Background
Atrioventricular septal defect (AVSD) is a common prenatal diagnosis with great heterogeneity. Few data guide counseling about outcomes and types of repair. The aim of this study was to describe predictors of survival and repair type in prenatally diagnosed atrioventricular septal defect.
Methods
A retrospective review of fetuses diagnosed with AVSD was conducted.
Results
Of 106 fetuses with AVSDs, outcome data were available in 88, 66 with situs solitus and 22 with heterotaxy. Overall mortality was 52%, with 66% surviving to birth, 58% to 1 month, and 47% to 1 year. There was increased mortality in patients with heterotaxy ( P = .02). In patients with situs solitus, there was increased mortality in those with unbalanced AVSDs ( P < .01). The presence of a chromosomal abnormality did not affect mortality ( P = .34). In pregnancies with intent to continue ( n = 60), 97% of patients survived to birth, 86% to 1 month, and 69% to 1 year. The presence of heterotaxy ( P < .01) or other complex intracardiac disease ( P < .01) was associated with single-ventricle repair. In patients with unbalanced AVSDs who underwent surgery, two-ventricle repair was uncommon (29%), but it was performed more often in those with restrictive ventricular septal defects (71%). In the 16 patients with unbalanced AVSDs who underwent multiple studies, 12 showed no change in left atrioventricular valve/right atrioventricular valve ratio from first to the final fetal echocardiographic study; in three patients, this ratio worsened, and in one patient, it improved.
Conclusions
Prenatal diagnosis of AVSD should prompt assessment for chromosomal abnormalities, additional cardiac anomalies, and heterotaxy syndrome. Heterotaxy, unbalanced AVSD, and the need for single-ventricle repair are associated with a poor prognosis. Left atrioventricular valve/right atrioventricular valve ratio early in gestation does not always predict this ratio at delivery.
Atrioventricular septal defects (AVSDs) are among the most common congenital cardiac lesions detected prenatally. These defects can be identified prenatally, as early as the first trimester. The term “AVSD” implies underdevelopment of the endocardial cushions, which form the origins of the atrioventricular septum and atrioventricular valves (AVV). Patients with AVSDs present with various combinations including, a common AVV and deficiency of the primum atrial septum and inlet ventricular septum. An unbalanced AVSD forms when the common AVV opens predominately toward the larger ventricle, with the contralateral ventricle being hypoplastic. Fetuses with both balanced and unbalanced AVSDs differ in their associations with additional congenital cardiac defects and chromosomal disorders. AVSD may be associated with heterotaxy syndrome, an embryologic lateralization disorder that results in abnormal formation of the left-right axis of the body and includes visceral defects including polysplenia or asplenia, liver and gallbladder abnormalities, and intestinal malrotation. Patients with heterotaxy also often have complex congenital heart disease, including venous anomalies, AVSDs, and conotruncal defects. There are two main types of heterotaxy: left isomerism (bilateral right-sidedness associated with asplenia) and right isomerism (bilateral left-sidedness associated with polysplenia); both are associated with AVSD.
Early reports of fetal diagnosis of AVSD have suggested that the outlook for fetuses diagnosed in utero is considerably worse than would be expected from postnatal surgical series. However, few data are available to guide prenatal counseling or predict postnatal outcomes. We report data from a large series of fetuses with AVSDs evaluated at our institution. We describe the characteristics and outcomes of the group and the effects of intracardiac and extracardiac features on prognosis and management, for the group as a whole at the time of diagnosis and for the subgroup with intent to continue pregnancy.
Methods
The echocardiography database at Children’s National Medical Center was searched for patients diagnosed with AVSD over a 10-year period (2001–2011). Patients selected were assigned at least one of the following diagnostic codes: common AVV, primum atrial septal defect, atrioventricular canal type ventricular septal defect (VSD), and partial, transitional, or unbalanced AVSD. Additional information on associated cardiac defects, genetic findings, surgical repair, and outcomes was gathered from echocardiographic reports, prenatal screening forms, and postnatal records. This study was approved by the institutional review board.
Fetal Measurements
All fetal echocardiograms were retrospectively reviewed by a single physician (A.Z.B.), who was unaware of the prenatal or postnatal diagnoses. The echocardiogram closest to 20 weeks’ gestation was reviewed. For the subgroup of patients with unbalanced AVVs, all studies were reviewed. Gestation age was calculated using the fetal biophysical parameters of biparietal diameter, abdominal circumference, and femur length according to the standard protocol.
