Abstract
Purpose
Congenital esophageal atresia requires neonatal surgery. Although short- and medium-term outcomes of thoracoscopic and open surgery have been investigated, long-term outcomes remain unclear. This study aimed to compare the long-term outcomes of these approaches using a Japanese national inpatient database.
Methods
We identified neonates who underwent open or thoracoscopic surgery for congenital esophageal atresia between April 2016 and March 2022. Patients with prior palliative surgery were excluded. Propensity score overlap weighting analyses were used to compare the outcomes between the groups. The primary outcome was the long-term (1 year after definitive surgery) anastomotic strictures. Secondary outcomes included long- and medium-term (within 30 days to 1 year after definitive surgery) gastroesophageal reflux and medium-term anastomotic strictures.
Results
Among 395 patients, 67 underwent thoracoscopic surgery and 328 underwent open surgery. Propensity score overlap weighting analyses revealed no significant differences in long-term anastomotic stricture (5.8% vs. 8.7%; risk difference (RD), − 2.9%; 95% confidence interval (CI), − 10.9 to 5.1), long-term gastroesophageal reflux (2.9% vs. 3.0%; RD, -0.1%; 95% CI − 7.1 to 4.9), medium-term anastomotic stricture (29.4% vs. 18.8%; RD, 10.6%; 95% CI − 2.5 to 23.7), or medium-term gastroesophageal reflux (4.9% vs. 6.0%; RD, − 1.1; 95% CI −7 .1 to 4.9).
Conclusions
Long-term outcomes did not differ significantly between thoracoscopic and open surgery for congenital esophageal atresia.
Keywords: Congenital esophageal atresia, Thoracoscopic surgery, Open surgery, Neonatal surgery, Long-term outcome
Introduction
Congenital esophageal atresia is reported to occur in approximately 1 in 2500–4000 live births [1] and requires surgical intervention during the neonatal period. Since the first report of thoracoscopic surgery for congenital esophageal atresia in 2000 [2], the number of institutions that perform thoracoscopic surgery for this condition has gradually increased.
Several observational studies and meta-analyses which compared surgical outcomes between thoracoscopic and open surgery for esophageal atresia have reported a similar proportion of postoperative complications, such as anastomotic leakage, anastomotic stricture, and gastroesophageal reflux requiring fundoplication [3–7]. However, these studies evaluated the surgical outcomes without adjusting for patient background.
A previous study using propensity score overlap weighting analysis revealed no significant difference in complications between thoracoscopic and open surgery for esophageal atresia. However, this study did not evaluate long-term complications [8]. Similarly, another study that used propensity score-matching analysis evaluated only short-term outcomes (within 30 days after definitive surgery) [9]. Thus, long-term outcomes adjusted for patient backgrounds remain uninvestigated.
We aimed to compare the long-term surgical outcomes of thoracoscopic and open surgery for congenital esophageal atresia while accounting for differences in patient background.
Materials and methods
Data source
This nationwide retrospective cohort study used data from the Diagnosis Procedure Combination database of Japan. More than 1200 hospitals voluntarily contributed to the database, which includes data on approximately 8,000,000 inpatients annually. All 82 academic hospitals in Japan participated in the database [10]. Approximately 90% of hospitals with neonatal intensive care units were included in this database [10, 11]. The database includes the following data: unique hospital identifiers, patient age, body weight, body height, sex, diagnoses, comorbidities at admission, and complications after admission recorded as text data in Japanese and the International Classification of Diseases, Tenth Revision (ICD-10) codes [12]. Previous studies have shown that the validity of diagnoses, procedural records, and operative information in the databases is high [13, 14].
This study was conducted in accordance with the STROCSS criteria [15]. The requirement for informed consent was waived because the patient data were anonymized. This study was approved by our institutional review board (Approval number: 3501-(5), May 19, 2021).
Patient selection
In Japan, thoracoscopic surgery for congenital esophageal atresia was covered by public health insurance in April 2016. We identified patients who underwent definitive surgery (thoracoscopic or open surgery) for congenital esophageal atresia during their initial hospitalization between April 2016 and March 2022. We excluded patients who (i) underwent palliative surgery (gastrostomy, exploratory thoracotomy, or laparotomy) prior to definitive surgery, and (ii) underwent definitive surgery after 1 week of age without palliative surgeries. These patients were excluded because they were considered to have an atypical clinical course or more complicated conditions, which could introduce bias in comparing surgical outcomes. Additionally, only a small number of patients aged 7–14 days underwent definitive surgery without prior palliative procedure. Including them would have had minimal impact on group sizes (only five additional patients in the open surgery group and none in the thoracoscopic group).We categorized eligible patients into thoracoscopic and open surgery groups. We also recorded all rehospitalizations for each eligible patient for up to 6 years after initial discharge.
