The umbilical vein in the human fetus has a non‐linear growth pattern across gestation

Abstract

Introduction and Objectives

Estimation of the umbilical venous blood flow volume relies on the diameter of the vein, which has been reported to be reduced in severely growth restricted fetuses. However, there is only limited information on the growth pattern of this vessel in the normal human fetus. The aim of this study is to examine the growth pattern of the umbilical vein across gestation in low‐risk human pregnancies.

Methods

In a prospective, longitudinal ultrasound study of 136 low‐risk pregnancies, the internal diameter of the intra‐amniotic portion of the umbilical vein was measured at 18, 26 and 34 weeks of gestation. To investigate the growth pattern of the venous diameter, the ratios of diameters at 26 weeks to diameters at 18 weeks (ratio 1) and the ratios at 34 to 26 weeks (ratio 2) were also calculated. A paired‐sample t‐test was performed to compare the means of the two ratios.

Results

The mean diameter of the umbilical vein at 18 weeks was 2.8 mm (SD 0.40), 5.8 mm (SD 0.67) at 26 weeks and 7.6 mm (SD 0.98) at 34 weeks. The mean of ratio 1 was 2.06 (95% CI 2.01–2.14), which was significantly higher than ratio 2 (mean 1.33, 95% CI 1.29–1.36), P < 0.001.

Conclusion

The umbilical venous diameter demonstrates a non‐linear growth pattern between 18 and 34 weeks of gestation. The diameter doubled in size between 18 and 26 weeks but grew at a slower rate between 26 and 34 weeks of gestation. This study provides new data on the normal growth pattern of the umbilical vein by identifying a period of gestation where its growth is accelerated.

Introduction

Human fetuses receive their supply of nutrients and oxygen from a single umbilical vein. Therefore, the estimation of umbilical venous flow volume is a reasonable representation of the volume of blood supplied to the placenta by the uterine arteries. The human studies of umbilical venous flow volume emerged in literature in early 1980s1, 2, 3, 4, 5 following the introduction of the non‐invasive Doppler techniques by Gill et al. in late 1970s.6, 7 In recent years, the methods have been refined to provide more detailed physiological information in order to use the venous flow volume as a clinical tool in the management of high‐risk pregnancies.8, 9, 10, 11, 12, 13

The growth in the internal diameter of the umbilical vein is a major contributing factor to the flow volume. There are various normal reference charts available for umbilical venous diameter and flow velocity.14, 15, 16, 17, 18 Barbera and colleagues14 examined the relative contribution of vein diameter and flow velocity to the increase in umbilical venous flow volume by comparing the growth slopes of both parameters vs. gestation. They concluded that since the slope for the diameter is steeper, it is the greater contributor to the increase in the umbilical venous flow volume compared to the flow velocity. The growth of umbilical venous diameter has been reported as steady16 or exponential across the gestation.14 Examining the limited graphs available for the umbilical venous diameter, gives an impression that the slope of the curve is flatter in the third trimester of pregnancy compared to the second trimester. The aim of present study is to re‐examine the growth pattern of the umbilical vein in appropriately grown fetuses, to determine the changes in the growth rate during gestation.

Materials and methods

Study sample

This study has used the ultrasound data from a sub‐set of pregnant women recruited into the Peel Child Health Study (PCHS, https://www.peelchildhealthstudy.com.au). The inclusion criteria were: low‐risk singleton pregnancy of 18 weeks' gestation or less, confirmed gestational age by ultrasound and living in the Peel region of Western Australia. The exclusion criteria were: multiple pregnancies, abnormal karyotype and pre‐existing maternal disease that can have a significant effect on the fetoplacental circulation, such as type I diabetes or thrombophilia. For timely completion of this project, the period of ultrasound data collection for this study was set for 18 months upon receiving the required ethical permissions. The aim was to include the data from at least 100 women meeting the selection criteria who attended for all three ultrasound examinations. Following the final approval from the Human Research Ethic Committees, data collection commenced in March 2010 and was completed by September 2011. Only 170 PCHS women in total had all their three ultrasound examinations at the radiology provider. Four women had twin gestations and were excluded. Of the remaining 166, only those whose ultrasound assessments were completed in the specified data collection period were included, which resulted in the final sample size of 136 and a total number of 408 ultrasound assessments.

Ethical permission was obtained from the Human Research Ethics Committees of the academic institutions overseeing the PCHS (Murdoch University, permit No. 2007/238 and the University of Western Australia, approval No. RA/4/1/2414). Informed written consent was obtained from each participant prior to recruitment. Full set of required ultrasound measurements in the three intervals were obtained from all 136 women included in this study.

Ultrasound examinations

Three ultrasound examinations were carried out at 18, 26 and 34 weeks' gestation according to PCHS protocols. Two experienced operators carried out the examinations using a single ultrasound scanner: Toshiba Xario‐XG (SSA‐680A) (Toshiba Medical System Corporation, Tokyo, Japan) equipped with PVT375BT convex transducer with grey‐scale frequency range of 1.9–6.0 MHz and the mid frequency of 3.5 MHz. An assessment of fetal biometry, amniotic fluid index and placental localisation were performed at each visit.

