DOI: 10.1002/pd.4466
REVIEW
Early detection of fetal cardiac abnormalities: how effective is it and how should we manage these patients? Sally-Ann B. Clur1,2* and Caterina M. Bilardo3,4 1
Department of Pediatric Cardiology of the Emma Children’s Hospital, Academic Medical Centre, Amsterdam, The Netherlands The Centre for Congenital Heart Anomalies Amsterdam-Leiden (CAHAL), Amsterdam, The Netherlands 3 Department of Obstetrics and Gynecology, Academic Medical Centre, Amsterdam, The Netherlands 4 Department of Obstetrics and Gynecology, University Medical Centre Groningen, University of Groningen, Groningen, The Netherlands *Correspondence to: Sally-Ann B. Clur. E-mail: [email protected] 2
ABSTRACT Congenital heart defects (CHDs), the most commonly occurring congenital malformations, cause significant mortality and morbidity. With the recognition of early markers for CHD and the development of better ultrasound resolution, interest has turned toward performing a screening anomaly scan, including the heart, together with the nuchal scan. It is also possible, with adequate skill and training, to competently perform an echocardiogram 99th percentile had a sensitivity of 31% and specificity of 98.7%, with a positive predictive value of 24 for the identification of major CHDs, whereas a sensitivity of 37% and specificity of 96.6% were found when using the 95th percentile of NT measurement cutoff.18 Wald et al. in a later meta-analysis of seven studies defined the relationship between an enlarged NT [cutoff 1.7 multiples of the median at a false–positive rate (FPR) of 5%] and specific CHDs as ‘likely to benefit’ from prenatal diagnosis. They found an increased NT in 52% of CHDs they considered likely to benefit from prenatal diagnosis.20 The difference between the sensitivity of the increased NT for CHDs found by Hyett et al. initially17 and the subsequent researchers has been partially explained by the differences in the study designs (prospective vs retrospective), study populations examined (high vs low risk), cutoffs used in the definition of an increased NT, gestational ages at time of NT measurement, and in the definition of major CHD.26 The measurement of NT has since been standardized and optimized, and NT screening has been implemented in many countries. This has resulted in closer examination of firsttrimester hearts and earlier detection of major CHDs. Ten years later, Makrydimas’s group published an updated metaanalysis including 20 studies, this time also reporting on a subgroup of studies where the NT was measured by Fetal Medicine Foundation-certified operators. The pooled sensitivity and specificity of the NT >95th percentile for major CHD in this subgroup were 45.6% and 94.7%. respectively, and for a NT >99th percentile 21% and 99.2% respectively, with a positive likelihood ratio of 30.5.27 Studies have also shown that the risk for CHD increases with increasing NT measurement28–30 without a particular preference for one CHD above another.28 In counseling couples about the risk of finding a CHD in a chromosomally normal fetus with an enlarged NT, the results of a pooled analysis of 11 studies may be helpful26 (Figure 1).
