Herman A, Maymon R, Dreazen E, Caspi E, Bukovsky I, Weinraub Z
Department of Obstetrics and Gynecology, Assaf Harofeh Medical Center, Zerifin, Israel.
Ultrasound Obstet Gynecol. 1998 Apr;11(4):266-70. doi: 10.1046/j.1469-0705.1998.11040266.x.
The objective of this study was to assess whether nuchal translucency image magnification contributes to the repeatability of caliper placement. Twenty-seven women undergoing first-trimester ultrasound screening at 11-14 weeks were examined by two qualified examiners. Fetal nuchal translucency thickness was measured twice by each examiner on a regular-sized image and subsequently on the same still image magnified. Intraobserver repeatability coefficients of the regular-sized images were 0.28 mm and 0.34 mm for examiners A and B, respectively. Identical intraobserver repeatability coefficients of 0.28 mm were attained for both examiners on the magnified images. Interobserver repeatability of regular-sized image measurements showed a significant difference (p < 0.01) but the mean difference (+/- standard deviation) of 0.10 (0.18) mm was negligible. Similarly, interobserver repeatability of the magnified-image measurements yielded a significant difference (p < 0.02), but again with a small mean difference of 0.13 (0.26) mm. There were also significant differences when regular-sized images and magnified images were compared for both examiners: 0.08 (0.16) mm for examiner A (p < 0.02) and 0.10 (0.17) mm for examiner B (p < 0.01). Our results demonstrate that nuchal translucency image magnification does not contribute to the reproducibility of the measurement. Despite significantly smaller mean values obtained from the magnified images, compared to the regular-sized measurements, those differences do not justify modification of the criteria for caliper placement on magnified images. Blind repeated measurements on a regular-sized and/or magnified image are recommended as a tool for self-assessment, quality control and training.
本研究的目的是评估颈部半透明图像放大是否有助于卡尺放置的可重复性。27名在11至14周接受孕早期超声筛查的女性由两名合格的检查人员进行检查。每位检查人员在常规尺寸图像上对胎儿颈部半透明厚度进行两次测量，随后在放大的同一静态图像上进行测量。检查人员A和B在常规尺寸图像上的观察者内重复性系数分别为0.28毫米和0.34毫米。两位检查人员在放大图像上均获得了相同的0.28毫米观察者内重复性系数。常规尺寸图像测量的观察者间重复性显示出显著差异（p<0.01），但平均差异（±标准差）为0.10（0.18）毫米可忽略不计。同样，放大图像测量的观察者间重复性也产生了显著差异（p<0.02），但平均差异同样较小，为0.13（0.26）毫米。对于两位检查人员，将常规尺寸图像和放大图像进行比较时也存在显著差异：检查人员A为0.08（0.16）毫米（p<0.02），检查人员B为0.10（0.17）毫米（p<0.01）。我们的结果表明，颈部半透明图像放大无助于测量的可重复性。尽管与常规尺寸测量相比，放大图像获得的平均值明显较小，但这些差异并不足以证明修改放大图像上卡尺放置标准的合理性。建议对常规尺寸和/或放大图像进行盲法重复测量，作为自我评估、质量控制和培训的工具。