Outcomes and covariates
The primary outcome was long-term anastomotic stricture. The secondary outcomes were long-term gastroesophageal reflux, medium-term anastomotic stricture, and gastroesophageal reflux, anastomotic leakage, short-term anastomotic stricture, duration of anesthesia, reoperation, in-hospital mortality, duration of postoperative mechanical ventilation, length of postoperative neonatal intensive care unit (NICU) stay, length of postoperative hospital stay, and total cost of hospitalization. We defined anastomotic stricture as the need for operative dilation using any type of dilating method (including balloon dilation and bougienage) after definitive surgery, based on a previous study [16]. We defined gastroesophageal reflux as the requirement for fundoplication after definitive surgery. “Short-term” was defined as requiring initial dilation within 30 days after definitive surgery, as in a previous study [10]. “Medium-term” and “long-term” were defined as requiring an initial procedure (dilation or fundoplication) within 30 days to 1 year and 1 year or later after definitive surgery, respectively. Anastomotic strictures and gastroesophageal reflux are likely to occur within the first year after definitive surgery for congenital esophageal atresia [17, 18]. We defined anastomotic leakage as both a recorded diagnosis of anastomotic leak after admission and the need for continuous suction drainage for 3 weeks after definitive surgery based on a previous study [16]. We defined the currency exchange rate as 140 Japanese yen per 1 US dollar.
The covariates included sex, age, birth weight, gestational age (in week), and age at surgery (in days), congenital heart disease (ICD-10 codes: Q20–26), congenital malformations of the respiratory system (ICD-10 codes: Q30–34), congenital malformations of the digestive system (ICD-10 codes: Q38–45), congenital malformation and deformation of the musculoskeletal system disease (ICD-10 codes: Q65–79), trisomy 21 (ICD-10 codes: Q90), trisomy 18 (ICD-10 codes: Q91.0–3), and trisomy 13 (ICD-10 codes: Q91.4–7), birth asphyxia, use of vasopressor (at the day of surgery), and transfusion (on the day of surgery), hospital volume, and fiscal year.
Body weight was categorized into the following three groups based on the World Health Organization definition of low birth weight [19]: < 1500 g, 1500–2500 g, and > 2500 g. Gestational age was categorized into two groups: < 37 weeks and ≥ 37 weeks. Hospital volume was defined as the average annual number of definitive operations for esophageal atresia performed at each hospital over 6 years and was categorized into tertile: 0 > case and < 1 case, ≥ 1 case and < 2 cases, and ≥ 2 cases. Patent ductus arteriosus was defined based on ICD-10 codes and operations for congenital heart disease [20]. We defined mild birth asphyxia as neonatal resuscitation for first-degree birth asphyxia (Apgar scores of 4–6) and severe birth asphyxia as neonatal resuscitation for second-degree birth asphyxia (Apgar scores of 0–3) [21].
Statistical analysis
We used Fisher’s exact test and the chi-squared test to compare the proportions of categorical variables, and the Mann–Whitney U test to compare the medians of continuous variables.
To compare the perioperative outcomes between the groups, we conducted a propensity score overlap weighting analysis [22, 23]. First, we calculated the propensity scores using a logistic regression model with patients undergoing thoracoscopic surgery as the dependent variable. The independent variables included sex, age at surgery, gestational age, birth weight, congenital heart disease, trisomy 18, vasopressor use, blood transfusion use, mild and severe birth asphyxia, congenital malformation (respiratory, digestive, and musculoskeletal systems), hospital volume, and fiscal year. Patients in the thoracoscopic surgery group were weighted according to the probability of not undergoing thoracoscopic surgery (1-propensity score), and those in the open surgery group were weighted according to the probability of undergoing thoracoscopic surgery (propensity score). Propensity score overlap weighting analysis attempts to mimic the important attributes of randomized clinical trials, such as the target population, covariate balance, and precision, which are potential limitations of conventional propensity score methods of inverse probability of treatment weighting and matching [24]. We calculated standardized differences to examine the balance of baseline characteristics between the groups before and after propensity score overlap weighting. A standardized difference of less than 0.1 was considered a negligible imbalance between the two groups [25].
We considered a significant level as P < 0.05 for all statistical tests, with all reported P values being two-sided. Statistical analyses were conducted using Stata/SE 18.0 (Stata Corp., College Station, TX, USA).
Results
We identified 625 patients who underwent definitive surgery for congenital esophageal atresia during their initial hospitalization between April 2016 and March 2022. We excluded 171 patients who underwent palliative surgery before definitive surgery and 59 patients who underwent definitive surgery after 1 week of age. Of the eligible 395 patients, 67 underwent thoracoscopic surgery and 328 underwent open surgery (Fig. 1).
