In depth

The Umbilical Cord and Stillbirth

Richard M. Pauli, M.D., Ph.D.

All have navels, save Adam and Eve.

While superficially simple in both structure and function, the umbilical cord, of course, serves as a critical lifeline to the developing fetus. On occasion a variety of intrinsic or accidental processes involving the umbilical cord can result in intrauterine fetal death. This is a brief review of what is known about the umbilical cord particularly as it relates to stillbirth.

Embryology, Anatomy and Function. The definitive umbilical cord begins to form in the 5th week of embryonic life. It is derived from early embryonic structures -- the yolk sac and the allantois -- which otherwise are, for the most part, vestigial structures in placentated mammals. Initially two arteries and two veins arise from the primitive allantois, but by around the 6th week of gestation the right umbilical vein is obliterated.

The umbilical cord is structurally simple. It is composed of two arteries, one vein, tiny rudiments of the allantois and the omphalomesenteric ducts (see below) and Wharton's jelly. The arteries shunt blood to the placenta while the single vein returns newly oxygenated blood to the embryo/fetus from the placenta. Wharton's jelly seems to be a substitute for the adventitia of other arteries; that is, the umbilical arteries have no adventitial support and this gelatinous connective tissue serves to provide a flexible structure which at the same time prevents accidental compression of the arteries.

Developmental Anomalies. Leftovers of the structures from which the umbilical cord is derived may persist anomalously within the cord. The allantois gives rise to the cloaca and part of the bladder; in addition it extends into the body stalk eventually giving rise to a portion of the cord. Abnormal persistence of

the allantoic connection between the bladder and the umbilical cord is a patent urachus. It is present in about 1 in every 1000 newborns. The yolk sac is incorporated into the midgut (as well as being pinched off by the developing body stalk which develops into the umbilical cord). Failure to completely pinch off this communication allows an omphalomesenteric duct to persist through which there is a gut to umbilicus communication. It occurs in about 1 in every 20,000 babies. Neither of these are associated with stillbirth.

The intestines normally enter into the umbilical cord at around 4 weeks gestation and reenter the abdomen only at about 10 weeks gestation. That is, cord herniation is a normal developmental stage. Failure of reentry results in an omphalocele, while failure of fascial closure after reentry causes an umbilical hernia.

Single Umbilical Artery. In all deliveries, around 1 in 200 cords will be found to have only one umbilical artery (range in various studies: 0.27% to 1.13%, with the best controlled study showing a rate of 0.44%). The presence of a single umbilical artery (SUA) is associated both with an increased risk for other malformations and an increased risk of intrauterine death.

Many studies have asked how frequently infants with SUA have significant birth defects. Estimates range from 7% to 48%. Overall these data cluster around a 20-25% incidence of significant birth defects, or around a seven fold increase compared with babies with two umbilical arteries. SUA does not cause these other malformations. Rather, aplasia of an umbilical artery should be considered a minor malformation and, like other minor malformations, it is frequently seen in association with other more significant anomalies, including, for example, caudal regression, the VATER association, Lower Mesodermal Defects sequence, trisomy 13 and trisomy 18.

A greater proportion of infants with SUA die in utero. This is not meant to imply that SUA is directly causal. Rather, it seems that most often associated abnormalities are more directly linked to the stillbirth of infants with SUA. Nonetheless, SUA is about 10-15 times as frequent in stillborns as in liveborns (with the best controlled study showing 95 of 877 of stillborns, or 10.8%, having SUA; note, however, that two-thirds of these had other anomalies).

Length and Size. Leonardo DaVinci was the first to study the length of the umbilical cord, concluding that its length equaled the length of the fetus throughout gestation. This remains a roughly accurate rule of thumb. At term the average umbilical cord is 59-60 cm long (slightly shorter in plural births; no significant difference among races). Of course, this varies with gestation.

Umbilical Cord Length by Gestational Age

At term a cord shorter than 35 cm or longer than 85 cm is considered to be abnormal.

Both short and long cords may be associated with intrauterine problems. A cord less than about 32 cm may increase the likelihood of placental abruption (since this is the average distance from fundus to vulva). However, there is no demonstrable association between a short cord per se and the frequency of stillbirth. There is evidence that shorter than normal cords are correlated with abnormal neurologic outcomes in liveborns. No one has yet sorted out whether intrinsic neurologic abnormality results in short cords (the so-called stretch hypothesis of cord growth) or if intrapartum complications of the short cord are causing the neurologic disability. Long cords (greater than 85 or 90 cm) may increase the

probability of cord entanglement, prolapse and true knot development. Perhaps surprisingly, then, epidemiologic data indicate that such long cords are not particularly frequent among stillborns.

Standards are also available for cord diameter (1.0 to 2.0 cm at term), cord circumference (2.3-5.1 cm at term) and periumbilical skin length (4-19 mm).

Cord Insertion. The site of insertion of the cord into the placenta is determined at implantation. It may be central, eccentric, marginal or velamentous. None of the first three have any significance with respect to stillbirth. A velamentous insertion is one in which the umbilical vessels run in between the amnion and chorion before entering the placenta. Around 0.5% of cords have such a velamentous insertion. One might suspect that there would be very high risk for compression of the vessels and for rupture during labor. In fact, the vast majority of infants (even those with vasa previa) are delivered without incident. Nonetheless occasionally (3 of the first 1000 in our series) stillbirth arises because of vessel rupture and exsanguination. Others have estimated that between 2% and 8% of pregnancies complicated by velamentous insertion will end in stillbirth.

