j o u r n a l o f s u r g i c a l r e s e a r c h x x x ( 2 0 1 4 ) 1 e4
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.JournalofSurgicalResearch.com
Comparing characteristics and outcomes in infants with prenatal and postnatal diagnosis of esophageal atresia Sara C. Fallon, MD,a,b Cecilia G. Ethun, MD,a,b Oluyinka O. Olutoye, MD, PhD,a,b,c Mary L. Brandt, MD,b Timothy C. Lee, MD,a,b Stephen E. Welty, MD,a,d Rodrigo Ruano, MD, PhD,a,c and Darrell L. Cass, MDa,b,c,* a
Texas Children’s Fetal Center, Texas Children’s Hospital, Houston, Texas Michael E. DeBakey Department of Surgery, Baylor College of Medicine, Houston, Texas c Department of Obstetrics and Gynecology, Baylor College of Medicine, Houston, Texas d Division of Neonatology, Department of Pediatrics, Baylor College of Medicine, Houston, Texas b
article info
abstract
Article history:
Background: Previous studies of infants with esophageal atresia (EA) suggest those diag-
Received 4 January 2014
nosed prenatally have worse outcomes because of a higher incidence of associated
Received in revised form
anomalies. The purpose of this study was to compare characteristics and outcomes of
18 March 2014
infants with EA diagnosed after fetal center evaluation to those diagnosed postnatally.
Accepted 21 March 2014
Methods: The records of all neonates treated for EA at our institution from 2002e2012 were
Available online xxx
reviewed. Infants with a prenatal diagnosis of EA were compared with those postnatally diagnosed using chi-square and Student t-test as appropriate.
Keywords:
Results: Of 91 patients treated with EA during the study period, 15 (16%) were diagnosed
Esophageal atresia
prenatally at our fetal center. Although those prenatally diagnosed had a higher incidence
Tracheoesophageal fistula
of pure EA and polyhydramnios, the gestational age and birth weight in that group were
Prenatal diagnosis
similar to those diagnosed postnatally. There were no differences in outcomes between
Fetal MRI
groups with regard to the incidence of major cardiac anomalies, surgical complications,
Fetal center
hospital length of stay, and survival. Conclusions: Treatment at a tertiary care center provides excellent outcomes for all infants with EA, despite an 80% frequency of concurrent anomalies. Prenatal diagnosis of EA and attentive obstetric management of polyhydramnios decrease the risk for prematurity and prematurity-associated morbidity. ª 2014 Elsevier Inc. All rights reserved.
1.
Introduction
Prenatal diagnosis and fetal management of infants with esophageal atresia (EA) and possible tracheoesophageal fistula (TEF) have evolved as imaging, and perinatal care has
improved. Theoretically, prenatal diagnosis improves patient management by ensuring delivery at a tertiary care hospital with immediate access to neonatal surgical services, by early detection of associated anomalies that may impact outcome, and by preparing the parents for a possible protracted
* Corresponding author. Texas Children’s Fetal Center, Texas Children’s Hospital, Division of Pediatric Surgery, 6701 Fannin, Suite 1210, Houston, TX 77030. Tel.: þ1 832 822 3135; fax: þ1 832 8253141. E-mail address: [email protected] (D.L. Cass). 0022-4804/$ e see front matter ª 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jss.2014.03.068
2
j o u r n a l o f s u r g i c a l r e s e a r c h x x x ( 2 0 1 4 ) 1 e4
neonatal stay [1]. Advances in ultrasound technology and the use of fetal magnetic resonance imaging (MRI) have improved prenatal diagnostic capabilities [1,2]; yet, the rate of prenatal diagnosis for EA continues to range from 12%e74% [3e6]. On fetal imaging, EA is suspected with combined ultrasonographic findings of an absent stomach bubble and polyhydramnios, although the positive predictive value of these findings remains low [7]. The use of fetal MRI can improve the predictive value significantly by improved characterization of the fetal esophagus and detection of an upper pouch sign, which is dependent on active fetal swallowing [2,7]. Whether prenatal diagnosis results in improved patient outcomes compared with those infants diagnosed postnatally remains unclear, as the literature on this topic is sparse and conflicting. Further confounding the issue is the hypothesis that the two patient groups may be fundamentally different, as those infants that have associated anomalies that decrease survival are more likely to be diagnosed prenatally compared with those with isolated EA and/or TEF in which the ultrasonographic abnormalities may be subtle. One study found that patients with a prenatal diagnosis had decreased survival, as this group had a higher rate of associated anomalies, a known predictor of increased mortality in EA and/or TEF [5]. Another study found improved survival in the prenatally diagnosed group [6]. We hypothesized that the prenatally diagnosed patients referred to our fetal center would have a higher rate of associated anomalies than those diagnosed postnatally. We also hypothesized that the group diagnosed prenatally in our fetal center would have similar survival and outcomes to those diagnosed after birth.
