To evaluate the detection rate of congenital heart defects (CHD) in a non-selected population and to follow outcome after diagnosis.
To evaluate the detection rate of congenital heart defects (CHD) in a non-selected population and to follow outcome after diagnosis.
All 30 149 fetuses/newborns that were scheduled to deliver at our hospital between February 1991 and December 2001 were registered prospectively. Of these, 29 460 (98%) fetuses had a prenatal ultrasound scan at our center. The routine fetal examination at approximately 18 weeks' gestation included the four-chamber view and the great arteries of the fetal heart. The follow-up period was 2–13 years.
Of 97 major CHDs, 55 (57%) were detected prenatally, 16% (9/55) prior to, 66% (36/55) at and 18% (10/55) after the routine scan. Forty-four percent (19/43) of the isolated CHDs, 67% (36/54) of those with associated malformations and 48% (11/23) of the isolated ductal-dependent CHDs were detected. Thirty-eight percent (37/97) had an abnormal karyotype. Of the 55 major CHDs detected, 44% (24) of the pregnancies with lethal/serious fetal malformations were terminated, 15% (8) died in utero, 42% (23) were born alive and 27% (15) were still alive after 2 years. Of the 42 CHDs detected postnatally, 2% (1) were terminated for other reasons, 98% (41) were born alive and 81% (34) were still alive after 2 years.
Prenatal detection of CHD is still a challenge, with a 57% detection rate only. Isolated defects are detected less frequently. The overall outcome suggests that the most severe defects are detected with the present screening setting; only 27% of the babies with major CHDs detected were still alive after 2 years. Data from long-term follow-up will be of importance for the counseling process. Copyright © 2006 ISUOG. Published by John Wiley & Sons, Ltd.
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
The use of ultrasonography has revolutionized prenatal care, and a high number of major fetal malformations are now detected prior to birth. Despite persistent attempts to improve cardiac evaluation during the last 20 years1, 2, the prenatal detection rate of major heart defects has not shown the same improvement as the detection of malformations in other fetal systems. Paradoxically, the heart is one of few organs where published data now show improvement in neonatal morbidity and mortality when defects are prenatally detected3, 4.
In many countries ultrasound examinations are offered to the total pregnant population. Since the majority of congenital heart defects (CHDs) are found among non-selected pregnancies5–7 this could potentially result in a high detection rate of CHD. Several factors seem to play an important role for the optimal imaging of the fetal heart and the consequent detection of heart defects. The number of ultrasound scans during the pregnancy, the gestational age at the time of examination, the experience of the ultrasound operator and the time allocated for the scan all play a role, as well as fetal and maternal factors8, 9.
In Norway, the second-trimester scan at approximately 18 weeks' gestation is the only ultrasound examination offered to all pregnant women. As 98% of the pregnant population receive this examination10, we need to focus on this second-trimester scan in order to detect the majority of the CHDs in the population.
The purpose of this study was to evaluate the detection rate of CHDs in a non-selected population following the introduction of analysis of the four-chamber view and the great arteries to the second-trimester fetal examination. We also wanted to understand the pattern of prenatal recognition of heart defects in an effort to further improve the detection rate. Cases were followed from the time of prenatal diagnosis into the postnatal period to obtain a better understanding of the outcome, in order to improve the accuracy of counseling.
The total population of 30 149 fetuses was collected from a well-defined area consisting of the city of Trondheim and eight surrounding communities. Of these, 29 460 (98%) had a prenatal ultrasound examination performed at our center during the study period from February 1991 to December 2001 and thus comprised the study population. More than 97% of the pregnant women residing in the defined area have a routine ultrasound examination and deliver at Trondheim University Hospital, thus the population is non-selected. The second-trimester routine scans were scheduled at 18 completed weeks of gestation based on the last menstrual period or early clinical assessment. Nurses/midwives, with 3 months to 17 years of experience following basic obstetric ultrasound training, performed the scans. Ultrasound examinations prior to the routine scan were done by physicians with ultrasound experience and performed on clinical indications only.
