AMERICAN JOURNAL OF PERINATOLOGY/VOLUME 8, NUMBER 6
November 1991
CONGENITAL HEART DISEASE AND FETAL THORACOABDOMINAL ANOMALIES: ASSOCIATIONS IN UTERO AND THE IMPORTANCE OF CYTOGENETIC ANALYSIS Mark Fogel, M.D., Joshua A. Copel, M.D., Mark T. Cullen, M.D., John C. Hobbins, M.D., and Charles S. Kleinman, M.D.
ABSTRACT
The association of congenital heart disease (CHD) with extracardiac malformations (ECMs) has recently been reviewed. These data form the basis for a previous study from our laboratory,1 and for most of the daily counseling that occurs when fetal anomalies are detected. Information reported in pediatric data bases may have an important bias, since any association uncovered between CHD and ECMs will be dependent on the infant surviving to be diagnosed. Even studies seeking to capture all newborns, or even stillbirths, may contain such a bias, for losses of late second or early third trimester fetuses may not appear. In previous work we have reported a substantially greater association between fetal heart disease and chromosome abnormalities than had been appreciated from the pediatric literature.2 The present study was undertaken to determine if a similar shift exists toward a greater association of ECMs and
Downloaded by: Universite Laval. Copyrighted material.
We examined the frequency with which congenital heart disease (CHD) and cytogenetic abnormalities were found associated with omphalocele, gastroschisis, duodenal atresia and posterior diaphragmatic hernias. We performed fetal echocardiograms on 80 patients with these diagnoses and found congenital heart disease in 13 of 37 with omphalocele (35%), 2 of 17 with gastroschisis (12%), 4 of 15 with duodenal atresia (27%), and 2 of 11 with posterior diaphragmatic hernia (18%). Karyotypes were obtained in 74 and were abnormal in 24 (32%). Although most fetuses with these extracardiac malformations and abnormal karyotypes had associated CHD, many did not. Normal karyotypes were found in 69% of fetuses with CHD and omphalocele, and 50% of fetuses with CHD and duodenal atresia. We conclude that CHD may be present in fetuses with extracardiac malformations whether or not the karyotype is normal and that the prenatal evaluation of fetuses with these lesions should include both karyotype and fetal echocardiography. Although karyotypes play an important role in prenatal diagnosis, they are not predictive of normal cardiac structure when normal in the abnormalities studied. Even when the karyotype is normal in the presence of these abnormalities, fetal echocardiography is indicated.
CHD in in utero studies. The hypothesis to be tested was that the true incidence of cardiovascular malformations associated with these ECMs was higher prenatally than is found in the postnatal literature. We were further interested in examining whether abnormal karyotypes were more prominent in those fetuses with combinations of CHD and ECMs than was usually associated with either alone.
MATERIALS AND METHODS
Between January 1, 1985, and December 31, 1988, we performed fetal echocardiograms on all fetuses with suspected ECMs who were referred to the Yale Perinatal Unit for ultrasound studies. In addition, all fetuses referred for fetal echocardiography for standard indications3 underwent full Level
Yale Fetal Cardiovascular Center, Departments of Obstetrics & Gynecology, Pediatrics, and Diagnostic Imaging, Yale University School of Medicine, New Haven, Connecticut Reprint requests: Dr. Copel, Department of Obstetrics & Gynecology, Yale University School of Medicine, 333 Cedar St., PO Box 3333, New Haven, CT 06510-8063 Copyright © 1991 by Thieme Medical Publishers, Inc., 381 Park Avenue South, New York, NY 10016. All rights reserved.
411
AMERICAN JOURNAL OF PERINATOLOGY/VOLUME 8, NUMBER 6
RESULTS
During the study period we performed fetal echocardiograms on 1665 fetuses of 1626 mothers at the Yale Fetal Cardiovascular Center. A total of 225 ECMs was seen, of which 80 had one or more of the ECMs under study (Table 1).
Table 1.
