用视觉印象或横断面容积成像测量头围:一项比较研究
该研究在巴西Paulo Niemeyer州立脑研究所选取同时接受卷尺头围测量和容积成像检查的儿童,由两名放射科医生分别用视觉印象和影像直接测量评估头围。卷尺与影像测量的一致性在视觉印象为52.3-72.7%(kappa 0.27-0.55),CT≥95.5%、MRI≥93.1%(kappa分别≥0.92、≥0.88)。结论认为横断面容积成像所得头围值与卷尺测量(金标准)相似,重复性和再现性高;视觉印象对头大小的判断不如影像测量并绘制于世界卫生组织标准生长曲线准确。
要点
- CT和MRI横断面容积成像测量头围与卷尺测量高度一致(CT一致率≥95.5%,MRI≥93.1%)。
- 视觉印象判断头大小的准确性低于影像测量并对照世界卫生组织标准生长曲线。
- 影像测量头围的组内和组间重复性高(ICC≥0.996)。
核心结论
- 横断面容积成像获得的头围值与卷尺测量(金标准)相似,具有高重复性和再现性;视觉印象对头大小的判断不如影像测量并绘制于世界卫生组织标准生长曲线准确。
研究设计与样本
- 在巴西Paulo Niemeyer州立脑研究所选取同时接受卷尺头围测量和容积成像检查的儿童,由两名放射科医生用视觉印象和影像直接测量进行分类,采用与卷尺测量法相似的解剖参数和适当的头生长曲线比较绝对値。
诊断标准与关键数值
- 卷尺与影像测量的一致率:视觉印象52.3-72.7%(kappa 0.27-0.55),CT≥95.5%(kappa≥0.92),MRI≥93.1%(kappa≥0.88)。
- 组内评估:ICC≥0.997,绝对技术误差测量0.20-0.31 cm,相对技术误差测量0.49-0.73%。
- 组间评估:ICC≥0.996,绝对技术误差测量0.24-0.29 cm,相对技术误差测量0.58-0.70%。
- 两种方法均值差异无统计学意义(P>0.05),Pearson相关系数≥0.99,Lin一致性相关系数≥0.99。
局限性
- 原文未报告局限性。
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
BACKGROUND: Head circumference (HC) correlates with brain growth and development. However, radiologists typically use their visual impressions to evaluate head dimensions, which is a purely subjective assessment.
OBJECTIVE: The first objective of this paper was to analyze whether it is possible to obtain HC measurements by CT and MRI cross-sectional volumetric imaging using parameters similar to the tape measurement method, which is considered the gold standard. The second objective was to determine the most accurate method for characterizing head size: visual impression or imaging measurement.
MATERIALS AND METHODS: Children who underwent HC measurement by tape and with volumetric imaging exams were selected from the Paulo Niemeyer State Brain Institute, Brazil. Two radiologists classified the children's heads using their visual impressions and by direct measurement on imaging. We used anatomical parameters similar to the tape measurement method and appropriate head growth charts and compared the absolute values obtained.
RESULTS: The concordance between tape and imaging ranged 52.3-72.7% for visual impression (kappa 0.27-0.55), ≥ 95.5% for CT and ≥ 93.1% for MRI (kappa ≥ 0.92 and ≥ 0.88, respectively). Intraclass correlation coefficients ≥ 0.997 and ≥ 0.996, absolute technical error measurements of 0.20-0.31 cm and 0.24-0.29 cm, and relative technical error measurements of 0.49-0.73% and 0.58-0.70% were evidenced in the intra- and inter-rater evaluations, respectively. The means between the methods were not significantly different (P > 0.05), with high values of Pearson correlation coefficient (≥ 0.99) and Lin concordance correlation coefficient (≥ 0.99).
CONCLUSION: Head circumference values obtained by cross-sectional volumetric imaging are similar to those obtained using the tape measurement method (gold standard), with high repeatability and reproducibility. Head size characterization by visual impression is less accurate than head circumference measurement by imaging and plotting the values obtained on World Health Organization normative charts.
引用
- Mourão, Corrêa, Correia, et al. Measuring head circumference using visual impressions or cross-sectional volumetric imaging: a comparison. Pediatr Radiol. 2022. PMID: 35829777. doi:10.1007/s00247-022-05448-x
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。