Background
Advances in fetal echocardiography have improved recognition of congenital heart disease (CHD). Imaging protocols have been developed that predict delivery room (DR) risk and anticipated postnatal level of care (LOC). The aim of this study was to determine the utility of fetal echocardiography in the perinatal management of CHD.
Methods
A retrospective analysis of fetal and postnatal records was conducted. The anticipated LOC was assigned by fetal echocardiography (LOC 1, nursery consult/outpatient follow-up; LOC 2, stable in DR with transfer to cardiac hospital; LOC 3 or 4, DR instability/urgent intervention needed). Prenatal diagnoses and LOC assignment were compared with postnatal diagnoses, treatment, and short-term outcomes.
Results
From 2004 to 2012, 8,101 fetuses were evaluated; 7,405 were normal. Of 696 with CHD, 101 terminated, 40 died in utero, and 37 received palliative care. LOC was assigned in the remaining 518. Of 219 LOC 1, 195 (89%) had postnatal follow-up. Only two required transfer for intervention (LOC 1 sensitivity, 0.9; LOC 1 positive predictive value, 0.99). Of 260 assigned LOC 2, 229 (88%) had follow-up. Of these, 200 (87%) were transferred for surgery or intervention. The median time to admission was 195 min. Twenty-two patients (10%) assigned LOC 2 did not require intervention; however, seven (all with D-transposition of the great arteries) required catheter intervention before surgery. Hospital survival was 86% (LOC 2 sensitivity, 0.97; LOC 2 positive predictive value, 0.87). All LOC 3 and 4 patients had follow-up. Thirty-four (87%) needed urgent intervention, with 100% DR and 87% hospital survival (LOC 3 and 4 sensitivity, 0.83; LOC 3 and 4 positive predictive value, 0.87).
Conclusions
Fetal echocardiography enables accurate postnatal risk stratification in CHD, with the exception of D-transposition of the great arteries. LOC 1 assignment facilitated outpatient follow-up; LOC 2 assignment facilitated transfer for intervention. LOC 3 and 4 patients underwent stabilizing intervention or surgery with good short-term outcomes. Given the inability to predict need for intervention in D-transposition of the great arteries, all such patients should be assigned as LOC 3 or 4. Fetal echocardiography with LOC assignment should be used in the planning of postnatal care in CHD.
Highlights
- •
Fetal echocardiography can accurately diagnosis CHD and enable planning for postnatal care.
- •
Risk-adjusted postnatal LOC in the DR and in the perinatal period can be determined on the basis of fetal echocardiographic findings.
- •
Postnatal care planning improves perinatal care in newborns with complex CHD.
Improvements in fetal cardiac imaging have led to an increased number of newborns diagnosed prenatally with congenital heart disease (CHD). High-resolution fetal echocardiography with analysis of serial studies to assess disease severity and progression has enabled the prediction of clinical course in utero and during delivery. As a consequence, specialists in fetal medicine are now being asked to manage fetuses as distinct patients, with the transition to postnatal life an important part of individualized care. Available data suggest that fetal diagnosis may improve outcomes in severe CHD by allowing the planning of specialized care to prevent the postnatal hemodynamic instability that occurs during transition. For most newborns, however, only routine resuscitation is required in the delivery room (DR), thus allowing stabilization in the local hospital and either outpatient follow-up or nonurgent hospital transfer to a cardiac center. Care protocols at our institution for risk assessment at delivery have been developed on the basis of published data and clinical experience for fetuses with CHD. The primary aims of this study were to assess the accuracy of fetal echocardiography to correctly diagnose CHD and predict postnatal anticipated care and resuscitation in the DR and perinatal period, and to determine the effectiveness of risk-stratified DR care protocols for the postnatal management of newborns diagnosed in utero with CHD.
Methods
Since 2004 as part of clinical practice, fetuses diagnosed with CHD have been assigned an anticipated postdelivery “level of care” (LOC) with commensurate care plan determined by fetal echocardiographic findings suggestive of postnatal hemodynamic compromise and/or need for intervention or surgery in the DR or perinatal period ( Table 1 ). This study was a retrospective analysis of the LOC protocol created at our institution to “prospectively” determine risk for compromise and need for specialty care at delivery.
