Kumar Neil M, de Cesar Netto Cesar, Schon Lew C, Fritz Jan
From the *Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University School of Medicine; and †Department of Orthopedic Surgery, MedStar Union Memorial Hospital, Baltimore, MD.
Invest Radiol. 2017 May;52(5):310-316. doi: 10.1097/RLI.0000000000000350.
Long echo train length (ETL) is an often recommended but unproven technique to decrease metal artifacts on magnetic resonance imaging (MRI) scans. Therefore, we quantitatively and qualitatively assessed the effects of ETL on metal artifact on MRI scans using a cobalt-chromium-containing arthroplasty implant system.
Using a total ankle arthroplasty system implanted into a human cadaver ankle and a clinical 1.5 T MRI system, turbo spin echo (TSE) pulse sequences were acquired with ETL ranging from 3 to 23 and receiver bandwidth (BW) from 100 to 750 Hz/pixel, whereas effective echo time and spatial resolution were controlled. A compressed sensing slice encoding for metal artifact correction TSE prototype pulse sequence was used as reference standard. End points included the total implant-related artifact area and implant-related signal void areas. Two raters evaluated the overall image quality and preference across varying BW and ETL. Two-factor analysis of variance, Friedman test, Kruskal-Wallis test, and Pearson correlation were used. P values of less than 0.05 were considered statistically significant.
The total implant-related artifact area ranged from 0.119 for compressed sensing slice encoding for metal artifact correction (BW, 600 Hz/pixel; ETL, 3) to 0.265 for TSE (BW, 100 Hz/pixel; ETL, 23). Longer ETL significantly increases the total implant-related artifact area (P = 0.0004), whereas it decreased with increasing BW (P < 0.0001). Implant-related signal void areas were not significantly affected by larger echo train length, but reduced with higher BW (P < 0.0001). Readers had a significant preference for images with high BW and short ETL (P < 0.0001).
High receiver BW is the most effective parameter for reduction of arthroplasty implant-induced metal artifact on MRI scans, whereas in contradiction to prevalent notions, long echo trains fail to reduce implant-related metal artifacts, but in fact cause degradation of image quality around the implant with resultant larger appearing total metal artifacts.
长回波链长度（ETL）是一种常被推荐但未经证实的减少磁共振成像（MRI）扫描中金属伪影的技术。因此，我们使用含钴铬的关节置换植入系统，对ETL在MRI扫描中对金属伪影的影响进行了定量和定性评估。
使用植入人体尸体踝关节的全踝关节置换系统和临床1.5T MRI系统，采集回波链长度（ETL）范围为3至23且接收带宽（BW）为100至750Hz/像素的快速自旋回波（TSE）脉冲序列，同时控制有效回波时间和空间分辨率。使用用于金属伪影校正的压缩感知切片编码TSE原型脉冲序列作为参考标准。终点指标包括与植入物相关的总伪影面积和与植入物相关的信号缺失区域。两名评估者对不同带宽和ETL下的整体图像质量和偏好进行评估。采用双因素方差分析、Friedman检验、Kruskal-Wallis检验和Pearson相关性分析。P值小于0.05被认为具有统计学意义。
与植入物相关的总伪影面积范围从用于金属伪影校正的压缩感知切片编码的0.119（带宽600Hz/像素；ETL 3）到TSE的0.265（带宽100Hz/像素；ETL 23）。较长的ETL显著增加了与植入物相关的总伪影面积（P = 0.0004），而随着带宽增加其减小（P < 0.0001）。与植入物相关的信号缺失区域不受较大回波链长度的显著影响，但随着带宽增加而减小（P < 0.0001）。读者对高带宽和短ETL的图像有明显偏好（P < 0.0001）。
高接收带宽是减少MRI扫描中关节置换植入物引起的金属伪影的最有效参数，而与普遍观点相反，长回波链未能减少与植入物相关的金属伪影，反而实际上导致植入物周围图像质量下降，从而使总的金属伪影看起来更大。