Complete AVSD was defined as a single AVV, a primum atrial septal defect, and a large (presumed pressure unrestrictive) inlet VSD. Partial AVSD was characterized by a primum atrial septal defect with an intact ventricular septum. Transitional AVSD was characterized by a primum ASD, a two-orifice AVV, and a small or moderate (presumed restrictive) inlet VSD. Because both partial and transitional AVSDs have similar physiology, with large atrial shunts and little or no ventricular shunts, data for these defects were combined in the analysis and labeled “partial/transitional AVSD.”
Quantitative measurements of the left AVV (LAVV) and right AVV (RAVV) were made in the four-chamber view at maximum diameter in diastole. An imaginary vertical line was drawn from the crest of the ventricular septum to the level of the AVV annulus in cardiac diastole, bisecting the valve, similar to the reported protocol for postnatal assessment of AVSD ( Figure 1 ). The ratio of the LAVV to the RAVV was calculated. In normal heart development, the LAVV/RAVV ratio is nearly equal at 16 to 20 weeks; however, as pregnancy advances, the ratio becomes smaller (progressive dilation of the right side), with a mean of 0.9 at term. We defined patients with balanced AVSDs as those with LAVV/RAVV ratios of 0.85 to 1.15. This range represents approximately 1 standard deviation (0.124) of the mean for published normal values. An LAVV/RAVV ratio < 0.85 indicated that the AVV was directed more to the right ventricle (right-dominant AVSD [R-AVSD]), and an LAVV/RAVV ratio > 1.15 indicated that the AVV was more directed to the left ventricle (left-dominant AVSD [L-AVSD]). Quantitative measurements of right and left ventricular lengths were also made in the four-chamber view at maximum diastole. Measurement was taken from the AVV orifice at the midcavitary point of each ventricle to that ventricle’s respective apex. In normal heart development, the ratio of left ventricular length to right ventricular length remains fairly constant at 1:1 (at 17 weeks’ gestation) to 1.1:1 (at term).
Associated Anomalies
For all fetal echocardiograms reviewed, the VSD was classified as large, small/moderate, or not present. If the VSD in the four-chamber view was larger than the aortic diameter, it was classified as large; otherwise, it was classified as small/moderate. A qualitative assessment of AVV regurgitation was noted and graded as none, mild/moderate, or severe. If AVV regurgitation was determined in the four-chamber view to extend to the posterior wall of the atrium, it was classified as severe; otherwise, it was classified as mild/moderate. Associated cardiac malformations and noncardiac defects including conotruncal anomalies, aortic coarctation, and/or semilunar valve atresia were recorded. Because of varied and complex combinations of intracardiac and extracardiac anomalies present in patients with heterotaxy syndrome, these fetuses were compared as a group with those with situs solitus. Fetuses with situs solitus who had associated structural cardiac malformations were labeled as having complex disease. The fetal karyotype determined prenatally, postnatally, or after termination of pregnancy was recorded whenever results were available or otherwise recorded as unknown.
Outcomes Data
Terminations and intrauterine deaths were noted. For live births, gestation age at delivery, survival to surgery, type of surgical repair (either single-ventricle [SV] or two-ventricle [2V] repair), and clinical status to date were determined from retrospective reviews of the babies’ medical records. SV repair was defined as any palliative surgery in which the resultant physiology was that of an SV supplying both the pulmonary and systemic circulation or in which an SV supplied the systemic circulation with passive flow to the pulmonary circulation. A 2V repair was defined as any surgery that restored a normal circulation, with the right ventricle supporting the pulmonary circulation, the left ventricle supporting the systemic circulation, and no intracardiac shunting.