| LOC | Definition | Example CHD | Prenatal planning | Delivery | DR recommendations |
|---|---|---|---|---|---|
| 1 | CHD without physiologic instability in first weeks of life |
| Arrange outpatient cardiology evaluation | Spontaneous vaginal delivery | Routine DR management |
| 2 | CHD with physiologic stability in DR but requiring postnatal intervention/surgery before discharge |
| Create plan of care for DR stabilization and neonatal management by local hospital with transport to CNMC | Spontaneous vaginal delivery vs induction near term | Neonatologist in DR; initiate PGE at low dose for ductal-dependent lesions |
| 3 | CHD with expected instability requiring immediate specialty care in DR before anticipated stabilizing intervention/surgery |
| Create plan of care to include specialized CNMC team in DR and interventional/surgical team on standby | Planned induction usually at 39 wk with “bailout” C/S if necessary for care coordination | Neonatologist and CNMC specialists in DR; stabilizing medications predetermined by care plan |
| 4 | CHD with expected instability requiring immediate specialty care and urgent intervention/surgery in DR to improve chance of survival |
| Create multidisciplinary plan of care to include CNMC delivery if possible with specialized care team in DR and interventional/surgical team ready | Planned C/S at CNMC usually at 38–39 wk or sooner if there is evidence of fetal distress or hydrops Maternal risk determined by obstetrician (only low-risk women deliver at CNMC) | Specialized care team in DR Stabilizing medications/equipment predetermined by care plan |
Subjects
Records of all fetuses diagnosed with CHD and evaluated in the Fetal Heart Program at Children’s National Medical Center (CNMC) from November 2004 to March 2012 were reviewed. Patients were identified by searching the fetal cardiac database. Fetal and postnatal echocardiograms and maternal and neonatal electronic medical records were reviewed, and data were collected after patient identifiers were removed. The review was approved with a waiver of the requirement for informed consent by the hospital institutional review board.
Fetal Determinants of LOC
Fetal echocardiographic criteria for disease severity and LOC assignment were determined using published data when available ( Table 2 ) and multidisciplinary discussion including experts in obstetrics, neonatology, and cardiology in all others. Results of the last fetal echocardiographic examination before delivery were used for final delivery planning.
| CHD diagnosis | Fetal echocardiographic findings | DR recommendation |
|---|---|---|
| VSD or AVSD (shunt lesions) Mild valve abnormalities | Isolated cardiac defect with normal FO and DA flow, normal or minimal flow disturbances at valves, and normal heart function (LOC 1) | Routine care, hospital or telemedicine consult |
| Pulmonary atresia, HLHS, other single ventricles, or cyanotic TOF (ductal-dependent lesions) | Ductal-dependent pulmonary circulation (LOC 2):
| Initiation of prostaglandin |
| HLHS and variants with intact or restrictive atrial septum |
| Initiation of prostaglandin; plan for immediate intervention to decompress left atrium |
| D-TGA and variants with restrictive atrial septum |
| Initiation of prostaglandin If RFO, plan for immediate balloon atrial septostomy in DR If ductal flow abnormal, consider pulmonary hypertension therapy |
| TOF/APV | Cardiac dysfunction (LOC 3) Hydrops fetalis (LOC 4) Lung findings suggestive of compression or fluid trapping (LOC 4) | Specialized ventilation Consider primed ECMO circuit |
| Severe Ebstein’s anomaly | Cardiac dysfunction (LOC 3) Hydrops fetalis (LOC 4) | Consider measures to decrease pulmonary resistance and support cardiac output |
| Unstable tachyarrhythmias | Uncontrolled with associated heart failure (LOC 3) Hydrops fetalis (LOC 4) | Consider early delivery if gestational age appropriate Cardioversion or medical therapy in DR |
| CHB | Low ventricular rate with heart failure (LOC 3) Hydrops fetalis (LOC 4) | Consider early delivery if gestational age appropriate Consider chronotrope vs pacing in DR |
Fetuses with mild cardiac disease, such as isolated shunt lesions, mild valve abnormalities, and benign arrhythmias requiring only telemedicine consultation or outpatient cardiology follow-up, were assigned to LOC 1. Fetuses with CHD requiring DR stabilization by the neonatologist on site with subsequent hospital transfer for subspecialty care including catheter intervention or surgery were assigned to LOC 2. These babies most often had ductal-dependent circulation requiring the initiation of a prostaglandin infusion before transfer to our hospital for catheter or surgical intervention. Ductal-dependent flow was determined on the basis of the finding of reversed flow documented by color Doppler in either the ductus arteriosus (ductal-dependent pulmonary flow) or foramen ovale and/or distal aortic arch (ductal-dependent systemic flow) ( Figures 1 a and 1b). Other diagnoses assigned to LOC 2 included nonsustained tachycardias or arrhythmias controlled in utero requiring postnatal treatment, cardiomyopathies without heart failure, and any defect in which prediction of a stable postnatal transition was uncertain. Fetuses with severe CHD requiring planned and coordinated delivery with a specialized CNMC