Due to quite wide variations in volume flow, related to changes in fetal state, the following factors were considered to minimise the potential errors relating to changes in venous diameter or flow velocity over time:

Measurements of the umbilical vein was conducted in fetal quiescence, this is due to the fact that during high amplitude fetal respiratory movements, the flow velocity can increase to 54% above the level during apnoea.19 Variations in umbilical venous diameter along the length of the cord is well documented20 and can contribute to variation in the estimated flow volume. Despite the difficulties in standardising the measurements site, those obtained in the free loop have been shown to be reproducible14, 16, 21 as long as they are performed away from the extremes of the umbilical vein that is, away from the fetal and placental insertion sites.16 This method of diameter meassurement has been shown to have an intra‐observer and inter‐observer coefficients of variation of 3.3 and 2.9% respectively.14

Ultrasound scan technique

The intra‐amniotic portion of the umbilical cord was examined for quantitative assessment of the umbilical flow volume. Measurements of the umbilical vein diameter were obtained during fetal quiescence. The free floating section of the cord at approximately mid‐way between the fetal and the placental cord insertions was imaged. This allowed for the variations in the cross section of the umbilical vein along the length of the cord.20 A longitudinal section of a straight portion of the vein was obtained13, 22 and the image was magnified to occupy approximately more than 30% of the image, without altering the resolution.15, 21 Care was taken to avoid segments of the cord compressed by any fetal parts, as this can compress the umbilical vein and alter its internal diameter. The diameter was measured perpendicularly from inner wall to inner wall across the vessel lumen (Figure 1). The vessel diameter was measured where the cross sectional perimeter of the vein had a circular shape to minimise errors in calculating the cross sectional area of the vessel.23 The average of three measurements was recorded.14, 21

Figure 1.

The position of measurement callipers for the measurement of the internal diameter of the umbilical vein.

To measure the flow velocity, the scanning plane was tilted at 90° and the sample gate positioned to cover the entire lumen of the vein (Figure 2). The flow velocity measurement was obtained with the slight tilt of the transducer to provide Doppler angles of 15° or less. The flow velocity was measured in a long section of the vein where the lumen of the vein was linear for at least approximate length of 3 times its internal diameter.24 This method ensured that the flow waveform had a parabolic profile, a prerequisite for accurate estimation of flow volume. The Doppler beam angle was set close to zero (where possible) or at less than 15°.12, 16, 25 Angle correction was applied when necessary. The waveforms were obtained continuously for 3–5 s. The waveforms were enveloped for automatic calculation of the time‐averaged maximum flow velocity (Figure 3). The images provided are chosen as the best example of the measurements they were aimed to depict and they were not from the same fetus.

Figure 2.

The position of Doppler sample gate for measurement of umbilical venous flow velocity.

Figure 3.

The umbilical venous flow waveforms and its measured velocity.

The velocity values were recorded as means of three different measurements.12 The flow volume was then calculated using the mean internal diameter and time‐averaged flow velocity in the formula below:

$$Q(\text{mL/min})={\text{Velocity}}_{\mathrm{mean}}(\text{cm/s})\times \mathrm{\pi }{(\text{diameter}(\text{cm})/2)}^2\times 60$$

The inter‐observer and intra‐observer agreements were tested in 18 participants. Each operator measured the internal diameter of the umbilical vein and its time‐averaged maximum velocity three times. The second operator then repeated the same measurements for each fetus. The two operators were blinded to each other's measurements. Birth data were also collected for all women following delivery.

Statistical analysis

Statistical Package for Social Sciences for Windows (PASW Statistics for Windows, Version 21.0: SPSS Inc., Chicago, IL, USA) was used for all statistical analyses. ANOVA tests for repeated measures were carried out for venous diameter, velocity and flow volume. There were statistically significant differences in means of above three values between the three intervals (P < 001).

To investigate differences in the growth pattern of the umbilical venous diameter across gestation, the ratios of the diameter measured at the second ultrasound assessment (26 weeks gestation) to those measured at the first ultrasound assessment (18 weeks gestation) were obtained for each woman (ratio 1). This process was repeated for the ratio of diameter at the third ultrasound assessment (34 weeks' gestation) to the second ultrasound assessment (ratio 2). The mean, standard deviation, standard error and the 95% confidence interval of the mean for those ratios were also obtained. A paired‐sample t‐test was performed to test the significance of the mean difference between the ratios.

Measure of agreement between operators

To measure the agreement between two ultrasound operators (labelled as observers A and B) intra‐class correlation coefficient (ICC) tests were performed. To assess the inter‐observer reliability, a two‐way random model (women and observers conceived as random samples), single‐measure ICC for absolute agreement was calculated.21 To assess the intra‐observer reliability, a two‐way mixed model (women conceived as random samples) single‐measurement ICC for absolute agreement was calculated for observer A and B.