Fig. 1.
Flow diagram. Flow diagram for patients who underwent thoracoscopic or open surgery for esophageal atresia
The baseline characteristics of the patients before and after propensity score overlap weighting are presented in Table 1. Before overlap weighing, patients in the thoracoscopic surgery group had higher birth weight, underwent surgery at a higher-volume hospital, and were less likely to have congenital heart disease compared to those in the open surgery group.
Table 1.
Baseline characteristics of all patients and propensity score-weighted patients
| All patients | Propensity-score overlap weighted patients | |||||||
|---|---|---|---|---|---|---|---|---|
| Thoracoscopic surgery | Open surgery |
Standardized difference | Thoracoscopic surgery | Open surgery | Standardized difference | |||
| n = 67 | n = 328 | |||||||
| Sex, male, n(%) | 36 | (53.7) | 174 | (53.0) | 0.01 | (52.6) | (52.6) | < 0.01 |
| Age at surgery, n(%) | ||||||||
| 0 day | 7 | (10.4) | 82 | (25.0) | 0.51 | (13.3) | (13.3) | < 0.01 |
| 1 day | 35 | (52.2) | 140 | (42.7) | (57.8) | (57.8) | ||
| 2 days | 15 | (22.4) | 61 | (18.6) | (15.6) | (15.6) | ||
| 3 days | 6 | (9.0) | 22 | (6.7) | (9.2) | (9.2) | ||
| 4 days | 1 | (1.5) | 11 | (3.4) | (2.3) | (2.3) | ||
| 5 days | 1 | (1.5) | 6 | (1.8) | (1.7) | (1.7) | ||
| 6 days | 0 | 0.0 | 5 | (1.5) | 0.0 | 0.0 | ||
| 7 days | 2 | (3.0) | 1 | (0.3) | 0.0 | 0.0 | ||
|
Preterm gestational age (< 37 weeks), n(%) |
17 | (25.4) | 76 | (23.2) | 0.05 | (27.5) | (27.5) | < 0.01 |
| Birth weight, n(%) | ||||||||
| < 1500 | 0 | 0.0 | 6 | (1.8) | 0.29 | 0.0 | 0.0 | < 0.01 |
| 1500–2500 | 26 | (38.8) | 147 | (44.8) | (44.4) | (44.4) | ||
| > 2500 | 37 | (55.2) | 155 | (47.3) | (55.6) | (55.6) | ||
| missing | 4 | (6.0) | 20 | (6.1) | 0.0 | 0.0 | ||
| Birth asphyxia, n(%) | 4 | (6.0) | 33 | (10.1) | 0.15 | (7.0) | (7.0) | < 0.01 |
| Severe birth asphyxia, n(%) | 5 | (7.5) | 21 | (6.4) | 0.04 | (8.1) | (8.1) | < 0.01 |
| Congenital malformation, n(%) | ||||||||
| Heart disease | 5 | (7.5) | 61 | (18.6) | 0.34 | (9.9) | (9.9) | < 0.01 |
| Digestive system | 12 | (17.9) | 67 | (20.4) | 0.06 | (19.8) | (19.8) | < 0.01 |
| Respiratory system | 3 | (4.5) | 24 | (7.3) | 0.12 | (5.8) | (5.8) | < 0.01 |
| Musculoskeletal system | 6 | (9.0) | 23 | (7.0) | 0.07 | (8.8) | (8.8) | < 0.01 |
| Trisomy 21 | 0 | 0.0 | 0 | 0.0 | 0.00 | 0.0 | 0.0 | < 0.01 |
| Trisomy 18 | 0 | 0.0 | 7 | (2.1) | 0.21 | 0.0 | 0.0 | < 0.01 |
| Trisomy 13 | 0 | 0.0 | 0 | 0.0 | 0.00 | 0.0 | 0.0 | < 0.01 |
| Vasopressor use, n(%) | 15 | (22.4) | 78 | (23.8) | 0.03 | (25.0) | (25.0) | < 0.01 |
| Transfusion, n(%) | 3 | (4.5) | 26 | (7.9) | 0.14 | (5.8) | (5.8) | < 0.01 |
|
Annual hospital volume for congenital esophageal atresia, n(%) | ||||||||
| 0 > , < 1 case | 6 | (10.2) | 111 | (36.2) | 0.57 | (15.8) | (15.8) | < 0.01 |
| ≥ 1, < 2 cases | 23 | (39.0) | 116 | (37.8) | (33.3) | (33.3) | ||
| ≥ 2 cases | 30 | (50.8) | 80 | (26.1) | (50.9) | (50.9) | ||
| Fiscal year, n(%) | ||||||||
| 2016 | 11 | (16.4) | 52 | (15.9) | 0.44 | (15.8) | (15.8) | < 0.01 |
| 2017 | 8 | (11.9) | 81 | (24.7) | (16.4) | (16.4) | ||
| 2018 | 10 | (14.9) | 68 | (20.7) | (15.8) | (15.8) | ||
| 2019 | 12 | (17.9) | 47 | (14.3) | (15.8) | (15.8) | ||
| 2020 | 13 | (19.4) | 38 | (11.6) | (18.7) | (18.7) | ||
| 2021 | 13 | (19.4) | 42 | (12.8) | (17.5) | (17.5) | ||
Data are presented as mean (SD) for continuous variables and n (%) for categorical variables
