A normally formed umbilical cord contains two umbilical arteries and one umbilical vein. A single umbilical artery (SUA) is the most common anatomical abnormality of the umbilical cord. It is found in 0.08% to 1.90% of all pregnancies [
The policy at University Hospitals Case Medical Center, Cleveland, Ohio, following a diagnosis of SUA in a fetus by the MFM physicians has been mandatory referral to the pediatric cardiology unit specializing in fetal and pediatric cardiac echocardiography in order to rule out any cardiac anomalies. The referral occurs regardless of whether heart or other congenital anomalies are detected. The rationale for this is that cardiac anomalies are more difficult to detect by ultrasound study due to the complexity of the human heart, especially during the gestational age of 18–20 weeks, when the great majority of pregnant women are referred for comprehensive fetal study. Pediatric cardiologists typically prefer to study the fetus at 22–24 weeks once the fetal heart and vessels are larger. Thus, within many tertiary-level pediatric hospitals, the expertise of these highly trained and skilled physicians is applied to study the fetal heart as needed.
As ultrasound technology improved and sophisticated ultrasound machines became available, many sonographers and sonologists became highly skilled in diagnosing congenital anomalies including the cardiac anomalies. More than 20 years ago Buskens et al. [
We were unable to identify prior studies in which patients were referred for pediatric fetal echocardiogram following diagnosis of a noncardiac anomaly (SUA), in order to compare cardiac anomaly diagnostic accuracy antenatally and confirmed with postpartum neonatal echocardiography between MFM physicians and pediatric cardiologists. In this study we took advantage of the mandatory referral policy in our institution to compare the degree of agreement between MFM physicians and pediatric cardiologists in their diagnosis of cardiac anomalies in fetuses.
The computerized archiving and reporting system of the Fetal Imaging Unit at University Hospitals Case Medical Center, Cleveland, Ohio, between January 1999 and October 2008 was searched using the words “single umbilical artery” to identify all fetuses diagnosed with a SUA. The initial MFM performed fetal ultrasound study report in which SUA was diagnosed and the pediatric echocardiography report of all cases included (antenatal and neonatal studies) were reviewed. Finally, the delivery and the newborns records were reviewed to confirm the diagnosis of SUA and the presence of or absence of a cardiac anomaly following delivery.
During the study period, the MFM physicians used ACUSON Sequoia ultrasound machine with ACUSON V5—Vector array transducer with a frequency range of 3.5/4.0/5.0 MHz and three types of General Electric ultrasound machines: Voluson 730, Voluson 730 Expert, and Voluson E8. The frequencies of the GE transducers were 4–8 MHz and 2–5 MHz. The pediatric cardiologists used various models of ACUSON Sequoia 512. The frequencies of the transducers were Acuson C7—Curved array with a frequency range of 7.0/5.0 MHz, and Acuson V5—Vector array with a frequency range of 3.5/4.0/5.0 MHz.
Data collected included race, gender, age, gestational age at diagnosis, ultrasound imaging findings, fetal echocardiogram findings, findings at delivery along with fetus survival, and neonatal echocardiogram findings. The pathology reports of the placenta and umbilical cord including the cord vessels were reviewed and entered into the database.
The second trimester ultrasound study by the MFM physicians was performed according to guidelines published by the American Institute of Ultrasound in Medicine [
In order for a case to be included in the current study, the following criteria were used: (1) detection of a single umbilical artery by ultrasound study was performed by MFM physician in the Fetal Imaging Unit and delivery at University Hospitals Case Medical Center, Cleveland, Ohio; (2) follow-up fetal heart study was performed antenatally in the echocardiography lab at Rainbow Babies and Children Hospital Case Medical Center and neonatal confirmation of the cardiac anomaly following delivery was available; and (3) postpartum confirmation of a single umbilical artery was done by clinical inspection and histopathologic examination. The data analyzed included only live births. Cases of fetal death and termination of pregnancy due to aneuploidy or anomalous fetuses were excluded.
Subjects were divided into four groups. Group A consisted of fetuses with an isolated SUA and no additional congenital anomalies. Group B consisted of fetuses with SUA and congenital cardiac anomaly only. Group C consisted of fetuses with SUA as well as multiple congenital anomalies without cardiac anomaly. Group D consisted of fetuses with SUA and multiple congenital malformations including cardiac anomaly.
