Pediatric & Perinatal Pathology Associates, PSC

Umbilical Cord Insertion
Perinatal autopsy lectures
Perinatal Blog
Wordpress blog

Umbilical cord insertion


Bottom line


Marginal or velamentous insertion


 Whether caused by placental migration or misalignment of the embryo and placenta at implantation, umbilical cords sometimes miss the placental surface. Those cords with a marginal insertion are often referred to as Battledore, while those with visible vessels in the membranes are referred to as velamentous (1% of insertions). There are two potential sets of complications. The best documented is the risk of vessel rupture with vasa previa when velamentous vessels cross over the cervical os. Velamentous vessels may also occur with a normal insertion of the umbilical cord and they occur between any placental lobes. The second set of complications is that the vessels are exposed to injury or thrombosis, and possible decreased blood flow evidenced by lower birth weight. It is not immediately obvious why they would be more vulnerable to injury than vessels on the surface of the placenta as long as the membranes are attached to the wall of the uterus.

Velamentous insertions are much more common in twins, even passing through the septum between the twins. The evidence that they are more common with fetal malformation is debatable and requires further clarification.


Furcata or interposed insertion:


In these insertions the vessels leave Wharton’s jelly before reaching the surface of the placenta and course through the amniotic cavity. In furcata the vessels appear naked with a “fork like” appearance. Interposed vessels have a thin fold of membrane draped over the vessel forming a tent anchored to the fetal surface. While I can’t formally explain the physics, twisting the umbilical cord twists and collapses these free vessels. This type of insertion could lead to occlusion of fetal blood flow and fetal death or asphyxial injury. The distance of uncovered vessels can vary from millimeters to centimeters, and the shorter instances may not be vulnerable to twist. The longer free vessel examples are rare, and the published literature is anecdotal without a large population study.


Severed velamentous vessels


An infant can hemorrhage from torn velamentous vessels with varying amounts and rapidity. Most such cases are associated with vasa previa. Whether such deaths might be preventable by testing vaginal hemorrhage for fetal blood cells or by other interventions is controversial. Velamentous vessels are commonly severed during Cesarean section or the third stage of labor, and are not significant. We inspect all velamentous vessels. If there is a torn segment, this is sampled histologically to look for fibrin or acute inflammation which indicates longer duration. We also correlate any pallor or loss of blood in the placental vessels and any increase in nucleated red cells.


Literature review


Risks and associations other than vasa previa from non-central umbilical cord insertion


            A prospective study of 1000 placentas failed to demonstrate a significant association of the location of the umbilical cord insertion with low birth weight, intrauterine fetal death, threatened abortion, premature delivery, fetal malformation or intrauterine fetal hypoxia (not further defined)1. They lumped marginal and velamentous cords into a single “peripheral insertion” category and did no analyze them separately. This category had 72 placentas, approximately 7%, while only 1% would have been expected to be velamentous.


A clinical study of velamentous insertion in 12,750 high risk patients found 216 cases, excluding twins and malformations or aneuploidy2. There was a significant but small increase in velamentous insertions in primigravidas, and not surprisingly a decreased incidence with prior Cesarean sections. (A primigravida by definition could not have had a prior Cesarean section, but this is not discussed in the paper). Of the pregnancy and delivery variables, there were 7 abruptions in the velamentous insertions (3% versus .75% P <.001). There was also a statistical increase in bloody amniotic fluid. The authors state that there were no cases of vasa previa. It strains credulity to believe that none of those abruptions were misdiagnoses of cases with vasa previa and fetal hemorrhage. In a logistic regression that hopefully compensated for the increases incidence of preeclampsia in primigravidas, there was a significant but small increase in infants delivered before 37 weeks of gestation, and of those with a birth weight <2,500 g. Of more concern is an increase in 5 minute Apgar scores less than 7 (assuming prematurity is controlled for), and abnormal FHR patterns, and borderline significance for venous pH less than 7.15. Were these cases of vasa previa with hemorrhage or compression of vessels that accounted for the differences between groups. In 48 cases with fetal Doppler studies, all abnormalities were believed due to preeclampsia. There was no direct evidence of flow abnormalities from the velamentous vessels per se. Only one velamentous cord insertion of 80 cases was detected by routine ultrasound (done from 1989 to 1993).


A review of 15,865 singleton deliveries, using the clinical record of the obstetrician’s observation of the delivered placenta, found 77 (0.5%) velamentous cord insertions3. There were 268 twins with 5 velamentous insertions (2%). Of 82 velamentous insertions, 3 were vasa previa (0.02% of deliveries). The mothers with velamentous insertion were more likely to be older than 35 years, more likely to be primiparous, and less likely to have a prior Cesarean section. (The last two variables are not independent!). Of the outcome variables, there was a significant increase in meconium stained fluid, blood stained fluid, abruption placenta, variable decelerations, and intervention with forceps or vacuum extractor. No prenatal ultrasound examinations including the cases with vasa previa had diagnosed velamentous cord insertion.


