- Email: [email protected]
Outcome of fetuses diagnosed with atrioventricular septal defect
Outcome of Fetuses Diagnosed With Atrioventricular Septal Defect MARIE-FRANCE N. DELISLE, MD, GEORGE G. S. SANDOR, MD, FRANCINE TESSIER, MD, AND DUNCAN F. FARQUHARSON, MD Objective: To quantify the association of prenatally diagnosed atrioventricular septal defect with Down syndrome and to evaluate its impact on obstetric and neonatal outcomes. Methods: Charts of 42 cases of atrioventricular septal defect diagnosed by fetal echocardiography from July 1985 to July 1997 were reviewed for prenatal history and outcome data (pregnancy outcome, pathologic confirmation, postnatal echocardiographic findings, and neonatal outcome). Statistical analysis was done using Fisher exact test and odds ratios. Results: The mean gestational age at diagnosis was 26 weeks. Four cases could not be confirmed antenatally on repeat echocardiograms and were excluded. Reasons for referral of the remaining 38 fetuses included an abnormal four-chamber view in 76%. Twenty-two fetuses (58%) had abnormal karyotypes: 19 trisomy 21, one trisomy 18, one trisomy 13, and one mosaicism. The cardiac lesions were isolated in 20 fetuses (53%). After excluding cases of termination, ten of 12 fetuses (83%) with Down syndrome survived, compared with seven of 13 (54%) with normal karyotypes (P ⴝ .125). The odds of trisomy 21 were 16 times higher (95% confidence interval 3.0, 85.3) in fetuses with isolated cardiac lesions compared with those with associated cardiac anomalies. Conclusion: Prenatal diagnosis of atrioventricular septal defect was associated with a 58% risk of aneuploidy (mainly trisomy 21). Down syndrome fetuses with this cardiac anomaly appeared to have a better survival rate than fetuses with normal karyotypes. Our sample did not have enough power to show a statistically significant difference. When an isolated atrioventricular septal defect was diagnosed prenatally, the odds of trisomy 21 were significantly higher than when other associated cardiac lesions were diagnosed. This information should be considered in prenatal counseling for atrioventricular septal defect. (Obstet Gynecol 1999;94:763–7. © 1999 by The American College of Obstetricians and Gynecologists.)
From the Division of Pediatric Cardiology and Division of MaternalFetal Medicine, Children’s and Women’s Health Centre of British Columbia, Vancouver, British Columbia, Canada.
VOL. 94, NO. 5, PART 1, NOVEMBER 1999
The morbidity and mortality associated with atrioventricular septal defect are well described in the pediatric cardiology and surgery literature.1–3 In recent years, the reliability of fetal echocardiography for the prenatal diagnosis of atrioventricular septal defect has steadily improved.4 –7 The incidence of Down syndrome associated with an atrioventricular septal defect detected prenatally is still not clearly defined in the literature, ranging from 30% to 82%.1,4,8 Little is known about the prenatal and postnatal effects of atrioventricular septal defect. The purpose of this retrospective study was to quantify the association between atrioventricular septal defect diagnosed antenatally and Down syndrome and to evaluate the effect of trisomy 21 on the outcomes of fetuses with atrioventricular septal defect to assist with prenatal counseling.
Materials and Methods From July 1985 to July 1997, more than 3000 fetal echocardiograms were done at the Centre of Prenatal Diagnosis and Treatment at British Columbia Women’s Hospital. We reviewed all 42 cases of atrioventricular septal defect diagnosed prenatally. Charts were reviewed to obtain demographic data including maternal age, parity, gestational age at diagnosis, reason for referral, and fetal sex. Investigational and outcome data reviewed included associated cardiac and noncardiac anomalies, karyotype, pathologic confirmation, postnatal echocardiographic findings, and obstetric and neonatal outcomes. Survival was defined as being alive at the time of our data collection, which represented a minimum of 8 months. Survival percentages were calculated for our total population and for subgroups of fetuses according to their karyotype results. The denominator of the survival percentages excluded cases of elective terminations. Descriptive data are presented as mean ⫾ standard deviation. Fisher exact test and odds ratios were used for statistical analysis.
0029-7844/99/$20.00 PII S0029-7844(99)00389-0
763
Table 1. Demographic Data Characteristic
Normal karyotype
Trisomy 21
Mean maternal age (y)* n Mean gestational age at diagnosis (wk)* n Fetal sex (number of males) n
31.2 ⫾ 6.3 16 27.1 ⫾ 7.0 16 9 15
32.7 ⫾ 6.8 19 25.9 ⫾ 7.8 19 6 17
* Mean ⫾ standard deviation.