Follow-up rates for presence of a pediatric cardiac study were compared between Groups A and D by use of chi-square test for frequency data. Percent agreement and 95% confidence intervals (by group and overall) for positive findings were calculated for those subjects who had both a MFM cardiac study and a pediatric cardiac study performed. The kappa statistic was calculated to describe interrater reliability and account for agreement beyond chance [
The study was approved by the Institutional Review Board (IRB) of University Hospitals Case Medical Center, Cleveland, Ohio. The IRB waived HIPAA authorization or consent required for access to and use of patients records since this study was a retrospective review.
A total of 39,942 pregnant women were studied between January 1999 and October 2008 by MFM physicians in the Fetal Imaging Unit at University Hospitals Case Medical Center, Cleveland, Ohio. In 376 (0.94%) cases, the ultrasound study noted SUA. Of those 376 fetuses, 182 (48.4%) met the inclusion criteria (Table
Follow-up rates once MFM performed fetal ultrasound study have diagnosed SUA and fetal echocardiogram performed by pediatric cardiologist.
Group | MFM antenatal study | Follow-up pediatric fetal cardiac study | Follow-up rate (%) | 95% confidence interval |
---|---|---|---|---|
A | 253 | 114 | 45.0 | 38.9, 51.4 |
B | 12 | 11 | 91.7 | 59.8, 99.6 |
C | 45 | 24 | 53.3 | 38.0, 68.1 |
D | 66 | 33 | 50.0 | 37.5, 62.4 |
Total |
|
|
|
|
Chi-square = 10.6;
The follow-up rate of fetal echocardiogram in the Pediatric Cardiology Unit varied between groups (Table
The182 fetuses that met all inclusion criteria were divided into four groups as specified in Table
Comparison of fetuses with cardiac anomaly found by MFM physician versus pediatric cardiologist for Groups A through D (
Positive cardiac study FindingsMFM physicians | Positive cardiac study FindingsPediatric cardiologists | Agreement % | 95% Confidence interval | |
---|---|---|---|---|
Group A
|
0 | 2 | 98.2% (112/114) | 93.2, 99.7 |
Group B
|
11 | 10 | 90.8% (10/11) | 57.2, 99.5 |
Group C
|
0 | 0 |
100% (24/24) | 82.3, 100 |
Group D
|
33 | 25 | 75.8% (25/33) | 57.4, 88.3 |
Overall |
|
|
|
|
The specific cardiac abnormalities found, including the additional cardiac anomalies noted by the pediatric cardiologists, are listed in Tables
Cardiac abnormalities detected by MFM physicians and pediatric cardiologists for fetuses in Group B.
Case | MFM performed comprehensive fetal ultrasound study | Cardiac diagnoses via pediatric fetal echocardiogram | Additional abnormal cardiac findings on pediatric fetal and neonatal echocardiogram |
---|---|---|---|
1 | Hypoplastic left heart, mitral stenosis, tricuspid insufficiency, pericardial effusion, and bradycardia | Hypoplastic left heart, mitral stenosis, tricuspid insufficiency, pericardial effusion, and bradycardia | Dilated coronary sinus with left superior vena cava returning to coronary sinus |
2 | Double-outlet right ventricle w/VSD | Double-outlet right ventricle w/VSD | Interrupted aortic arch |
3 | VSD w/pericardial effusion | VSD w/pericardial effusion | |
4 | Hypoplastic right ventricle and pulmonary artery, single pulmonary vein, and a VSD | Hypoplastic right ventricle and pulmonary artery, single pulmonary vein, and a VSD | ASD with atrial septal aneurysm |
5 | VSD | VSD | Coarctation of the aorta |
6 | Hypoplastic left heart | Hypoplastic left heart | Restrictive foramen ovale |
7 | Hypoplastic left ventricle, aortic stenosis/atresia, mitral stenosis/atresia, pulmonary stenosis/atresia, abnormal location of ductus venosus | Hypoplastic left ventricle, aortic stenosis/atresia, mitral stenosis/atresia, pulmonary stenosis/atresia, abnormal location of ductus venosus | Fibroelastosis |
8 | Small VSD | Unremarkable study | Unremarkable study |
9 | Hypoplastic left heart | Hypoplastic left heart | |
10 | Tetralogy of Fallot | Tetralogy of Fallot | |
11 | Hypoplastic right heart | Hypoplastic right heart | Subaortic stenosis |
Cardiac abnormalities detected by MFM physician versus pediatric cardiologist performed fetal echocardiogram for Group D.