The routine fetal examinations were scheduled at 30-min intervals. The gestational age was based on the fetal biparietal diameter at the routine examination between 16 and 22 weeks or the biparietal diameter and/or crown–rump length prior to the routine examination. All data from the ultrasound examinations were prospectively registered. In addition to regular fetal biometry, the detailed anatomical survey included the four-chamber view and views of the great arteries of the heart. The heart structures to be seen were: two atria of equal size; two ventricles of equal size; moderator band in the anterior right ventricle; tricuspid valve closer to the apex than mitral valve; intact ventricle septum; crossing of the ascending aorta with main pulmonary artery; aortic arch and ductal arch. From January 1995 a new scan was offered at approximately 20 weeks' gestation if a proper four-chamber view had not been obtained at the routine scan. The 20-week scan included assessment of the four-chamber view, the great arteries and the venous return, and was considered a part of the routine scan when data were evaluated.
When a heart defect was diagnosed or suspected the physicians at the center performed a detailed anatomical survey, in most cases in the presence of a pediatric cardiologist. Fetal karyotyping was offered in all cases and during the first part of the study period it was mainly performed by fetal blood sampling, but later by amniocentesis and fluorescence in situ hybridization analysis (FISH)11. When a karyotype was not obtained, it was considered as probably normal. During the study period we did not routinely check for 22q11 microdeletion.
The heart defects were classified as major or minor. The classification was done retrospectively after a final diagnosis was achieved either by postnatal echocardiogram or by autopsy. The majority of the spontaneous losses after 22 weeks were autopsied. A perinatal pathologist performed all the postmortem examinations12. When postnatal echocardiogram or autopsy were not available, the prenatal video recordings and hard copies were scrutinized to ensure the correct heart diagnosis.
The defects were classified as major when surgical repair was likely to be required because of gross structural complexity of functional significance, e.g. transposition of the great arteries (TGA), hypoplastic left heart syndrome (HLHS), atrioventricular septal defect (AVSD), coarctation of the aorta, and large ventricular septal defects (VSD)13. A heart defect was classified as minor when no intervention was likely to be required, e.g. mild pulmonary stenosis, mild aortic stenosis, small VSDs, small atrial septal defects. Arrhythmias with a structurally normal heart, patent ductus arteriosus, atrial septal defects < 3 mm and absent inferior vena cava with azygos continuation were excluded from this study. Rhabdomyomas were categorized as tumors and thus excluded as heart defects.
The major heart defects were classified into the following groups: isolated if a normal karyotype was present with absence of extracardiac malformations; abnormal karyotype if there was a chromosome aberration with or without extracardiac malformations; and associated malformations if extracardiac malformations were present with normal karyotype.
The most common major heart defects (AVSD, TGA, HLHS, coarctation of aorta) were grouped as simple or complex. A simple heart defect was defined as one without additional cardiac defects, e.g. TGA or AVSD. A heart defect was defined as complex when additional heart defects were present, e.g. TGA with associated single ventricle or HLHS with additional coarctation of the aorta.
A perinatal team at Trondheim University Hospital (obstetrician, nurse/midwife, pediatric cardiologist, social worker) supported the parents and the fetus/newborn from the time of diagnosis of the heart defect until after termination or delivery. (In Norway termination of pregnancy may be approved until approximately 22 weeks' gestation.) When the pregnancy was continued, a specialist pediatric cardiac nurse became a part of the perinatal team. When immediate intervention after birth was expected, the planned delivery took place at the National Hospital in Oslo, where the neonatal cardiac surgical service is centralized in Norway.
Within the first 24 h after birth, all newborns were clinically examined by a pediatrician. If a heart defect was suspected, the child was referred to the pediatric cardiologist for an echocardiogram. Congenital heart defects detected after discharge from the hospital were referred to the Pediatric Cardiology Unit at our hospital. The health-care program in Norway ensures all children have regular physical examinations at child health centers at 6 weeks, 6 and 12 months and at 2, 4 and 6 years. The children included in this study were followed for a minimum of 2 years and a maximum of 13 years. ‘Healthy’ was defined according to the New York Heart Association classification system as a person with cardiac disease without limitations of physical activity or objective evidence of cardiovascular disease14.