Statistical Analysis
Statistical analysis was performed on data obtained for the AVSD group as a whole, and then data were compared between those with heterotaxy and those with situs solitus. After this primary analysis, a secondary analysis was conducted to determine predictors of outcomes and repair in continuing pregnancies. Pearson’s correlation coefficient was used to determine the relationship between the ratio of left ventricular length and right ventricular length and the LAVV/RAVV ratio. For comparison of subgroups within the AVSD population, 95% confidence intervals were calculated using the method of proportions. Categorical variables were analyzed using Fisher’s exact tests, and continuous variables were analyzed using independent-samples two-tailed t tests. Survival in continuing pregnancies was subject to Kaplan-Meier analysis, with censoring of surviving individuals at their last known follow-up date and with an additional event time to describe prenatal terminations and spontaneous intrauterine deaths. Differences in survival between subgroups were compared using separate Kaplan-Meier analyses and tested for statistical significance using log-rank tests. Statistical analyses were performed using MedCalc for Windows version 12.0 (MedCalc Software, Mariakerke, Belgium). In all statistical tests, P values < .05 were regarded as significant. Numbers in the subgroups with R-AVSD and L-AVSD were small, and therefore results are reported as percentages (without statistical analysis). These percentages are included given that these findings contain important information on outcome trends.
To determine intrarater reliability, 30 studies were blindly reviewed by a second physician (J.I.P.). The intraclass correlation coefficient was used to determine the absolute agreement in LAVV/RAVV ratio between reviewers.
Results
From 2001 to 2011, 106 fetuses were diagnosed with AVSD at Children’s National Medical Center. The mean gestational age at the time of the first fetal echocardiographic study was 22.5 ± 4.6 weeks (range, 14–35 weeks). Demographics and associated intracardiac and extracardiac anomalies of the study population are shown in Table 1 . For the group, LAVV/RAVV ratio was correlated linearly with the ratio of left ventricular length to right ventricular length ( R = 0.77, P < .01); thus, all analyses were made using only the LAVV/RAVV ratio to describe balance versus unbalance. For the studies blindly reviewed by two physicians, absolute agreement between calculated LAVV/RAVV ratios was good (intraclass correlation coefficient, 0.96; 95% confidence interval, 0.92–0.98).
| Variable | Heterotaxy ( n = 24) | Situs solitus ( n = 82) | P |
|---|---|---|---|
| GA at presentation (wks) | 22.1 (21.3–23.0) | 22.8 (20.8–24.8) | .49 |
| GA at birth (wks) | 37.5 (36.7–38.2) | 37.0 (36.0–38.0) | .50 |
| Balanced AVSD | 10 (42%) | 56 (68%) | .02 |
| Mild to moderate AVVR | 4 (17%) | 13 (16%) | 1.00 |
| Severe AVVR | 1 (4%) | 1 (1.5%) | .41 |
| Restrictive VSD | 4 (17%) | 17 (21%) | .78 |
| Aortic coarctation | 1 (4%) | 16 (20%) | .07 |
| Conotruncal anomaly | 16 (67%) | 15 (18%) | <.001 |
| Male gender ∗ | 5 (21%) | 20 (24%) | 1.00 |
| 2V repair † | 4 (17%) | 32 (39%) | .05 |
| Overall mortality ‡ | 16 (73%) | 33 (50%) | .03 |
| Mortality in pregnancies with intent to continue § | 5 (46%) | 11 (22%) | .15 |
† Available for 51 patients who underwent surgical repair or palliation.
‡ Available for 88 patients with confirmed outcomes (heterotaxy in 22, situs solitus in 66).
§ Available for 60 patients with pregnancy with intent to continue (heterotaxy in 11, situs solitus in 49).
Genetic information was available for 76 fetuses, 18 with heterotaxy (76%) and 58 (71%) with situs solitus. In total, 31 abnormalities were identified. Only one patient with heterotaxy had a chromosomal abnormality, a deletion of one copy of chromosome 1. Thirty fetuses with situs solitus (51%) had genetic abnormalities. Of these, chromosomal abnormalities were significantly more likely in those with balanced AVSDs (53% balanced vs 2% unbalanced, P < .01). The majority of these patients had trisomy 21 (80%). Full karyotype results are displayed in Table 2 .