Abnormal ductus venous flow and tricuspid regurgitation Current data on >130 000 unselected pregnancies shows a modest sensitivity of an increased NT alone as a screening tool for major CHD21,31; however, the sensitivity increases when other markers related to cardiac (dys)function, such as an Prenatal Diagnosis 2014, 34, 1–11
Figure 1 Relationship between nuchal translucency thickness and congenital heart disease risk; a meta-analysis of 12 studies. CHD, congenital heart disease; NT, nuchal translucency. Adapted from Clur et al.26 abnormal DV flow or TR are found at 11 to 14 weeks.26,31–39 Abnormal DV flow (absent or reversed A-wave during atrial contraction) was initially observed in association with cardiac dysfunction in the second and third trimesters.40,41 Matias et al. were the first to describe an association between an increased NT, CHD, and abnormal DV flow.42 Maiz et al. (2008) in a meta-analysis of seven studies, including 600 chromosomally normal fetuses with an NT ≥95th percentile, found that an abnormal DV flow pattern at 11 to 14 weeks’ gestation in the presence of an NT ≥3.5 mm was associated with a threefold increased risk of CHDs.34 The finding of a normal DV flow pattern in this setting halved the CHD risk. A recent meta-analysis reported that an abnormal DV A-wave can reveal 83% of CHD cases in the presence of an increased NT and 19% when the NT is normal, with an FPR of 20% and 4%, respectively.36 A study of 40 000 singleton pregnancies with normal chromosomes at 11 to 14 weeks showed that screening for CHD using an increased NT and the reversed A-wave of the DV can detect 47.1% of major CHD with an FPR of 6.7%.37 Instead of evaluating the DV flow in terms of the A-wave (positive, absent, reversed), our group suggested using the DV pulsatility index (PIV) and found an abnormal DVPIV in twothirds of the fetuses with an increased NT, normal karyotype, and CHDs.33 The sensitivity of an abnormal DVPIV for CHDs was 70% with a specificity of 62%. The DVPIV multiples of the median significantly influenced the risk of a CHD (odds ratio 2.4) at any cutoff point suggesting that the DVPIV can be used to increase the specificity of the NT in the identification of CHDs.33 Borrell et al. assessed the best method to combine the NT and DV Doppler in the detection of major CHD in euploid fetuses and demonstrated different detection rates (DRs) depending on the cutoffs of NT or DVPIV used.43 The mechanism of the abnormal DV flow in the presence of CHD is unclear, but the predominance of right-sided obstructive lesions44–46 and atrioventricular septal defects (AVSD) with atrioventricular (AV)-valve regurgitation and hypoplastic left heart syndrome (HLHS)33 suggests that right heart/atrial volume and/or pressure overload or diastolic dysfunction may be responsible. © 2014 John Wiley & Sons, Ltd.
CHD detection and management in the first trimester
Tricuspid regurgitation is frequently observed in 11-week to 14-week fetuses with trisomies47 but may also be found in euploid fetuses with CHDs.39,48–50 The mechanism is not clear but may be related to the reduced diastolic function and still high afterload at this gestational age. Faiola et al. reported an eight times increased risk of CHDs when TR is found at 11 to 13 + 6 weeks’ gestation.38 A more recent study of 40 990 fetuses found an NT >95th percentile, TR, and reversed A-wave in the DV flow in 35.3%, 32.9%, and 28.2% of the 85 fetuses with major CHD, respectively, and in 4.8%, 1.3%, and 2.1% of those without CHD.39 Any one of the three markers was found in 57.6% of the fetuses with CHD (95% CI 47–67.6%) and in 8% of those with normal hearts (95% CI 7.7–8.2%).39 In conclusion, an abnormal DVPIV in the first trimester can detect about 70% of the major CHD and is an indication for referral for specialized fetal echocardiography, even when the NT is normal. The same applies for TR. The risk of CHD increases with increasing NT measurement,28–30 and is further increased in the presence of an abnormal DV flow and/or TR, but is reduced if these findings are absent.39