During the study period, the ultrasound machines used were Hitachi EUB-415 and EUB-6000 with a 5-MHz and a 3.5-MHz curvilinear transducer (Tokyo, Japan), VingMed CFM 800 with a 5-MHz mechanical sector transducer and VingMed System Five with a 3.5-MHz curvilinear transducer (Horten, Norway).
Pre- and postnatal data were prospectively registered in a computer database. Groups were compared in 2 × 2 contingency tables using Pearson's Chi-square or Fisher's exact test. The level of significance was set at P < 0.05.
In the study population of 29 460 fetuses, 430 had a congenital heart defect. Of those, 97 (23%) were major and 333 (77%) were minor. The incidences of major and minor heart defects were 3.3 and 11.3/1000, respectively, making a total of 14.6/1000 fetuses with heart defects. Figure 1 shows the spectrum of major heart defects in this non-selected population.
The sex distribution among the 97 major heart defects was 50 (52%) females, 47 (48%) males and among the 333 minor heart defects it was 178 (53%) females, 155 (47%) males. The median maternal age for the total population was 28.8 (range, 15–53) years, and for the CHD population it was 28.5 (range, 16–48) years. The median gestational age at the time of the diagnosis of the major heart defects was 18.0 (range, 11–39) weeks.
Of the 97 major heart defects, 55 (57%) were detected prenatally. Forty-six percent of all the major defects in the study population were detected prior to 22 gestational weeks. Table 1 lists the detected and undetected major heart defects and the time of detection. Of the 14 defects classified as complex, 11 (79%) were detected prenatally compared to 22/40 (55%) of the CHDs classified as simple (P = 0.12). Of the 42 major CHDs not detected prenatally, one was detected by autopsy and 6/41 (15%) in those born alive were detected after discharge from the hospital.
|Time of detection|
|Early||Routine scan||Late||Total||Not detected|
|Coarctation of aorta, total||9||9||3||33||1||11||4||44||5||56|
|Coarctation of aorta, simple||7||7||3||43||3||43||4||57|
|Coarctation of aorta, complex||2||2||1||50||1||50||1||50|
|Ventricular septal defect||9||9||2||22||1||11||3||33||6||67|
|Tetralogy of Fallot||7||7||1||14||2||29||3||43||4||57|
|Tricuspid atresia/critical stenosis||2||2||1||50||1||50||1||50|
|Common arterial trunk||1||1||1||100||1||100|
|Aortic valve stenosis||1||1||1||100|
Table 2 shows the 9/97 (9%) major heart defects detected prior to the routine scan, indications for the early scan, findings and outcomes. An AVSD was seen in 4/9 (44%) of the cases detected prior to the routine scan and represented 19% of all AVSDs in the population. The three with HLHS detected early comprised 50% (3/6) of the HLHS detected prenatally and 30% of the defect in the total population (Table 1).
|GA||Indication early scan||Defect||Associated findings||Karyotype||Outcome|
|10||Bleeding||AVSD||NT 3 mm||Trisomy 18||TOP week 13|
|Moderate general edema|
|11||Bleeding||Critical tricuspid stenosis||Omphalocele||Normal||TOP week 16|
|12||Bleeding||Primum ASD||NT 14 mm||45,X0||TOP week 18|
|General edema 3–4 mm|
|12||Previous children CHD||HLHS||NT 4.5 mm||Normal||IUFD week 37|
|12||Genetic counseling (age)||HLHS, complex||Situs inversus||Normal||TOP week 15|
|13||Gestational age||HLHS||Multiple defects||Trisomy 13||TOP week 17|
|Genetic counseling (age)|
|13||Previous chromosome aberration||AVSD||NT 6 mm||Trisomy 21||IUFD week 15|
|15||Previous spontaneous abortion||AVSD, complex||Balanced translocation||Live birth, surgery|
|Died 6 months old|
|16||Trisomy 21 diagnosed by amniocentesis (age)||AVSD||NT 5–6 mm||Trisomy 21||TOP week 17|
Of the 10 major heart defects detected after the routine scan, six were missed at 18 weeks' gestation, while four affected pregnancies whose first scan at our institution was after 29 weeks' gestation. The late scans were done for a variety of clinical reasons. Of the six missed at the time of the routine scan, four (67%) would have had a normal four-chamber view (two with tetralogy of Fallot, one with TGA and one with pulmonary valve stenosis).