2.
Methods
2.1.
Fetal center management
At the Texas Children’s Hospital Fetal Center, all patients referred for suspected EA undergo a multidisciplinary evaluation. If ultrasonographic findings are convincingly suspicious for EA (absent stomach bubble, polyhydramnios, and esophageal pouch sign), a fetal MRI and fetal echocardiography are obtained to confirm the diagnosis and evaluate for associated anomalies. In addition to the initial consultations with maternal fetal medicine and pediatric surgery, cardiology, cardiac surgery, and genetic consultation are obtained as needed.
2.2.
Study design
After approval by the Institutional Review Board of Baylor College of Medicine (H-32260), all patients born with EA and/or TEF treated at our institution from January 2002eDecember 2012 were identified by International Classification of Diseases, Ninth Revision code. Patients who received initial surgery at another center but had subsequent treatment at our facility were not included. Patients with prenatally suspected EA with subsequently terminated pregnancies were also excluded. Patients who received prenatal care at an outside facility and for which a prenatal diagnosis of EA could not be confirmed were categorized in the postnatal diagnosis group.
A retrospective chart review was performed. Patient characteristics and surgical outcomes of the patients who received a prenatal diagnosis of EA and/or TEF and treated at the Texas Children’s Fetal Center were compared with those who were diagnosed with EA and/or TEF after birth. The primary outcome was survival. Secondary outcomes included the type of anomaly, the incidence of other congenital abnormalities, surgical complication rates (including anastomotic leak, recurrent fistula, pneumothorax requiring intervention, infection, and reoperation), hospital length of stay, time to full feeds, time of mechanical ventilation, and duration of oxygen requirement. Neonates with EA and/or TEF are treated in a standardized fashion in our institution with early surgical repair once stabilized. Those with EA and no TEF usually will undergo gastrostomy with delayed primary esophageal repair at 2e4 mo based on radiologic demonstration of less than one vertebral body distance between the proximal esophagus and distal esophagus. Statistical analysis included chi-square for categorical variables and Student t-test for continuous variables.
3.
Results
3.1.
Patient characteristics
We identified 15 patients with a confirmed prenatal diagnosis of EA and 76 with a postnatal diagnosis. All except one (93%) prenatally diagnosed patient were born at our institution compared with 26% of infants postnatally diagnosed. The rates of polyhydramnios were higher in the prenatally diagnosed group (Table 1). Patients with pure EA had a higher rate of polyhydramnios (100%) compared to those with a fistula (45%). Pregnancies were otherwise reported as being uncomplicated, except for a 12% rate of gestational diabetes mellitus in the postnatally diagnosed patients. Patients in each group were born at similar mean gestational ages, with similar mean birth weights (Table 1). There was a significantly higher incidence of pure EA in the prenatally diagnosed group (47% versus 11%; P ¼ 0.003) and conversely a lower incidence of EA with
Table 1 e Patient characteristics and demographics.
Male Prenatal care Polyhydramnios GA at birth (wk) Preterm birth (GA < 38 wk) Birth weight (g) Pure EA (type A) EA with TEF (types B, C, and D) Other anomalies VACTERL association Major cardiac anomalies GA ¼ gestational age.
Prenatal diagnosis, n ¼ 15 (%)
Postnatal diagnosis, n ¼ 76 (%)
P value
8 (53) 15 (100) 14/15 (93) 36.5 2.3 9 (60)
36 (47) 69 (89) 38/69 (55) 35.8 3.0 50 (66)
0.78 0.59 0.007 0.35 0.77
2390 474 7 (47) 8 (53)
2347 787 8 (11) 66 (88)
0.84 0.003 0.003
12 (80) 4 (27) 10 (67)
54 (82) 26 (34) 46 (61)
1.0 0.77 0.78
3
j o u r n a l o f s u r g i c a l r e s e a r c h x x x ( 2 0 1 4 ) 1 e4
TEF (53% versus 88%; P ¼ 0.003). The presence of associated anomalies was similar between groups (80% versus 82%; P ¼ 1.0) as was the incidence of the VACTERL association (27% versus 34%; P ¼ 0.78) and major cardiac anomalies (defined as those which required either medical or surgical treatment; 67% versus 61%; P ¼ 0.78).
