Search :     
Articles » Cord » 2-vessel cord
1992-06-17-07 Umbilical vessels: visualization © Richards
Umbilical vessels: visualization

Douglas S. Richards, MD, S. Gail Allen, RDMS, Margaret A. White, RDMS, Debbie R. Perez, RN


MESH Umbilical arteries-abnormalities, -pathology, -physiopathology; Umbilical-Cord-pathology BDE 2500 ICD9 747.5 CDC 747.500

Address correspondence to: Douglas S. Richards, MD, Department of Obstetrics and Gynecology, University of Florida College of Medicine, Box 100294, Gainesville, Florida, 32610-0294 Ph: 904-392-2894; Fax: 904-392-6994


A single umbilical artery is associated with an increased risk of pregnancy complications. In about 10-20% of pregnancies with a single umbilical artery, there is a fetal malformation1. Intrauterine growth retardation and an increased risk of perinatal mortality have also been associated with a single umbilical artery1,2. Because of the important implications of this finding, a simple technique for visualization of two umbilical arteries as part of a basic ultrasound examination is desirable. When specifically looked for, a single umbilical artery can be identified with gray scale imaging in a high percentage of cases after 16 weeks3. It has been found in others" experience that, even in a targeted ultrasound examination, a single umbilical artery is often missed4,5. Because of such factors as tortuosity of the cord and reverberation artifact within the amniotic fluid, it is not always easy to obtain a satisfactory cross sectional view of the cord within the amniotic cavity to document the presence of two arteries. We have noted, as have others3,6, that during an axial survey of the fetal abdomen, the umbilical arteries are easily seen merging toward the midline in the lower abdomen, before exiting through the umbilical cord.

We designed a study to determine whether it is easier to image two umbilical arteries with images through the fetal lower abdomen or with cross-sectional images of a free cord loop in the amniotic fluid. We also studied the effect of maternal factors, which might influence the ease with which a three-vessel cord is identified.

Material and methods

Two hundred-fifty patients referred for prenatal ultrasound exams were studied. All exams were done on General Electric RT 3200 machines with 3.5 and 5.0 mHz curvilinear transducers. The only reason for exclusion was the presence of a fetal anomaly. Prior to entering the exam room, the sonographer was given a randomization card, which determined which technique was to be used. Immediately upon starting the examination, the sonographer recorded the start time. He or she then obtained an image documenting two umbilical arteries by either the free cord loop or abdomen technique. The free cord loop technique used a cross section of the umbilical cord floating in the amniotic cavity, showing two arteries and the larger umbilical vein.

The abdomen technique involved scanning axially through the fetus to the cord insertion, then angling the trans­ducer toward the bladder, at which point the umbilical arteries could be seen side by side (fig. 1).



Fig. 1: A typical image obtained by the abdominal technique of vessel identification. The two umbilical arteries (a) are seen along the bladder and then exiting the lower abdomen side by side.

When an appropriate image was obtained, a picture was taken to document the finish time. The time required to document the presence of two arteries was then calculated by comparing the start and finish times. Although this study design does not duplicate the usual flow of an obstetrical exam, it was chosen because it gives clear endpoints, without confounding from “accidental” visualization of umbilical arteries before timing is begun.

 Other data collected on each patient included the patient"s weight (her recalled weight from her last clinic visit), abdominal wall thickness (measured from the skin surface to the interior uterine wall), gestational age (the best estimate based on last menstrual period, prior to ultrasound exams, and/or the current ultrasound), and amniotic fluid index (the sum of the maximum vertical pocket measured in each uterine quadrant). The effect of the method used, effects of other independent variables, and interaction between variables are determined by multiple regression analysis. Because the time required was not normally distributed, a logarithmic transformation was made of this dependent variable prior to regression analysis. Proportions were compared by Chi Square.


Of the 250 patients enrolled, seven were excluded because of incomplete data recording. Patient characteristics are shown in Table 1.

Table 1: Patient characteristics (mean + 1 standard deviation)


Abdominal view (n=125)

Amniotic fluid view (n=118)


Maternal weight (lbs)

158 + 35

157 + 38

157 + 37

Maternal abdominal wall thickness (mm)

25 + 9

23 + 7

24 + 8

Amniotic fluid index (cm)

14 + 4

13 + 4

13 + 4

Weeks gestation

26 + 7

26 + 6

26 + 7


There was no statistically significant difference between the two study groups for any of these characteristics. There were twelve patients in the abdomen group and three patients in the free cord loop group in whom the cord vessels could not be satisfactorily imaged in 60 seconds (statistically significant difference, p=.04). These twelve patients from the abdomen group tended to be obese (mean weight 169 lbs.) and were found across the spectrum of gestational ages. There was a statistically significant effect of abdominal wall thickness but not maternal weight on the log of time required (p=.03 for wall effect). Paradoxically, visualization became more difficult at the lower extreme of wall thickness (fig. 2, top).