Results Four of the cases reviewed in our series were excluded from analysis because follow-up fetal echocardiograms did not confirm the initially suspected diagnosis of atrioventricular septal defect. These four pregnancies resulted in live-born infants. Two had ventricular septal defects and two had normal postnatal echocardiograms. In the other 38 cases, the atrioventricular septal defect was confirmed by postmortem examination or postnatal echocardiography. Demographic data are presented in Table 1. The average gestational age at diagnosis was 26 weeks (range 16 –38). The indications for referral are summarized in Table 2. Failure to identify four normal cardiac chambers was the primary indication for referral in 29 fetuses (76%). Other noncardiac findings led to the referral in seven cases (18%), including cystic hygroma (three), cerebral anomalies (one), polyhydramnios (one), and fetal growth restriction (one); one was a known case of trisomy 21. Table 3 is a case-by-case listing of our 38 cases of atrioventricular septal defect subdivided by the karyotype results. It includes information on associated cardiac anomalies, fetal sex, cardiac surgery, and obstetric and neonatal outcomes. The atrioventricular septal defect was an isolated cardiac lesion at the fetal echocardiogram in 20 cases. All of the other fetuses had associated cardiac anomalies diagnosed antenatally: isomerism in nine (six right, three left), transposition of
Table 2. Indications for Referral and Results of Karyotype Abnormal karyotype (n ⫽ 22)
Reason for referral Abnormal four-chamber view Bradyarrhythmia Cystic hygroma Cerebral anomalies Known trisomy 21 Polyhydramnios Fetal growth restriction
764 Delisle et al
Normal karyotype Trisomy 21 (n ⫽ 19) (n ⫽ 16) 12 2 1
Other (n ⫽ 3)
15
2
2 1 (Trisomy 13) 1
1 1
Atrioventricular Septal Defect
the great arteries in seven, and tetralogy of Fallot in two. Two fetuses had complete heart block, and both of them had left atrial isomerism and normal chromosomes. They both died shortly after birth. Twelve of 18 fetuses (67%) with atrioventricular septal defect associated with other cardiac anomalies had normal karyotypes. Among the nine fetuses with atrial isomerism, eight had a normal karyotype. Various noncardiac anomalies also were detected prenatally. There were two cases of holoprosencephaly; one fetus had trisomy 13 and was electively delivered at 23 weeks. The other one (case no. 33) had a normal karyotype but multiple other anomalies (polyhydramnios, horseshoe kidney, and diaphragmatic eventration) and died shortly after birth. There was one case (no. 12) of biliary atresia in a fetus with a normal karyotype. One fetus with Down syndrome had duodenal atresia that was surgically corrected. Three fetuses (case nos. 3, 21, and 27) also had some growth restriction, but none were delivered preterm for that reason. There were eight cases of asplenia or polysplenia syndrome. Of these, four died in the neonatal period or infancy, one was electively terminated at 24 weeks, and three survived. Except for one case of mosaicism, all of the other fetuses had normal karyotypes. One infant was diagnosed with VATER association (vertebral defects, imperforate anus, tracheoesophageal fistula, and radial and renal dysplasia) (case no. 18) and successfully underwent several operations. Chromosomal anomalies were detected in 22 fetuses (58%): 19 had trisomy 21, one had trisomy 18, one had trisomy 13, and one had mosaicism (46XY/47XY ⫹19q). These results are included in Table 3. Within the group of 15 women whose fetuses were diagnosed with atrioventricular septal defect before viability (24 weeks), 11 (73%) opted to terminate their pregnancies. Thus, seven of the eight cases (88%) of trisomy 21 detected before 24 weeks and three of the six cases (50%) with normal karyotypes were terminated. These 11 pregnancies represented 29% of our total population. Of the 27 cases in which the pregnancy was continued, one fetus with trisomy 13 died in utero and 26 fetuses were born alive. Of these, nine infants (three with trisomy 21 and six with normal karyotypes) died in the neonatal period or during infancy (one during cardiac surgery) (Table 4). There were 17 survivors (63%), ten with trisomy 21 and seven with normal karyotypes. After excluding the cases of pregnancy termination, ten of 12 fetuses (83%) with Down syndrome survived, compared with seven of 13 (54%) with normal karyotypes. These survival rates were compared using Fisher exact test. The difference was not