The crude outcomes for all patients are shown in Table 2. No significant difference was observed in the proportions of long-term anastomotic stricture (4.5% vs. 8.5%, P = 0.26), long-term gastroesophageal reflux (3.0% vs 2.4%, P = 0.80), medium-term anastomotic stricture (29.9% vs 23.2%, P = 0.25), medium-term gastroesophageal reflux (7.5% vs.4.6%, P = 0.33), the average duration of anesthesia (249.5 (95% confidential interval (CI) 219.4–279.6) min vs. 277.8 (265.4–290.2) min, P = 0.07), anastomotic leakage (7.5% vs. 6.7%, P = 0.82), short-term anastomotic stricture (3.0% vs. 3.1%, P = 0.98), reoperation (4.5% vs. 2.7%, P = 0.45), in-hospital mortality (1.5% vs. 3.7%, P = 0.37), the average duration of postoperative mechanical ventilation (54.5 (28.4–80.5) days vs. 44.4 (29.8–58.9) days, P = 0.56), the average length of postoperative NICU stay (26.5 (22.2–30.7) days vs. 24.4 (23.3–25.5) days, P = 0.18), the average length of postoperative hospital stay (83.6 (59.1–108.1) days vs. 86.8 (70.6–102.9) days, P = 0.87), and the average total cost of hospitalization (70,053 (60,163–79,943) US dollars vs. 67,831 (61,553–74,110) US dollars, P = 0.76). No short-term gastroesophageal reflux was observed in either the thoracoscopic or open surgery groups.
Table 2.
Crude outcomes of all patients
| Outcomes | Thoracoscopic surgery | Open surgery | P-value | ||
|---|---|---|---|---|---|
| n = 67 | n = 328 | ||||
| Long-term | |||||
| AS, n(%) | 3 | (4.5) | 28 | (8.5) | 0.26 |
| GER, n(%) | 2 | (3.0) | 8 | (2.4) | 0.80 |
| Middle-term | |||||
| AS, n(%) | 20 | (29.9) | 76 | (23.2) | 0.25 |
| GER, n(%) | 5 | (7.5) | 15 | (4.6) | 0.33 |
| Perioperative | |||||
| Duration of anesthesia, min(SD) | 249.5 | (123.3) | 277.8 | (114.4) | 0.07 |
| Anastomotic leakage, n(%) | 5 | (7.5) | 22 | (6.7) | 0.82 |
| Short-term AS, n(%) | 2 | (3.0) | 10 | (3.1) | 0.98 |
| Reoperation, n(%) | 3 | (4.5) | 9 | (2.7) | 0.45 |
| In-hospital mortality, n(%) | 1 | (1.5) | 12 | (3.7) | 0.37 |
| Postoperative ventilatory duration, days(SD) | 54.5 | (105.9) | 44.4 | (133.1) | 0.56 |
| Postoperative NICU stay, days (SD) | 26.5 | (17.3) | 24.4 | (10.0) | 0.18 |
| Postoperative hospital stay, days (SD) | 83.6 | (100.5) | 86.8 | (148.5) | 0.87 |
| Total cost of hospitalization, USD (SD) | 70,053 | (40,547) | 67,831 | (57,798) | 0.76 |
Data are presented as mean (SD) for continuous measures, and n (%) for categorical measures
AS anastomotic stricture, GER gastroesophageal reflux, NICU neonatal intensive care unit, USD US dollars, SD standard deviation
In the overlap weight analysis, all baseline characteristics were well-balanced (Table 1). No significant difference was observed between the groups in long-term anastomotic stricture (5.8% vs. 8.7%; risk difference, (RD) −2.9%; 95% CI −10.9 to 5.1), long-term gastroesophageal reflux (RD, − 0.1%; 95% CI − 5.3 to 5.1), medium-term anastomotic stricture (RD, 10.6%, 95% CI − 2.5 to 23.7), medium-term gastroesophageal reflux (RD, − 1.1, 95% CI − 7.1 to 4.9), mean duration of anesthesia (difference, -19.9 min; 95% CI − 55.7 to 16.0), anastomotic leakage (RD, 0.6%; 95% CI − 8.0 to 9.1), short-term anastomotic stricture (RD, −0.4, 95% CI −6.0 to 5.2), reoperation (RD, 3.8%; 95% CI −2.0 to 9.6), in-hospital mortality (RD, 1.2%; 95% CI − 1.7 to 4.1), mean duration of postoperative mechanical ventilation (difference, 14.7 days; 95% CI − 20.8 to 50.3), mean length of postoperative NICU stay (difference, 3.5 days; 95% CI −0.7 to 7.7), mean length of postoperative hospital stay (difference, − 2.6 days, 95% CI − 38.3 to 33.2), or mean total cost of hospitalization (difference, 3495 US dollars; 95% CI − 10,613 to 17,603) (Table 3).