Results

The median age of women was 30 years. Ninety‐three per cent were Caucasian. There were 32% nulliparous women and approximately 10% were smokers. Birth data were available for all 136 participants. All fetuses were live born and had a three vessel umbilical cord. There were 73 male (54%) and 63 female neonates (46%). The mean birth weight and gestation at birth were 3487.7 ± 484.4 g and 38.9 ± 1.2 weeks respectively.

The mean internal diameter of the umbilical vein at 18 weeks was 2.8 mm (SD 0.4), which increased to 5.8 mm (SD 0.7) at 26 weeks and 7.6 mm (SD 1.0) at 34 weeks' gestation. The percentage increase in the diameter between 18 and 26 weeks' gestation was approximately 107% compared with only 31% from 26 to 34 weeks' gestation (Figure 4). The mean flow velocity measured 14.7 cm/s (SD 2.7) at 18 weeks' gestation which increased slightly to 16.7 cm/s (SD 3.6) at 26 weeks and 19.5 cm/s (SD 3.6) at 34 weeks' gestation. The increase in the flow velocity between 18 and 26 weeks' gestation was approximately 14% compared to 17% between 26 and 34 weeks, which unlike the diameter showed a smaller increase in the first 8 weeks compared to the second 8 weeks (Figure 5). The flow volume increased from 28.7 mL/min (SD 10.3) at 18 weeks to 131.5 mL/min (SD 38.1) at 26 weeks and 269 cm/s (SD 82.4) at 34 weeks' gestation (Figure 6).

Figure 4.

Mean (95% CI) umbilical venous diameter at three stages of gestation.

Figure 5.

Mean (95% CI) umbilical venous flow velocity at three stages of gestation.

Figure 6.

Mean (95% CI) umbilical venous flow volume at three stages of gestation.

The mean of ratio 1 was 2.06 (SD 0.37, 95% CI 2.01–2.14). Ratio 2 measured with mean of 1.33 (SD 0.19, 95% CI 1.29–1.36). The mean difference between the ratios was 0.75 (SD 0.48, 95% CI 0.67–0.83). The paired‐sample t‐test to determine the significance of the mean difference between the ratios yielded a P < 0.001 and revealed a significant difference in growth of the diameter between the two periods across the gestation. Therefore, the growth in umbilical venous diameter from 18 to 26 weeks was proved to be significantly different to the growth between 26 and 34 weeks gestation.

Looking at the means of ratio 1 and 2, and their respective confidence intervals, the doubling of the size of diameter was observed between 18 and 26 weeks but only an increase by a factor of 1.3 was noted between 26 and 34 weeks. This statistically confirmed the non‐linear growth pattern of the umbilical venous diameter across 16 weeks of gestation (18–34 weeks) as observed in Figure 4.

The inter‐observer ICC for the venous diameter was 0.940 (95% CI 0.850–0.977). The intra‐observers ICCs were 0.953 (95% CI 0.901–0.980) for operator A and 0.985 (95% CI 0.969–0.994) for observer B respectively.

Discussion

The umbilical venous diameter demonstrates a non‐linear growth pattern between 18 and 34 weeks of gestation. The diameter doubled in size between 18 and 26 weeks but grew at a slower rate between 26 and 34 weeks of gestation. Previously published growth charts for umbilical venous diameter did not reveal similar differential growth rate as majority of available graphs had their earliest assessment at around 22 weeks gestation, hence this accelerated growth between 18 and 26 weeks could not be clearly observed.

Our results indicate an almost perfect reliability for venous diameter measurements. The inter‐observer ICC coefficients for venous diameter in our study were higher than those reported by Fernandez et al.21 (0.94 vs. 0.65). Both observers in our study showed an almost perfect reliability with intra‐observer ICC correlation coefficients of above 0.9 which were again higher when compared to those reported by Fernandez et al. (0.7).

Umbilical venous diameter and hence its cross sectional area are reported to be reduced in small for gestational age fetuses with some as early as in the second trimester9, 13 which can contribute to the reduction in the flow volume. While some argue that diameter alone could be a predictor of changes in flow volume,13, 14 others have reported a reduction in flow velocity and only a minimal change in the diameter of the vessel in small for gestational age fetuses.11 Given the consequences on perfusion and metabolism of fetal liver, a reduction in the venous size especially when measured earlier in pregnancy could potentially be a useful early prognostic tool for intrauterine growth restriction (IUGR) in the second trimester. This may either occur in the absence of abnormal arterial Doppler indices or may precede their onset and can serve as an earlier indicator of abnormal intra uterine environment.

Conclusion

The findings of this study suggest that rapid increase in umbilical venous diameter in the second trimester is required for normal fetal development. Further larger longitudinal studies are warranted to assess the value of including umbilical venous measurements in routine ultrasound studies of fetal growth.

Author contributions

The principal and corresponding author and co‐authors Professor Jan Dickinson and Associate Professor Eugen Mattes have equally contributed to study design, methodology and analyses of results. The co‐author Associate Professor Peter Jacoby advised on the study design and was responsible for the statistical analyses of the results. All named authors have equally contributed to the drafting of the article and approve the final version submitted for publication.

Disclosure

All authors declare no conflicts of interest.