ACCURACY OF CHD DETECTION Early fetal anomaly scan The first case descriptions of CHD diagnosed in the first trimester with the use of transvaginal echocardiography appeared in the early 1990’s.51,52 After the first report in 1998,53 showing that transabdominal echocardiography left), (B) left ventricular outflow tract view with color Doppler showing the small aorta, (C) right ventricular outflow tract view showing the enlarged main pulmonary artery and ductus arteriosus, (D) sagittal section showing the disassociation between the umbilical vein and hepatic vein due to the absence of a ductus venosus, (E) sagittal section of one of the twins showing the increased nuchal translucency, and (F) sagittal section showing the coarctation of the aorta and the retrograde flow in the distal aortic arch (arrow head). Ao, aorta; CoA, coarctation of the aorta; LV, left ventricle; PA, pulmonary artery; NT, nuchal translucency; RV, right ventricle; VCI, inferior caval vein; v hep, hepatic vein; v umb, umbilical vein scanning provides sufficient resolution thereafter.61,63,67 The detection of the 4CV and three-vessel view can be enhanced by approximately 5% by the addition of the transvaginal approach.5 Moon-Grady et al. compared EFEC
REVIEW
Early detection of fetal cardiac abnormalities: how effective is it and how should we manage these patients? Sally-Ann B. Clur1,2* and Caterina M. Bilardo3,4 1
Department of Pediatric Cardiology of the Emma Children’s Hospital, Academic Medical Centre, Amsterdam, The Netherlands The Centre for Congenital Heart Anomalies Amsterdam-Leiden (CAHAL), Amsterdam, The Netherlands 3 Department of Obstetrics and Gynecology, Academic Medical Centre, Amsterdam, The Netherlands 4 Department of Obstetrics and Gynecology, University Medical Centre Groningen, University of Groningen, Groningen, The Netherlands *Correspondence to: Sally-Ann B. Clur. E-mail: [email protected] 2
ABSTRACT Congenital heart defects (CHDs), the most commonly occurring congenital malformations, cause significant mortality and morbidity. With the recognition of early markers for CHD and the development of better ultrasound resolution, interest has turned toward performing a screening anomaly scan, including the heart, together with the nuchal scan. It is also possible, with adequate skill and training, to competently perform an echocardiogram 99th percentile had a sensitivity of 31% and specificity of 98.7%, with a positive predictive value of 24 for the identification of major CHDs, whereas a sensitivity of 37% and specificity of 96.6% were found when using the 95th percentile of NT measurement cutoff.18 Wald et al. in a later meta-analysis of seven studies defined the relationship between an enlarged NT [cutoff 1.7 multiples of the median at a false–positive rate (FPR) of 5%] and specific CHDs as ‘likely to benefit’ from prenatal diagnosis. They found an increased NT in 52% of CHDs they considered likely to benefit from prenatal diagnosis.20 The difference between the sensitivity of the increased NT for CHDs found by Hyett et al. initially17 and the subsequent researchers has been partially explained by the differences in the study designs (prospective vs retrospective), study populations examined (high vs low risk), cutoffs used in the definition of an increased NT, gestational ages at time of NT measurement, and in the definition of major CHD.26 The measurement of NT has since been standardized and optimized, and NT screening has been implemented in many countries. This has resulted in closer examination of firsttrimester hearts and earlier detection of major CHDs. Ten years later, Makrydimas’s group published an updated metaanalysis including 20 studies, this time also reporting on a subgroup of studies where the NT was measured by Fetal Medicine Foundation-certified operators. The pooled sensitivity and specificity of the NT >95th percentile for major CHD in this subgroup were 45.6% and 94.7%. respectively, and for a NT >99th percentile 21% and 99.2% respectively, with a positive likelihood ratio of 30.5.27 Studies have also shown that the risk for CHD increases with increasing NT measurement28–30 without a particular preference for one CHD above another.28 In counseling couples about the risk of finding a CHD in a chromosomally normal fetus with an enlarged NT, the results of a pooled analysis of 11 studies may be helpful26 (Figure 1).