Table 3.
Outcome analysis after propensity score overlap weighting
| Outcomes | Thoracoscopic surgery | Open surgery | Risk difference | (95% CI) | P value | ||
|---|---|---|---|---|---|---|---|
| Long-term | |||||||
| AS, % | 5.8 | 8.7 | − 2.9 | (− 10.9–5.1) | 0.48 | ||
| GER, % | 2.9 | 3.0 | − 0.1 | (− 5.3–5.1) | 0.97 | ||
| Middle-term | |||||||
| AS, % | 29.4 | 18.8 | 10.6 | (− 2.5–23.7) | 0.11 | ||
| GER, % | 4.9 | 6.0 | − 1.1 | (− 7.1–4.9) | 0.71 | ||
| Perioperative | |||||||
| Duration of anesthesia, min(SD) | 254.9 | (120.1) | 274.7 | (107.9) | − 19.9 | (− 55.7–16.0) | 0.28 |
| Anastomotic leakage, % | 8.2 | 7.7 | 0.6 | (− 8.0–9.1) | 0.90 | ||
| Short-term AS, % | 2.9 | 3.3 | − 0.4 | (− 6.0–5.2) | 0.88 | ||
| Reoperation, % | 5.0 | 1.2 | 3.8 | (− 2.0–9.6) | 0.20 | ||
| In-hospital mortality, % | 1.5 | 0.3 | 1.2 | (− 1.7–4.1) | 0.43 | ||
| Postoperative ventilatory duration, days (SD) | 58.3 | (108.8) | 43.5 | (153.5) | 14.7 | (− 20.8–50.3) | 0.42 |
| Postoperative NICU stay, days(SD) | 26.7 | (16.4) | 23.1 | (9.2) | 3.5 | (− 0.7–7.7) | 0.10 |
| Postoperative hospital stay, days(SD) | 86.4 | (103.0) | 89.0 | (170.8) | − 2.6 | (− 38.3–33.2) | 0.89 |
| Total cost of hospitalization, USD (SD) | 70,955 | (41,224) | 67,460 | (65,459) | 3495 | (− 10,613–17,603) | 0.63 |
Data are presented as mean (SD) for continuous measures and % for categorical measures
AS anastomotic stricture, GER gastroesophageal reflux, NICU neonatal intensive care unit, USD US dollars, SD standard deviation
Discussion
In this study, we compared the long-term outcomes of thoracoscopic and open surgery for congenital esophageal atresia using a national inpatient database in Japan. Propensity score overlap weighting analyses revealed no significant differences in long-term outcomes between the two groups. This is the first report to compare the proportions of long-term anastomotic strictures and gastroesophageal reflux between thoracoscopic and open surgery, with adjustment for patient characteristics.
The current studies found no significant difference in the proportion of short- and medium-term outcomes, such as anastomotic leakage and anastomotic stricture within 1 year after definitive surgery. These findings are consistent with previous research [3–8]. The exclusion of patients who underwent palliative procedures before definitive surgery may have contributed to these findings by limiting the inclusion of cases with a long-gap esophageal atresia, which is a known risk factor for anastomotic leakage and strictures [26, 27].
Our findings revealed no significant differences in long-term anastomotic strictures or gastroesophageal reflux between groups. One possible reason for this is that we excluded patients who underwent palliative procedures prior to definitive surgery owing to a long gap. A long gap is strongly associated with anastomotic stricture and gastroesophageal reflux. Patients requiring staged repair are more likely to develop these complications because a long gap increases the anastomotic tension associated with an anastomotic leak and subsequent anastomotic strictures [17, 28]. Higher tension at the anastomotic site can retract the lower esophagus upward and induce anatomical changes at the gastroesophageal junction [29]. Our findings revealed no significant difference in the proportion of anastomotic strictures or gastroesophageal reflux between groups. These findings indicate that thoracoscopic and open surgeries could have similar impacts on the tension at the anastomotic site related to anastomotic stricture, and anatomical and functional changes at the gastroesophageal junction associated with gastroesophageal reflux.