Twists and Supertwists. Most umbilical cords show a twist. This coiling probably provides greater strength while retaining necessary flexibility. Presumably this coiling arises because the longer umbilical vein twists around the umbilical arteries. Of the 95-97% of cords which are coiled, 87% show a left handed twist (like a left-handed screw). The only twist-associated process which seems relevant to stillbirth is absence of any coiling. Various series have shown frequencies of 10-18% for untwisted cords in stillborns (around 4 times the rate in liveborns). This doesn't mean that lack of twisting is causal of intrauterine death. In fact, no one has any idea what the sequence of events is that connects untwisted cords and stillbirth.

Supertwisting (superimposition of a second level of coiling) is probably rarely or never of significance. In stillborns, movement after death can cause twisting at the fetal end of the cord resulting in such a supertwist which is not of any pathologic relevance.

Constrictions and Knots. A constriction or coarctation of the cord is frequently observed near the fetal skin surface in stillborns (below, top). In the past it's been thought that this caused the stillbirth. In fact, however, such constrictions are usually simply a loss of Wharton's jelly and are rarely genuine. Pathologic assessment usually shows no compromise of the vessels and no obstruction of flow. Likely these arise as a secondary consequence of death in utero.

False knots (aneurysmal blebs) of the cord are never significant. True knots (left, bottom), on the other hand, may sometimes be tight enough to cause obstruction of blood flow and in that manner result in stillbirth. Just because a true knot is present does not mean that this is the cause of a stillbirth, however. Between 0.3% and 2.1% of all cords in liveborn babies have true knots. Pulsation within the cord will usually keep knots loose and clinically insignificant. Even a tight true knot doesn't in itself mean that a baby has died because of the knot. After death, pulsations cease and so what was a loose and inconsequential knot premortem may be tight and apparently important at the time of delivery. If obstruction has occurred prior to death then there will be secondary evidence of these abnormalities of flow: the diameter of the cord on the placental side of the knot will increase; on the fetal side the diameter will often diminish; there may be evidence for old hemorrhage at the sight of the knot; histologic examination will show venous distention on the placental side and arterial hemorrhage on the fetal side. Only when such pathologic changes are present should a true knot be considered causal. In our series, 7 of the first 1000 assessments suggested that intrauterine death was caused by such true knots.

Vessel Rupture and Vessel Thrombosis. Rupture of an umbilical vessel is rare -- estimated to occur about once in every 6000 deliveries. It is more frequent with velamentous insertion. When it does occur, it usually results in catastrophic exsanguination and a fetal mortality rate of about 50%. Three of the first 1000 infants we helped assess had exsanguination from a ruptured cord vessel.

Umbilical vein thrombosis is seen in about one in every 1300 deliveries. It is much more common in stillbirths. However, it rarely is seen in isolation and is most frequently a post mortem artifact. Artery thrombosis is extremely rare.

Entanglements. Wrappage of the cord around the neck (nuchal cord) is common. Single loops are present in around 20% of all births, double loops in about 2% and triple loops in

approximately 0.2%. Population studies do not demonstrate any general increase in mortality associated with such wrappage. Stillbirth can only be attributed to a nuchal cord if the entanglement is so tight that there

is obstruction of flow through the cord. (Obviously, death won't arise just because the part entangling the cord is compressed). Associated findings which support the notion that an entanglement is the cause of intrauterine fetal death include deep grooving of the skin plus evidence for premortem obstruction of blood flow in the cord (as discussed related to true knots). Six of 1000 evaluated babies in our series died secondary to nuchal compression of the cord.

The Umbilical Cord and Stillbirth. Pressed to find some explanation for a baby's death, medical personnel often are tempted to attribute death to the most obvious feature at hand -- a twisted, knotted, short, long, or constricted umbilical cord. In fact, about 30% of fetal death reports submitted to the State list placental and cord processes as etiologic in stillbirths. We have previously shown that this is a profound overestimate and that only around 1/3 of those instances actually have placental or cord based processes. In the first 1000 infants we assessed, 10.9% of stillbirths could be attributed to placental or cord problems of which 28 were abnormalities or accidents involving the cord itself.

Cord Associated Causes of Stillbirth in the First
1000 Referrals to WiSSP

Recommended Cord Evaluation in Stillborns. Although only about one in every 40 stillbirths arises because of cord processes, nonetheless the umbilical cord should be routinely evaluated in every stillborn including notation of the number of vessels, total length, and explicit note whenever any of the following are present: nuchal entanglement, velamentous insertion, areas of constriction, vessel rupture, true knots. If death might have been secondary to obstruction of flow through the umbilical vessels (true knot, nuchal cord etc.) then, in addition, circumference of the cord on the fetal and on the placental side of the proposed area of obstruction should be measured, and histologic study of areas proximal to, at and distal to the area of apparent obstruction should be completed .

Further Reading*

Kaplan C (1993): Placental pathology for the nineties. Pathol annual 28:15-72.

Lacro RV, Jones KL, Benirschke K (1987): The umbilical cord twist: Origin, direction, and relevance. Am j obstet gynecol 157:833-838.

McLennan H, Price E, Urbanska M, Craig N (1988): Umbilical cord knots and encirclements. Aust NZ J Obstet Gynaecol 28:116-119.

Naeye RL (1987): Functionally important disorders of the placenta, umbilical cord, and fetal membranes. Hum Pathol 18:680-691.

Naeye RL (1985): Umbilical cord length: Clinical significance. J Pediat 107:278-281.

Strong TH, Elliott JP, Radin TG (1993): Non-coiled umbilical blood vessels: A new marker for the fetus at risk. Obstet gynecol 81:409-411.

*Copies of these and other relevant articles are available for personal use by request from WiSSP.