Nearly 1 in 100 liveborn fetuses are diagnosed with single umbilical artery (SUA)1–3 and yet no consensus exists regarding the clinical relevance of this finding.2,4 Clinicians have long suspected an association between SUA and findings of intrauterine growth restriction (IUGR), preterm delivery, stillbirth, and congenital anomalies. Several studies have detected associations between SUA and adverse fetal outcomes to varying degrees. Some show not only an increased incidence of prematurity and IUGR in SUA babies but also a higher incidence of renal anomalies.5,6 Those findings are refuted by other studies that show no increased risk of IUGR or congenital malformations associated with single umbilical artery.7–9 Many of these studies were limited in their ability to detect the association between SUA and these rare outcomes by small sample size and incomplete follow-up, making it difficult to draw any meaningful conclusions.
We aimed to improve upon the existing published data regarding SUA and its associated findings. We hypothesized that single umbilical artery is associated with fetal growth restriction, fetal demise, as well as renal and cardiac anomalies.
MATERIALS AND METHODS
We performed a retrospective cohort study to estimate the relationship between single umbilical artery and adverse pregnancy outcomes. Our cohort was composed of all consecutive singleton pregnancies undergoing routine anatomic survey between 1990 and 2007 at our tertiary care center. The study was conducted using the institutional perinatal database. Before study initiation, approval by the Washington University School of Medicine human studies review board was obtained.
Women participating in the study had their demographic information, history, and pregnancy outcomes entered into a prenatal database. Since creation of the prenatal database in 1988, all of the patients seen in the prenatal diagnosis center have been followed by dedicated pregnancy outcome coordinators. To achieve complete follow-up information on pregnancy outcome, each patient was given a standardized form at the first visit to be completed after delivery, which detailed pregnancy outcome. The standardized follow-up sheet also included details about pregnancy complications, delivery indications, and neonatal outcomes, including chromosomal and structural abnormalities. All of this information was then entered into the database. If the form was not returned within four weeks of expected date of delivery, the patient received a phone call from an outcome coordinator. If the patient cannot be reached after delivery, the referring physician was then contacted. For patients delivering within our network, outcome information was extracted from the electronic medical record. Gestational age was determined by the first day of the woman's last menstrual period (LMP). If the LMP-estimated due date was consistent (±5 days in the first trimester, ±14 days in the second trimester, and ±21 days in the third trimester) with the due date obtained from growth measurements at the first ultrasound examination then the due date was not changed. If the due dates by LMP and first ultrasound examination were not consistent, then the ultrasound-obtained due date was used to define gestational age. Maternal demographics, past obstetric history, indications for the ultrasound visit, findings from the ultrasound examination or any testing performed, and outcome of the pregnancy were all collected and stored in the database. Patients with incomplete follow-up data were excluded from this study.
Evaluation of single umbilical artery was performed as part of every routine anatomic survey. We defined the primary exposure as the presence of only one umbilical artery identified at that survey. We compared the incidence of adverse pregnancy outcomes in our two study populations: those with a single umbilical artery versus those with two umbilical arteries. Our primary outcomes were IUGR (less than 10th percentile for gestational age10), renal anomalies, and cardiac anomalies. We also studied the association between SUA and spontaneous preterm birth (both before 34 weeks and 37 weeks).
The incidence of SUA in our study population was estimated. Descriptive statistics were used to describe and compared the baseline characteristics of the two study groups (SUA and DUA). Student's t test was used for continuous variables, χ2 for categorical variables, or Fisher exact test for rare categorical variables. Univariable analysis was used to estimate the relativerisk of each outcome of interest (IUGR [less than 10th percentile], renal anomalies, and cardiac anomalies). We performed stratified analyses to identify potentially confounding factors. Logistic regression analyses were used to better estimate the relationship between SUA and our defined outcomes while adjusting for potentially confounding effects. Factors identified by the univariable analyses, as well as those with biological plausibility or historically reported to be associated with the outcomes of interest, were considered in the logistic regression analysis. Year of examination was considered categorically in increments of 4 years, to account for the long duration of the study period. Backward selection was used to reduce the number of variables in the model by assessing the magnitude of change in the effect size of the other covariates. Differences in the explanatory models were tested using the likelihood ratio test or Wald test.11 All variables that were statistically significant were included in the final models. A subgroup analysis to estimate the association between SUA and IUFD, stratified by the presence or absence of IUGR was also performed. Additionally, the individual incidence of the various renal anomalies found in the SUA and DUA groups were estimated and compared with χ2 analysis. All statistical analyses were performed using STATA 10.0 (special edition, StataCorp, College Station, TX).
Of 72,373 singleton pregnancies, 64,047 (88.5%) had complete pregnancy outcome information and were included in this analysis. A total of 392 cases of SUA (0.61%) were diagnosed at anatomic survey. Of these, 281 singleton pregnancies had an isolated SUA and 111 singleton pregnancies had associated malformations (28.3%).