No. Omphalocele Gastroschisis Duodenal atresia Diaphragmatic hernia
37 17 15 11
Omphalocele
Forty-one omphaloceles were diagnosed prenatally over the study period; however, only 37 (90%) were actually found to have omphalocele postnatally. Three of the four misdiagnoses had gastroschisis and are included in the gastroschisis category of this study. In each of these the location of the defect was lateral to the umbilicus, excluding the possibility of a ruptured omphalocele. The fourth false-positive diagnosis of omphalocele had a cyst of the base of the umbilical cord due to an arteriovenous malformation. That fetus has also been excluded from further consideration here. The final study group is comprised of the 37 confirmed cases of omphalocele. Thirteen (35%) had CHD as detailed in Table 2. The most common type of CHD was ventricular septal defect (VSD) in four (31%), followed by double-outlet right ventricle in three (23%). Complete pentalogy of Cantrell was seen in five (omphalocele, diaphragmatic hernia, sternal defect, pericardial-peritoneal connection, and intrinsic heart disease), and incomplete variations of the pentalogy were seen in two others (total 7 of 37, 19%). There were two fetuses with complete atrioventricular septal defects (CAVSD), both of whom had trisomy 18. Of particular interest, one set of monochorionic diamniotic twins with trisomy 18 were discordant for the type of structural cardiac lesion present (hypoplastic right heart syndrome and CAVSD) and one had anencephaly in addition to the omphalocele. Table 2 details the types of structural cardiac abnormalities seen, as well as the karyotype results. Karyotypes were available from 33 of the 37 omphalocele patients (89%) and of those, five were abnormal (15%, Table 2). Three of these abnormalities were trisomy 18, all with CHD (two CAVSD and one tricuspid atresia and hypoplastic right heart). All of the pentalogy syndrome fetuses had normal chromosomes. One fetus with mosaic translocation of chromosome 1 onto the Y chromosome had omphalocele with sirenomelus and hydrocephalus but no cardiac disease. An inversion of chromosome 9 was also detected, but there was no deletion evident, suggesting that this was a variant of normal. This fetus had a single ovary in addition to the omphalocele, but no CHD. A fetus with trisomy 13 had hydranencephaly and a VSD. Of the omphalocele patients who had no car-
Extracardiac Malformations
CHD' * (%)
AK/# 1'(* (%)
13 2 4 2
5/33 2/16 7/15 4/10
(35) (12) (27) (18)
(15) (13) (47) (40)
CHD & AKI CHD (%) 4/13 1/2 2/4 2/2
(31) (50) (50) (100)
NL K* &CHDI NL K (%)
9/27 1/14 2/8 0/6
412 *CHD: congenital heart disease; AK: abnormal karyotype; # K: number in whom karyotype performed; NL K: normal karyotype.
(33) (7) (25) (0)
Downloaded by: Universite Laval. Copyrighted material.
II ultrasound examination as a part of the fetal echocardiogram. We reviewed all the hospital and private physician records of mothers who presented to the Yale Fetal Cardiovascular Center over the study period whose fetuses were identified to have any of the following: gastroschisis, omphalocele, diaphragmatic hernia, or duodenal atresia. These conditions were chosen because they were the most common prenatal diagnoses encountered, and too few fetuses were encountered with other ECMs to permit meaningful evaluation of any associations. Scans were performed between 16 and 40 weeks' gestation and included a Level II scan as well as fetal echocardiogram, the latter of which was reviewed by both an obstetrician and pediatric cardiologist with extensive experience in fetal echocardiography. Standard tomographic views of the heart were sought (four chamber, short axis ventricular, short axis great arterial, long axis left ventricular, pulmonary arteryductus, and aortic arch). Level II scans were performed with GE model RT3600 (General Electric, Milwaukee, WI) and echocardiograms with HewlettPackard model 77020AC (Hewlett-Packard, Andover, MA) ultrasound machines using 5 or 3.5 MHz transducers. Genetic evaluations were carried out before delivery by amniocentesis or fetal blood sampling, or after delivery by analysis of peripheral blood and examination by experienced dysmorphologists. If the patient was delivered at Yale-New Haven Hospital, the medical records of both mother and child, as well as genetics records, were reviewed. If the fetus was aborted or the neonate died postnatally, autopsy records were examined, when available. Those not delivering at Yale-New Haven Hospital were followed by contacting the referring physician. All neonates suspected of having congenital heart disease were examined by chest radiographs, electrocardiograms, two-dimensional ultrasound, and cardiac catheterization, as appropriate, to confirm the prenatal diagnosis.
