文献速递骨肌外周血管技术与安全CT

全膝关节假体患者超高分辨率光子计数CT血管成像中迭代金属伪影消除预设的比较

Comparison of Iterative Metal Artifact Reduction Presets In Ultra-high-resolution Photon-counting CT Angiography of Patients with Total Knee Endoprosthesis.
PubMed 文献 · Acad Radiol · 2026-07-01

本研究纳入25例全膝关节置换术后行下肢超高分辨率光子计数CT血管成像的患者共31侧下肢,比较无IMAR及7种IMAR预设的图像质量。结果显示IMARDental对高密度及低密度伪影的抑制均最强,腘动脉可评估性主观评分最高。

要点

  • 膝关节假体患者下肢UHR-PCCTA中,IMARDental预设对腘动脉区域金属伪影抑制效果最佳。
  • IMARDental使高密度伪影降至932.5 HU,低密度伪影降至-97.5 HU,优于其他预设。
  • 各重建方案图像噪声无显著差异(中位10.6-11.3 HU),不影响伪影选择决策。

核心结论

  • IMARDental尽管名称不直观,但在全膝关节假体患者下肢UHR-PCCTA中提供最强的金属伪影抑制,腘动脉可评估性最佳。

研究设计与样本

  • 回顾性研究,纳入2024年8月至2025年8月间25例膝关节置换术后行下肢UHR-PCCTA的患者,共31侧下肢。
  • 每例数据分别以无IMAR及7种IMAR预设重建,客观图像质量通过伪影最明显区域的ROI测量评估。
  • 主观图像质量由7名放射科医生采用强制选择配对比较工具评估,共完成6076次配对比较。

关键影像表现/征象

  • IMARDental重建的高密度伪影中位值为932.5 HU,低于无IMAR的4300.1 HU及其他预设的3117.9-3377.1 HU。
  • IMARDental重建的低密度伪影中位值为-97.5 HU,优于其他预设的-567.9至-1523.3 HU及无IMAR的-3319.7 HU。
  • 所有重建方案的图像噪声中位值为10.6-11.3 HU,组间无显著差异。

诊断标准与关键数值

  • IMARDental对高密度伪影的降低显著优于无IMAR及其他6种预设(均p<0.001)。
  • IMARDental对低密度伪影的降低显著优于无IMAR及其他预设(均p≤0.022)。
  • 腘动脉可评估性主观配对比较中IMARDental偏好概率为0.92(95%CI 0.90-0.93)。

临床应用与阅片要点

  • 全膝关节假体患者下肢UHR-PCCTA中应优先选用IMARDental预设以优化腘动脉血管显示。
  • IMAR预设名称与其适用部位可能不匹配,选择重建参数时不应仅依据名称判断。
  • IMARDental在降低伪影的同时不增加图像噪声,可用于常规临床重建流程。

局限性

  • 本研究为单中心回顾性设计,样本量较小。
  • 研究仅评估了特定膝关节假体类型,结果对其他假体类型的普适性尚不明确。

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)。

英文原文(摘要)· 点击展开

RATIONALE AND OBJECTIVES: Ultra-high-resolution (UHR) photon-counting CT angiography (PCCTA) has become the reference standard for non-invasive assessment of peripheral artery disease. In the presence of total knee endoprosthesis, however, metal artifacts hamper visualization of the popliteal vasculature. This study investigated the effect of the different iterative metal artifact reduction (IMAR) settings in patients with knee joint replacement.

MATERIALS AND METHODS: Twenty-five patients with knee arthroplasty (74.0 ± 8.5 years; 12 women) undergoing lower extremity UHR-PCCTA between August 2024 and August 2025 were included, yielding 31 lower extremities for analysis. Each dataset was reconstructed without IMAR and with the seven available IMAR presets ("Extremity", "Hip", "Pacemaker", "Lung coils", "Shoulder", "Dental", and "Neuro coils"). Objective image quality was assessed by region-of-interest measurements in the most pronounced hyperdense and hypodense artifacts. Subjective image quality was evaluated by seven radiologists using a forced-choice pairwise comparison tool, totaling 6076 pairwise comparisons.

RESULTS: Compared with non-IMAR datasets (median 4300.1 HU) and the other six IMAR presets (3117.9 - 3377.1 HU), application of IMARDental resulted in the strongest reduction of hyperdense artifacts (932.5 HU; all p<0.001). Similarly, IMARDental (-97.5 HU) provided superior reduction of hypodense artifacts versus datasets reconstructed with the other IMAR presets (-567.9 - -1523.3 HU) and without IMAR (-3319.7 HU; all p≤0.022). Image noise was comparable across all reconstructions (median 10.6 - 11.3 HU; all p>0.05). Subjective evaluation of popliteal artery assessability favored IMARDental with the highest preference probability in pairwise comparison to the other IMAR and non-IMAR data (0.92, 95% confidence interval 0.90-0.93).

CONCLUSION: Despite its misleading name, IMARDental provides the strongest reduction of metal artifacts in UHR-PCCTA of the leg, allowing for the best assessability of the popliteal vasculature in patients with total knee endoprosthesis.

来源:Acad Radiol,查看原文。版权归原作者所有。

引用

  1. Huflage, Kunz, Wech, et al. Comparison of Iterative Metal Artifact Reduction Presets In Ultra-high-resolution Photon-counting CT Angiography of Patients with Total Knee Endoprosthesis. Acad Radiol. 2026. PMID: 42386454. doi:10.1016/j.acra.2026.06.012

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。