Fig. 2: Relationship between the maternal abdominal wall thickness, the gestational age and the amniotic fluid index and the time required to visualize umbilical arteries by both techniques. The number of patients in each group is indicated above the bar. 

Three patients in the free cord loop group in whom the cord could not be imaged in 60 seconds tended to be thin (mean weight 129 lbs) and were scanned at fairly early gestational ages (14, 21, and 27 weeks).

 There was minimal overall difference between the two methods on the time required (median 17 seconds for abdominal versus 16 seconds for free cord loop); however, there was a highly significant interaction between gestational age and the method used (p=0.0001). The abdominal technique was slightly easier before 22 weeks (fig 2, middle), but by late gestation the free cord loop technique had a clear advantage. It was significantly easier to visualize the cord with increased amniotic fluid volumes by both methods (fig. 2, bottom, p=.03).


A single umbilical artery is found in 0.7% of deliveries1. This entity is found in 2.5% of spontaneously aborted fetuses7, is five times more frequent in low birth weight infants, and five times more common in infants of diabetics2. Bryan and Kohler found that when a single umbilical artery was present, there was a 16% prematurity rate and a 34% rate of intrauterine growth retardation1. Approximately twenty percent of infants with a single umbilical artery have another malformation1.

In light of the association of a single umbilical artery with perinatal complications, it would seem advantageous to identify this condition with prenatal ultrasound. The prenatal diagnosis of a single umbilical artery was first reported by Jassani in 19808. With improved equipment resolution and operator experience since then, the diagnosis of a single umbilical artery is no longer considered remarkable, but is frequently missed, even with targeted ultrasound examinations done for high risk conditions. In 450 patients scanned for intrauterine growth retardation or suspected anomalies, Hermann found 6 cases of a single umbilical artery but missed three. In a review of 107 fetuses with CNS abnormalities, Nyberg identified six cases in which a two-vessel cord was seen prospectively, six cases in which the finding was apparent in a retrospective review of films, and eight cases in which the finding was missed entirely4.

Lee et al assessed their ability to visualize a three-vessel cord using the standard technique (cross section of free loop in the amniotic fluid) between 15 and 20 weeks3. Two arteries could be convincingly demonstrated in 66% of cases at 15 weeks, and in 97% at 18 weeks. It appears that in the great majority of cases beyond the early second trimester, two umbilical arteries can be seen if the sonographer makes the attempt to visualize them. It is likely that in the majority of cases where a single umbilical artery is missed, the sonographer was not specifically attempting to image the cord vessels. This point is supported by the series reported by Nyberg, in which eight cases were not identified prospectively but could be identified by retrospective review of ultrasound films4.

If identification of a three-vessel cord is to become part of a basic ultrasound exam, appropriate images must be easily obtained. We had noted that on axial views of the fetal pelvis the umbilical arteries could be easily seen joining in the lower abdomen. We hypothesized that it would be less time- consuming to identify two arteries in this view than to obtain a satisfactory cross-sectional image in the amniotic fluid. We thought that this might be especially true in situations where the amniotic fluid volume was abnormally increased or diminished. Lee et al found that gestational age, obesity, and amniotic fluid volume influenced their ability to visualize three-vessel cords3, so we studied the effect of these factors as well. We did not evaluate the ease of a third technique described by Jeanty6, in which the diameters of the fetal iliac vessels are compared.

There were more women in the abdomen group for whom two arteries could not be visualized. As expected, it was more difficult to visualize the cord vessels with increased thickness of the maternal abdominal wall. Overall maternal weight was not helpful as an independent predictor of difficulty. We found that it was increasingly difficult to image the vessels as amniotic fluid volume decreased, but we encountered no difficulty at the upper end of the fluid range. It should be noted that none of our patients had marked oligohydramnios or polyhydramnios, however. Statistical analysis showed that with increasing gestational age the umbilical arteries were significantly more difficult to image in the fetal pelvis than in free cord loops. For exams done at earlier gestational ages, there was little difference between the two techniques.


1. Bryan EM, Kohler HG: The missing umbilical artery. Arch Dis Child 49:844, 1974.

2. Froehlich LA, Fujikura, T: Significance of a single umbilical artery. Am J Obstet Gynecol 94:274, 1966.

3. Lee W, Rice M, Kirk JS, et al: Single umbilical artery: visualization. The Fetus 1:7475-1, 1991.

4. Nyberg, DA, Shepard T, Laurence A, et al: Significance of a single umbilical artery in fetuses with central nervous system malformations. J Ultrasound Med 7:265, 1988.

5. Herrmann UJ, Sidiropoulos D: Single umbilical artery: Prenatal findings. Prenat Diagn 8:275, 1988.

6. Jeanty P: Fetal and funicular vascular anomalies: Identification with prenatal ultrasound. Radiology 173:367, 1989.

7. Fox H, Elston CW: Pathology of the placenta. London, Saunders, 1978.

8. Jassani MN, Brennan JN, Merkatz IR: Prenatal diagnosis of single umbilical artery by ultrasound. JCU 8:447, 1980.

Help Support :