Current Practice and Outcomes of Thoracoscopic Esophageal Atresia and Tracheoesophageal Fistula Repair: A Multi-institutional Analysis in Japan

Introduction

Lobe et al. ¹ in 1999 and Rothenberg ² in 2000 first reported thoracoscopic repair of esophageal atresia (EA) and EA/tracheoesophageal fistula (TEF). Since then, there have been many reports describing the operative techniques and the outcomes of thoracoscopic repair of EA and EA/TEF. However, most of the previous reports included small numbers of patients other than a few large private series^3–5 and a multi-institutional analysis. ⁶ Because the number of cases completed by individual practitioners is limited, there is a wide variety of the technique and management in each series. In order to better understand the current practice and outcomes of thoracoscopic repair of EA/TEF, a multi-institutional analysis was conducted among seven Japanese institutes that perform advanced laparoscopic and thoracoscopic procedures in infants and children. All of the co-authors belong to these institutes.

Materials and Methods

A survey was sent to the seven Japanese institutes that perform thoracoscopic repair of EA/TEF. The survey was composed of demographics, exclusion criteria, operative technique, postoperative management, and outcomes of the patients who underwent thoracoscopic repair of EA/TEF. The outcomes included the number of operations that were staged or converted to a thoracotomy and the number of patients who required fundoplication. The complications included anastomotic leakage, anastomotic stricture requiring at least one dilatation, recurrent TEF, tracheomalacia requiring conservative or surgical management, atelectasis, and pneumonia. Because of a wide variety of the operative techniques and the clinical course, infants who had EA without TEF were excluded from this study.

Results

All seven institutes responded to the survey. A low birth weight (five institutes: <2.5 kg in one, <2.0 kg in three, and <1.8 kg in one institute), major cardiac anomalies (three institutes), compromised physiologic status (three institutes: pulmonary hypertension in one, hemodynamic instability due to cardiac anomaly in two institutes), and long gap between proximal and distal esophagus (three institutes) were identified as exclusion criteria for thoracoscopic repair of EA/TEF (Table 1). Although a unified criterion for long gap is not present, the term “long gap” was used when the length of gap was more than 2–3.5 vertebral bodies in this series. The operation was uniformly performed via an intrapleural approach in the 0–45° prone position to allow the lung to fall away from the posterior mediastinum.

Anesthetic management and operative procedures are shown in Table 2. Preoperative bronchoscopy was routinely performed in six (85.7%) institutes, and single-lung ventilation using a bronchial blocker was performed in two (28.6%) institutes. The azygos vein was divided using a vessel-sealing device or electric cautery in six institutes (85.7%). The TEF was occluded with suture ligature in four institutes (57.1%) and clips in the remaining three (42.9%) institutes. Anastomosis was performed using the extracorporeal knot-tying technique with 5-0 to 6-0 absorbable sutures in four institutes (57.1%) and the intracorporeal technique in three institutes (42.9%). In order to facilitate anastomosis, stay sutures were used in three (43%) institutes. When the gap was too long for primary anastomosis, the thoracoscopic procedure was converted to the Foker procedure in three institutes, excessive mobilization in two institutes, and open thoracotomy in one institute. A gastrostomy was placed in two institutes (28.6%) routinely. Patients were left intubated and paralyzed for 3–7 days postoperatively in four institutes (57.1%).

In total, 58 patients underwent thoracoscopic repair of EA/TEF at the seven institutes (Table 3). The 58 cases were distributed among the seven institutes as follows: 16, 13, 12, 7, 5, 4, and 1 case, respectively. Fifty-two (89.7%) of the patients underwent successful thoracoscopic repair. Six (10.3%) operations were converted to open thoracotomy because of a long gap (n=4), right aortic arch (n=1), and intraoperative instability (n=1). One operation was staged because of the patient's low birth weight (1.2 kg).