Obstetrics & Gynecology
Table 3. Atrioventricular Septal Defect in 38 Fetuses Case Normal karyotype (n ⫽ 16) 1* 2* 4* 6 8 10 12* 14* 18 19 22 23 24 28* 29* 33 Abnormal karyotype (n ⫽ 22) 3 5 7 9 11 13 15 16 17 20* 21 25 26 27 30 31 32 34 35 36 37 38
Other cardiac anomalies RAI, TGA, PulmST LAI, PulmST, CB, HLH RAI, TGA, PulmST RAI, TGA, PulmA, HLH TGA, PulmA, HLH CA, HLH LAI LAI, TGA, CB CA
TGA PulmA, RAI RAI
CA
RAI
HA TO
TGA TO
Karyotype 46, 46, 46, 46, 46, 46, 46, 46, 46, 46, 46, 46, 46, 46, 46, 46,
XY XX XY XY XY XY XX XX XX XY XX N XX XY XY XY
47, XY, ⫹18 47, XX, ⫹21 47, XY, ⫹21 47, XX, ⫹21 47, XX, ⫹21 47, XX, ⫹21 47, XY, ⫹21 47, XX, ⫹21 47, XY, ⫹21 MOS 47, XY, ⫹21 47, XX, ⫹21 47, XX, ⫹21 47, XX, ⫹21 47, XY, ⫹21 47, XY, ⫹21 47, XX, ⫹13 47, XX, ⫹21 47, XX, ⫹21 47, ⫹21 47, XX, ⫹21 47, ⫹21
Cardiac surgery Yes No Yes No No Yes No Yes Yes Yes Yes Yes No No No Yes Yes Yes
No Yes
Yes No Yes Yes Yes Yes Yes
Outcome ID (3 mo) NND (1 mo) Alive NND (2 d) TA NND (18 d) Alive NND (1 mo) Alive Alive Alive Alive TA Alive TA NND (1 d) IUD (38 wk) TA ID (2 mo) NND (3 d) TA Alive Alive TA TA NND (28 wk) Alive TA TA Alive Alive Alive TA Alive TA Alive Alive Alive
RAI ⫽ right atrial isomerism; TGA ⫽ transposition of great arteries; PulmST ⫽ pulmonary stenosis; ID ⫽ infant death; LAI ⫽ left atrial isomerism; CB ⫽ cardiac block; HLH ⫽ hypoplasia of left heart; NND ⫽ neonatal death; PulmA ⫽ pulmonary atresia; TA ⫽ termination of pregnancy; CA ⫽ coarctation of aorta; IUD ⫽ intrauterine death; MOS ⫽ mosaicism; HA ⫽ hypoplastic aorta; TO ⫽ tetralogy of Fallot. * Asplenia or polysplenia syndrome.
statistically significant (P ⫽ .125), most likely because of the small sample size and limited power. We also compared survival rates in relation to whether the cardiac lesion was isolated. Overall, 11 of 20 fetuses (55%) survived with isolated cardiac lesions, compared with six of 18 (33%) with nonisolated lesions. After excluding the cases of termination, 11 of 14 fetuses (79%) with isolated cardiac lesions survived, compared with six of 13 fetuses (46%) with nonisolated cardiac lesions. This difference was not statistically significant (P ⫽ .120). Atrioventricular septal defect was an isolated cardiac lesion in 20 cases, of which 16 fetuses (80%) had
VOL. 94, NO. 5, PART 1, NOVEMBER 1999
chromosomal abnormalities. Of the total cases of trisomy 21, 84% had an isolated atrioventricular septal defect, compared with only 25% of the cases with a normal karyotype. This difference was statistically significant (P ⬍ .01). The odds of having a fetus with Down syndrome were 16 times higher (95% confidence interval 3.0, 85.3) in the group with isolated cardiac lesions compared with the group with associated cardiac anomalies. Nine of the ten survivors (90%) with Down syndrome had isolated cardiac lesions. Of the seven survivors with normal karyotypes, only two (29%) had isolated cardiac lesions.
Delisle et al
Atrioventricular Septal Defect
765
Table 4. Obstetric and Neonatal Outcomes of Atrioventricular Septal Defects Diagnosed Prenatally Characteristic of cardiac lesion Isolated (n ⫽ 20)
Karyotype 16 Trisomy 21
4 Normal karyotype
Nonisolated (n ⫽ 18)
3 Trisomy 21 12 Normal karyotype
3 Other aneuploidy
Outcome 5 9 2 1 2 1 2 1 2 5 5 1 2
TA Alive NND ⫹ ID TA Alive NND TA Alive TA Alive NND ⫹ ID TA NND ⫹ IUD
TA ⫽ termination of pregnancy; NND ⫽ neonatal death; ID ⫽ infant death; IUD ⫽ intrauterine death.