3.2.
Patient outcomes
Overall, survival was 80% in infants diagnosed prenatally and 88% for those diagnosed postnatally at a median length of follow-up of 7.9 y (range, 0.47e10.2 y). Survival in patients with isolated EA and/or TEF was 100% in both groups (Table 2). The time to death was similar in each group (median day, 24 [range, 1e287] versus 24 [range, 0e450]; P ¼ 0.8). Eight patients were unable to undergo repair of their EA and/or TEF because of the finding of nonsurvivable genetic or cardiac conditions. This did not differ between groups with two patients in the prenatal group and six patients in the postnatal group expiring before the repair could be completed (13% versus 8%; P ¼ 0.6). There was no statistically significant difference in postsurgical complication rates (46% versus 31%; P ¼ 0.34); specifically, anastomotic leak and recurrence rates were similar between groups. Although the difference was not statistically significant, the median length of stay in the prenatal diagnosis group was nearly twice that of the postnatally diagnosed group (92 versus 52 d; P ¼ 0.2). Regarding long-term outcomes and sequelae from the underlying disease process, stricture rates requiring dilation procedures were similar (50% versus 41%; P ¼ 0.75). There was no statistically significant difference in fundoplication rates, although the prenatal rate was double the postnatal group’s rate (25% versus 12%; P ¼ 0.36). Time to initial oral feeding was higher in the prenatally diagnosed group (median, 144 versus 30 d; P ¼ 0.009), which was mostly because of the higher rate of pure EA in that group. Similarly, the rate of gastrostomy tube insertion was higher in the prenatal group (73% versus 49%; P ¼ 0.09), a reflection of the standard therapy for this anomaly. The rates of tracheomalacia (33% versus 29%; P ¼ 0.7) and tracheostomy (13% versus 11%; P ¼ 0.7) were unchanged in the prenatal versus postnatal groups, respectively.
4.
Discussion
The management of EA and TEF has progressed to the point that survival in infants with isolated disease is >90%, and in some reports, it approaches 100%, as seen in the current patient series [7]. However, infants with EA continue to experience significant morbidity, such as postsurgical complications (leak, stricture, and recurrence), prolonged need for supplemental feeding, respiratory difficulties, and failure to thrive. Our data suggest that prenatal diagnosis does not reduce the incidence of morbidity in patients with EA and/or TEF. However, prenatal diagnosis does facilitate optimal obstetric management. Our study found a much higher rate of polyhydramnios in the prenatally diagnosed group, which could predispose these mothers to preterm labor, thus exposing the infant to morbidity related to prematurity. Prenatal management allowed for this higher risk group to be born at a similar gestational age and birth weight with an equivalent prematurity rate, which may have had a positive effect on outcomes that our sample size was too small to detect. The difference in our prenatal and postnatal groups with respect to the rate of pure EA (47% versus 11%) is attributable to the presence of polyhydramnios. Fetuses with EA and no TEF will have polyhydramnios (100% in our series) and an absent stomach bubble in nearly all cases, which will increase the likelihood of prenatal detection and fetal center referral, as demonstrated in the present series. In some instances, inhaled amniotic fluid in a fetus with a TEF will pass through the trachea and fistula tract and enter the stomach, which limits the progression to polyhydramnios (45% in our series). Although the primary outcome of survival was similar between the two comparison groups, there were some differences in outcomes, particularly related to hospital stay and time to first oral feeds that favored the postnatally diagnosed group. Such differences can likely be attributed to the higher rate of pure EA without fistula in the prenatal group, as these patients have a planned delay before repair is attempted. Surgical complication rates were similar in the two groups, in contrast to other studies, which have shown a higher leak and stricture rate in patients with pure EA [8].
Table 2 e Postnatal and postoperative outcomes. Prenatal diagnosis, n ¼ 15 (%) Survival, all patients Survival, no other anomalies Surgical complications Anastomotic leak Recurrent fistula Postoperative pneumothorax Esophageal dilation Fundoplication Time to initial oral feeds (d) Median length of hospital stay (d) Median follow-up (y) * y
12 3 6/13 5 1 2 6/12 3/12 144 92 2.05
(80) (100) (46)* (33) (8) (15) (50)y (25) (3e857) (12e232) (0.46e7.66)
Excludes patients who died before surgery and those with missing postoperative data. Excludes patients who died.