November 1991
FETAL CARDIAC A N D EXTRACARDIAC ANOMALIES/Fogel, et al. Table 2. Patient 1 2 3a 3b 4 5 6 7 8 9 10 11 12 13 14
Congenital Heart Disease and Karyotypes in Omphalocele
Type of CHD*
Karyotype
Other ECMs
Double-outlet right ventricle Hypoplastic left heart syndrome Complete atrioventricular septal defect Hypoplastic right heart Complete atrioventricular septal defect Ventricular septal defect Transposition of the great arteries, ventricular septal defect, pulmonic stenosis Double-outlet right ventricle Atrial septal defect, ventricular septal defect Double-outlet right ventricle Ectopia Tetralogy of Fallot, ectopia Ventricular septal defect, ectopia None None
Normal Normal Trisomy Trisomy Trisomy Trisomy Normal
Cleft lip None MCA* Anencephaly MCA MCA None
18 18 18 13
Not available Normal Normal Normal Normal Normal Complex mosaic* Inv 9*
? Amniotic bands Cantrell pentalogy Cantrell pentalogy Cantrell pentalogy Cantrell pentalogy Cantrell pentalogy Sirenomelus, hydrocephaly Single ovary
diac disease and a normal karyotype, two had associated anterior diaphragmatic hernias without having the full pentalogy syndrome, and three others had an associated imperforate anus. One of those had the omphalocele, exstrophy of the bladder, imperforate anus, spinal defects (OEIS) syndrome. In all, four patients (11% of the group) had an abnormal karyotype and an associated cardiac lesion, and only one fetus had an abnormal karyotype in this group without CHD. Abnormal chromosomes were present in 31% of all fetuses in whom cardiac lesions were found in association with omphalocele. Thirty percent of fetuses with omphalocele and normal karyotypes had CHD. Only one of the 37 (3%) fetuses with omphalocele had a false-positive fetal echocardiogram. That fetus was thought to have a small VSD from the fetal echocardiogram. Distortion of the cardiac position and anatomy by the omphalocele were thought to be responsible for the abnormal ultrasound appearance of the heart. None of the fetuses had false-negative fetal echocardiograms. Gastroschisis
Seventeen patients with gastroschisis were seen during the study period, including three thought prenatally to have omphalocele. Two of the 17 (12%) were found to have cardiac disease (Table 3). One fetus had a CAVSD, and another had a simple VSD. One patient had situs inversus in addition to the gastroschisis and a structurally normal heart. Karyotypes were obtained in 16 of the 17 (94%), and a normal clinical examination was present in the remaining neonate. There was one abnormal karyotype (6%): trisomy 18 in a fetus also found to have a CAVSD. One of the 16 fetuses with gastroschisis and a normal karyotype had CHD (6%), a perimembranous VSD. There were no false-negative or falsepositive diagnoses of CHD in this group.
Table 3. Congenital Heart Disease and Karyotypes in Gastroschisis Patient
Type of CHD
Karyotype
Other ECMs
1
Complete atrioventricular septal defect Ventricular septal defect
Trisomy 18
MCA*
Normal
None
2
*MCA: multiple congenital anomalies (stigmata of trisomy syndrome).
Diaphragmatic Hernia
Eleven posterior diaphragmatic hernias were detected. Two had associated CHD (18%): one aortic atresia with a VSD, and the other CAVSD with coarctation of the aorta. Ten karyotypes were obtained (91%) of which four were abnormal (40%), including both of the patients with CHD. The four abnormal karyotypes encountered included one Turner syndrome, one trisomy 18 with inversion of chromosome 1, who had multiple congenital anomalies (but a normal heart), one trisomy 21 who also had CHD (a VSD with aortic atresia), and a patient with CAVSD and coarctation of the aorta who had multiple congenital anomalies and a duplication of chromosome 1 (Table 4). Duodenal Atresia
Duodenal atresia was suspected in 16 patients prenatally, although one of the mothers was ethanol intoxicated at the time fetal gastric distention was observed, and the apparent double bubble resolved on repeat scan in 48 hours. That fetus had multiple stigmata of the fetal alcohol syndrome, including CHD, but is not considered further here. There were, then, 15 patients with the prenatal diagnosis of duodenal atresia, of whom four had 413
Downloaded by: Universite Laval. Copyrighted material.
*MCA: multiple congenital anomalies (stigmata of trisomy syndrome); Complex mosaic: karyotype 46,XY/46,X,ter(Y)t(1; Y)(q21 ;q12); Inv 9: Karyotype 46,XX inv (9q).