Discussion Many articles on atrioventricular septal defect in the postnatal period have been published in the pediatric cardiology or surgery literature.1,2,4,7 Many authors have described outcomes, survival rates, surgical techniques and results, and associations with other anomalies or aneuploidy.4,9,10 However, little is known about the outcome of atrioventricular septal defect diagnosed antenatally. Machado et al4 reported a series of 29 cases in 1988. They had 14 cases of aneuploidy, of which nine were trisomy 21. Their prenatal incidence of trisomy 21 associated with a diagnosis of atrioventricular septal defect was 31%. Of the 15 pregnancies that went to term, there were only four survivors, for a rate of 27%. In 1994, Allan et al8 reported on 177 cases of atrioventricular septal defect diagnosed prenatally. Of the 35 cases with chromosomal anomalies, 62% were terminated, 20% were intrauterine deaths, 11% were neonatal deaths, and 3% were deaths in infancy or childhood. Thirty-seven of the fetuses that were not terminated survived. Our results differ from those of Machado et al4 and Allan et al.8 More than half (58%) of our 38 cases of atrioventricular septal defect had abnormal karyotypes, compared with 30 –35% in those series. Trisomy 21 represented 86% of these anomalies, half of our total number of atrioventricular septal defects. Machado et al4 and Allan et al8 observed survival rates of less than 40%, but ours was 63%. None of the fetuses in our series were hydropic at diagnosis or at delivery (including the one with left isomerism and intrauterine heart block). This finding may partly account for our better overall survival rate compared with other series. Some of the discrepancy in
766 Delisle et al
Atrioventricular Septal Defect
survival also may be explained by the years covered by the retrospective reviews: 1980 –1986 in Machado et al,4 1980 –1993 in Allen et al,8 and 1985–1997 in our series. We disagree with Machado et al,4 who qualified the prognosis associated with a prenatal diagnosis of atrioventricular septal defect as being very poor (27% in their series). After excluding the terminations of pregnancy, our overall survival rate was 63%. Although this survival rate appears clinically good, it clearly tends to improve in specific situations (association with trisomy 21 [83%] and isolated cardiac lesions [79%]). The differences in survival rates between fetuses with Down syndrome and those with normal karyotypes in our series were not statistically significant. The same finding applies to the difference in survival rates between isolated and nonisolated cardiac lesions. This is most likely due to our small sample size and limited power; however, we believe this difference remains clinically important and should be part of antenatal counseling. It is very important that any associated cardiac or noncardiac anomalies be ruled out by thorough ultrasonographic examinations and fetal echocardiograms when atrioventricular septal defect is diagnosed. Clearly identifying isolated versus nonisolated atrioventricular septal defect is an important factor in prenatal counseling. Our results confirm in the antenatal population the already described association between postnatal isolated atrioventricular septal defect and aneuploidy. In more than 80% of our cases of trisomy 21, the cardiac lesion was isolated, compared with only 25% when the karyotype was normal. When an isolated cardiac lesion of atrioventricular septal defect was diagnosed prenatally, there was a significantly higher chance of it being associated with trisomy 21 than with a normal karyotype. Prenatal counseling in cases of atrioventricular septal defect should include the 58% risk of aneuploidy and 50% risk of trisomy 21. When the cardiac lesion is isolated, the risk increases to 80%. Although these differences are not statistically significant, the survival percentages appear to be better in the Down syndrome group and the isolated cardiac lesion group. Clinical situations involving isolated cardiac lesions associated with trisomy 21 were clearly the most common in our center.
References 1. Tweddell JS, Litwin SB, Berger S, Friedberg DZ, Thomas JP, Frommelt PC, et al. Twenty-year experience with repair of complete atrioventricular septal defects. Ann Thorac Surg 1996;62:419 – 24. 2. Rizzoli G, Mazzucco A, Maizza F, Daliento L, Rubino M, Tursi V, et al. Does Down syndrome affect prognosis of surgically managed
Obstetrics & Gynecology
3.
4. 5.
6.
7.
8.
atrioventricular canal defects? J Thorac Cardiovasc Surg 1992;104: 945–53. Marino B. Atrioventricular septal defect—anatomic characteristics in patients with and without Down’s syndrome. Cardiol Young 1992;2:308 –10. Machado MVL, Crawford DC, Anderson RH, Allen LD. Atrioventricular septal defect in prenatal life. Br Heart J 1988;59:352–5. Allan LD, Crawford DC, Anderson RH, Tynan MJ. Echocardiographic and anatomical correlations in fetal congenital heart disease. Br Heart J 1984;52:542– 8. Allan LD, Crawford DC, Anderson RH, Tynan MJ. Spectrum of congenital heart disease detected echocardiographically in prenatal life. Br Heart J 1985;54:523– 6. Gembruch U, Knopfle G, Chatterjee M, Bald R, Redel DA, Fodisch HJ, et al. Prenatal diagnosis of atrioventricular canal malformations with up-to-date echocardiographic technology: Report of 14 cases. Am Heart J 1991;121:1489 –97. Allan LD, Sharland GK, Milburn A, Lockhart SM, Groves AMM, Anderson RH, et al. Prospective diagnosis of 1006 consecutive cases of congenital heart disease in the fetus. J Am Coll Cardiol 1994;23:1452– 8.