Postnatal diagnosis, n ¼ 76 (%) 66 22 19/61 15 3 4 27/66 8/66 30 52 7.41
(88) (100) (31)* (25) (5) (7) (41)y (12) (7e760) (7e289) (0.79e10.21)
P value 0.44 1.0 0.34 0.31 0.52 0.26 0.75 0.36 0.009 0.20 0.002
4
j o u r n a l o f s u r g i c a l r e s e a r c h x x x ( 2 0 1 4 ) 1 e4
The limited literature that discusses possible benefits to prenatal diagnosis for EA reports conflicting results. de Jong et al. found decreased survival (70% versus 88%; P ¼ 0.05) and an increased rate of associated anomalies (80% versus 59%; P ¼ 0.04) in the prenatal versus postnatal groups, respectively [5]. In contrast, Brantberg et al., who described outcomes for 38 infants with EA and additional anomalies, showed improved survival (67%) with prenatal diagnosis of EA compared with postnatal diagnosis (27%). In a study by Brantberg et al., overall survival was 90% for 10 infants with isolated EA but only 45% for those with EA and additional anomalies [6]. As this study does not report the specific associated anomalies found in each group, it is difficult to tell if the difference in outcomes was related to the underlying patient group characteristics, as hypothesized in the de Jong et al. study. In our present series, overall survival for those with EA and additional anomalies was 80%. As the outcomes in the current literature come to discrepant conclusions, and the ability of the fetal center to accurately diagnose the condition prenatally has been improving, further multicenter prospective study is needed to convincingly address this question. Limitations to the present study include a small sample size in the prenatally diagnosed group. There may be differences in outcomes that were too small to detect with our smaller patient population, particularly with respect to survival. Likewise, as this is a specific fetal center population for our comparison group, the generalizability of findings from this study is likely limited to tertiary referral centers. Further study including data from multiple centers would allow for more definitive conclusions. As the prenatally diagnosed group represents a slightly more contemporary group of patients than the postnatally diagnosed patients, the amount of observed follow-up time for the prenatal group is shorter and so there may be a detection bias for some of the long-term complication rates reported. In conclusion, these data can be used to improve counseling during prenatal consultation for patients with suspected EA. Despite previous reports that a prenatal diagnosis portends decreased survival and an increased rate of other congenital anomalies, our data show overall survival is the same, and patients do not appear to have a higher rate of other congenital anomalies. Although prenatally diagnosed patients have a higher rate of pure EA, the rate of concurrent anomalies is also equivalent.
Acknowledgment S.F. and D.C. contributed to the conception and design, data collection and analysis, and manuscript drafting and revision; C.E. to the conception and design, data collection and analysis, and manuscript drafting; O.O. to the conception and design and manuscript drafting and revision; M. B., S.W., and R.R. to the concept and design and manuscript revision; T.L. to the conception and design, data collection and analysis, and manuscript revision.
Disclosure The authors have no sources of support or disclosures to report with respect to the preparation of this manuscript.
references
[1] Houfflin-Debarge V, Bigot J. Ultrasound and MRI prenatal diagnosis of esophageal atresia: effect on management. J pediatr gastroenterol nutr 2011;52(Suppl 1):S9. [2] Ethun CG, Fallon SC, Cassady CI, et al. Fetal MRI improves diagnostic accuracy in patients referred to a fetal center for suspected esophageal atresia. J Pediatr Surg; 2014. In press. [3] Seo J, Kim do Y, Kim AR, et al. An 18-year experience of tracheoesophageal fistula and esophageal atresia. Korean j pediatr 2010;53:705. [4] Pedersen RN, Calzolari E, Husby S, Garne E. Oesophageal atresia: prevalence, prenatal diagnosis and associated anomalies in 23 European regions. Arch dis child 2012;97:227. [5] de Jong EM, de Haan MA, Gischler SJ, et al. Pre- and postnatal diagnosis and outcome of fetuses and neonates with esophageal atresia and tracheoesophageal fistula. Prenatal diagn 2010;30:274. [6] Brantberg A, Blaas HG, Haugen SE, Eik-Nes SH. Esophageal obstruction-prenatal detection rate and outcome. Ultrasound obstetrics gynecol: official j Int Soc Ultrasound Obstetrics Gynecol 2007;30:180. [7] Kunisaki SM, Foker JE. Surgical advances in the fetus and neonate: esophageal atresia. Clin perinatology 2012;39:349. [8] Engum SA, Grosfeld JL, West KW, et al. Analysis of morbidity and mortality in 227 cases of esophageal atresia and/or tracheoesophageal fistula over two decades. Arch Surg 1995; 130:502.