There was one false positive diagnosis in the total series. At a 42-week scan for overdue pregnancy, coarctation of the aorta was suspected owing to dilated right side of the heart and a possible aneurysm of ductus arteriosus. Postnatal echocardiography showed a normal heart.
Table 3 shows the detection rate for isolated and ductal-dependent CHDs and those with abnormal karyotype or associated malformations. Of the isolated major heart defects missed prenatally, 14/24 (58%) would be expected to have an abnormal four-chamber view at the second-trimester routine examination.
|Ductal dependent||Abnormal karyotype or ass. malf.|
|Coarctation of aorta, total||9||9||4||44||2||50||5||56||4||80|
|Ventricular septal defect||9||9||3||33||6||67||3||50|
|Tetralogy of Fallot||7||7||2||29||5||71||3||60|
|Tricuspid atresia/critical stenosis||2||2||1||50||1||50||1||100|
|Common arterial trunk||1||1||1||100||1||100|
|Aortic valve stenosis||1||1||1||100||1||100|
There were 333 minor heart defects in the study population (Table 4). Of these, 12 (4%) were detected prenatally (Table 5). Of the 321 not detected prenatally, 11 (3%) were detected by autopsy and 72/310 (23%) of those born alive were detected after discharge from the hospital. Isolated VSDs and ASDs constituted 70% (226/321) of the minor CHDs not detected prenatally (Table 4).
|CHD diagnosis||Not detected prenatally|
|Total present||Abnormal karyotype||Normal karyotype Associated malf.|
|VSD, muscular and ASD secundum||14||4.2||14||100|
|Pulmonary valve stenosis||12||3.6||12||100||2||17|
|Aortic valve stenosis||11||3.3||11||100||1||9||2||18|
|VSD, perimembranous and ASD secundum||9||2.7||9||100||3||33||3||33|
|VSD, perimembranous and VSD, muscular||6||1.8||6||100||3||50||1||17|
|Pulmonary artery stenosis, peripheral||5||1.5||5||100||1||20|
|Pulm. art. stenosis, peripheral, and pulmonary valve stenosis||2||0.6||2||100||1||50|
|Coarctation of aorta, mild||2||0.6||2||100||1||50|
|ASD secundum and partial APVR||1||0.3||1||100|
|Aortic valve stenosis and ASD secundum||1||0.3||1||100|
|Pulmonary valve stenosis and VSD, muscular||1||0.3||1||100|
|Pulmonary valve stenosis and ASD secundum||1||0.3||1||100|
|Tricuspid valve malformation||1||0.3||1||100|
|Coarctation of aorta, mild, and VSD, perimembranous||1||0.3||1||100||1||100|
|Aortic hypoplasia, mild||1||0.3|
|Left ventricle hypoplasia, mild||1||0.3|
|Tri 18||Unbalanced translocation||Associated malformations|
|VSD, perimembranous||9||75.0||9||100||8||89||1||11||4 TOP, 5 IUFD|
|Aortic hypoplasia, mild||1||8.3||1||100||Live birth|
|LV hypoplasia, mild||1||8.3||1||100||1||100||TOP|
Fetal karyotype was obtained prenatally in 47/55 (85%) of the cases with major heart defects detected and postnatally in two (trisomy 21; 22q11 deletion); in one case the amniotic fluid cells failed to grow, in two fetal death occurred prior to the performance of the amniocentesis and three women refused amniocentesis. Amniocentesis was performed for all the 12 minor heart defects detected. Chromosome aberrations for all major CHDs are listed in Table 6 and for the CHDs detected prior to the routine scan in Table 2. Tables 7 and 8 show the number of chromosome aberrations and associated malformations for the major CHDs detected prenatally and postnatally, respectively. Of the CHDs grouped as complex and simple, 2/14 (14%) and 21/40 (53%) had abnormal karyotype, respectively (P = 0.02) (Tables 7 and 8). Five (50%) of the 10 CHDs detected after the routine scan had a chromosomal aberration.
