Deppe Dominik, Ziegeler Katharina, Hermann Kay Geert A, Proft Fabian, Poddubnyy Denis, Radny Felix, Makowski Marcus R, Muhle Maximilian, Diekhoff Torsten
Department of Radiology, Charité-Universitätsmedizin Berlin, Campus Mitte, Humboldt-Universität zu Berlin, Freie Universität Berlin, Charitéplatz 1, 10117 Berlin, Germany.
Department of Gastroenterology, Infectiology and Rheumatology, Charité-Universitätsmedizin Berlin, Humboldt-Universität zu Berlin, Freie Universität Berlin, Charitéplatz 1, 10117 Berlin, Germany.
Diagnostics (Basel). 2023 Feb 18;13(4):776. doi: 10.3390/diagnostics13040776.
To assess the ability of low-dose dual-energy computed tomography (ld-DECT) virtual non-calcium (VNCa) images for detecting bone marrow pathologies of the sacroiliac joints (SIJs) in patients with axial spondyloarthritis (axSpA). : Sixty-eight patients with suspected or proven axSpA underwent ld-DECT and MRI of the SIJ. VNCa images were reconstructed from DECT data and scored for the presence of osteitis and fatty bone marrow deposition by two readers with different experience (beginner and expert). Diagnostic accuracy and correlation (Kohen's k) with magnetic resonance imaging (MRI) as the reference standard were calculated for the overall and for each reader separately. Furthermore, quantitative analysis was performed using region-of-interest (ROI) analysis. : Twenty-eight patients were classified as positive for osteitis, 31 for fatty bone marrow deposition. DECT's sensitivity (SE) and specificity (SP) were 73.3% and 44.4% for osteitis and 75% and 67.3% for fatty bone lesions, respectively. The expert reader achieved higher diagnostic accuracy for both osteitis (SE 93.33%; SP: 51.85%) and fatty bone marrow deposition (SE: 65%; SP: 77.55%) than the beginner (SE: 26.67%; SP: 70.37% for osteitis; SE: 60%; SP: 44.9% for fatty bone marrow deposition). Overall correlation with MRI was moderate (r = 0.25, = 0.04) for osteitis and fatty bone marrow deposition (r = 0.25, = 0.04). Fatty bone marrow attenuation in VNCa images (mean: -129.58 HU; ±103.61 HU) differed from normal bone marrow (mean: 118.84 HU, ±99.91 HU; < 0.01) and from osteitis (mean: 172 HU, ±81.02 HU; < 0.01) while osteitis did not differ significantly from normal bone marrow ( = 0.27). : In our study, low-dose DECT failed to detect osteitis or fatty lesions in patients with suspected axSpA. Thus, we conclude that higher radiation might be needed for DECT-based bone marrow analysis.
评估低剂量双能计算机断层扫描（ld-DECT）虚拟去钙（VNCa）图像检测轴向脊柱关节炎（axSpA）患者骶髂关节（SIJ）骨髓病变的能力。：68例疑似或确诊axSpA的患者接受了SIJ的ld-DECT和MRI检查。从DECT数据重建VNCa图像，并由两位经验不同的阅片者（初学者和专家）对骨炎和脂肪骨髓沉积的存在情况进行评分。分别计算总体以及每位阅片者与作为参考标准的磁共振成像（MRI）的诊断准确性和相关性（科恩kappa值）。此外，使用感兴趣区域（ROI）分析进行定量分析。：28例患者骨炎分类为阳性，31例脂肪骨髓沉积分类为阳性。DECT对骨炎的敏感性（SE）和特异性（SP）分别为73.3%和44.4%，对脂肪性骨病变的敏感性和特异性分别为75%和67.3%。与初学者相比，专家阅片者对骨炎（SE 93.33%；SP：51.85%）和脂肪骨髓沉积（SE：65%；SP：77.55%）均获得了更高的诊断准确性（骨炎的SE：26.67%；SP：70.37%；脂肪骨髓沉积的SE：60%；SP：44.9%）。骨炎和脂肪骨髓沉积与MRI的总体相关性中等（骨炎r = 0.25，P = 0.04；脂肪骨髓沉积r = 0.25，P = 0.04）。VNCa图像中的脂肪骨髓衰减（平均值：-129.58 HU；±103.61 HU）与正常骨髓（平均值：118.84 HU，±99.91 HU；P < 0.01）和骨炎（平均值：172 HU，±81.02 HU；P < 0.01）不同，而骨炎与正常骨髓无显著差异（P = 0.27）。：在我们的研究中，低剂量DECT未能检测出疑似axSpA患者的骨炎或脂肪性病变。因此，我们得出结论，基于DECT的骨髓分析可能需要更高的辐射剂量。
