Background
The aim of this study was to determine the feasibility of performing complete early fetal echocardiography (FE) at <17 weeks of gestation with comparison with standard FE in the midtrimester (17–23 weeks).
Methods
Fetal echocardiograms obtained in pregnancies studied at <17 weeks at the University of California, San Francisco, over a 5-year period were retrospectively reviewed. FE was considered complete if anatomic details could be assessed (systemic and pulmonary venous connections and atrial, ventricular, and septal [four-chamber sweeps], outflow and great artery, branch pulmonary artery, and arch anatomy) and if color and pulsed Doppler evaluations of the inferior vena cava, pulmonary veins, ventricular inflows and outflows, umbilical artery and vein, and ductus venosus were demonstrated.
Results
One hundred thirty-nine pregnancies were assessed by early FE at <17 weeks transabdominally during the study period (median gestational age, 14.0 weeks; range, 12-0/7–16-6/7 weeks). Additional transvaginal imaging was performed in 14 of 139 (10%) of early fetal echocardiographic studies. One hundred thirteen pregnancies were assessed using both early and later, standard (>17 weeks) FE. Of these, complete fetal echocardiograms were obtained in 27 early (24%; 95% confidence interval [CI], 17%–33%) and 76 later (67%; 95% CI, 58%–75%) exams. In most early exams, color and pulsed Doppler interrogation of the pulmonary veins was unsuccessful. If pulmonary vein Doppler assessment was excluded, complete studies were performed in 80 early exams (71%; 95% CI, 62%–78%) and 97 standard midtrimester exams (86%; 95% CI, 78%–91%). On early FE, heart disease was suspected in 20 pregnancies, and although no major congenital heart disease was missed, in four pregnancies, ventricular septal defects were found only on later FE or after birth.
Conclusions
Early FE yields nearly complete information (exclusive of pulmonary venous interrogation) in the majority of patients.
Second-trimester fetal echocardiography (FE) performed only after 17 weeks of gestation is a well-established technique for the detection and definition of complex cardiac malformation before birth. Most forms of structural congenital heart disease (CHD), however, are present by the end of the first trimester and may even progress well before the standard midtrimester exam. The detection of CHD beyond 17 weeks provides little time for further testing and decision making regarding termination versus continuation of pregnancy in most countries where standard termination is not offered at later gestational ages, and termination of pregnancy in the middle of the second trimester carries a higher risk than if performed earlier in gestation. Finally, pregnancies at risk for fetal CHD are increasingly identified earlier in gestation using first-trimester screening, particularly with the implementation of ultrasound assessment of nuchal translucency.
Despite the potential benefits of early FE, it is not routinely practiced in most fetal and pediatric cardiology programs internationally. This may at least in part be due to a lack of training in early FE. It is also likely due to the paucity of data regarding the details of the examination possible in earlier gestations. Since the early 1990s, many groups have reported their experience with imaging of the fetal heart at 10 to 16 weeks, but most have focused on imaging of basic fetal cardiac anatomy, including the four chambers and ventricular outflow tracts. With technological advances in imaging and further experience, several investigators have documented successful prenatal detection of structural and functional abnormalities and arrhythmias at this early gestational age. None of these studies, however, have specifically evaluated the anatomic and functional details possible at <17 weeks with what is currently the standard approach in the midtrimester.
At our center, FE as early as 12 weeks of gestation has been offered for high-risk pregnancies since 2004, with confirmatory FE at or after 18 weeks gestation in most cases. We hypothesized that early FE (≤16-6/7 weeks) provides detailed structural and functional information at a similar frequency as mid-second-trimester (≥17 weeks) FE.
Methods
A search of the database of the Fetal Cardiovascular Program at the University of California, San Francisco, was conducted to retrospectively identify all patients evaluated over a consecutive 5-year period (2004–2008) who underwent FE at <16-6/7 weeks; these patients were included in the descriptive analysis. Additionally, all patients undergoing repeat FE at ≥17 weeks were included in subsequent statistical analysis. This study was approved by the institutional review board at our institution (H41370-32744-01).
Echocardiography
All fetal echocardiographic studies were performed by American Registry for Diagnostic Medical Sonography–certified cardiac sonographers and/or board-certified pediatric cardiologists with expertise in FE.
All studies were performed using a Siemens Sequoia ultrasound system (Siemens Medical Solutions USA, Inc., Mountain View, CA). Studies performed before 17 weeks were done using an 8C4 or 10V4 phased-array transducer or an 8C4 curved-array transducer for transabdominal imaging and an Acuson EV-8C4 transvaginal transducer (Siemens Medical Solutions USA, Inc.) for transvaginal imaging. The highest frequency transducer was typically used for transabdominal imaging at the onset of the study, but if sufficient resolution could not be achieved because of the distance of the fetus from the maternal abdominal wall, a lower transducer frequency was used. If image quality was still insufficient, transvaginal imaging was attempted. Studies done after 17 weeks were performed using a 6C2 or an 8C4 curved-array transducer, per our protocol at the time for obstetric scanning. Transabdominal imaging at all gestational ages was performed using our laboratory’s protocol on the basis of guidelines established by the American Society of Echocardiography. This protocol included anatomic overview of the uterus, fetal number and lie, biometric examination, cardiac imaging and sweeps, Doppler examination incorporating color and pulsed-wave Doppler, and examination of heart rate and rhythm. Two-dimensional (2D) axial and sagittal 2-sec to 4-sec sweeps were used for demonstration of visceroatrial situs, systemic (superior and inferior vena cava and hepatic veins) and pulmonary venous connections (at least one left and one right pulmonary vein), atrial and ventricular arrangement and size, atrioventricular valves, outflow tracts, semilunar valves, and great artery size, arrangement, and patency. Color Doppler was used to demonstrate pulmonary venous return to the left atrium and atrioventricular and semilunar patency and to assess for regurgitation. Color Doppler interrogation of the ductal and aortic arches was performed in the sagittal plane. Pulsed-wave Doppler interrogation of the umbilical artery and vein, ductus venosus, hepatic and inferior vena cava, tricuspid and mitral valves, outflows, right upper or left-sided pulmonary vein, and ductal and aortic arches was also performed. Figures 1 through 4 and Videos 1–7 (available at www.onlinejase.com ) illustrate imaging and scoring examples.