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.JournalofSurgicalResearch.com
Comparing characteristics and outcomes in infants with prenatal and postnatal diagnosis of esophageal atresia Sara C. Fallon, MD,a,b Cecilia G. Ethun, MD,a,b Oluyinka O. Olutoye, MD, PhD,a,b,c Mary L. Brandt, MD,b Timothy C. Lee, MD,a,b Stephen E. Welty, MD,a,d Rodrigo Ruano, MD, PhD,a,c and Darrell L. Cass, MDa,b,c,* a
Texas Children’s Fetal Center, Texas Children’s Hospital, Houston, Texas Michael E. DeBakey Department of Surgery, Baylor College of Medicine, Houston, Texas c Department of Obstetrics and Gynecology, Baylor College of Medicine, Houston, Texas d Division of Neonatology, Department of Pediatrics, Baylor College of Medicine, Houston, Texas b
article info
abstract
Article history:
Background: Previous studies of infants with esophageal atresia (EA) suggest those diag-
Received 4 January 2014
nosed prenatally have worse outcomes because of a higher incidence of associated
Received in revised form
anomalies. The purpose of this study was to compare characteristics and outcomes of
18 March 2014
infants with EA diagnosed after fetal center evaluation to those diagnosed postnatally.
Accepted 21 March 2014
Methods: The records of all neonates treated for EA at our institution from 2002e2012 were
Available online xxx
reviewed. Infants with a prenatal diagnosis of EA were compared with those postnatally diagnosed using chi-square and Student t-test as appropriate.
Keywords:
Results: Of 91 patients treated with EA during the study period, 15 (16%) were diagnosed
Esophageal atresia
prenatally at our fetal center. Although those prenatally diagnosed had a higher incidence
Tracheoesophageal fistula
of pure EA and polyhydramnios, the gestational age and birth weight in that group were
Prenatal diagnosis
similar to those diagnosed postnatally. There were no differences in outcomes between
Fetal MRI
groups with regard to the incidence of major cardiac anomalies, surgical complications,
Fetal center
hospital length of stay, and survival. Conclusions: Treatment at a tertiary care center provides excellent outcomes for all infants with EA, despite an 80% frequency of concurrent anomalies. Prenatal diagnosis of EA and attentive obstetric management of polyhydramnios decrease the risk for prematurity and prematurity-associated morbidity. ª 2014 Elsevier Inc. All rights reserved.
1.
Introduction
Prenatal diagnosis and fetal management of infants with esophageal atresia (EA) and possible tracheoesophageal fistula (TEF) have evolved as imaging, and perinatal care has
improved. Theoretically, prenatal diagnosis improves patient management by ensuring delivery at a tertiary care hospital with immediate access to neonatal surgical services, by early detection of associated anomalies that may impact outcome, and by preparing the parents for a possible protracted
* Corresponding author. Texas Children’s Fetal Center, Texas Children’s Hospital, Division of Pediatric Surgery, 6701 Fannin, Suite 1210, Houston, TX 77030. Tel.: þ1 832 822 3135; fax: þ1 832 8253141. E-mail address: [email protected] (D.L. Cass). 0022-4804/$ e see front matter ª 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jss.2014.03.068
2
j o u r n a l o f s u r g i c a l r e s e a r c h x x x ( 2 0 1 4 ) 1 e4
neonatal stay [1]. Advances in ultrasound technology and the use of fetal magnetic resonance imaging (MRI) have improved prenatal diagnostic capabilities [1,2]; yet, the rate of prenatal diagnosis for EA continues to range from 12%e74% [3e6]. On fetal imaging, EA is suspected with combined ultrasonographic findings of an absent stomach bubble and polyhydramnios, although the positive predictive value of these findings remains low [7]. The use of fetal MRI can improve the predictive value significantly by improved characterization of the fetal esophagus and detection of an upper pouch sign, which is dependent on active fetal swallowing [2,7]. Whether prenatal diagnosis results in improved patient outcomes compared with those infants diagnosed postnatally remains unclear, as the literature on this topic is sparse and conflicting. Further confounding the issue is the hypothesis that the two patient groups may be fundamentally different, as those infants that have associated anomalies that decrease survival are more likely to be diagnosed prenatally compared with those with isolated EA and/or TEF in which the ultrasonographic abnormalities may be subtle. One study found that patients with a prenatal diagnosis had decreased survival, as this group had a higher rate of associated anomalies, a known predictor of increased mortality in EA and/or TEF [5]. Another study found improved survival in the prenatally diagnosed group [6]. We hypothesized that the prenatally diagnosed patients referred to our fetal center would have a higher rate of associated anomalies than those diagnosed postnatally. We also hypothesized that the group diagnosed prenatally in our fetal center would have similar survival and outcomes to those diagnosed after birth.