When comparing women carrying a fetus with SUA to those with DUA, the groups were statistically similar with respect to mean maternal age and gravidity. The incidence of preeclampsia and gestational diabetes was also statistically similar between the two groups. However, there were some differences. Women with SUA were more likely to be smokers and have chronic hypertension and preexisting diabetes, while women with two umbilical arteries were more likely to be of African-American race (Table 1).
Single umbilical artery was associated with an over 2-fold increased risk in IUGR compared to pregnancies with DUA after adjusting for smoking, gestational diabetes, African-American race, and preeclampsia (adjusted odds ratio [aOR] 2.1, 95% CI 1.6–2.7, P<.01). After excluding all fetuses with known anomalies other than SUA, there was still an almost twofold increased risk of IUGR (aOR 1.9, 95% CI 1.4–2.5). Single umbilical artery was associated with a more than 20-fold increased risk of cardiac anomalies (RR 18.9, 95% CI 12.9–27.5) after adjusting for pregestational diabetes (aOR 20.3, 95% CI 13.6–30.5, P<.01) when compared with DUA (Table 2). Examination year did not influence the results and therefore did not remain in the models. The diagnosis of SUA also confers an increased risk of spontaneous preterm delivery before both 34 weeks (aOR 3.1, 95% CI 2.1–4.8, P<.01) and before 37 weeks (aOR 2.2, 95% CI 1.6–2.9) (Table 3).
Single umbilical artery was associated with a nearly threefold increased risk of renal anomalies even after adjusting for advanced maternal age and fetal male sex (aOR 3.0 95% CI 1.9–4.9, P<.01). The distribution and type of renal anomaly varied by number of umbilical arteries, but the most common type of renal anomaly was pelvic dilation in each group (Fig. 1).
Finally, to better estimate the relationship between SUA and stillbirth, we excluded all fetuses with known anomalies other than SUA. There were four cases of antepartum stillbirth in the SUA group and 441 in the DUA group (1.5% compared with 0.8%, RR 1.9, 95% CI 0.7–5.1, P=.18). All four cases of stillbirths in the SUA group were also IUGR.
We found that the presence of SUA detected at routine anatomic survey is associated with an increased risk for IUGR, even in the absence of other anomalies. We also found a significantly increased risk of cardiac anomalies and a threefold increased risk for renal anomalies in fetuses with SUA compared with DUA. Our findings differ slightly from some recent reports in the literature.9,12
A case-control study by Predanic et al12 compared 141 singleton pregnancies with SUA to 141 control pregnancies. Their study found no increase in IUGR in the SUA compared to the DUA (7.1% compared with 4.8%, P=.75). However, 57 SUA pregnancies were excluded from the study because the diagnosis of SUA was not verified at the time of delivery. This left their study with limited power to draw significant conclusions about the relationship between SUA and IUGR. Bombrys et al9 performed a case-control study comparing 297 pregnancies with SUA to 297 pregnancies with DUA to estimate the association with IUGR. The difference in risk of IUGR was not statistically significant between the SUA and DUA groups (13.7% compared with 13.9%, P=.93). The interpretation of their study results may be limited, secondary to possible confounding factors that were not adjusted for. An examination of their maternal demographic information revealed that a larger percentage of their SUA population was Caucasian, compared to their DUA population (76% compared with 46%, P<.001). Additionally, the average maternal age of the SUA group was significantly higher than that of the DUA group (26.8±6.2 compared with 24.9±5.9, P<.05). These significant demographic differences between the comparison populations may represent confounding factors that helped bias their results toward the null hypothesis.
In contrast to those two studies, we found an association between SUA and IUGR while adjusting for relevant confounding factors, even when excluding other anomalies. Based on the current American Congress of Obstetrics and Gynecology guidelines for management of IUGR,13 it would be reasonable to consider serial growth ultrasound exams in pregnancies complicated by SUA to screen for IUGR.
When examining renal anomalies associated with SUA, as in those with two umbilical arteries, the most common type of renal anomaly seen with single umbilical artery was pelvic dilation. The second most common finding in both study groups was hydronephrosis. These findings were consistent with the previously published data describing the distribution of renal anomalies associated with SUA.2,13 In the study done by Doornebal and colleagues2 hydronephrosis was the most common finding on renal ultrasound of SUA fetuses. Their study found that routine renal ultrasound screening of all SUA fetuses was unnecessary, given that the most common finding would be hydronephrosis, a diagnosis which bears little clinical relevance and would have minimal impact on the outcome of the pregnancy.
Lastly, because all four cases of antepartum stillbirth in the SUA group were also IUGR, we thought it possible that SUA may be associated with an increased risk of fetal death in the setting of IUGR. Given that we were limited in power to refine our risk estimate of the association between SUA and IUFD (n=4) we could only hypothesize that restricted growth could be the pathway to fetal death and not single umbilical artery itself.