Case | MFM performed comprehensive fetal ultrasound study | Cardiac diagnoses via pediatric fetal echocardiogram | Additional abnormal cardiac findings on pediatric fetal and neonatal echocardiogram |
---|---|---|---|
1 | VSD, double-outlet right ventricle | VSD, double-outlet right ventricle | Subaortic stenosis |
2 | VSD with echogenic intracardiac focus | Echogenic intracardiac focus only | |
3 | Tetralogy of Fallot with pulmonary artery hypoplasia | Tetralogy of Fallot with pulmonary artery hypoplasia | |
4 | VSD, pentalogy of Cantrell, absent ductus venosus | VSD, pentalogy of Cantrell, absent ductus venosus | Complete common atrio-ventricular canal defect |
5 | VSD | VSD | Mild left heart hypoplasia |
6 | VSD, hypoplastic aortic arch, and aortic stenosis | VSD, hypoplastic aortic arch, and aortic stenosis | |
7 | Mildly hypoplastic right ventricle w/small pericardial effusion | Unremarkable study | |
8 | Partial atrioventricular septal defect | Partial atrioventricular septal defect | Mitral valve regurgitation |
9 | Mildly hypoplastic left ventricle w/small VSD | Unremarkable study | |
10 | Hypoplastic left atrium w/mitral stenosis or atresia and VSD, suspected segmental stenosis in the inferior vena cava | Hypoplastic left atrium w/mitral stenosis, VSD | Absent renal to hepatic inferior vena cava segment with azygous continuation |
11 | Hypoplastic left heart with VSD | Hypoplastic left heart w/VSD | |
12 | VSD narrow and elongated LVOT | VSD and elongated LVOT | Atrioventricular septal defect |
13 | Mild pulmonic insufficiency w/mild pulmonary artery dilation | Unremarkable study | |
14 | Hypoplastic left ventricle | Hypoplastic left ventricle | Small VSD and coarctation of the aorta |
15 | Overriding aorta with VSD | Overriding aorta with VSD | |
16 | Cardiomegaly | Unremarkable study | |
17 | Unremarkable study | Biventricular hypertrophy, small pericardial effusion | |
18 | Atrioventricular canal defect | Atrio-ventricular canal defect | |
19 | Right sided cardiomegaly | Unremarkable study | |
20 | Endocardial cushion defect, aortic arch hypoplasia, and aortic stenosis | Endocardial cushion defect, aortic arch hypoplasia, and aortic stenosis | |
21 | Hypoplastic left heart | Hypoplastic left heart | |
22 | Cardiac fibroelastosis with severe biventricular hypertrophy and bradycardia | Cardiac fibroelastosis w/severe biventricular hypertrophy and bradycardia | |
23 | Atrioventricular septal defect | Unremarkable study | |
24 | Hypoplastic left ventricle with double right ventricle outlet | Hypoplastic left ventricle with double right outlet ventricle outlet | |
25 | Biventricular hypertrophy | Biventricular hypertrophy | |
26 | Mild left ventricular hypoplasia | Mild left ventricular hypoplasia | |
27 | VSD, pulmonary stenosis/atresia | VSD, pulmonary stenosis/atresia | |
28 | Dextroversion of the heart w/situs solitus | Dextroversion of the heart w/situs solitus | |
29 | Unbalanced AV canal, hypoplastic left heart with outflow tract anomalies (dilated pulmonary artery, nonvisualized aortic tract, truncus not excluded) | Unbalanced AV canal with abnormal outflow tract versus hypoplastic left heart with outflow tract anomalies (dilated pulmonary artery, nonvisualized aortic tract, truncus not excluded), aortic stenosis, hypoplastic ascending aortic arch | |
30 | Narrow aortic outflow | Unremarkable study | |
31 | VSD | VSD | Mild right atrium dilation |
32 | Pericardial effusion | Pericardial effusion | |
33 | Single right ventricle with dextrocardia | Single right ventricle with dextrocardia |
Agreement between MFM physicians and pediatric cardiologists in all groups combined was 94% (171/182) (95% CI [89.2, 96.8]) as seen in Table
Presence (+) or absence (−) of cardiac abnormality by test: fetal echocardiogram performed by MFM versus pediatric cardiologists.