2.
Methods
2.1.
Fetal center management
At the Texas Children’s Hospital Fetal Center, all patients referred for suspected EA undergo a multidisciplinary evaluation. If ultrasonographic findings are convincingly suspicious for EA (absent stomach bubble, polyhydramnios, and esophageal pouch sign), a fetal MRI and fetal echocardiography are obtained to confirm the diagnosis and evaluate for associated anomalies. In addition to the initial consultations with maternal fetal medicine and pediatric surgery, cardiology, cardiac surgery, and genetic consultation are obtained as needed.
2.2.
Study design
After approval by the Institutional Review Board of Baylor College of Medicine (H-32260), all patients born with EA and/or TEF treated at our institution from January 2002eDecember 2012 were identified by International Classification of Diseases, Ninth Revision code. Patients who received initial surgery at another center but had subsequent treatment at our facility were not included. Patients with prenatally suspected EA with subsequently terminated pregnancies were also excluded. Patients who received prenatal care at an outside facility and for which a prenatal diagnosis of EA could not be confirmed were categorized in the postnatal diagnosis group.
A retrospective chart review was performed. Patient characteristics and surgical outcomes of the patients who received a prenatal diagnosis of EA and/or TEF and treated at the Texas Children’s Fetal Center were compared with those who were diagnosed with EA and/or TEF after birth. The primary outcome was survival. Secondary outcomes included the type of anomaly, the incidence of other congenital abnormalities, surgical complication rates (including anastomotic leak, recurrent fistula, pneumothorax requiring intervention, infection, and reoperation), hospital length of stay, time to full feeds, time of mechanical ventilation, and duration of oxygen requirement. Neonates with EA and/or TEF are treated in a standardized fashion in our institution with early surgical repair once stabilized. Those with EA and no TEF usually will undergo gastrostomy with delayed primary esophageal repair at 2e4 mo based on radiologic demonstration of less than one vertebral body distance between the proximal esophagus and distal esophagus. Statistical analysis included chi-square for categorical variables and Student t-test for continuous variables.
3.
Results
3.1.
Patient characteristics
We identified 15 patients with a confirmed prenatal diagnosis of EA and 76 with a postnatal diagnosis. All except one (93%) prenatally diagnosed patient were born at our institution compared with 26% of infants postnatally diagnosed. The rates of polyhydramnios were higher in the prenatally diagnosed group (Table 1). Patients with pure EA had a higher rate of polyhydramnios (100%) compared to those with a fistula (45%). Pregnancies were otherwise reported as being uncomplicated, except for a 12% rate of gestational diabetes mellitus in the postnatally diagnosed patients. Patients in each group were born at similar mean gestational ages, with similar mean birth weights (Table 1). There was a significantly higher incidence of pure EA in the prenatally diagnosed group (47% versus 11%; P ¼ 0.003) and conversely a lower incidence of EA with
Table 1 e Patient characteristics and demographics.
Male Prenatal care Polyhydramnios GA at birth (wk) Preterm birth (GA < 38 wk) Birth weight (g) Pure EA (type A) EA with TEF (types B, C, and D) Other anomalies VACTERL association Major cardiac anomalies GA ¼ gestational age.
Prenatal diagnosis, n ¼ 15 (%)
Postnatal diagnosis, n ¼ 76 (%)
P value
8 (53) 15 (100) 14/15 (93) 36.5 2.3 9 (60)
36 (47) 69 (89) 38/69 (55) 35.8 3.0 50 (66)
0.78 0.59 0.007 0.35 0.77
2390 474 7 (47) 8 (53)
2347 787 8 (11) 66 (88)
0.84 0.003 0.003
12 (80) 4 (27) 10 (67)
54 (82) 26 (34) 46 (61)
1.0 0.77 0.78
3
j o u r n a l o f s u r g i c a l r e s e a r c h x x x ( 2 0 1 4 ) 1 e4
TEF (53% versus 88%; P ¼ 0.003). The presence of associated anomalies was similar between groups (80% versus 82%; P ¼ 1.0) as was the incidence of the VACTERL association (27% versus 34%; P ¼ 0.78) and major cardiac anomalies (defined as those which required either medical or surgical treatment; 67% versus 61%; P ¼ 0.78).
