First‐trimester assessment of umbilical vein diameter using the semiautomated system for nuchal translucency measurement

Abstract

We read with great interest the articles recently published in this journal on the possibility of obtaining semiautomated measurements of nuchal translucency (NT)1-3. These studies demonstrate that, in the range of NT thickness between 1.5 and 3 mm, this new technique significantly improves the reproducibility of measurements. Although NT measurement is the cornerstone of the first-trimester ultrasound examination, it is also possible during this same gestational period to quantify absolute umbilical vein (UV) flow4, 5, and there is evidence that a reduction in UV flow is associated with subsequent impaired fetal growth5. However, the quality of ultrasound measurement of UV diameter is crucial in the calculation of absolute blood flow. Indeed, small errors in measurement of UV diameter may result in large errors in absolute flow computation, since vessel cross-sectional area is proportional to the square of the diameter. Since during the first trimester UV size is in the same range as that of NT, we tested the hypothesis that the software developed to semiautomatically measure NT may also be useful for measurement of UV diameter. To this end, after obtaining written informed consent from the mothers, we examined 50 consecutive fetuses undergoing the 11 + 0 to 13 + 6 weeks' ultrasound scan using an Aloka Prosound Alpha 7 machine with a 6-MHz transabdominal probe (Aloka, Tokyo, Japan). Using a previously reported technique6 the image of the fetal abdomen was magnified and the UV was visualized by perpendicularly insonating its intra-abdominal portion. Images were then stored on the hard disk of the ultrasound equipment. In each set of 50 images, two investigators (G.R. and A.C.) measured the UV diameter offline twice manually and twice with the semiautomated method (Figure 1). The two series of measurements were obtained 1 week apart. Manual measurements were obtained by placing the electronic calipers of the ultrasound equipment inner to inner at the point of maximal size of the UV. For the automated assessment the measurement box was placed on a rectilinear portion of the intra-abdominal UV, and the system automatically calculated the maximum diameter between superior and inferior borders (Figure 1). For each investigator the intraobserver reliability of the two manual and the two semiautomated measurements was assessed by the within-operator SD, calculated as the SD of the differences between the two measurements divided by √2. Interclass correlation coefficients (ICC) were used to assess interobserver reliability. Transverse view of the fetal abdomen at 12 + 5 weeks showing the measurement box placed over the umbilical vein (UV). The automated system draws two lines on the inner borders of the umbilical vein and measures the maximum UV diameter. For both investigators the within-operator SD values were significantly reduced with the semiautomated method (G.R.: manual, 0.111 (95% CI, 0.091–0.131) vs. semiautomated, 0.047 (95% CI, 0.036–0.059), P < 0.0001; A.C.: manual, 0.123 (95% CI, 0.091–0.154) vs. semiautomated, 0.049 (95% CI, 0.013–0.084), P < 0.0001). An improvement in the interobserver reliability was also seen when using the semiautomated method (manual ICC 0.883 vs. semiautomated ICC 0.957). In conclusion, the semiautomated system developed for NT measurement, when applied to UV diameter measurement, substantially reduces the within- and between-observer variation when compared with the traditional manual method. However, the process of UV measurement is greatly affected by the quality of image acquisition, a factor that requires appropriate sonographer training. Although these difficulties cannot be overcome by automated techniques7, their use may greatly enhance the clinical applications of UV flow assessment by trained operators. G. Rizzo* , A. Capponi* , E. Aiello , D. Arduini , * Fetal Medicine Center, Genoma Molecular Genetics Laboratory, Rome, Italy, Department of Obstetrics and Gynecology, Università di Roma ‘Tor Vergata’, Rome, Italy, Department of Obstetrics and Gynecology, Ospedale G.B. Grassi, Rome, Italy