|Tri 21||Tri 18||Tri 13||45,X0||Other|
|Coarctation of aorta||9||9||5||56||3||60||2||40|
|Ventricular septal defect||9||9||4||44||1||25||3||75|
|Tetralogy of Fallot||7||7||3||43||1||33||1||33||1||33|
|Tricuspid atresia/critical stenosis||2||2|
|Aortic valve stenosis||1||1|
|Common arterial trunk||1||1||1||100||1||100|
|Abnormal karyotype||Associated malform.||TOP||IUFD||Born alive||Surgery||Died||Alive > 2 years||Total died/TOP|
|Coarctation of aorta, total||4||7||4||100||2||50||1||25||1||25||1||100||4||100|
|Tetralogy of Fallot||3||5||2||67||1||33||1||33||1||33||1||33||1||100||1||33||2||67|
|Pulm. valve stenosis||1||2||1||100||1||100|
|Common arterial trunk||1||2||1||100||1||100||1||100||1||100|
|Abnormal karyotype||Associated malform.||TOP||Born alive||Surgery||Died||Alive > 2 years||Total died/TOP|
|Coarctation of aorta, total||5||12||1||20||5||100||5||100||5||100|
|Tetralogy of Fallot||4||10||1||25||1||25||4||100||4||100||4||100|
|Aortic valve stenosis||1||2||1||100||1||100||1||100|
The number of fetuses with associated malformations including abnormal karyotype was significantly higher among those prenatally detected 36/55 (65%) than those postnatally detected 18/42 (43%) (P = 0.03). The same numbers for those with abnormal karyotype only were 26/55 (47%) and 11/42 (26%), respectively (P = 0.03).
There was postnatal verification of the diagnoses of CHD in 94 (97%) of the 97 major defects and in 331 (99%) of the 333 minor defects. All five cases not verified postnatally by an autopsy had an abnormal karyotype (three trisomy 18, two trisomy 21) (Table 9). In 49 (94%) of the 52 major heart defects detected and postnatally verified the main pre- and postnatal diagnoses were in agreement. In the remaining three cases (6%) the pre- and postnatal heart findings were not in complete agreement but the discrepancies did not alter the outcome. An abnormal karyotype was present in all these cases (two trisomy 18 and one Turner syndrome with general hydrops fetalis). Among the 10 minor heart defects detected and postnatally verified, the main pre- and postnatal heart diagnosis was in agreement in seven (70%) fetuses. In the remaining three cases, the pre- and postnatal heart findings were not in complete agreement but the discrepancies did not alter the outcome (two trisomy 18 and one lissencephaly type II).
|GA||Prenatal findings||Prenatal heart diagnosis||Karyotype||Outcome|
|Major congenital heart defects|
|11||NT 3 mm|
|Moderate general edema||AVSD||Trisomy 18||TOP week 13|
|13||NT 6 mm|
|Cystic hygroma, skin edema||AVSD||Trisomy 21||IUFD week 15|
|18||Twin pregnancy (1 healthy)||AVSD||Trisomy 21||IUFD week 25|
|Minor congenital heart defects|
|15||Omphalocele||VSD, perimembranous||Trisomy 18||Stillbirth week 35|
|Trisomy 18 features|
|19||Twin pregnancy (1 healthy)||VSD, perimembranous||Trisomy 18||Stillbirth week 40|
|Trisomy 18 features|
The outcomes of the major cases detected and not detected prenatally are described in Tables 7 and 8, respectively. Figure 2 shows the broad effect of the outcome consequences. Table 10 shows the clinical status of the terminated fetuses. One (2%) of the 42 with a major heart defect not detected prenatally was terminated owing to prenatal findings of arthrogryposis multiplex, narrow thorax and cleft palate. The heart defect, primum ASD, was found during autopsy. Intrauterine fetal death occurred in 8/55 (15%) of the fetuses with a major defect detected, but in none of the 42 fetuses with a defect that was not detected (P = 0.01).
|Turner syndrome (general hydrops fetalis)||1||4||Lethal|
|46,X, + mar (conjoined twins)||1||4||Lethal|
Among the fetuses with a major heart defect detected, 23/55 (42%) were born alive compared to 41/42 (98%) of those not detected prenatally (P < 0.001). When the terminated pregnancies were excluded, the corresponding numbers were 23/31 (74%) and 41/41 (100%), respectively (P = 0.001).
