Baumgarten Justus, Happel Christian, Ackermann Hanns, Grünwald Frank
Nuklearmedizin. 2017;56(4):132-138. doi: 10.3413/Nukmed-0869-16-12. Epub 2018 Jan 4.
A supplementary diagnostic tool in the assessment of cold thyroid nodules is scintigraphic imaging withTc-MIBI. Aim of this study was to investigate the validity of this tool by determining the intra- and interobserver agreement in the assessment of cold thyroid nodules in Tc-MIBI scintigrams.
A retrospective study with 284 patients (16-85 years of age, 194 women, 90 men) was performed. They had at least one cold nodule and from each of whom were available at least oneTc-MIBI and the Tc- pertechnetate image. Eight physicians, active in nuclear medicine, reviewed the sctinti- grams twice in a random order. They were asked if they considered the combination a match, a mismatch, or inconclusive, and if the early or delayed image was more significant or if there was no difference.
Intraobserver agreement ranged from κ = 0.56 (moderate) to κ = 0.78 (substantial). Interobserver agreement ranged from κ = 0.44 to κ = 0.53 (moderate). Interobserver agreement for observers with more than 5 years of work experience in nuclear medicine ranged from κ=0.61 to κ = 0.70 (substantial), for observers with 2-5 years from κ = 0.53 (moderate) to κ = 0.61 (substantial) and for observers with < 2 years from κ = 0.47 to κ = 0.61. "No difference" was chosen in 70 resp. 77 % of all cases in session 1 resp. 2. The early image was preferred in 26 resp. 20 %, and the delayed one in 3 resp. 4 % of all cases.
The values of interobserver agreement of all eight observers show that the assessment of Tc-MIBI scintigrams is subject to a certain variance. Hence, they ought to be finally assessed by observers with at least 5 years of work experience.
用锝-甲氧基异丁基异腈（Tc-MIBI）闪烁显像作为评估甲状腺冷结节的辅助诊断工具。本研究的目的是通过确定观察者内部和观察者之间对Tc-MIBI闪烁显像中甲状腺冷结节评估的一致性，来研究该工具的有效性。
对284例患者（年龄16 - 85岁，女性194例，男性90例）进行回顾性研究。他们至少有一个冷结节，且每人至少有一张Tc-MIBI和高锝酸盐图像。八名从事核医学的医生以随机顺序对闪烁显像进行了两次评估。询问他们是否认为这种组合是匹配、不匹配或不确定的，以及早期或延迟图像是否更具意义或是否没有差异。
观察者内部一致性范围为κ = 0.56（中等）至κ = 0.78（高度）。观察者之间的一致性范围为κ = 0.44至κ = 0.53（中等）。在核医学领域有超过5年工作经验的观察者之间的一致性范围为κ = 0.61至κ = 0.70（高度），有2 - 5年工作经验的观察者之间的一致性范围为κ = 0.53（中等）至κ = 0.61（高度），工作经验不足2年的观察者之间的一致性范围为κ = 0.47至κ = 0.61。在第一次和第二次评估中，分别有70%和77%的所有病例选择了“无差异”。在所有病例中，分别有26%和20%的病例更倾向于早期图像，3%和4%的病例更倾向于延迟图像。
所有八名观察者的观察者间一致性值表明，对Tc-MIBI闪烁显像的评估存在一定差异。因此，最终应由至少有5年工作经验的观察者进行评估。