Although no significant differences in long-term anastomotic stricture and gastroesophageal reflux were observed between thoracoscopic and open surgery for esophageal atresia in this study, a comparison of outcomes between thoracoscopic and open surgery for esophageal atresia may be premature. Thoracoscopic surgery for esophageal atresia has not been widely performed, and the expertise in thoracoscopic surgery may be insufficient due to the rarity of esophageal atresia. Thoracoscopic surgery for esophageal atresia was included under public health insurance in Japan in 2016, and our study period spanned from 2016 to 2022. A previous study reported a steep learning curve for thoracoscopic surgery (approximately 10–20 cases) for esophageal atresia [30]. Moreover, the level of expertise of surgeon performing thoracoscopic surgery may differ from those performing open surgery. Most surgeons who perform thoracoscopic surgery are certified expert surgeons, whereas those who perform open surgery are likely to be surgical trainees [31]. In the future, as thoracoscopic surgery for congenital esophageal atresia is more widely adopted, further studies adjusting for both patient backgrounds and surgeon levels will be required.
However, our study had certain limitations. First, we could not detect the type of congenital esophageal atresia (Gross classification) or gap length in the database. To align patient backgrounds, we excluded patients who underwent palliative surgery (such as gastrostomy) before definitive surgery and those who underwent definitive surgery after 1 week of age. Second, we were unable to follow up with patients across different hospitals in this database. Therefore, we may have underestimated the proportion of patients with anastomotic strictures and gastroesophageal reflux. Third, the Diagnosis Procedure Combination database did not include the results of tests for anastomotic stricture and gastroesophageal reflux, such as esophagogastrography, endoscopy, and pH-monitoring tests. Thus, we were unable to evaluate the severity of anastomotic stricture and gastroesophageal reflux. In addition, we could not define the anastomotic leakage owing to the lack of data on esophagograms and patients’ symptoms. Fourth, we could not detect conversions from thoracoscopic to open surgery. The proportion of conversions varied among previous studies (3.2–53%) [3–7]. Therefore, we may have underestimated the effects of thoracoscopic surgery. Fifth, the database did not provide information on other important outcomes, such as thoracic deformity, including scoliosis, which has been reported as a long-term complication of open thoracotomy [32, 33].
Conclusion
No significant differences were observed in the long-term outcomes between thoracoscopic surgery and open surgery for congenital esophageal atresia. Using a national inpatient database in Japan and applying propensity score overlap weighting analyses, we provided the first adjusted comparison of these long-term outcomes. Our findings suggest that both surgical approaches may have similar impacts on anastomotic tension and anatomical changes at the gastroesophageal junction.
Author contributions
Conceptualization: Yoshitsugu Yanagida; Formal analysis and investigation: Yoshitsugu Yanagida; Writing—original draft preparation: Yoshitsugu Yanagida; Writing—review and editing; Shotaro Aso, Michimasa Fujiogi, Kaori Morita, Mai Kutsukake, Naohiro Takamoto, Kiyohide Fushimi, Jun Fujishiro, Hideo Yasunaga; Funding acquisition: Kiyohide Fushimi, Hideo Yasunaga; Resources: Kiyohide Fushimi, Hideo Yasunaga; Supervision: Shotaro Aso.
Funding
Open Access funding provided by The University of Tokyo. This work was supported by grants from the Ministry of Health, Labour and Welfare, Japan (23AA2003 and 24AA2006).
Data availability
No datasets were generated or analysed during the current study.
Declarations
Conflict of interest
The authors declare no competing interests.
Ethical approval
This study was conducted in accordance with the STROCSS criteria. This study was approved by our institutional review board (approval number: 3501-(5), May 19, 2021).