AMERICAN JOURNAL OF PERINATOLOGY/VOLUME 8, NUMBER 6 Table 4. Patient
November 1991
Congenital Heart Disease and Karyotypes in Posterior Diaphragmatic Hernia
Type of CHD
Karyotype
Other ECMs*
Ventricular septal defect, aortic atresia Complete atrioventricular septal defect, coarctation of aorta None None
Trisomy 21 46,XYdup(1)(q12;q23) Trisomy 18 45,X
Trisomy 21 MCA+ MCA 45,X
CHD (27%). Three were recognized prenatally, including one each with an atrial septal defect (ASD) (as a component of a partial atrioventricular septal defect), a VSD, and CAVSD. A false-negative fetal echocardiogram occurred in a patient with normal chromosomes, an annular pancreas, malrotation of the gut, and a perimembranous VSD. There were no false-positive diagnoses of CHD in this group. All 15 patients with duodenal atresia had karyotypes obtained, and seven (47%) were abnormal. There were five cases of trisomy 21 and two of trisomy 13. Two of the trisomy 21 fetuses had CHD (40%). The apparent absence of CHD in the two fetuses with trisomy 13 was confirmed by autopsy. Two additional patients with CHD had normal karyotypes (both fetuses had simple VSDs); therefore 50% of those with CHD and duodenal atresia had normal chromosomes (Table 5). One of these fetuses had the vertebral, anal, tracheal, esophageal, renal (VATER) sequence. DISCUSSION
The current study was undertaken to examine whether the designated fetal thoracoabdominal defects were associated with CHD more often than has been previously reported in pediatric studies. We hoped to reduce ascertainment bias by scanning all fetuses with these prenatal diagnoses, and so include many of those that might be lost to pediatric studies due to intrauterine or early neonatal death. Our previous finding that chromosomal abnormalities were present in 38% of fetuses with CHD, compared with 6% reported in pediatric series,4 furthermore, led us to speculate that if more CHD was associated
Table 5.
414
with these ECMs in utero, the common thread might be a parallel increase in chromosomal abnormalities. Previous studies have suggested that there is a strong association between omphalocele and CHD, varying from 18 to 52%. Table 6 summarizes 11 studies incorporating 580 patients with omphalocele, with a weighted mean incidence of CHD of 24%. Our finding of CHD in 35% is not significantly different by chi-square test. Thus, the prenatal diagnosis of omphalocele had no impact on the assessment of risk for CHD, when compared to pediatric diagnoses. Omphalocele has been associated with trisomy 13 and 18.7>8>10>12>15 In the current series there were three fetuses with trisomy 18, and one with trisomy 13. Two of the fetuses with abnormal chromosomes had other ECMs. The fetus with trisomy 13 had hydranencephaly. One of twins, both with trisomy 18 and CHD, was anencephalic. An additional ECM, a cleft lip, was found in one of the fetuses with a normal karyotype. The large number of fetuses with the pentalogy of Cantrell (5 of 37 [14%]), reduced the frequency of related aneuploidy, since the pentalogy is not associated with abnormal chromosomes. If these five fetuses are removed from consideration, CHD was present in eight of the fetuses with omphalocele, and four of these had abnormal chromosomes. One of the normal fetuses was thought on pathologic examination to have amniotic band syndrome (fetal disruption sequence), so the final estimate of the risk of aneuploidy in the fetuses with both CHD and omphalocele (excluding pentalogy of Cantrell and amniotic band syndrome) in the present study is 4 of 7 (57%). Few of the studies in Table 6 provide sufficient
Congenital Heart Disease and Karyotypes in Duodenal Atresia
Patient
Type of CHD
Karyotype
Other ECMs
1 2 3 4 5 6 7 8 9
Atrial septal defect Complete atrioventricular septal defect None None None None None Ventricular septal defect Ventricular septal defect
Trisomy Trisomy Trisomy Trisomy Trisomy Trisomy Trisomy Normal Normal
Trisomy 2 1 * Trisomy 2 1 * Trisomy 2 1 * Trisomy 21 * Trisomy 2 1 * Trisomy 13* Trisomy 13* Annular pancreas, malrotation of gut None
21 21 21 21 21 13 13
*AII fetuses with karyotype abnormalities demonstrated characteristic external appearance of syndrome at birth.
Downloaded by: Universite Laval. Copyrighted material.
*ln each case of chromosome abnormality, the fetus or neonate demonstrated multiple external stigmata of the chromosome syndrome. +MCA: multiple congenital anomalies (stigmata of trisomy syndrome).
FETAL CARDIAC AND EXTRACARDIAC ANOMALIES/Fogel, et al. Table 6.
Association of Congenital Heart Disease and Chromosome Abnormalities in Omphalocele
Reference
No. Patients
No. CHD (%)
No.