Transvaginal scans were generally done in the same exam room; the patient was first asked to empty her bladder and then positioned with a bolster under her hips to tilt the pelvis, allowing more anterior imaging. The probe was then inserted in standard fashion, and multiple sagittal plane (in reference to the mother) images were taken. Sagittal and axial images through the fetus were attempted in each by rotating up to 180° and tilting the transducer to the mother’s left and right and anteriorly and posteriorly, to allow evaluation of abdominal situs, four-chamber view, outflows, three-vessel view, and transverse views of the arches with 2D and color Doppler. Significant activity of the fetus at this early gestation also contributed to greater success of imaging of specific cardiac structures.
A retrospective review of the stored Digital Imaging and Communications in Medicine images was performed by two fetal cardiologists (L.K.H., A.M.-G.) for the presence or absence of the specific anatomic features listed in Table 1 . In addition, completeness of the individual studies was assessed as follows: 2D assessment complete if systemic veins, four-chamber, right and left outflows, and arches were shown; Doppler complete if color and/or pulsed-wave of inflows and outflows, plus color of arches, pulsed-wave of cord and systemic veins, and pulsed-wave or color of pulmonary veins (inferior vena cava Doppler was optional as long as the ductus venosus was assessed). Studies were considered complete if both 2D and Doppler criteria for completeness were met.
| 2D anatomic details |
| Systemic veins (inferior and right and/or left superior vena cavae) |
| Pulmonary veins (at least one left and one right pulmonary vein connecting with left atrium) |
| Four-chamber view (both atria, both ventricles, and ventricular septum visualized) |
| LV outflow (outflow, aortic valve, ascending aorta) |
| RV outflow (outflow, semilunar valve, main pulmonary artery) |
| Three-vessel view |
| Branch pulmonary arteries |
| Aortic arch (morphology and relationship with trachea) |
| Ductal arch (morphology and relationship with trachea) |
| Color Doppler interrogation |
| Pulmonary veins |
| LV inflow |
| RV inflow |
| LV outflow tract, aortic valve, ascending aorta |
| RV outflow tract, pulmonary valve, main pulmonary artery |
| Aortic arch |
| Ductal arch (ductus arteriosus) |
| Pulsed Doppler interrogation |
| Umbilical artery |
| Umbilical vein |
| Ductus venosus |
| Inferior vena cava |
| Pulmonary vein |
| RV inflow |
| LV inflow |
| RV outflow |
| LV outflow |
Statistical Analysis
The presence of each of the scored parameters is expressed in terms of frequencies or percentages.
Results
During the study period, 139 pregnancies met the inclusion criteria; 113 patients underwent both early and late FE. Sixteen fetuses underwent early FE only, without later echocardiography (most because the results of the first scan were normal and follow-up was not obtained by the referring clinicians) and are excluded from analysis. Eight had termination of pregnancy before a second scan (all abnormal results), and there were two with spontaneous intrauterine fetal demise before a second scan (one with pentalogy of Cantrell, one with cardiomyopathy).
All echocardiographic studies were performed for clinical indications, including pregestational diabetes (35%), family history of significant CHD (28%), increased nuchal translucency (16%), fetal arrhythmia (8%), teratogen exposure (4%), and others (9%), including gestational diabetes (diet controlled, White classification A1), assisted reproduction, multiple gestation, or suspected cardiac abnormality on screening ultrasound. Among the 113 study patients with two exams, FE performed from 12 to 16-6/7 weeks (median, 14 weeks) constituted the “early” studies, and FE performed from 17 to 23 weeks (median, 19 weeks) constituted “standard” studies ( Figure 5 ).
In 125 of 139 early studies (90%), the entire study was performed using exclusively a transabdominal approach. In 14 pregnancies (10%), transvaginal ultrasound had also been performed by the clinician after initial transabdominal images had been obtained, usually in the setting of suboptimal transabdominal imaging. All 113 later studies were performed transabdominally.
Early Versus Late FE
Figure 6 A documents the frequency with which 2D images of specific cardiovascular structures were obtained in the early versus midgestation fetal echocardiographic examinations. In all but one case, the four-chamber view was obtained and ventricular systolic function qualitatively assessed at both early and later gestational ages. Color and pulsed-wave Doppler studies were performed in all cases, with successful demonstration of specific blood flows and Doppler spectral signals in most ( Figures 6 B and 6 C). Pulmonary venous anatomy by 2D and color and pulsed-wave Doppler interrogation was difficult to demonstrate in the early scans, with improvement in the later studies. This was not true of the remainder of the parameters studied: ≥80% success was achieved for each individual 2D parameter and >90% success for color and ≥74% for pulsed-wave Doppler whether evaluated early or at the standard gestational age.