VOL. 94, NO. 5, PART 1, NOVEMBER 1999
9. Carmi R, Boughman JA, Ferencz C. Endocardial cushion defect: Further studies of “isolated” versus “syndromic” occurrence. Am J Med Genet 1992;43:569 –74. 10. Marino B, Vairo U, Corno A, Nava S, Guccione P, Calabro R, et al. Atrioventricular canal in Down syndrome. Am J Dis Child 1990; 144:1120 –2.
Reprints are not available.
Received January 12, 1999. Received in revised form April 15, 1999. Accepted May 5, 1999.
Copyright © 1999 by The American College of Obstetricians and Gynecologists. Published by Elsevier Science Inc.
Delisle et al
Atrioventricular Septal Defect
767
From the Division of Pediatric Cardiology and Division of MaternalFetal Medicine, Children’s and Women’s Health Centre of British Columbia, Vancouver, British Columbia, Canada.
VOL. 94, NO. 5, PART 1, NOVEMBER 1999
The morbidity and mortality associated with atrioventricular septal defect are well described in the pediatric cardiology and surgery literature.1–3 In recent years, the reliability of fetal echocardiography for the prenatal diagnosis of atrioventricular septal defect has steadily improved.4 –7 The incidence of Down syndrome associated with an atrioventricular septal defect detected prenatally is still not clearly defined in the literature, ranging from 30% to 82%.1,4,8 Little is known about the prenatal and postnatal effects of atrioventricular septal defect. The purpose of this retrospective study was to quantify the association between atrioventricular septal defect diagnosed antenatally and Down syndrome and to evaluate the effect of trisomy 21 on the outcomes of fetuses with atrioventricular septal defect to assist with prenatal counseling.
Materials and Methods From July 1985 to July 1997, more than 3000 fetal echocardiograms were done at the Centre of Prenatal Diagnosis and Treatment at British Columbia Women’s Hospital. We reviewed all 42 cases of atrioventricular septal defect diagnosed prenatally. Charts were reviewed to obtain demographic data including maternal age, parity, gestational age at diagnosis, reason for referral, and fetal sex. Investigational and outcome data reviewed included associated cardiac and noncardiac anomalies, karyotype, pathologic confirmation, postnatal echocardiographic findings, and obstetric and neonatal outcomes. Survival was defined as being alive at the time of our data collection, which represented a minimum of 8 months. Survival percentages were calculated for our total population and for subgroups of fetuses according to their karyotype results. The denominator of the survival percentages excluded cases of elective terminations. Descriptive data are presented as mean ⫾ standard deviation. Fisher exact test and odds ratios were used for statistical analysis.
0029-7844/99/$20.00 PII S0029-7844(99)00389-0
763
Table 1. Demographic Data Characteristic
Normal karyotype
Trisomy 21
Mean maternal age (y)* n Mean gestational age at diagnosis (wk)* n Fetal sex (number of males) n
31.2 ⫾ 6.3 16 27.1 ⫾ 7.0 16 9 15
32.7 ⫾ 6.8 19 25.9 ⫾ 7.8 19 6 17
* Mean ⫾ standard deviation.