November 1991
CONGENITAL HEART DISEASE AND FETAL THORACOABDOMINAL ANOMALIES: ASSOCIATIONS IN UTERO AND THE IMPORTANCE OF CYTOGENETIC ANALYSIS Mark Fogel, M.D., Joshua A. Copel, M.D., Mark T. Cullen, M.D., John C. Hobbins, M.D., and Charles S. Kleinman, M.D.
ABSTRACT
The association of congenital heart disease (CHD) with extracardiac malformations (ECMs) has recently been reviewed. These data form the basis for a previous study from our laboratory,1 and for most of the daily counseling that occurs when fetal anomalies are detected. Information reported in pediatric data bases may have an important bias, since any association uncovered between CHD and ECMs will be dependent on the infant surviving to be diagnosed. Even studies seeking to capture all newborns, or even stillbirths, may contain such a bias, for losses of late second or early third trimester fetuses may not appear. In previous work we have reported a substantially greater association between fetal heart disease and chromosome abnormalities than had been appreciated from the pediatric literature.2 The present study was undertaken to determine if a similar shift exists toward a greater association of ECMs and
Downloaded by: Universite Laval. Copyrighted material.
We examined the frequency with which congenital heart disease (CHD) and cytogenetic abnormalities were found associated with omphalocele, gastroschisis, duodenal atresia and posterior diaphragmatic hernias. We performed fetal echocardiograms on 80 patients with these diagnoses and found congenital heart disease in 13 of 37 with omphalocele (35%), 2 of 17 with gastroschisis (12%), 4 of 15 with duodenal atresia (27%), and 2 of 11 with posterior diaphragmatic hernia (18%). Karyotypes were obtained in 74 and were abnormal in 24 (32%). Although most fetuses with these extracardiac malformations and abnormal karyotypes had associated CHD, many did not. Normal karyotypes were found in 69% of fetuses with CHD and omphalocele, and 50% of fetuses with CHD and duodenal atresia. We conclude that CHD may be present in fetuses with extracardiac malformations whether or not the karyotype is normal and that the prenatal evaluation of fetuses with these lesions should include both karyotype and fetal echocardiography. Although karyotypes play an important role in prenatal diagnosis, they are not predictive of normal cardiac structure when normal in the abnormalities studied. Even when the karyotype is normal in the presence of these abnormalities, fetal echocardiography is indicated.
CHD in in utero studies. The hypothesis to be tested was that the true incidence of cardiovascular malformations associated with these ECMs was higher prenatally than is found in the postnatal literature. We were further interested in examining whether abnormal karyotypes were more prominent in those fetuses with combinations of CHD and ECMs than was usually associated with either alone.
MATERIALS AND METHODS
Between January 1, 1985, and December 31, 1988, we performed fetal echocardiograms on all fetuses with suspected ECMs who were referred to the Yale Perinatal Unit for ultrasound studies. In addition, all fetuses referred for fetal echocardiography for standard indications3 underwent full Level
Yale Fetal Cardiovascular Center, Departments of Obstetrics & Gynecology, Pediatrics, and Diagnostic Imaging, Yale University School of Medicine, New Haven, Connecticut Reprint requests: Dr. Copel, Department of Obstetrics & Gynecology, Yale University School of Medicine, 333 Cedar St., PO Box 3333, New Haven, CT 06510-8063 Copyright © 1991 by Thieme Medical Publishers, Inc., 381 Park Avenue South, New York, NY 10016. All rights reserved.
411
AMERICAN JOURNAL OF PERINATOLOGY/VOLUME 8, NUMBER 6
RESULTS
During the study period we performed fetal echocardiograms on 1665 fetuses of 1626 mothers at the Yale Fetal Cardiovascular Center. A total of 225 ECMs was seen, of which 80 had one or more of the ECMs under study (Table 1).
Table 1.