Our study offered several strengths. Our large sample size allowed us to study a relatively rare diagnosis and its association with rare but clinically important outcomes. Additionally, the comprehensive database allowed us to access complete pregnancy follow-up information that was obtained in a prospective manner, as well as data on patient demographics and history. This allowed us to estimate the relationship between single umbilical artery and multiple outcomes of interest, while adjusting for known confounders. Although the anomaly ultrasounds were performed by a number of sonographers and maternal fetal medicine fellows, all final ultrasound SUA diagnoses were made by dedicated maternal-fetal medicine ultrasound attendings. However, our study was not without weaknesses. One potential weakness was that we used ultrasound diagnosis of SUA at midtrimester screening rather than pathologic diagnosis at the time of delivery. However, we would offer that this potential misclassification would likely bias our results toward the null hypothesis.14 And more importantly, clinicians have only ultrasound diagnosis of SUA on which to base management plans and counseling, thus increasing the clinical utility of our study. A second weakness, inherent to retrospective cohort studies is the potential for missing data. While our database was complete with respect to potentially confounding variables, we were lacking follow up information on 11.5% of our patients. However, sensitivity analyses revealed that the patients with missing outcome data who were excluded from this study were statistically similar to those included with respect to baseline and exposure data, making this potential source of selection bias less likely (data not shown, but available on request). Lastly, while our cohort is, to our knowledge, the largest used to study SUA and pregnancy outcomes, we were still limited in our ability to refine the relationship estimates between SUA and IUFD and cardiac anomalies and their specific subtypes due to the rare occurrence of those outcomes in our cohort.
In conclusion, our study found a significant increase in the risk of fetal growth restriction, and renal and cardiac anomalies in pregnancies complicated by single umbilical artery detected at routine ultrasound compared to those with normal anatomy. These findings can be used to counsel women whose pregnancies are affected by the diagnosis of SUA, and to help guide appropriate antenatal surveillance. Specifically, we would recommend that women whose fetus is diagnosed with SUA at routine anatomic survey undergo detailed anatomic evaluation of the renal and cardiac systems. It also seems reasonable to screen these pregnancies for the development of IUGR with serial ultrasound assessments of estimated fetal weight.
1. Parilla BV, Tamura RK, MacGregor SN, Beibel LJ, Sabbagha RE. The clinical significance of a single umbilical artery as an isolated finding on prenatal ultrasound. Obstet Gynecol 1995;85:570–2.
2. Doornebal N, de Vries TW, Bos AF, de Vries NK. Screening infants with an isolated single umbilical artery for renal anomalies: nonsense? Early Hum Dev 2007;83:567–70.
3. Callen PW. Ultrasonography in obstetrics and gynecology. 4th ed. Philadelphia (PA): WB Saunders; 2000.
4. Budorick NE, Kelly TF, Dunn JA, Scioscia AL. The single umbilical artery in a high-risk patient population: what should be offered? J Ultrasound Med 2001;20:619–27; quiz 628.
5. Leung AK, Robson WL. Single umbilical artery: a report of 159 cases. Am J Dis Child 1989;143:108–10.
6. Catanzarite VA, Hendricks SK, Maida C, Westbrook C, Cousins L, Schrimmer D. Prenatal diagnosis of the two-vessel cord: implications for patient counselling and obstetric management. Ultarasound Obstet Gynecol 1995;5:98–105.
7. Wiegand S, McKenna DS, Croom C, Ventolini G, Sonek JD, Neiger R. Serial sonographic growth assessment in pregnancies complicated by an isolated single umbilical artery. Am J Perinatol 2008;25:149–52.
8. Thummala MR, Raju T, Langenberg P. Isolated single umbilical artery anomaly and the risk for congenital malformations: a meta-analysis. J Pediatr Surg 1998;33:580–5.
9. Bombrys AE, Neiger R, Hawkins S, Sonek J, Croom C, McKenna D, et al. Pregnancy outcome in isolated single umbilical artery. Am J Perinatol 2008;25:239–42.
10. Battaglia FC, Lubchenco LO. A practical classification of newborn infants by weight and gestational age. J Pediatr 1967;71:159–63.
11. Hosmer D, Lemeshow S. Applied logistic regression, 2000. New York (NY): John Wiley & Sons; 2000.
12. Predanic M, Perni SC, Friedman A, Chervenak FA, Chase ST. Fetal growth assessment and neonatal birth weight in fetuses with an isolated single umbilical artery. Obstet Gynecol 2005;105:1093–7.
13. American College of Obstetricians and Gynecologists. Intrauterine growth restriction. ACOG Practice Bulletin 12. Washington, DC: ACOG; 2000.
© 2010 by The American College of Obstetricians and Gynecologists. Published by Wolters Kluwer Health, Inc. All rights reserved.
14. Jones TB, Sorokin Y, Bhatia R, Zador IE, Bottoms SF. Single umbilical artery: accurate diagnosis? Am J Obstet Gynecol 1993;169:538–40.