            A study of 148 placentas measured by computer the distance of the insertion from the center of mass of the placenta as well as the shortest distance to the margin and the longest4. The study tested whether the scatter of measurements were consistent with an attempt by the body to target the center of the placenta, or consistent with relative migration of the placenta after the insertion was centrally placed (trophotropism). The findings were more consistent with the former, that is insertions were random about the center. The author expected that if the eccentricity of the insertion were due to placental migration then the displacement of the cord insertion would have been along the long axis of the placenta. The study does not disprove such migration from having some effect especially in twins or low implantations of the placenta. The mathematics is difficult to follow and I can not independently confirm the conclusions, and the details of how the program was done are not included in this 1968 study. The data fit very tightly an equation (longest distance to margin from insertion) over (the shortest distance from the insertion) equals the (distance between the insertion and the center of mass) over (the shortest distance from the insertion to the margin) plus .  This equation holds for circles and some sets of convex surfaces. I suspect that this striking result is just a consequence of the procedure of determining and measuring these distances, but I am not sure.


A retrospective study of 183 “products of conception,” 6 of 18 (33%) specimens from 9 to 12 weeks of gestation had a velamentous cord insertion, 9 of 34 (26%) specimens  from 13 to 16 weeks of gestation and in 7 of 63 (11%) specimens from 17 to 38 weeks of gestation5. “Of the 28 specimens with velamentous insertion of the cord, the placenta was distinctly formed in 10.”  7 of the 28 (25%) velamentous insertion specimens had malformation, compared to 12 of 155 (8%) with normal insertion, which was a significant difference. Only one velamentous insertion was in a twin. The author distinguishes insertion velamentosa from interposition velamentosa. The former has an arterial anastomosis before branching in the membranes; the latter has unbranching arteries in the membranes that anastomose in the placenta. The author argues that the increased incidence of velamentous insertion is evidence that the insertion is a detriment to survival. He gives a detailed discussion of possible mechanisms of development of velamentous insertion.  He postulates that the difficulty of maintaining circulation may be the cause of the higher incidence of malformation. [He does not give credence to the possibility that placental growth might center some early cases of apparent membranous insertion.]


A Finnish study comparing 355 patients with elevated serum screening human chorionic gonadotropin to 4935 with normal values found a doubling of the incidence of velamentous cord insertion, as well as approximately a doubling of the risk of preeclampsia6. The 95% confidence interval is 1.47 – 4.69. Another Finnish study of 170 cases of placental abruption found a doubling of velamentous umbilical cord insertion7. A third Finnish study found an increased incidence (7.5%) of velamentous cord insertion in women with placenta previa8. There was no biological rationale offered for these associations.


Vasa Previa


            Two cases of fetal death from vasa previa, one with a torn velamentous vein, and the other with intact velamentous vessels associated with a bilobed placenta were presented with a review of the medical literature from 1801-19519. The author suggests looking at the blood from vaginal bleeding for nucleated red cells in hemorrhage occurring before or at the rupture of membranes and associated with fetal heart rate abnormalities. The fetal mortality was greater than 50% and the lesion was seldom diagnosed prior to delivery of the placenta. Subsequent reviews have added very little new information.

            Three cases of vasa previa presented in 1965 demonstrated fetal death following delivery from a ruptured vessel, fetal survival without hemorrhage but with cord compression pattern of the fetal heart rate, and survival from hemorrhage that was diagnosed as vasa previa by the Apt test prior to delivery10.

            A comprehensive review of vasa previa was associated with a case report in 198011. Vasa previa is defined as vessels unsupported by placenta or Wharton’s jelly that cross over the cervical os in front of the presenting part. The author notes that Torrey defined two kinds of hemorrhage from velamentous vessels, those with bare vessels crossing the cervical os, i.e. vasa previa, and those that rupture without crossing over the os. Once fetal hemorrhage has occurred the fetal prognosis is poor with a mortality greater than 50%. Neither tests for fetal blood in vaginal blood, nor amnioscopy were often cited in cases in which antepartum or intrapartum diagnosis was made. Palpation of the vessels was often cited. There have been many associated finding with velamentous insertion and vasa previa including single umbilical artery and abnormal form or location of the placenta. 


A case report of vasa previa with deep variable decelerations leading to Cesarean section demonstrated long, large, unbranched velamentous vessels that were seen over the cervical os at Cesarean section12.


A report of 4 cases of vasa previa emphasized that in 2 cases the vessel rupture and hemorrhage can occur prior to rupture of membranes13.  Two of the cases were due to velamentous vessels between placental lobes.