At 2 years of age, 15/55 (27%) of the children with a major CHD detected and 34/42 (81%) of those not detected prenatally were still alive (P < 0.001). The outcomes after 2 years when the terminated pregnancies were excluded were 15/31 (48%) and 34/41 (83%), respectively (P = 0.01). Of those who survived until 2 years of age, all were still alive at a median follow-up of six (range, 2–13) years.
Of the 55 major heart defects detected prenatally, 45 (82%) were detected prior to 22 weeks' gestation when termination of pregnancy was an option. Among the 24/45 (53%) who opted for a termination, 21 (87%) had associated malformations including abnormal karyotype (Table 10), compared to 10/21 (48%) of those who continued their pregnancy (P = 0.01).
Table 5 shows the outcome of the 12 fetuses with minor heart defects detected prenatally. The one live birth had a possible genetic defect and a syndrome resulting in psychomotor disability. Of the 321 fetuses with minor defects not detected, 310 (97%) were alive after a follow-up period of 2–13 years, seven (2%) were terminated for other reasons (three abnormal karyotype, four serious defects in the central nervous system) and four (1%) died (one intrauterine fetal death, three premature deliveries of multiple pregnancies).
Isolated ventricular septal defects without extracardiac malformations totaled 56% (188/333) of the minor heart defects (Table 4). Of those, 162 (86%) were muscular and 26 (14%) were perimembranous. During the first year of postnatal life, 77/162 (48%) of the isolated muscular and 3/26 (12%) of the isolated perimembranous VSDs closed spontaneously.
The present study includes data from a large, well-defined, non-selected population. The strength of this study is the prospective collection of data from the time of diagnosis through to a follow-up period of 2–13 years. Inclusion of pathological results, postnatal echocardiography and long-term follow-up ensured accurate ascertainment of population data and outcome that makes it possible to draw prognostic consequences necessary for the counseling process. The incidence of major CHDs (3.3/1000 fetuses) was consistent with data from other non-selected populations6, 15, 16.
Several publications address the detection rate of CHDs based on the four-chamber view and the great arteries6, 16–19. Comparison between these studies might be difficult owing to differences in population selection, study design, early obstetric management and heart classification. Our data show that inclusion of the great arteries in addition to the four-chamber view improved the total prenatal detection rate of major heart defects from 39%15 to 57%. The detection rate at the second-trimester fetal scan improved from 26%15 to 37% (Table 1). Although these results seem similar to those of Carvalho et al.6, who detected 38% of the major CHDs in their non-selected population at the 18–23-week anomaly scan, a comparison seems difficult since 80% of the women in their study population had an early scan detecting 20% of the major CHDs. In addition, the anomaly scan was carried out slightly later than our 17–20-week gestation routine scan, making it easier to obtain the heart structures9.
An abnormal four-chamber view is the most common indication for fetal echocardiography and thus contributes to the detection of CHDs5, 20, 21. In the present study 59% of the isolated CHDs not detected prenatally were expected to have an abnormal four-chamber view. On the other hand, we detected 47% of the isolated CHDs with a normal four-chamber view, indicating that the four-chamber view is not the only trigger for detection. Wong et al.22 compared findings at the routine scan between tertiary versus non-tertiary institutions and found a significant difference in the detection rate when the four-chamber view was abnormal (78% versus 47%, respectively) compared to when it was normal (17% versus 7%, respectively). Those findings are slightly different from ours and suggest that there are several factors involved when the four-chamber view is evaluated, such as experience of the operator, gestational age, fetal position, maternal factors and the time factor8, 23.
AVSD was the most frequently diagnosed fetal cardiac abnormality (71%) in our study, as has also been the case in other studies22, 24–26. The abnormal four-chamber view is the characteristic sonographic feature of AVSD; interestingly the same characteristic view found in 58% of the isolated defects that we overlooked. On the other hand, 90% of the AVSDs in our population had associated malformations. As we detected significantly more of the major CHDs with associated malformations than the isolated CHDs (Table 3), this might be a reason for the high detection rate of AVSD22, 25, 27.