Results Four of the cases reviewed in our series were excluded from analysis because follow-up fetal echocardiograms did not confirm the initially suspected diagnosis of atrioventricular septal defect. These four pregnancies resulted in live-born infants. Two had ventricular septal defects and two had normal postnatal echocardiograms. In the other 38 cases, the atrioventricular septal defect was confirmed by postmortem examination or postnatal echocardiography. Demographic data are presented in Table 1. The average gestational age at diagnosis was 26 weeks (range 16 –38). The indications for referral are summarized in Table 2. Failure to identify four normal cardiac chambers was the primary indication for referral in 29 fetuses (76%). Other noncardiac findings led to the referral in seven cases (18%), including cystic hygroma (three), cerebral anomalies (one), polyhydramnios (one), and fetal growth restriction (one); one was a known case of trisomy 21. Table 3 is a case-by-case listing of our 38 cases of atrioventricular septal defect subdivided by the karyotype results. It includes information on associated cardiac anomalies, fetal sex, cardiac surgery, and obstetric and neonatal outcomes. The atrioventricular septal defect was an isolated cardiac lesion at the fetal echocardiogram in 20 cases. All of the other fetuses had associated cardiac anomalies diagnosed antenatally: isomerism in nine (six right, three left), transposition of
Table 2. Indications for Referral and Results of Karyotype Abnormal karyotype (n ⫽ 22)
Reason for referral Abnormal four-chamber view Bradyarrhythmia Cystic hygroma Cerebral anomalies Known trisomy 21 Polyhydramnios Fetal growth restriction
764 Delisle et al
Normal karyotype Trisomy 21 (n ⫽ 19) (n ⫽ 16) 12 2 1
Other (n ⫽ 3)
15
2
2 1 (Trisomy 13) 1
1 1
Atrioventricular Septal Defect
the great arteries in seven, and tetralogy of Fallot in two. Two fetuses had complete heart block, and both of them had left atrial isomerism and normal chromosomes. They both died shortly after birth. Twelve of 18 fetuses (67%) with atrioventricular septal defect associated with other cardiac anomalies had normal karyotypes. Among the nine fetuses with atrial isomerism, eight had a normal karyotype. Various noncardiac anomalies also were detected prenatally. There were two cases of holoprosencephaly; one fetus had trisomy 13 and was electively delivered at 23 weeks. The other one (case no. 33) had a normal karyotype but multiple other anomalies (polyhydramnios, horseshoe kidney, and diaphragmatic eventration) and died shortly after birth. There was one case (no. 12) of biliary atresia in a fetus with a normal karyotype. One fetus with Down syndrome had duodenal atresia that was surgically corrected. Three fetuses (case nos. 3, 21, and 27) also had some growth restriction, but none were delivered preterm for that reason. There were eight cases of asplenia or polysplenia syndrome. Of these, four died in the neonatal period or infancy, one was electively terminated at 24 weeks, and three survived. Except for one case of mosaicism, all of the other fetuses had normal karyotypes. One infant was diagnosed with VATER association (vertebral defects, imperforate anus, tracheoesophageal fistula, and radial and renal dysplasia) (case no. 18) and successfully underwent several operations. Chromosomal anomalies were detected in 22 fetuses (58%): 19 had trisomy 21, one had trisomy 18, one had trisomy 13, and one had mosaicism (46XY/47XY ⫹19q). These results are included in Table 3. Within the group of 15 women whose fetuses were diagnosed with atrioventricular septal defect before viability (24 weeks), 11 (73%) opted to terminate their pregnancies. Thus, seven of the eight cases (88%) of trisomy 21 detected before 24 weeks and three of the six cases (50%) with normal karyotypes were terminated. These 11 pregnancies represented 29% of our total population. Of the 27 cases in which the pregnancy was continued, one fetus with trisomy 13 died in utero and 26 fetuses were born alive. Of these, nine infants (three with trisomy 21 and six with normal karyotypes) died in the neonatal period or during infancy (one during cardiac surgery) (Table 4). There were 17 survivors (63%), ten with trisomy 21 and seven with normal karyotypes. After excluding the cases of pregnancy termination, ten of 12 fetuses (83%) with Down syndrome survived, compared with seven of 13 (54%) with normal karyotypes. These survival rates were compared using Fisher exact test. The difference was not
Obstetrics & Gynecology
Table 3. Atrioventricular Septal Defect in 38 Fetuses Case Normal karyotype (n ⫽ 16) 1* 2* 4* 6 8 10 12* 14* 18 19 22 23 24 28* 29* 33 Abnormal karyotype (n ⫽ 22) 3 5 7 9 11 13 15 16 17 20* 21 25 26 27 30 31 32 34 35 36 37 38
Other cardiac anomalies RAI, TGA, PulmST LAI, PulmST, CB, HLH RAI, TGA, PulmST RAI, TGA, PulmA, HLH TGA, PulmA, HLH CA, HLH LAI LAI, TGA, CB CA
TGA PulmA, RAI RAI
CA
RAI
HA TO
TGA TO
Karyotype 46, 46, 46, 46, 46, 46, 46, 46, 46, 46, 46, 46, 46, 46, 46, 46,
XY XX XY XY XY XY XX XX XX XY XX N XX XY XY XY
47, XY, ⫹18 47, XX, ⫹21 47, XY, ⫹21 47, XX, ⫹21 47, XX, ⫹21 47, XX, ⫹21 47, XY, ⫹21 47, XX, ⫹21 47, XY, ⫹21 MOS 47, XY, ⫹21 47, XX, ⫹21 47, XX, ⫹21 47, XX, ⫹21 47, XY, ⫹21 47, XY, ⫹21 47, XX, ⫹13 47, XX, ⫹21 47, XX, ⫹21 47, ⫹21 47, XX, ⫹21 47, ⫹21
Cardiac surgery Yes No Yes No No Yes No Yes Yes Yes Yes Yes No No No Yes Yes Yes
No Yes
Yes No Yes Yes Yes Yes Yes
Outcome ID (3 mo) NND (1 mo) Alive NND (2 d) TA NND (18 d) Alive NND (1 mo) Alive Alive Alive Alive TA Alive TA NND (1 d) IUD (38 wk) TA ID (2 mo) NND (3 d) TA Alive Alive TA TA NND (28 wk) Alive TA TA Alive Alive Alive TA Alive TA Alive Alive Alive
RAI ⫽ right atrial isomerism; TGA ⫽ transposition of great arteries; PulmST ⫽ pulmonary stenosis; ID ⫽ infant death; LAI ⫽ left atrial isomerism; CB ⫽ cardiac block; HLH ⫽ hypoplasia of left heart; NND ⫽ neonatal death; PulmA ⫽ pulmonary atresia; TA ⫽ termination of pregnancy; CA ⫽ coarctation of aorta; IUD ⫽ intrauterine death; MOS ⫽ mosaicism; HA ⫽ hypoplastic aorta; TO ⫽ tetralogy of Fallot. * Asplenia or polysplenia syndrome.
