T-SMART断层融合成像、CT与X线摄影对髋关节假体周围结构观察的对比研究

A comparative study of tomosynthesis,computed tomography and X-ray in imaging the structures around hip prostheses

目的：比较tomosynthesis-shimadzumetalartefactreductiontechnology（T-SMART）断层融合成像（ tomosynthesis，TOMOS ）、计算机 X 线断层扫描（ computed tomography，CT ）与数字化 X 线摄影（ digital radiography，DR）显示髋关节假体及其周围结构的能力。方法连续选取进行单侧或双侧髋关节置换术（ total hip arthroplasty，THA ），且已行术后 DR、CT 及 TOMOS 三项影像学复查的患者20例，男12例，女8例，年龄36～82岁，平均年龄61.0岁。调阅并分析这20个病例的TOMOS、CT及DR影像资料，由3名高年资主治医师读片，并进行主观评分，评分标准为：假体周围结构观察清晰，图像评为“优”，记2分；因轻微伪影或受限于图像分辨率等原因，图像清晰度受影响，但基本不影响观察，评为“中”，记1分；因金属伪影大等因素，图像清晰度差，严重影响观察，评为“差”，记0分。评分结果以广义估计方程进行统计学分析。结果应用TOMOS观察髋臼（含髋关节间隙）的图像质量频率分布为：2分55%，1分40%，0分5%，观察股骨上段假体周围组织的图像质量频率分布为：2分75%，1分25%，0分5%；应用CT观察髋臼（含髋关节间隙）的图像质量频率分布为：2分5%，1分60%，0分35%，观察股骨上段假体周围组织的图像质量频率分布为：2分5%，1分55%，0分40%，TOMOS的评分高于CT，差异有统计学意义（ P＜0.001），应用DR检查观察髋臼（含髋关节间隙）的图像质量频率分布为：2分25%，1分75%，0分0%；观察股骨上段假体周围组织的图像质量频率分布为：2分50%，1分50%，0分0%，DR的评分高于CT，差异有统计学意义（P＝0～0.0077）。结论髋关节假体TOMOS成像质量比CT图像好，又克服了DR不能分层薄层观察的缺点，在髋关节假体复查中有应用价值。

Objective To compare the capacity of tomosynthesis （ TOMOS ）, computed tomography （ CT ） and digital radiography （ DR ） in imaging hip prostheses and their surrounding structures.Methods A total of 20 consecutive patients receiving either unilateral or bilateral total hip arthroplasty （ THA ） were enrolled, including 12 males and 8 females. Their mean age was 61.0 years old （ range： 36-82 years ）. All the patients underwent TOMOS, CT and DR after THA. The TOMOS, CT and DR images of the 20 patients were reviewed and subjectively scored by 3 senior attending radiologists. The evaluation standard was stated as following. The clear image of the structures around the prostheses was rated as “excellent”, with the score of 2 points. The sharpness of the image was affected by minor artifacts or lower resolution, but the basic observation wasn’t affected. Such images were rated as “mid”, with the score of 1 point. If the image deifnition was seriously affected by metal artifacts or other factors, it was rated as “poor”, with the score of 0 point. The scores calculated would be statistically analysed via Generalized Estimating Equations.Results The score frequency distribution of TOMOS imaging for the acetabulum （ including the hip joint space ） was 55% （ 2 points ）, 40% （ 1 point ）, and 5% （ 0 point ）. The score frequency distribution of TOMOS imaging for the proximal femoral peri-prothesis region was 75% （ 2 points ）, 25% （ 1 point ）, and 0% （ 0 point ）. The score frequency distribution of CT imaging for the acetabulum （ including the hip joint space ） is 5% （ 2 points ）, 60%（ 1 point ）, and 35% （ 0 point ）. The score frequency distribution of CT imaging for the proximal femoral peri-prothesis region was 5% （ 2 points ）, 55% （ 1 point ）, and 40% （ 0 point ）. TOMOS was superior to CT in image quality, and the differences in the scores were statistically signigicant （P＆lt;0.001 ）. The score frequency distribution of DR imaging for the acetabulum （ including the hip joint space ） was 25% （ 2 points ）, 75% （ 1 point ）, and 0% （ 0 point ）. The score frequency distribution of DR imaging for the proximal femoral peri-prothesis region was 50% （ 2 points ）, 50%（ 1 point ）, and 0% （ 0 point ）. The image deifnition of DR was better than that of CT, and the differences in the scores were statistically signigicant （P=0-0.0077 ）.Conclusions The TOMOS imaging quality of hip prostheses is better than that of CT, with the advantage of layering and thin-slice observation over DR. Thus, TOMOS is a promising imaging modality in the follow-up after THA .