Kyung Nyeo, Kim Dong Wook
Department of Radiology, Gyeongsang National University School of Medicine and Gyeongsang National University Changwon Hospital, Changwon, Republic of Korea.
Department of Radiology, Busan Paik Hospital, Inje University College of Medicine, Busan, Republic of Korea.
PLoS One. 2017 Nov 7;12(11):e0186024. doi: 10.1371/journal.pone.0186024. eCollection 2017.
The purpose of this study was to investigate the clinical feasibility of spectral analyses using dual-layer detector spectral computed tomography (CT) in acute intracerebral haemorrhage (ICH).
We retrospectively reviewed patients with acute ICH who underwent CT angiography on a dual-layer detector spectral CT scanner. A spectral data analysis was performed to detect contrast enhancement in or adjacent to acute ICH by using spectral image reconstructions including monoenergetic (MonoE), virtual noncontrast (VNC), and iodine overlay fusion images. We also acquired a spectral plot to assess material differentiation within lesions.
Among the 30 patients, the most common cause of acute ICH was chronic hypertension (18/30, 60%) followed by trauma (5/30, 16.7%), brain tumour (3/30, 10%), Moyamoya disease (2/30, 6.7%), and haemorrhagic diathesis from anticoagulation therapy (2/30, 6.7%). Of 30 patients, 13 showed suboptimal iodine suppression in the subcalvarial spaces on VNC images compared with true noncontrast images. The CT angiographic spot sign within the acute ICH was detected in four patients (4/30, 13.3%). All three tumours were metastatic and included lung cancer (n = 2) and hepatocellular carcinoma (n = 1) which showed conspicuous delineation of an enhancing tumour portion in the spectral analysis. Spectral analyses allowed the discrimination of acute haemorrhage and iodine with enhanced lesion visualization on the MonoE images obtained at lower keVs (less than 70 keV) and spectral plot.
Even though the image quality of VNC is perceived to be inferior, it is feasible to evaluate acute ICH in clinical settings using dual-layer detector spectral CT. The MonoE images taken at lower keVs were useful for depicting contrast enhancing lesion, and spectral plot might be helpful for material differentiation in patients with acute ICH.
本研究旨在探讨使用双层探测器光谱计算机断层扫描（CT）进行光谱分析在急性脑出血（ICH）中的临床可行性。
我们回顾性分析了在双层探测器光谱CT扫描仪上接受CT血管造影的急性ICH患者。通过使用包括单能（MonoE）、虚拟平扫（VNC）和碘叠加融合图像在内的光谱图像重建，进行光谱数据分析以检测急性ICH内或其附近的对比增强。我们还获取了光谱图以评估病变内的物质区分。
在30例患者中，急性ICH最常见的病因是慢性高血压（18/30，60%），其次是创伤（5/30，16.7%）、脑肿瘤（3/30，10%）、烟雾病（2/30，6.7%）以及抗凝治疗导致的出血素质（2/30，6.7%）。30例患者中，与真正的平扫图像相比，13例在VNC图像上显示颅下间隙碘抑制欠佳。4例患者（4/30，13.3%）在急性ICH内检测到CT血管造影斑点征。所有3个肿瘤均为转移瘤，包括肺癌（n = 2）和肝细胞癌（n = 1），在光谱分析中显示增强的肿瘤部分有明显的轮廓。光谱分析能够在较低keV（小于70 keV）获得单能图像及光谱图上增强病变可视化的情况下区分急性出血和碘。
尽管VNC的图像质量被认为较差，但使用双层探测器光谱CT在临床环境中评估急性ICH是可行的。在较低keV下拍摄的单能图像有助于描绘对比增强病变，光谱图可能有助于急性ICH患者的物质区分。