(+) Fetal echocardiogram | (−) Fetal echocardiogram | ||
---|---|---|---|
(+) US imaging study | 35 | 8 | 43 |
(−) US imaging study | 3 | 136* | 139 |
Total number |
|
|
|
Overall agreement between MFM and pediatric cardiology for all groups 171/182 (94.0%; 95% CI
Data for all 182 fetuses regarding the presence or absence of cardiac abnormality as determined by the MFM physicians versus the “gold standard” fetal and neonatal echocardiogram performed by pediatric cardiologists are listed in Table
In this study, we took advantage of the mandatory referral policy for pediatric cardiac echocardiography following diagnosis of SUA in a fetus by the MFM physicians to compare the degree of agreement between MFM physicians and pediatric cardiologists. Since the study period spans over 9 years, it involves multiple physicians performing ultrasound studies. Thus, the comparison between the two imaging services is the summation of expertise of multiple participating sonographers and physicians.
The incidence of cardiac abnormalities among fetuses with SUA was 20.3% (37/182), well in accord with previously published studies that reported incidence ranging from 1% to 32% [
Our data suggests that in our institution once an isolated single umbilical artery (i.e., no other abnormality) has been diagnosed by MFM physician, the risk of completely underdiagnosing a significant congenital cardiac anomaly is low 1.1% (2/182). Of the 35 fetuses with confirmed congenital cardiac abnormalities by pediatric cardiologists (gold standard) 33 (94.2%) fetuses were first diagnosed by the MFM physicians, similar to that reported by Gossett et al. study of 1/18 (95%) [
In 8 fetuses, the MFM physicians diagnosed a form of congenital cardiac abnormality which could not be confirmed by the pediatric cardiologists. The “false positive” diagnosis by the MFM physician of a ventricular septal defect might be explained by the fact that in 74% of pregnancies in which an isolated fetal ventricular septal defect was diagnosed the defect resolved spontaneously before birth [
An important fact uncovered by this study is that pediatric echocardiogram is more likely to diagnose additional cardiac and vascular abnormalities beyond those already identified by the MFM physicians. Most of the additional abnormal findings were not detected by the pediatric cardiologists in the antenatal fetal echocardiogram but rather only in the postdelivery neonatal echocardiography. The overall higher rate of diagnosis by the pediatric cardiologists might be explained by the fact that, in most cases, antenatal pediatric fetal echocardiogram was performed at about 22 weeks or later while the initial MFM study was performed at 16–21 weeks for the great majority of cases. One would expect that the optimal technical study terms in the early neonatal period would result in better diagnosis when performed by experienced and skilled pediatric cardiologist as has been previously reported [
An important finding of our study is the difference in patients’ compliance with the recommendation for antenatal pediatric fetal echocardiogram follow-up. Compliance varied significantly among the different 4 groups and was mainly related to whether a cardiac defect or any other congenital anomalies were detected on the initial ultrasound imaging performed by the MFM physicians. It might reflect subject bias in accepting and complying with the recommendation for a follow-up visit with the pediatric cardiology service in view of their understanding of the severity of the abnormalities found. This was more pronounced in Group D, which consisted of fetuses with cardiac anomaly in addition to multiple other congenital malformations. This in turn could have introduced certain bias in the nature and or severity of cases followed up by the pediatric cardiologists.
Recently Trivedi et al. [
There are several limitations to our study. Despite a larger sample relative to most previously published retrospective studies [
While some of our study findings are similar to findings reported by other researchers [
In summary, our study reveals good agreement between MFM and pediatric cardiology physicians in detecting fetal cardiac anomalies in fetuses with SUA. Studies performed antenatally by MFM physicians and pediatric cardiologists are less likely to uncover the entire spectrum of cardiac abnormalities within each affected fetus. In addition, overdiagnosis of congenital cardiac anomaly by MFM physician is noted among fetuses with multiple other anomalies. Thus, follow-up study with pediatric cardiology may be warranted for all fetuses with SUA or when cardiac anomaly is suspected. The decision to consider mandatory follow-up by pediatric fetal echocardiogram should depend on proven expertise in each institution.
On behalf of the listed authors, Noam Lazebnik hereby states that none of the authors served as consultant, has a spouse who is a chairman, received a research grant, received lecture fees, holds a patent, has been reimbursed by for attending several conferences, and or received honoraria for writing promotional material from any companies that may have a financial interest in the information contained in the paper.
The authors declare that there is no conflict of interests regarding the publication of this paper. The authors have no commercial, proprietary, or financial interest in the products or companies described in this paper. The authors hereby declare that they do not have financial support or relationship that may pose conflict of interests according to the new guideline and confirm that the results of this paper have not been distorted by research funding or conflicts of interest.