3.2.
Patient outcomes
Overall, survival was 80% in infants diagnosed prenatally and 88% for those diagnosed postnatally at a median length of follow-up of 7.9 y (range, 0.47e10.2 y). Survival in patients with isolated EA and/or TEF was 100% in both groups (Table 2). The time to death was similar in each group (median day, 24 [range, 1e287] versus 24 [range, 0e450]; P ¼ 0.8). Eight patients were unable to undergo repair of their EA and/or TEF because of the finding of nonsurvivable genetic or cardiac conditions. This did not differ between groups with two patients in the prenatal group and six patients in the postnatal group expiring before the repair could be completed (13% versus 8%; P ¼ 0.6). There was no statistically significant difference in postsurgical complication rates (46% versus 31%; P ¼ 0.34); specifically, anastomotic leak and recurrence rates were similar between groups. Although the difference was not statistically significant, the median length of stay in the prenatal diagnosis group was nearly twice that of the postnatally diagnosed group (92 versus 52 d; P ¼ 0.2). Regarding long-term outcomes and sequelae from the underlying disease process, stricture rates requiring dilation procedures were similar (50% versus 41%; P ¼ 0.75). There was no statistically significant difference in fundoplication rates, although the prenatal rate was double the postnatal group’s rate (25% versus 12%; P ¼ 0.36). Time to initial oral feeding was higher in the prenatally diagnosed group (median, 144 versus 30 d; P ¼ 0.009), which was mostly because of the higher rate of pure EA in that group. Similarly, the rate of gastrostomy tube insertion was higher in the prenatal group (73% versus 49%; P ¼ 0.09), a reflection of the standard therapy for this anomaly. The rates of tracheomalacia (33% versus 29%; P ¼ 0.7) and tracheostomy (13% versus 11%; P ¼ 0.7) were unchanged in the prenatal versus postnatal groups, respectively.
4.
Discussion
The management of EA and TEF has progressed to the point that survival in infants with isolated disease is >90%, and in some reports, it approaches 100%, as seen in the current patient series [7]. However, infants with EA continue to experience significant morbidity, such as postsurgical complications (leak, stricture, and recurrence), prolonged need for supplemental feeding, respiratory difficulties, and failure to thrive. Our data suggest that prenatal diagnosis does not reduce the incidence of morbidity in patients with EA and/or TEF. However, prenatal diagnosis does facilitate optimal obstetric management. Our study found a much higher rate of polyhydramnios in the prenatally diagnosed group, which could predispose these mothers to preterm labor, thus exposing the infant to morbidity related to prematurity. Prenatal management allowed for this higher risk group to be born at a similar gestational age and birth weight with an equivalent prematurity rate, which may have had a positive effect on outcomes that our sample size was too small to detect. The difference in our prenatal and postnatal groups with respect to the rate of pure EA (47% versus 11%) is attributable to the presence of polyhydramnios. Fetuses with EA and no TEF will have polyhydramnios (100% in our series) and an absent stomach bubble in nearly all cases, which will increase the likelihood of prenatal detection and fetal center referral, as demonstrated in the present series. In some instances, inhaled amniotic fluid in a fetus with a TEF will pass through the trachea and fistula tract and enter the stomach, which limits the progression to polyhydramnios (45% in our series). Although the primary outcome of survival was similar between the two comparison groups, there were some differences in outcomes, particularly related to hospital stay and time to first oral feeds that favored the postnatally diagnosed group. Such differences can likely be attributed to the higher rate of pure EA without fistula in the prenatal group, as these patients have a planned delay before repair is attempted. Surgical complication rates were similar in the two groups, in contrast to other studies, which have shown a higher leak and stricture rate in patients with pure EA [8].
Table 2 e Postnatal and postoperative outcomes. Prenatal diagnosis, n ¼ 15 (%) Survival, all patients Survival, no other anomalies Surgical complications Anastomotic leak Recurrent fistula Postoperative pneumothorax Esophageal dilation Fundoplication Time to initial oral feeds (d) Median length of hospital stay (d) Median follow-up (y) * y
12 3 6/13 5 1 2 6/12 3/12 144 92 2.05
(80) (100) (46)* (33) (8) (15) (50)y (25) (3e857) (12e232) (0.46e7.66)
Excludes patients who died before surgery and those with missing postoperative data. Excludes patients who died.