statistically significant (P ⫽ .125), most likely because of the small sample size and limited power. We also compared survival rates in relation to whether the cardiac lesion was isolated. Overall, 11 of 20 fetuses (55%) survived with isolated cardiac lesions, compared with six of 18 (33%) with nonisolated lesions. After excluding the cases of termination, 11 of 14 fetuses (79%) with isolated cardiac lesions survived, compared with six of 13 fetuses (46%) with nonisolated cardiac lesions. This difference was not statistically significant (P ⫽ .120). Atrioventricular septal defect was an isolated cardiac lesion in 20 cases, of which 16 fetuses (80%) had
VOL. 94, NO. 5, PART 1, NOVEMBER 1999
chromosomal abnormalities. Of the total cases of trisomy 21, 84% had an isolated atrioventricular septal defect, compared with only 25% of the cases with a normal karyotype. This difference was statistically significant (P ⬍ .01). The odds of having a fetus with Down syndrome were 16 times higher (95% confidence interval 3.0, 85.3) in the group with isolated cardiac lesions compared with the group with associated cardiac anomalies. Nine of the ten survivors (90%) with Down syndrome had isolated cardiac lesions. Of the seven survivors with normal karyotypes, only two (29%) had isolated cardiac lesions.
Delisle et al
Atrioventricular Septal Defect
765
Table 4. Obstetric and Neonatal Outcomes of Atrioventricular Septal Defects Diagnosed Prenatally Characteristic of cardiac lesion Isolated (n ⫽ 20)
Karyotype 16 Trisomy 21
4 Normal karyotype
Nonisolated (n ⫽ 18)
3 Trisomy 21 12 Normal karyotype
3 Other aneuploidy
Outcome 5 9 2 1 2 1 2 1 2 5 5 1 2
TA Alive NND ⫹ ID TA Alive NND TA Alive TA Alive NND ⫹ ID TA NND ⫹ IUD
TA ⫽ termination of pregnancy; NND ⫽ neonatal death; ID ⫽ infant death; IUD ⫽ intrauterine death.