The isolated defects might be used to evaluate our true ability to examine the fetal heart, as the heart in those cases is the only affected organ. Although our data showed only a 44% detection rate of isolated heart defects, the results were better than those of other studies reporting between 11 and 23% detection25, 27. The data suggest that more focus needs to be put on the fetal heart itself since associated malformations seem to play an important role in the detection of CHDs.
Over the past few years, the 11–14 weeks' nuchal translucency screening has become routine in several countries, resulting in early detection of fetuses with chromosome aberrations and congenital heart defects6, 19. In our study, scans prior to the 18-week routine examination were performed on clinical indication only. Still, nine percent of all major defects detected were found prior to 17 weeks. Five of those nine detected prior to the routine scan had a chromosome aberration (Table 2), which supports nuchal translucency as a marker for chromosomal defects28, 29 and heart defects30–33. Although chromosomal defects were in the majority among those detected early, the nine percent detection rate of major heart defects implies a possible increase in the total detection rate of CHDs if the potential of the early scan is utilized in a systematic way and offered to the total population.
Eleven percent of the major CHDs were detected after the routine scan. It is well known that the nature of some heart defects makes them difficult to detect during the first half of pregnancy34–36. This makes it difficult to detect all CHDs at the second-trimester routine scan. Nevertheless, 82% of the detected major CHDs were detected prior to week 22, indicating that the increase in the total detection rate from 39%15 to 57% was due to diagnostic improvement in the first half of the pregnancy. However, the fact that 16% of the detected CHDs were diagnosed prior to the routine scan makes the true improvement in the detection rate at the routine fetal examination difficult to assess. Those nine early cases were serious defects and might have ended in spontaneous loss prior to the routine scan, thus, including those cases may have falsely improved our detection rate. As the incidence of major CHDs in our population has been unchanged over the last years, there is no reason to believe that the early cases represent a bias in our previous series and thus do not interfere with the detection rate15.
The main purpose of diagnosing major heart defects prenatally is to detect those whose long-term prognosis will benefit from such diagnosis. Despite improvements in detection rates, they are still disappointing. The four-chamber view itself is hypothesized to detect 50% of the major CHDs2, and by adding the great arteries one would expect the detection rate to increase even further. However, studies from unselected populations offering only one second-trimester scan during the pregnancy report detection rates between 19 and 61% when the great arteries are included16, 22, 24, 37, reaching the detection rate expected from the four-chamber view only. Studies offering an early scan and a second-trimester scan between 20 and 24 gestational weeks6, 38 show a prenatal detection rate as high as 75%. This clearly emphasizes the impact of the 11–14-week scan in detecting major CHDs and the advantage of performing the fetal heart examination after 20 weeks' gestation9, 39. Our data support this finding, as four out of the six CHDs with a normal four-chamber view missed at the 18-week routine scan were detected at a later gestation. In Norway, the upper limit for termination of pregnancy is 22 weeks, thus CHDs need to be detected prior to 22 weeks to allow the option of termination.
It has been shown that prenatal detection of ductal-dependent heart defects significantly improves the postnatal mortality3, thus it is important to detect fetuses with this type of CHD. We detected 48% of the ductal-dependent isolated CHDs, leaving more than one half undetected. As the four-chamber view in those CHDs is often normal at the routine scan (Table 3) an incorporation of the three-vessel view40 with color Doppler41 might be considered to optimize the evaluation of the great arteries. By using the three-vessel view it may be possible to improve detection of all CHDs with abnormalities of the outflow tracts and the great arteries42.
The outcome for the fetuses with a heart defect detected postnatally was significantly better than for those detected prenatally. The poorer outcome for the fetuses with prenatally detected CHDs was most probably owing to a greater degree of severity of heart defect43. This was clearly emphasized by the significant difference between the prenatally and postnatally detected cases regarding associated malformations, and the seriousness of the malformations in the terminated pregnancies in our series. Fifty-eight percent had associated lethal chromosome aberrations or had developed a life-threatening condition at an early gestational age, and the rest, except for the trisomy 21 cases, were severe anomalies requiring major surgery (Table 10). The absence of intrauterine deaths among those not detected prenatally may also indicate a less severe degree of heart defect. Even when the terminated pregnancies were excluded, the difference in outcome between the two groups was significant.