Abnormal ductus venous flow and tricuspid regurgitation Current data on >130 000 unselected pregnancies shows a modest sensitivity of an increased NT alone as a screening tool for major CHD21,31; however, the sensitivity increases when other markers related to cardiac (dys)function, such as an Prenatal Diagnosis 2014, 34, 1–11
Figure 1 Relationship between nuchal translucency thickness and congenital heart disease risk; a meta-analysis of 12 studies. CHD, congenital heart disease; NT, nuchal translucency. Adapted from Clur et al.26 abnormal DV flow or TR are found at 11 to 14 weeks.26,31–39 Abnormal DV flow (absent or reversed A-wave during atrial contraction) was initially observed in association with cardiac dysfunction in the second and third trimesters.40,41 Matias et al. were the first to describe an association between an increased NT, CHD, and abnormal DV flow.42 Maiz et al. (2008) in a meta-analysis of seven studies, including 600 chromosomally normal fetuses with an NT ≥95th percentile, found that an abnormal DV flow pattern at 11 to 14 weeks’ gestation in the presence of an NT ≥3.5 mm was associated with a threefold increased risk of CHDs.34 The finding of a normal DV flow pattern in this setting halved the CHD risk. A recent meta-analysis reported that an abnormal DV A-wave can reveal 83% of CHD cases in the presence of an increased NT and 19% when the NT is normal, with an FPR of 20% and 4%, respectively.36 A study of 40 000 singleton pregnancies with normal chromosomes at 11 to 14 weeks showed that screening for CHD using an increased NT and the reversed A-wave of the DV can detect 47.1% of major CHD with an FPR of 6.7%.37 Instead of evaluating the DV flow in terms of the A-wave (positive, absent, reversed), our group suggested using the DV pulsatility index (PIV) and found an abnormal DVPIV in twothirds of the fetuses with an increased NT, normal karyotype, and CHDs.33 The sensitivity of an abnormal DVPIV for CHDs was 70% with a specificity of 62%. The DVPIV multiples of the median significantly influenced the risk of a CHD (odds ratio 2.4) at any cutoff point suggesting that the DVPIV can be used to increase the specificity of the NT in the identification of CHDs.33 Borrell et al. assessed the best method to combine the NT and DV Doppler in the detection of major CHD in euploid fetuses and demonstrated different detection rates (DRs) depending on the cutoffs of NT or DVPIV used.43 The mechanism of the abnormal DV flow in the presence of CHD is unclear, but the predominance of right-sided obstructive lesions44–46 and atrioventricular septal defects (AVSD) with atrioventricular (AV)-valve regurgitation and hypoplastic left heart syndrome (HLHS)33 suggests that right heart/atrial volume and/or pressure overload or diastolic dysfunction may be responsible. © 2014 John Wiley & Sons, Ltd.
CHD detection and management in the first trimester
Tricuspid regurgitation is frequently observed in 11-week to 14-week fetuses with trisomies47 but may also be found in euploid fetuses with CHDs.39,48–50 The mechanism is not clear but may be related to the reduced diastolic function and still high afterload at this gestational age. Faiola et al. reported an eight times increased risk of CHDs when TR is found at 11 to 13 + 6 weeks’ gestation.38 A more recent study of 40 990 fetuses found an NT >95th percentile, TR, and reversed A-wave in the DV flow in 35.3%, 32.9%, and 28.2% of the 85 fetuses with major CHD, respectively, and in 4.8%, 1.3%, and 2.1% of those without CHD.39 Any one of the three markers was found in 57.6% of the fetuses with CHD (95% CI 47–67.6%) and in 8% of those with normal hearts (95% CI 7.7–8.2%).39 In conclusion, an abnormal DVPIV in the first trimester can detect about 70% of the major CHD and is an indication for referral for specialized fetal echocardiography, even when the NT is normal. The same applies for TR. The risk of CHD increases with increasing NT measurement,28–30 and is further increased in the presence of an abnormal DV flow and/or TR, but is reduced if these findings are absent.39
ACCURACY OF CHD DETECTION Early fetal anomaly scan The first case descriptions of CHD diagnosed in the first trimester with the use of transvaginal echocardiography appeared in the early 1990’s.51,52 After the first report in 1998,53 showing that transabdominal echocardiography left), (B) left ventricular outflow tract view with color Doppler showing the small aorta, (C) right ventricular outflow tract view showing the enlarged main pulmonary artery and ductus arteriosus, (D) sagittal section showing the disassociation between the umbilical vein and hepatic vein due to the absence of a ductus venosus, (E) sagittal section of one of the twins showing the increased nuchal translucency, and (F) sagittal section showing the coarctation of the aorta and the retrograde flow in the distal aortic arch (arrow head). Ao, aorta; CoA, coarctation of the aorta; LV, left ventricle; PA, pulmonary artery; NT, nuchal translucency; RV, right ventricle; VCI, inferior caval vein; v hep, hepatic vein; v umb, umbilical vein scanning provides sufficient resolution thereafter.61,63,67 The detection of the 4CV and three-vessel view can be enhanced by approximately 5% by the addition of the transvaginal approach.5 Moon-Grady et al. compared EFEC