No. Omphalocele Gastroschisis Duodenal atresia Diaphragmatic hernia
37 17 15 11
Omphalocele
Forty-one omphaloceles were diagnosed prenatally over the study period; however, only 37 (90%) were actually found to have omphalocele postnatally. Three of the four misdiagnoses had gastroschisis and are included in the gastroschisis category of this study. In each of these the location of the defect was lateral to the umbilicus, excluding the possibility of a ruptured omphalocele. The fourth false-positive diagnosis of omphalocele had a cyst of the base of the umbilical cord due to an arteriovenous malformation. That fetus has also been excluded from further consideration here. The final study group is comprised of the 37 confirmed cases of omphalocele. Thirteen (35%) had CHD as detailed in Table 2. The most common type of CHD was ventricular septal defect (VSD) in four (31%), followed by double-outlet right ventricle in three (23%). Complete pentalogy of Cantrell was seen in five (omphalocele, diaphragmatic hernia, sternal defect, pericardial-peritoneal connection, and intrinsic heart disease), and incomplete variations of the pentalogy were seen in two others (total 7 of 37, 19%). There were two fetuses with complete atrioventricular septal defects (CAVSD), both of whom had trisomy 18. Of particular interest, one set of monochorionic diamniotic twins with trisomy 18 were discordant for the type of structural cardiac lesion present (hypoplastic right heart syndrome and CAVSD) and one had anencephaly in addition to the omphalocele. Table 2 details the types of structural cardiac abnormalities seen, as well as the karyotype results. Karyotypes were available from 33 of the 37 omphalocele patients (89%) and of those, five were abnormal (15%, Table 2). Three of these abnormalities were trisomy 18, all with CHD (two CAVSD and one tricuspid atresia and hypoplastic right heart). All of the pentalogy syndrome fetuses had normal chromosomes. One fetus with mosaic translocation of chromosome 1 onto the Y chromosome had omphalocele with sirenomelus and hydrocephalus but no cardiac disease. An inversion of chromosome 9 was also detected, but there was no deletion evident, suggesting that this was a variant of normal. This fetus had a single ovary in addition to the omphalocele, but no CHD. A fetus with trisomy 13 had hydranencephaly and a VSD. Of the omphalocele patients who had no car-
Extracardiac Malformations
CHD' * (%)
AK/# 1'(* (%)
13 2 4 2
5/33 2/16 7/15 4/10
(35) (12) (27) (18)
(15) (13) (47) (40)
CHD & AKI CHD (%) 4/13 1/2 2/4 2/2
(31) (50) (50) (100)
NL K* &CHDI NL K (%)
9/27 1/14 2/8 0/6
412 *CHD: congenital heart disease; AK: abnormal karyotype; # K: number in whom karyotype performed; NL K: normal karyotype.
(33) (7) (25) (0)
Downloaded by: Universite Laval. Copyrighted material.
II ultrasound examination as a part of the fetal echocardiogram. We reviewed all the hospital and private physician records of mothers who presented to the Yale Fetal Cardiovascular Center over the study period whose fetuses were identified to have any of the following: gastroschisis, omphalocele, diaphragmatic hernia, or duodenal atresia. These conditions were chosen because they were the most common prenatal diagnoses encountered, and too few fetuses were encountered with other ECMs to permit meaningful evaluation of any associations. Scans were performed between 16 and 40 weeks' gestation and included a Level II scan as well as fetal echocardiogram, the latter of which was reviewed by both an obstetrician and pediatric cardiologist with extensive experience in fetal echocardiography. Standard tomographic views of the heart were sought (four chamber, short axis ventricular, short axis great arterial, long axis left ventricular, pulmonary arteryductus, and aortic arch). Level II scans were performed with GE model RT3600 (General Electric, Milwaukee, WI) and echocardiograms with HewlettPackard model 77020AC (Hewlett-Packard, Andover, MA) ultrasound machines using 5 or 3.5 MHz transducers. Genetic evaluations were carried out before delivery by amniocentesis or fetal blood sampling, or after delivery by analysis of peripheral blood and examination by experienced dysmorphologists. If the patient was delivered at Yale-New Haven Hospital, the medical records of both mother and child, as well as genetics records, were reviewed. If the fetus was aborted or the neonate died postnatally, autopsy records were examined, when available. Those not delivering at Yale-New Haven Hospital were followed by contacting the referring physician. All neonates suspected of having congenital heart disease were examined by chest radiographs, electrocardiograms, two-dimensional ultrasound, and cardiac catheterization, as appropriate, to confirm the prenatal diagnosis.