Consent to participate
The requirement for informed consent was waived because the patient data were anonymized.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Mortell AE, Azizkhan RG (2009) Esophageal atresia repair with thoracotomy: the Cincinnati contemporary experience. Semin Pediatr Surg 18:12–19. 10.1053/j.sempedsurg.2008.10.003 [DOI] [PubMed] [Google Scholar]
- 2.Rothenberg SS (2002) Thoracoscopic repair of tracheoesophageal fistula in newborns. J Pediatr Surg 37:869–872. 10.1053/jpsu.2002.32891 [DOI] [PubMed] [Google Scholar]
- 3.Elhattab A, Ali L, Rousseau V, Clermidi P, Michelet D, Farnoux C et al (2020) Thoracoscopy versus thoracotomy in the repair of esophageal atresia with distal tracheoesophageal fistula. J Laparoendosc Adv Surg Tech A 30:1289–1294. 10.1089/lap.2020.0642 [DOI] [PubMed] [Google Scholar]
- 4.Etchill EW, Giuliano KA, Boss EF, Rhee DS, Kunisaki SM (2021) Association of operative approach with outcomes in neonates with esophageal atresia and tracheoesophageal fistula. J Pediatr Surg 56:2172–2179. 10.1016/j.jpedsurg.2021.04.006 [DOI] [PubMed] [Google Scholar]
- 5.Yang S, Wang P, Yang Z, Li S, Liao J, Hua K et al (2021) Clinical comparison between thoracoscopic and thoracotomy repair of gross type C esophageal atresia. BMC Surg 21:403. 10.1186/s12893-021-01360-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Zou C, Dong J, Xu G, Xia R, Xiao Y, Li M et al (2023) Thoracoscopic versus Open repair for oesophageal atresia: a retrospective cohort study of 359 patients at a single center. J Pediatr Surg 58:2069–2074. 10.1016/j.jpedsurg.2023.05.002 [DOI] [PubMed] [Google Scholar]
- 7.Way C, Wayne C, Grandpierre V, Harrison BJ, Travis N, Nasr A (2019) Thoracoscopy vs. thoracotomy for the repair of esophageal atresia and tracheoesophageal fistula: a systematic review and meta-analysis. Pediatr Surg Int 35:1167–1184. 10.1007/s00383-019-04527-9 [DOI] [PubMed] [Google Scholar]
- 8.Marquart JP, Bowder AN, Bence CM, St Peter SD, Gadepalli SK, Sato TT et al (2023) Thoracoscopy versus thoracotomy for esophageal atresia and tracheoesophageal fistula: outcomes from the Midwest Pediatric Surgery Consortium. J Pediatr Surg 58:27–33. 10.1016/j.jpedsurg.2022.09.015 [DOI] [PubMed] [Google Scholar]
- 9.Castro P, Fall F, Pace D, Mack SJ, Rothstein DH, Devin CL et al (2024) Association of operative approach with postoperative outcomes in neonates undergoing surgical repair of esophageal atresia and tracheoesophageal fistula. J Pediatr Surg 59:161641. 10.1016/j.jpedsurg.2024.07.026 [DOI] [PubMed] [Google Scholar]
- 10.Yasunaga H (2024) Updated information on the diagnosis procedure combination data. Ann Clin Epidemiol 6:106–110 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Ishitsuka K, Matsui H, Michihata N, Fushimi K, Nakamura T, Yasunaga H (2015) Medical procedures and outcomes of Japanese patients with trisomy 18 or trisomy 13: analysis of a nationwide administrative database of hospitalized patients. Am J Med Genet A 167A:1816–1821. 10.1002/ajmg.a.37104 [DOI] [PubMed] [Google Scholar]
- 12.WHO (2018) International classification of diseases. World Health Organization. http://www.who.int/classifications/icd/en/ Accessed 22 June 2018.
- 13.Yamana H, Moriwaki M, Horiguchi H, Kodan M, Fushimi K, Yasunaga H (2017) Validity of diagnoses, procedures, and laboratory data in Japanese administrative data. J Epidemiol 27:476–482. 10.1016/j.je.2016.09.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Konishi T, Yoshimoto T, Fujiogi M, Yamana H, Tanabe M, Seto Y et al (2022) Validity of operative information in Japanese administrative data: a chart review-based analysis of 1221 cases at a single institution. Surg Today 52:1484–1490. 10.1007/s00595-022-02521-8 [DOI] [PubMed] [Google Scholar]
- 15.Mathew G, Agha R, STROCSS Group (2021) STROCSS 2021: strengthening the reporting of cohort, cross-sectional and case-control studies in surgery. Ann Med Surg (Lond) 72:103026. 10.1016/j.amsu.2021.103026 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Ishimaru T, Fujiogi M, Michihata N, Matsui H, Fushimi K, Kawashima H et al (2019) Impact of congenital heart disease on outcomes after primary repair of esophageal atresia: a retrospective observational study using a nationwide database in Japan. Pediatr Surg Int 35:1077–1083. 10.1007/s00383-019-04542-w [DOI] [PubMed] [Google Scholar]