| Situs | Type of AVSD | Genetics known | Chromosomal anomaly | Cases | Group | Termination | Neonatal death | Alive |
|---|---|---|---|---|---|---|---|---|
| Heterotaxy | Balanced | |||||||
| n = 10 | 8 (80%) | None | 0 | 0 | 0 | 0 | 0 | |
| Unbalanced | 10 (71%) | |||||||
| n = 14 | Chromosome 1 deletion | 1 | 2.50% | 1 | 0 | 0 | ||
| Situs solitus | Balanced | 42 (75%) | ||||||
| n = 56 | Trisomy 18 | 3 | 6.50% | 3 | 0 | 0 | ||
| Trisomy 21 | 24 | 52% | 1 | 3 | 21 | |||
| Smith-Lemli-Opitz syndrome | 1 | 2.10% | 0 | 1 | 0 | |||
| Unbalanced | 16 (62%) | |||||||
| n = 26 | Trisomy 21 | 2 | 5% | 0 | 1 | 1 |
Outcomes
Outcome results for the 106 patients identified prenatally are displayed in Figure 2 . Outcomes could not be determined for 18 fetuses (17%) whose mothers did not return for any further evaluation (two with heterotaxy and 16 with situs solitus). The number of patients lost to follow-up is much higher than the normal rate at our institution, suggesting that the lack of return was likely due to termination of the pregnancy. Because this could not be confirmed, however, these fetuses were excluded from analysis.
Overall mortality in the remaining 88 patients with known outcomes was 53%, including 28 terminated fetuses. Although the results did not meet the threshold for significance, there was a trend toward increased termination rates in fetuses with heterotaxy ( P = .06) and fetuses with unbalanced ventricles ( P = .06). The identification of a chromosomal abnormality did not affect termination rates (40% with chromosomal abnormalities vs 42% without, P = 1.00). The mean length of follow-up for women who continued with their pregnancies was 48.6 months (range, 2.7–122.5 months). Kaplan-Meier analysis is shown in Figure 3 . Sixty-six percent of fetuses survived to birth, 63% to 1 week, 58% to 1 month, and 47% to 1 year. There were no deaths after 1 year.
Heterotaxy Syndrome
Twenty-four cases of heterotaxy syndrome were identified. There were equal numbers with asplenia ( n = 12) and polysplenia ( n = 12). Four patients with polysplenia were noted to have systemic venous anomalies, and two (one intrauterine demise) were found to have complete heart block. Eight patients with asplenia were noted to have pulmonary venous anomalies (three with partial and five with total anomalous pulmonary venous return). Patients with heterotaxy compared with those with situs solitus were more likely to have unbalanced AVSDs ( P = .02) and associated conotruncal anomalies ( P < .01) and less likely to undergo 2V repair ( P = .05).
There was decreased survival in patients with heterotaxy (27% with heterotaxy vs 54% with situs solitus, P = .02; Figure 4 A). Although not powered for significance, it is notable that only one of the 12 (8%) patients with heterotaxy syndrome and unbalanced AVSDs survived, compared with eight of 22 (36%) of those with situs solitus and unbalanced AVSDs ( P = .10), suggesting that the presence of heterotaxy may have affected survival in patients with unbalanced AVSDs. There was no difference in survival between fetuses with heterotaxy and those with situs solitus when the AVSD was balanced and the ventricles were equal in size ( P = .60).
Situs Solitus
There were 82 fetuses with AVSDs and situs solitus. Fifty-six had balanced defects, and 26 had unbalanced defects. In patients with situs solitus, those with unbalanced AVSDs showed decreased survival (63% with balanced AVSDs vs 35% with unbalanced AVSDs survived, P < .01; Figure 4 B). Identification of a chromosomal abnormality did not affect mortality (38% with chromosomal abnormalities vs 33% with normal chromosomes died, P = .78).
Intent-to-Continue Analysis
A subanalysis was conducted on fetuses of mothers with intent to continue their pregnancies ( n = 60) to assess factors influencing postnatal outcomes and repair. Survival in utero to delivery was 97%. For live births, the mean duration of follow-up was 48.6 months (range, 2.7–122.5). Kaplan-Meier analysis of this group is shown in Figure 5 A. Eighty-six percent survived the neonatal period (first 30 days), and 69% survived the first year of life. For fetuses with isolated and uncomplicated AVSDs, Kaplan-Meier analysis is shown in Figure 5 B. Not included in this group are fetuses with heterotaxy, double outlet right ventricle (DORV), conotruncal anomalies, unbalanced AVVs, and major extracardiac anomalies. In these 33 patients, survival was significantly better than in the cohort as a whole (85% vs 50%, P < .01).
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