November 1991
FETAL CARDIAC A N D EXTRACARDIAC ANOMALIES/Fogel, et al. Table 2. Patient 1 2 3a 3b 4 5 6 7 8 9 10 11 12 13 14
Congenital Heart Disease and Karyotypes in Omphalocele
Type of CHD*
Karyotype
Other ECMs
Double-outlet right ventricle Hypoplastic left heart syndrome Complete atrioventricular septal defect Hypoplastic right heart Complete atrioventricular septal defect Ventricular septal defect Transposition of the great arteries, ventricular septal defect, pulmonic stenosis Double-outlet right ventricle Atrial septal defect, ventricular septal defect Double-outlet right ventricle Ectopia Tetralogy of Fallot, ectopia Ventricular septal defect, ectopia None None
Normal Normal Trisomy Trisomy Trisomy Trisomy Normal
Cleft lip None MCA* Anencephaly MCA MCA None
18 18 18 13
Not available Normal Normal Normal Normal Normal Complex mosaic* Inv 9*
? Amniotic bands Cantrell pentalogy Cantrell pentalogy Cantrell pentalogy Cantrell pentalogy Cantrell pentalogy Sirenomelus, hydrocephaly Single ovary
diac disease and a normal karyotype, two had associated anterior diaphragmatic hernias without having the full pentalogy syndrome, and three others had an associated imperforate anus. One of those had the omphalocele, exstrophy of the bladder, imperforate anus, spinal defects (OEIS) syndrome. In all, four patients (11% of the group) had an abnormal karyotype and an associated cardiac lesion, and only one fetus had an abnormal karyotype in this group without CHD. Abnormal chromosomes were present in 31% of all fetuses in whom cardiac lesions were found in association with omphalocele. Thirty percent of fetuses with omphalocele and normal karyotypes had CHD. Only one of the 37 (3%) fetuses with omphalocele had a false-positive fetal echocardiogram. That fetus was thought to have a small VSD from the fetal echocardiogram. Distortion of the cardiac position and anatomy by the omphalocele were thought to be responsible for the abnormal ultrasound appearance of the heart. None of the fetuses had false-negative fetal echocardiograms. Gastroschisis
Seventeen patients with gastroschisis were seen during the study period, including three thought prenatally to have omphalocele. Two of the 17 (12%) were found to have cardiac disease (Table 3). One fetus had a CAVSD, and another had a simple VSD. One patient had situs inversus in addition to the gastroschisis and a structurally normal heart. Karyotypes were obtained in 16 of the 17 (94%), and a normal clinical examination was present in the remaining neonate. There was one abnormal karyotype (6%): trisomy 18 in a fetus also found to have a CAVSD. One of the 16 fetuses with gastroschisis and a normal karyotype had CHD (6%), a perimembranous VSD. There were no false-negative or falsepositive diagnoses of CHD in this group.
Table 3. Congenital Heart Disease and Karyotypes in Gastroschisis Patient
Type of CHD
Karyotype
Other ECMs
1
Complete atrioventricular septal defect Ventricular septal defect
Trisomy 18
MCA*
Normal
None
2
*MCA: multiple congenital anomalies (stigmata of trisomy syndrome).
Diaphragmatic Hernia
Eleven posterior diaphragmatic hernias were detected. Two had associated CHD (18%): one aortic atresia with a VSD, and the other CAVSD with coarctation of the aorta. Ten karyotypes were obtained (91%) of which four were abnormal (40%), including both of the patients with CHD. The four abnormal karyotypes encountered included one Turner syndrome, one trisomy 18 with inversion of chromosome 1, who had multiple congenital anomalies (but a normal heart), one trisomy 21 who also had CHD (a VSD with aortic atresia), and a patient with CAVSD and coarctation of the aorta who had multiple congenital anomalies and a duplication of chromosome 1 (Table 4). Duodenal Atresia
Duodenal atresia was suspected in 16 patients prenatally, although one of the mothers was ethanol intoxicated at the time fetal gastric distention was observed, and the apparent double bubble resolved on repeat scan in 48 hours. That fetus had multiple stigmata of the fetal alcohol syndrome, including CHD, but is not considered further here. There were, then, 15 patients with the prenatal diagnosis of duodenal atresia, of whom four had 413
Downloaded by: Universite Laval. Copyrighted material.
*MCA: multiple congenital anomalies (stigmata of trisomy syndrome); Complex mosaic: karyotype 46,XY/46,X,ter(Y)t(1; Y)(q21 ;q12); Inv 9: Karyotype 46,XX inv (9q).