- 17.Nice T, Tuanama Diaz B, Shroyer M, Rogers D, Chen M, Martin C et al (2016) Risk factors for stricture formation after esophageal atresia repair. J Laparoendosc Adv Surg Tech A 26:393–398. 10.1089/lap.2015.0120 [DOI] [PubMed] [Google Scholar]
- 18.Rintala RJ (2017) Fundoplication in patients with esophageal atresia: patient selection, indications, and outcomes. Front Pediatr 5:109. 10.3389/fped.2017.00109 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.World Health Organization (2004). International statistical classification of diseases and related health problems, tenth revision, 2nd ed. World Health Organization. [PubMed]
- 20.Encinas JL, Luis AL, Avila LF, Martinez L, Guereta L, Lassaletta L et al (2006) Impact of preoperative diagnosis of congenital heart disease on the treatment of esophageal atresia. Pediatr Surg Int 22:150–153. 10.1007/s00383-005-1595-2 [DOI] [PubMed] [Google Scholar]
- 21.Japan Society of Obstetrics and Gynecology (2020) Guidelines for obstetrical practice in Japan. 2020 ed; 2020. (in Japanese), https://minds.jcqhc.or.jp/common/wp-content/plugins/pdfjs-viewer-shortcode/pdfjs/web/viewer.php?file=https://minds.jcqhc.or.jp/common/summary/pdf/c00572.pdf&dButton=false&pButton=false&oButton=false&sButton=true#zoom=auto&pagemode=none&_wpnonce=3b871a512b. Accessed 1 Mar 2025
- 22.Li F, Morgan KL, Zaslavsky AM (2019) Balancing covariates via propensity score weighting. J Am Stat Assoc 113:390–400. 10.1080/01621459.2016.1260466 [Google Scholar]
- 23.Thomas LE, Li F, Pencina MJ (2020) Overlap weighting: a propensity score method that mimics attributes of a randomized clinical trial. JAMA 323:2417–2418. 10.1001/jama.2020.7819 [DOI] [PubMed] [Google Scholar]
- 24.Austin PC (2011) An introduction to propensity score methods for reducing the effects of confounding in observational studies. Multivariate Behav Res 46:399–424. 10.1080/00273171.2011.568786 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Austin PC (2009) Balance diagnostics for comparing the distribution of baseline covariates between treatment groups in propensity-score matched samples. Stat Med 28:3083–3107. 10.1002/sim.3697 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Bourg A, Gottrand F, Parmentier B, Thomas J, Lehn A, Piolat C et al (2023) Outcome of long gap esophageal atresia at 6 years: a prospective case control cohort study. J Pediatr Surg 58:747–755. 10.1016/j.jpedsurg.2022.07.023 [DOI] [PubMed] [Google Scholar]
- 27.Ishikawa M, Tomita H, Ito Y, Tsukizaki A, Abe K, Shimotakahara A et al (2024) Analysis of gap length as a predictor of surgical outcomes in esophageal atresia with distal fistula: a single center experience. Pediatr Surg Int 40:99. 10.1007/s00383-024-05678-0 [DOI] [PubMed] [Google Scholar]
- 28.Newland N, Snajdauf J, Kokesova A, Styblova J, Hradsky O, Meusel I et al (2023) Anastomotic stricture prediction in patients with esophageal atresia with distal fistula. Pediatr Surg Int 39:136. 10.1007/s00383-023-05423-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Tovar JA, Fragoso AC (2013) Gastroesophageal reflux after repair of esophageal atresia. Eur J Pediatr Surg 23:175–181. 10.1055/s-0033-1347911 [DOI] [PubMed] [Google Scholar]
- 30.Okuyama H, Tazuke Y, Ueno T, Yamanaka H, Takama Y, Saka R et al (2018) Learning curve for the thoracoscopic repair of esophageal atresia with tracheoesophageal fistula. Asian J Endosc Surg 11:30–34. 10.1111/ases.12411 [DOI] [PubMed] [Google Scholar]
- 31.Al Tokhais T, Zamakhshary M, Aldekhayel S, Mandora H, Sayed S, AlHarbi K et al (2008) Thoracoscopic repair of tracheoesophageal fistulas: a case-control matched study. J Pediatr Surg 43:805–809. 10.1016/j.jpedsurg.2007.12.015 [DOI] [PubMed] [Google Scholar]
- 32.Bastard F, Bonnard A, Rousseau V, Gelas T, Michaud L, Irtan S et al (2018) Thoracic skeletal anomalies following surgical treatment of esophageal atresia. Lessons from a national cohort. J Pediatr Surg 53:605–609. 10.1016/j.jpedsurg.2017.07.013 [DOI] [PubMed] [Google Scholar]
- 33.Hattori K, Kawashima H, Ishimaru T, Yanagida Y, Miyake K, Iguchi M et al (2024) Musculoskeletal deformities after thoracoscopic versus conventional open repair for esophageal atresia. Asian J Surg 47:968–972. 10.1016/j.asjsur.2023.11.043 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
No datasets were generated or analysed during the current study.