Boldt et al.20 found the same results when the outcomes of 93 prenatally detected major CHDs were evaluated. Their 36% (10/28) termination rate of CHDs detected prior to 24 gestational weeks was lower than our 53%, while the rate of intrauterine fetal deaths was similar (8% and 15%, respectively). On the other hand, 57% (43/76) of their cases born alive died after birth, compared to 35% in our study. The outcome data after a median follow-up of 3.8 (range, 0.3–13) years showed 34 (37%) of the 93 cases prenatally detected alive. This is a similar poor result to our 27% alive at the age of 2 years. Despite the lower termination rate in the study of Boldt et al.20 the severity of the heart defects seems to make it difficult to prevent neonatal death in those prenatally detected cases. Interestingly, after a median follow-up of 6 years in the present study, there were no more deaths among the children who survived until 2 years of age, indicating that the first 2 years might be the most critical postnatal period. The long-term morbidity rate still needs to be evaluated to get a complete picture of the outcome.
Abnormal karyotype and extracardiac malformations are strongly associated with CHDs5, 6, 15, 22, 44. Among the major CHDs in our study, 38% had a chromosome aberration, a number slightly higher than the 25% reported from another non-selected population6, but in agreement with another large study45. During the study period we did not routinely check for 22q11 microdeletion, common in conotruncal heart defects46; this may have affected the reported incidence of chromosome aberrations in our series. Associated malformations seem to play an important role not only for prenatal detection of heart defects, as discussed, but also for the outcome20, 47. In our study as much as 65% of the CHDs detected had associated malformations (Table 7). This information is of importance when a heart defect is detected prenatally and the parents expect counseling about the prognosis48. The prognosis, and thus counseling, are strongly dependent on the presence or absence of chromosomal defects and/or associated malformations (Table 7). This demonstrates the importance of cytogenetic evaluation and data on long-term follow-up to optimize the prognostic counseling20, 49.
Recent data show improving results for newborns with a heart defect detected prenatally3, 4, 50, 51, not only in the preoperative period3, 52–55, but also improvement in the results from postnatal surgery56 and possibly long-term outcome54, 57. Therefore, there is reason to believe that some of the fetuses with heart defects missed prenatally, and that died postnatally, could have survived with prenatal diagnosis. As there is also reason to believe that this progression will continue through the years to come, it seems ethically correct to continue the effort to increase the prenatal detection rate of congenital heart defects. The importance of detecting major CHDs is supported by the neonatologists' efforts to detect those defects prior to discharge from hospital58, 59 and also by suggesting postnatal screening procedures such as echocardiography or pulse oximetry60–63.
Teaching and training are key issues when the prenatal detection of CHDs is discussed6–8, 23, 64. In the data presented here it seems difficult to recognize any pattern in the prenatal detection, as the detection rate was approximately 50% both for the isolated defects with normal or abnormal four-chamber view, and for the ductal-dependent defects, emphasizing the need for specific focus on the fetal heart during the ultrasound examination. As this implies extra resources, a scientific evaluation of the teaching and training aspects is currently being undertaken.
In conclusion, prenatal detection of fetal heart defects remains challenging. On the one hand, the overall detection rate is still inferior compared to those of other fetal organ defects. On the other hand, we are faced with the poor prognosis emphasizing that CHDs also represent therapeutic problems. In addition to solving the diagnostic and therapeutic problems we have to accept that future discussions concerning the fetal heart not only will focus on the detection rate of CHD, but also will involve difficult ethical problems, including decisions regarding continuation of the pregnancy. Data from long-term follow-up of children with prenatally diagnosed congenital heart defects will be of importance for the counseling process.
We wish to thank the nurses/midwives taking part in this study, Nancy Lea Eik-Nes for revising the manuscript, Pål Romundstad, MSc PhD, for help with the statistical analyses, and the publisher Sandvik Mor & Barn [Sandvik Mother & Child], Stavanger, Norway, for financial support.