AMERICAN JOURNAL OF PERINATOLOGY/VOLUME 8, NUMBER 6 Table 4. Patient
November 1991
Congenital Heart Disease and Karyotypes in Posterior Diaphragmatic Hernia
Type of CHD
Karyotype
Other ECMs*
Ventricular septal defect, aortic atresia Complete atrioventricular septal defect, coarctation of aorta None None
Trisomy 21 46,XYdup(1)(q12;q23) Trisomy 18 45,X
Trisomy 21 MCA+ MCA 45,X
CHD (27%). Three were recognized prenatally, including one each with an atrial septal defect (ASD) (as a component of a partial atrioventricular septal defect), a VSD, and CAVSD. A false-negative fetal echocardiogram occurred in a patient with normal chromosomes, an annular pancreas, malrotation of the gut, and a perimembranous VSD. There were no false-positive diagnoses of CHD in this group. All 15 patients with duodenal atresia had karyotypes obtained, and seven (47%) were abnormal. There were five cases of trisomy 21 and two of trisomy 13. Two of the trisomy 21 fetuses had CHD (40%). The apparent absence of CHD in the two fetuses with trisomy 13 was confirmed by autopsy. Two additional patients with CHD had normal karyotypes (both fetuses had simple VSDs); therefore 50% of those with CHD and duodenal atresia had normal chromosomes (Table 5). One of these fetuses had the vertebral, anal, tracheal, esophageal, renal (VATER) sequence. DISCUSSION
The current study was undertaken to examine whether the designated fetal thoracoabdominal defects were associated with CHD more often than has been previously reported in pediatric studies. We hoped to reduce ascertainment bias by scanning all fetuses with these prenatal diagnoses, and so include many of those that might be lost to pediatric studies due to intrauterine or early neonatal death. Our previous finding that chromosomal abnormalities were present in 38% of fetuses with CHD, compared with 6% reported in pediatric series,4 furthermore, led us to speculate that if more CHD was associated
Table 5.
414
with these ECMs in utero, the common thread might be a parallel increase in chromosomal abnormalities. Previous studies have suggested that there is a strong association between omphalocele and CHD, varying from 18 to 52%. Table 6 summarizes 11 studies incorporating 580 patients with omphalocele, with a weighted mean incidence of CHD of 24%. Our finding of CHD in 35% is not significantly different by chi-square test. Thus, the prenatal diagnosis of omphalocele had no impact on the assessment of risk for CHD, when compared to pediatric diagnoses. Omphalocele has been associated with trisomy 13 and 18.7>8>10>12>15 In the current series there were three fetuses with trisomy 18, and one with trisomy 13. Two of the fetuses with abnormal chromosomes had other ECMs. The fetus with trisomy 13 had hydranencephaly. One of twins, both with trisomy 18 and CHD, was anencephalic. An additional ECM, a cleft lip, was found in one of the fetuses with a normal karyotype. The large number of fetuses with the pentalogy of Cantrell (5 of 37 [14%]), reduced the frequency of related aneuploidy, since the pentalogy is not associated with abnormal chromosomes. If these five fetuses are removed from consideration, CHD was present in eight of the fetuses with omphalocele, and four of these had abnormal chromosomes. One of the normal fetuses was thought on pathologic examination to have amniotic band syndrome (fetal disruption sequence), so the final estimate of the risk of aneuploidy in the fetuses with both CHD and omphalocele (excluding pentalogy of Cantrell and amniotic band syndrome) in the present study is 4 of 7 (57%). Few of the studies in Table 6 provide sufficient
Congenital Heart Disease and Karyotypes in Duodenal Atresia
Patient
Type of CHD
Karyotype
Other ECMs
1 2 3 4 5 6 7 8 9
Atrial septal defect Complete atrioventricular septal defect None None None None None Ventricular septal defect Ventricular septal defect
Trisomy Trisomy Trisomy Trisomy Trisomy Trisomy Trisomy Normal Normal
Trisomy 2 1 * Trisomy 2 1 * Trisomy 2 1 * Trisomy 21 * Trisomy 2 1 * Trisomy 13* Trisomy 13* Annular pancreas, malrotation of gut None
21 21 21 21 21 13 13
*AII fetuses with karyotype abnormalities demonstrated characteristic external appearance of syndrome at birth.
Downloaded by: Universite Laval. Copyrighted material.
*ln each case of chromosome abnormality, the fetus or neonate demonstrated multiple external stigmata of the chromosome syndrome. +MCA: multiple congenital anomalies (stigmata of trisomy syndrome).
FETAL CARDIAC AND EXTRACARDIAC ANOMALIES/Fogel, et al. Table 6.
Association of Congenital Heart Disease and Chromosome Abnormalities in Omphalocele
Reference
No. Patients
No. CHD (%)
No.