care team in the DR were assigned to LOC 3 or 4, depending on the severity of the cardiac diagnosis and the anticipated complexity of the staffing needed for DR stabilization, taking into account the predicted likelihood that an urgent intervention would be required. In instances when specialized medical care was anticipated for stabilization and the need for an urgent catheterization or surgical intervention was a possibility but not a certainty (LOC 3), delivery at the hospital adjacent to CNMC was planned. If it was anticipated that an urgent interventional catheterization or surgical procedure would be required (LOC 4), every effort was made to deliver at CNMC in the cardiac operating room. Babies assigned to LOC 3 or 4 most often had hypoplastic left heart syndrome (HLHS) and a restrictive foramen ovale (RFO) or intact atrial septum (IAS) or D-transposition of the great arteries (D-TGA) and an RFO and/or abnormal ductus arteriosus. Other diagnoses included uncontrolled sustained tachyarrhythmias or any severe CHD with heart failure. Fetuses with HLHS and RFO or IAS were stratified into LOC 3 versus LOC 4 on the basis of pulmonary venous flow pattern. LOC 3 was assigned if the pulmonary vein forward/reversed velocity-time integral flow ratio was >3 and <5 and LOC 4 if this ratio was <3. Fetuses with D-TGA and any abnormality of the atrial septum (including a hypermobile, tethered, bowing, or intact atrial septum) and/or ductus arteriosus were assigned to LOC 3 versus LOC 4 on the basis of the specific features of the atrial septum and flow in the ductus arteriosus ( Figure 2 a and b). Complete heart block with low ventricular rate, uncontrolled severe tachycardias, and diseases expected to have compromise in the DR, such as severe tetralogy of Fallot (TOF) with absent pulmonary valve or severe Ebstein’s anomaly, were assigned to LOC 3 if cardiac dysfunction was present and to LOC 4 if hydrops fetalis was documented.
For each LOC, written recommendations were prepared for the obstetric and neonatal teams for DR and postnatal management, including delivery location, DR staff, neonatal care, cardiac-specific care, and either outpatient follow-up or hospital transfer to CNMC. Delivery letters summarizing the CHD diagnosis, expected clinical findings, recommendations, and important phone numbers were sent to the delivery hospital at 30 weeks and again near term. For LOC 1, a call was made to arrange outpatient follow-up if the baby had not been scheduled 1 week after the due date or was unexpectedly transferred for care. For LOC 2, 1 week before the expected delivery date, the local neonatologist was called to review the care plan, and the CNMC transport team was notified. For LOC 3 and 4, delivery planning included formulation of a detailed multisubspecialty care algorithm with predelivery simulation and postdelivery debriefing for all patients.
Decisions regarding timing, place, and mode of delivery were made considering family preference when possible and in consultation with the obstetrician and the neonatal team. For LOC 1, spontaneous delivery was most often planned. For LOC 2, decisions regarding spontaneous delivery versus need for elective induction near term (≥39 weeks) were made on the basis of family preference regarding place of delivery (local hospital vs facility closer to CNMC) and the opinion of the on-site physicians, who often favored a planned delivery date to help facilitate the coordination of care. For LOC 3 and 4, decisions regarding elective induction near term versus a planned cesarean section were based on the complexity of the staffing anticipated in the DR for stabilization of the infant.
Data Collection and Analysis
Fetal data collection included (1) cardiac diagnosis at final evaluation before delivery; (2) LOC assignment with proposed delivery plan including place of delivery, intervention anticipated at delivery, and recommendations for outpatient follow-up or hospital transfer; and (3) recorded delivery information including place of delivery, time from delivery to outpatient follow-up or admission to the cardiac intensive care unit, time to intervention or surgery, intervention or surgery performed, and survival to outpatient follow-up or discharge from hospital after intervention. For LOC 1 in which outpatient follow-up was recommended, return for follow-up was documented as the primary clinical end point. For LOC 2, in which stabilization by the delivery hospital with ambulance transport for definitive intervention was planned, time from birth to admission and survival to hospital discharge were documented as the primary clinical end points. For LOC 3 and 4 in which comprehensive delivery coordination of care with cardiology specialists in the DR was planned, DR survival, time from birth to intervention, and survival to hospital discharge were documented as the primary clinical end points.
All decisions regarding postnatal care, including catheterization or surgical intervention, were made by the clinical care team. For LOC 3 and 4 patients, decisions regarding DR intervention were made on the basis of clinical findings such as severe hypoxia, worsening acidosis, hemodynamic instability, unstable rhythm, and/or echocardiographic features suggesting the need for atrial septostomy in newborns with D-TGA or HLHS and RFO or IAS.