Table 4 shows the summary of the outcomes of the present study in comparison with the previous large series. In the present study, the body weight and age at operation varied widely. The operative time ranged from 115 to 428 minutes. There were no major intraoperative complications, with one death not related to EA/TEF. The gap distance between the proximal and distal esophagus ranged from zero to four vertebral bodies. Eleven patients (18.9%) suffered from anastomotic leakage, which healed following conservative management. Twenty-eight patients (48.2%) developed anastomotic stricture. Endoscopic dilatation was successfully in all cases except 1, which required re-anastomosis due to severe stricture unresponded to multiple dilatations. Recurrent TEF developed in three infants (5.2%), two of whom required surgical closure of the recurrent TEF. Among the 3 patients with recurrent TEF, the methods of TEF closure were using clips in 1 patient and suture ligatures in 2 patients. Lung complication developed in 11 patients (19.0%). Tracheomalacia requiring prolonged intubation developed in 4 patients (6.9%), and aortopexy was not performed in any patients. Thirteen patients with gastroesophageal reflux disease (22.4%) later required fundoplication.

Discussion

A recent International Pediatric Endosurgery Group survey showed that there appears to be variability in technical aspects of the operation as well as the postoperative management of patients with EA/TEF. ⁷ Because of the small number of cases being completed by individual practitioners, there is a need for multi-institutional studies to achieve data that can lead to best practices. This is the first study to examine the nationwide variability in thoracoscopic management of patients with EA/TEF in Japan.

In this series, the operation was uniformly performed via an intrapleural approach in a prone position with the right side slightly elevated. In contrast to the uniform operative approach, we discovered that practice variation is immense. For example, closure of TEF (three suture ligatures, four titanium clips), knot-tying technique (four extracorporeal, three intracorporeal), and use of stay suture (three use, four do not) are different in each institute. As the number of cases experienced at each institute is limited, the procedure is thought to be in the process of the best operation maneuver being searched for.

The handling of the long gap is also different. In three institutes, surgeons preferred the Foker procedure, although the details of the procedure vary in each institute (thoracoscopic or open procedure). ⁸ In two institutes, excessive mobilization is preferred. According to our experience, primary anastomosis is possible after excessive mobilization of the proximal and distal esophagus in most cases of EA/TEF. As the thoracoscopic approach could make deep neck dissection possible under direct vision, there was no vocal cord paralysis even after excessive dissection in the current series. This may be another advantage of the thoracoscopic approach.

Another variable is postoperative management. Although the routine use of paralysis and mechanical ventilation appeared to have little rationale currently, it seems to be reasonable that if the anastomosis is performed under significant tension, the patient remains mechanically ventilated with muscle relaxation for several days to avoid additional tension. A further study is required to determine whether postoperative ventilatory management is necessary or not.

In evaluating the outcomes of this current series, the overall mortality rate was similar to those of the previous reports (Table 4). The incidence of the leak, the recurrent TEF, and the later fundoplication also favorably compare with the previous thoracoscopic and open series.^6,9,10

All of the 11 patients with anastomotic leakage underwent operation in the three institutes that had more than 10 cases of thoracoscopic repair. In contrast, there were no patients with leakage in the remaining four institutes that had fewer than 10 cases of thoracoscopic repairs. Therefore, the number of cases in each institute is less likely to affect the outcome of the thoracoscopic repair of EA/TEF. Regarding the recurrent TEF, the methods of TEF closure were using clips in 1 patient and suture ligatures in 2 patients. This fact suggests that the methods of TEF closure are not related to the development of recurrent TEF.

Although 28 patients (48.2%) required esophageal dilatation at least once, only 1 patient required revision of anastomosis. In most cases of stricture, the patients responded well to the first dilatation. These data are consistent with the result of a recent meta-analysis that indicated that thoracoscopic repair of EA/TEF is as safe and feasible as open repair. ¹¹

Conclusions

In conclusion, considerable variability was observed among the seven institutes with respect to the operative indications, surgical technique, and postoperative management of thoracoscopic repair of EA/TEF, which can be safely performed with less surgical trauma by experienced endoscopic surgeons. Although postoperative stricture was common in this series, there were no major intraoperative complications. The identification of variance in this survey is the first step to conducting future studies to identify best practices. Standardizing the surgical technique and postoperative management may reduce the incidence of complications after thoracoscopic repair for EA/TEF.