Seven cases of vasa previa were reported from 35,000 deliveries14. They estimated 1% velamentous placentas which resulted in a calculated 1 vasa previa for every 50 velamentous cord insertions. Five of their cases were diagnosed prepartum, three did not have ruptured vessels. Two of the infants with hemorrhage and antepartum diagnosis died, but one with prompt transfusion survived.


Velamentous cord insertions in twins


A study of 38 diamniotic, monochorionic twins demonstrated not only increased velamentous cord insertions in comparison to singletons, but also statistically significantly increased, 7 of 11 (64%), in twin to twin transfusion syndrome (defined as polyhydramnios/ oligohydramnios without other detectable cause) compared to 5 of 27 (18%) diamniotic monochorionc twins without transfusion, P <.0115. The cord from the smaller twin was inserted in the septum in 3 of the transfusion cases. In another the smaller twin’s cord inserted into the marginal membranes, and in 3 the site was undocumented. In the twins without transfusion syndrome the cord inserted in the marginal membranes in 3, and the site was undocumented in two. The authors argue that pressure from the polyhydramnios on the velamentous vessels was a factor in unbalancing the twin circulations and at least increased the effect of twin to twin transfusion. [I do not understand why the increased pressure would not have the same effect on all placental surface vessels. Perhaps those over the placenta respond differently because of their branches to the intervillous space? The hypotheses that the correlation is accidental or that the mechanism is more indirect and complex have not been eliminated]


A retrospective study of 64 diamniotic, monochorionic twin placentas, there was a 9 fold relative risk of velamentous umbilical cord insertion compared to having normal insertions on both disks in infants with birth weight discordancy16. There was no increased risk for dichorionic twins with birth weight discordancy. [This study is published only in abstract]


1.         Uyanwah-Akpom P, Fox H. The clinical significance of marginal and velamentous insertion of the cord. Br J Obstet Gynaecol 1977;84(12):941-3.

2.         Heinonen S, Ryynanen M, Kirkinen P, Saarikoski S. Perinatal diagnostic evaluation of velamentous umbilical cord insertion: clinical, Doppler, and ultrasonic findings. Obstet Gynecol 1996;87(1):112-7.

3.         Eddleman KA, Lockwood CJ, Berkowitz GS, Lapinski RH, Berkowitz RL. Clinical significance and sonographic diagnosis of velamentous umbilical cord insertion. Am J Perinatol 1992;9(2):123-6.

4.         McLennan JE. Implications of the eccentricity of the human umbilical cord. Am J Obstet Gynecol 1968;101(8):1124-30.

5.         Monie IW. Velamentous Insertion of the Cord in Early Pregnancy. Am J Obstet Gynecol 1965;93:276-81.

6.         Heinonen S, Ryynanen M, Kirkinen P, Saarikoski S. Elevated midtrimester maternal serum hCG in chromosomally normal pregnancies is associated with preeclampsia and velamentous umbilical cord insertion. Am J Perinatol 1996;13(7):437-41.

7.         Toivonen S, Heinonen S, Anttila M, Kosma VM, Saarikoski S. Reproductive risk factors, Doppler findings, and outcome of affected births in placental abruption: a population-based analysis. Am J Perinatol 2002;19(8):451-60.

8.         Papinniemi M, Keski-Nisula L, Heinonen S. Placental ratio and risk of velamentous umbilical cord insertion are increased in women with placenta previa. Am J Perinatol 2007;24(6):353-7.

9.         Torrey WE, Jr. Vasa previa. Am J Obstet Gynecol 1952;63(1):146-52.

10.       Naftolin F, Mishell DR, Jr. Vasa previa. Report of 3 cases. Obstet Gynecol 1965;26(4):561-5.

11.       Kouyoumdjian A. Velamentous insertion of the umbilical cord. Obstet Gynecol 1980;56(6):737-42.

12.       Cordero DR, Helfgott AW, Landy HJ, Reik RF, Medina C, O'Sullivan MJ. A non-hemorrhagic manifestation of vasa previa: a clinicopathologic case report. Obstet Gynecol 1993;82(4 Pt 2 Suppl):698-700.

13.       Carp HJ, Mashiach S, Serr DM. Vasa previa: a major complication and its management. Obstet Gynecol 1979;53(2):273-5.

14.       Quek SP, Tan KL. Vasa Praevia. Aust N Z J Obstet Gynaecol 1972;12(3):206-9.

15.       Fries MH, Goldstein RB, Kilpatrick SJ, Golbus MS, Callen PW, Filly RA. The role of velamentous cord insertion in the etiology of twin-twin transfusion syndrome. Obstet Gynecol 1993;81(4):569-74.

16.       Hanley M, Shen-Schwarz S, Samulian J, Anath C, Lai-Ling Y, AM V. Placental cord insertion in relation to birth weight and placental weight discordancy n twin gestations. Am J Obstet Gynecol 1997;176:S134.



Enter supporting content here