Postnatal diagnosis, n ¼ 76 (%) 66 22 19/61 15 3 4 27/66 8/66 30 52 7.41
(88) (100) (31)* (25) (5) (7) (41)y (12) (7e760) (7e289) (0.79e10.21)
P value 0.44 1.0 0.34 0.31 0.52 0.26 0.75 0.36 0.009 0.20 0.002
4
j o u r n a l o f s u r g i c a l r e s e a r c h x x x ( 2 0 1 4 ) 1 e4
The limited literature that discusses possible benefits to prenatal diagnosis for EA reports conflicting results. de Jong et al. found decreased survival (70% versus 88%; P ¼ 0.05) and an increased rate of associated anomalies (80% versus 59%; P ¼ 0.04) in the prenatal versus postnatal groups, respectively [5]. In contrast, Brantberg et al., who described outcomes for 38 infants with EA and additional anomalies, showed improved survival (67%) with prenatal diagnosis of EA compared with postnatal diagnosis (27%). In a study by Brantberg et al., overall survival was 90% for 10 infants with isolated EA but only 45% for those with EA and additional anomalies [6]. As this study does not report the specific associated anomalies found in each group, it is difficult to tell if the difference in outcomes was related to the underlying patient group characteristics, as hypothesized in the de Jong et al. study. In our present series, overall survival for those with EA and additional anomalies was 80%. As the outcomes in the current literature come to discrepant conclusions, and the ability of the fetal center to accurately diagnose the condition prenatally has been improving, further multicenter prospective study is needed to convincingly address this question. Limitations to the present study include a small sample size in the prenatally diagnosed group. There may be differences in outcomes that were too small to detect with our smaller patient population, particularly with respect to survival. Likewise, as this is a specific fetal center population for our comparison group, the generalizability of findings from this study is likely limited to tertiary referral centers. Further study including data from multiple centers would allow for more definitive conclusions. As the prenatally diagnosed group represents a slightly more contemporary group of patients than the postnatally diagnosed patients, the amount of observed follow-up time for the prenatal group is shorter and so there may be a detection bias for some of the long-term complication rates reported. In conclusion, these data can be used to improve counseling during prenatal consultation for patients with suspected EA. Despite previous reports that a prenatal diagnosis portends decreased survival and an increased rate of other congenital anomalies, our data show overall survival is the same, and patients do not appear to have a higher rate of other congenital anomalies. Although prenatally diagnosed patients have a higher rate of pure EA, the rate of concurrent anomalies is also equivalent.
Acknowledgment S.F. and D.C. contributed to the conception and design, data collection and analysis, and manuscript drafting and revision; C.E. to the conception and design, data collection and analysis, and manuscript drafting; O.O. to the conception and design and manuscript drafting and revision; M. B., S.W., and R.R. to the concept and design and manuscript revision; T.L. to the conception and design, data collection and analysis, and manuscript revision.
Disclosure The authors have no sources of support or disclosures to report with respect to the preparation of this manuscript.
references
[1] Houfflin-Debarge V, Bigot J. Ultrasound and MRI prenatal diagnosis of esophageal atresia: effect on management. J pediatr gastroenterol nutr 2011;52(Suppl 1):S9. [2] Ethun CG, Fallon SC, Cassady CI, et al. Fetal MRI improves diagnostic accuracy in patients referred to a fetal center for suspected esophageal atresia. J Pediatr Surg; 2014. In press. [3] Seo J, Kim do Y, Kim AR, et al. An 18-year experience of tracheoesophageal fistula and esophageal atresia. Korean j pediatr 2010;53:705. [4] Pedersen RN, Calzolari E, Husby S, Garne E. Oesophageal atresia: prevalence, prenatal diagnosis and associated anomalies in 23 European regions. Arch dis child 2012;97:227. [5] de Jong EM, de Haan MA, Gischler SJ, et al. Pre- and postnatal diagnosis and outcome of fetuses and neonates with esophageal atresia and tracheoesophageal fistula. Prenatal diagn 2010;30:274. [6] Brantberg A, Blaas HG, Haugen SE, Eik-Nes SH. Esophageal obstruction-prenatal detection rate and outcome. Ultrasound obstetrics gynecol: official j Int Soc Ultrasound Obstetrics Gynecol 2007;30:180. [7] Kunisaki SM, Foker JE. Surgical advances in the fetus and neonate: esophageal atresia. Clin perinatology 2012;39:349. [8] Engum SA, Grosfeld JL, West KW, et al. Analysis of morbidity and mortality in 227 cases of esophageal atresia and/or tracheoesophageal fistula over two decades. Arch Surg 1995; 130:502.