Prenatal diagnoses and plans of care were compared with postnatal diagnoses and DR and postnatal events; records of unanticipated events or complications were reviewed to determine the accuracy of fetal diagnoses and prenatal LOC assignment. Contingency table analyses was implemented to determine if actual postnatal care, used as the gold standard, could be predicted by prenatal LOC assignment, computing sensitivity and positive predictive value (PPV) for each LOC. LOC groups were then divided in two ways to determine the sensitivity and specificity of LOC assignment to predict (1) those neonates requiring only cardiology consultation or outpatient follow-up (LOC 1) versus those requiring postnatal care at the cardiac hospital (LOC 2–4) and (2) those neonates needing standard neonatal DR care (LOC 1 and 2) versus those who would benefit from specialized cardiac care in the DR (LOC 3 and 4).
Results
From 2004 to 2012, 8,101 fetuses were evaluated; 7,405 were diagnosed as having normal hearts, and no postnatal follow-up was arranged, though three were unexpectedly admitted as neonates with CHD and one was seen as an outpatient. Of 696 with CHD, 15% terminated, 6% died in utero, and 5% were delivered and received palliative care before death. LOC was assigned in the remaining 518 fetuses with CHD and intent to treat. Of these, 463 had postnatal follow-up ( Figure 3 ). Extracardiac anomalies were found in 90 of the fetuses (19%) assigned a LOC and 64 (36%) of the fetuses who died in utero or received palliative care. Genetic abnormalities were confirmed in 97 of the fetuses (20%) assigned a LOC and 42 of the fetuses (24%) who died in utero or received palliative care.
Accuracy of Fetal Diagnosis
Diagnoses by LOC are listed in Table 3 . Overall, the exact fetal cardiac diagnosis matched the postnatal diagnosis in 391 of 463 patients who had postnatal evaluation and care (84% diagnostic accuracy). Differences in diagnoses occurred only in LOC 1 ( n = 43) and LOC 2 ( n = 29) patients and were minor and did not affect care in most cases. Common examples of minor discrepancies from pre- to postnatal diagnosis include (1) presence or description of ventricular septal defect (VSD), (2) presence or absence of mild ventricular hypertrophy, (3) presence or absence of minor valve abnormalities, (4) prediction of coarctation of the aorta, and (5) minor anatomic details of complex single ventricles. In only 13 patients (seven LOC 1 and six LOC 2) were there changes in the primary cardiac diagnosis that affected care (97% diagnostic accuracy for major cardiac defects), though in only four would the prenatal plan have involved more specialized intervention in the DR ( Table 4 ).
| LOC 1 | LOC 2 | LOC 3 or 4 |
|---|---|---|
| n = 219 | n = 260 | n = 39 |
| n = 195 with follow-up | n = 229 with follow-up | n = 39 with follow-up |
| AVSD ( n = 41) | Ductal-dependent SV ( n = 70) | D-TGA with RFO (n = 17) |
| VSD ( n = 39) | HLHS ( n = 48) | HLHS/variants with RFO/IAS ( n = 11) |
| TOF ( n = 30) | VSD with coarctation or IAA ( n = 19) | Critical AS with LV dysfunction ( n = 2) |
| TOF with PA ( n = 1) | Isolated coarctation ( n = 19) | CHB with low HR ( n = 2) |
| Possible coarctation ( n = 17) | TOF ( n = 17) | Ectopia cordis ( n = 2) |
| Non-ductal-dependent SV ( n = 13) | TOF with PA ( n = 10) | Conjoined twins with CHD ( n = 1) |
| RV hypertrophy ( n = 11) | Truncus arteriosus ( n = 8) | TOF/APV with CLE ( n = 1) |
| Dilated coronary sinus ( n = 9) | PA/IVS ( n = 8) | Complex PA with RFO ( n = 1) |
| Mild PS or AS ( n = 7) | D-TGA ( n = 6) | Complex arrhythmia with VT ( n = 1) |
| Irregular heart rhythm ( n = 6) | Unbalanced AVSD/coarctation ( n = 6) | |
| Myocardial tumor ( n = 6) | D-TGA with VSD ( n = 4) | |
| Cardiac malposition ( n = 4) | Ebstein’s anomaly ( n = 4) | |
| Atrial septal defect ( n = 4) | VSD ( n = 2) | |
| VSD/possible coarctation ( n = 2) | Controlled arrhythmias ( n = 3) | |
| Tricuspid regurgitation ( n = 2) | AS or PS/no heart failure ( n = 3) | |
| Small pericardial effusion ( n = 1) | Myocardial tumor ( n = 1) | |
| Vascular ring ( n = 1) | Dilated cardiomyopathy ( n = 1) | |
| Absent ductus venosus ( n = 1) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
