From the Institute of Diagnostic and Interventional Radiology, University Hospital Zurich and Faculty of Medicine, University of Zurich, Zurich.
Department of Diagnostic and Interventional Radiology, Lausanne University Hospital, University of Lausanne, Lausanne.
Invest Radiol. 2021 Mar 1;56(3):147-152. doi: 10.1097/RLI.0000000000000717.
The aims of this study were to test whether spectral photon-counting radiography (SPCR) is able to identify and distinguish different crystals associated with arthropathies in vitro and to validate findings in a gouty human third toe ex vivo.
Industry-standard calibration rods of calcium pyrophosphate, calcium hydroxyapatite (HA), and monosodium urate (MSU) were scanned with SPCR in an experimental setup. Each material was available at 3 different concentrations, and a dedicated photon-counting detector was used for SPCR, whereas validation scans were obtained on a clinical dual-energy computed tomography (DECT) scanner. Regions of interest were placed on SPCR images and consecutive DECT images to measure x-ray attenuation characteristics, including effective atomic numbers (Zeff). Statistical tests were performed for differentiation of Zeff between concentrations, materials, and imaging modalities. In addition, a third toe from a patient with chronic gouty arthritis was scanned with SPCR and DECT for differentiation of MSU from HA.
In both SPCR and DECT, significant differences in attenuation and Zeff values were found for different concentrations among (P < 0.001) and between different materials (P < 0.001). Overall, quantitative measurements of Zeff did not differ significantly between SPCR- and DECT-derived measurements (P = 0.054-0.412). In the human cadaver toe, gouty bone erosions were visible on standard grayscale radiographic images; however, spectral image decomposition revealed the nature and extent of MSU deposits and was able to separate it from bone HA by Zeff.
Identification and differentiation of different crystals related to arthropathies are possible with SPCR at comparable diagnostic accuracy to DECT. Further research is needed to assess diagnostic accuracy and clinical usability in vivo.
本研究旨在测试能谱光子计数射线照相术(SPCR)是否能在体外识别和区分不同的与关节病相关的晶体,并验证其在人类痛风第三趾的体外结果。
采用 SPCR 对工业标准焦磷酸钙、羟磷灰石(HA)和单钠尿酸盐(MSU)校准棒进行体外扫描。每种材料有 3 种不同浓度,专用的光子计数探测器用于 SPCR,而验证扫描则在临床双能 CT(DECT)扫描仪上进行。在 SPCR 图像和连续的 DECT 图像上放置感兴趣区,以测量 X 射线衰减特性,包括有效原子序数(Zeff)。采用统计学方法检验浓度、材料和成像方式之间 Zeff 的差异。此外,对慢性痛风性关节炎患者的第三趾进行 SPCR 和 DECT 扫描,以区分 MSU 和 HA。
在 SPCR 和 DECT 中,不同浓度间(P < 0.001)和不同材料间(P < 0.001)的衰减和 Zeff 值均存在显著差异。总体而言,SPCR 和 DECT 测量的 Zeff 值无显著差异(P = 0.054-0.412)。在人类尸体趾中,标准灰度射线照相图像可见痛风性骨侵蚀;然而,光谱图像分解揭示了 MSU 沉积物的性质和范围,并通过 Zeff 将其与骨 HA 区分开来。
SPCR 可识别和区分与关节病相关的不同晶体,与 DECT 的诊断准确性相当。需要进一步研究以评估体内诊断准确性和临床实用性。