Discussion Many articles on atrioventricular septal defect in the postnatal period have been published in the pediatric cardiology or surgery literature.1,2,4,7 Many authors have described outcomes, survival rates, surgical techniques and results, and associations with other anomalies or aneuploidy.4,9,10 However, little is known about the outcome of atrioventricular septal defect diagnosed antenatally. Machado et al4 reported a series of 29 cases in 1988. They had 14 cases of aneuploidy, of which nine were trisomy 21. Their prenatal incidence of trisomy 21 associated with a diagnosis of atrioventricular septal defect was 31%. Of the 15 pregnancies that went to term, there were only four survivors, for a rate of 27%. In 1994, Allan et al8 reported on 177 cases of atrioventricular septal defect diagnosed prenatally. Of the 35 cases with chromosomal anomalies, 62% were terminated, 20% were intrauterine deaths, 11% were neonatal deaths, and 3% were deaths in infancy or childhood. Thirty-seven of the fetuses that were not terminated survived. Our results differ from those of Machado et al4 and Allan et al.8 More than half (58%) of our 38 cases of atrioventricular septal defect had abnormal karyotypes, compared with 30 –35% in those series. Trisomy 21 represented 86% of these anomalies, half of our total number of atrioventricular septal defects. Machado et al4 and Allan et al8 observed survival rates of less than 40%, but ours was 63%. None of the fetuses in our series were hydropic at diagnosis or at delivery (including the one with left isomerism and intrauterine heart block). This finding may partly account for our better overall survival rate compared with other series. Some of the discrepancy in
766 Delisle et al
Atrioventricular Septal Defect
survival also may be explained by the years covered by the retrospective reviews: 1980 –1986 in Machado et al,4 1980 –1993 in Allen et al,8 and 1985–1997 in our series. We disagree with Machado et al,4 who qualified the prognosis associated with a prenatal diagnosis of atrioventricular septal defect as being very poor (27% in their series). After excluding the terminations of pregnancy, our overall survival rate was 63%. Although this survival rate appears clinically good, it clearly tends to improve in specific situations (association with trisomy 21 [83%] and isolated cardiac lesions [79%]). The differences in survival rates between fetuses with Down syndrome and those with normal karyotypes in our series were not statistically significant. The same finding applies to the difference in survival rates between isolated and nonisolated cardiac lesions. This is most likely due to our small sample size and limited power; however, we believe this difference remains clinically important and should be part of antenatal counseling. It is very important that any associated cardiac or noncardiac anomalies be ruled out by thorough ultrasonographic examinations and fetal echocardiograms when atrioventricular septal defect is diagnosed. Clearly identifying isolated versus nonisolated atrioventricular septal defect is an important factor in prenatal counseling. Our results confirm in the antenatal population the already described association between postnatal isolated atrioventricular septal defect and aneuploidy. In more than 80% of our cases of trisomy 21, the cardiac lesion was isolated, compared with only 25% when the karyotype was normal. When an isolated cardiac lesion of atrioventricular septal defect was diagnosed prenatally, there was a significantly higher chance of it being associated with trisomy 21 than with a normal karyotype. Prenatal counseling in cases of atrioventricular septal defect should include the 58% risk of aneuploidy and 50% risk of trisomy 21. When the cardiac lesion is isolated, the risk increases to 80%. Although these differences are not statistically significant, the survival percentages appear to be better in the Down syndrome group and the isolated cardiac lesion group. Clinical situations involving isolated cardiac lesions associated with trisomy 21 were clearly the most common in our center.
References 1. Tweddell JS, Litwin SB, Berger S, Friedberg DZ, Thomas JP, Frommelt PC, et al. Twenty-year experience with repair of complete atrioventricular septal defects. Ann Thorac Surg 1996;62:419 – 24. 2. Rizzoli G, Mazzucco A, Maizza F, Daliento L, Rubino M, Tursi V, et al. Does Down syndrome affect prognosis of surgically managed
Obstetrics & Gynecology
3.
4. 5.
6.
7.
8.
atrioventricular canal defects? J Thorac Cardiovasc Surg 1992;104: 945–53. Marino B. Atrioventricular septal defect—anatomic characteristics in patients with and without Down’s syndrome. Cardiol Young 1992;2:308 –10. Machado MVL, Crawford DC, Anderson RH, Allen LD. Atrioventricular septal defect in prenatal life. Br Heart J 1988;59:352–5. Allan LD, Crawford DC, Anderson RH, Tynan MJ. Echocardiographic and anatomical correlations in fetal congenital heart disease. Br Heart J 1984;52:542– 8. Allan LD, Crawford DC, Anderson RH, Tynan MJ. Spectrum of congenital heart disease detected echocardiographically in prenatal life. Br Heart J 1985;54:523– 6. Gembruch U, Knopfle G, Chatterjee M, Bald R, Redel DA, Fodisch HJ, et al. Prenatal diagnosis of atrioventricular canal malformations with up-to-date echocardiographic technology: Report of 14 cases. Am Heart J 1991;121:1489 –97. Allan LD, Sharland GK, Milburn A, Lockhart SM, Groves AMM, Anderson RH, et al. Prospective diagnosis of 1006 consecutive cases of congenital heart disease in the fetus. J Am Coll Cardiol 1994;23:1452– 8.
VOL. 94, NO. 5, PART 1, NOVEMBER 1999
9. Carmi R, Boughman JA, Ferencz C. Endocardial cushion defect: Further studies of “isolated” versus “syndromic” occurrence. Am J Med Genet 1992;43:569 –74. 10. Marino B, Vairo U, Corno A, Nava S, Guccione P, Calabro R, et al. Atrioventricular canal in Down syndrome. Am J Dis Child 1990; 144:1120 –2.
Reprints are not available.
Received January 12, 1999. Received in revised form April 15, 1999. Accepted May 5, 1999.
Copyright © 1999 by The American College of Obstetricians and Gynecologists. Published by Elsevier Science Inc.
Delisle et al
Atrioventricular Septal Defect
767