- Email: [email protected]
Slide tracheoplasty for the treatment of tracheoesophogeal fistulas
Journal of Pediatric Surgery 49 (2014) 910–914
Contents lists available at ScienceDirect
Journal of Pediatric Surgery journal homepage: www.elsevier.com/locate/jpedsurg
Slide tracheoplasty for the treatment of tracheoesophogeal fistulas Matthew J. Provenzano a, Michael J. Rutter a,⁎, Daniel von Allmen b, Peter B. Manning c, R. Paul Boesch d, Philip E. Putnam e, Angela P. Black a, Alessandro de Alarcon a a b c d e
Division of Pediatric Otolaryngology – Head and Neck Surgery, Cincinnati Children’s Hospital Medical Center, 3333 Burnet Avenue, Cincinnati, OH 45229–3026 Division of General and Thoracic Surgery, Division of Pediatric Surgery, Cincinnati Children’s Hospital Medical Center, 3333 Burnet Avenue, Cincinnati, OH 45229–3026 The Heart Center: Cardiology Cardiothoracic Surgery, St. Louis Children’s Hospital, One Children's Place, Suite 5S50, Saint Louis, MO 63110 Pediatric and Adolescent Medicine, Mayo Clinic, 201W Center St, Rochester, MN 55902 Division of Gastroenterology, Hepatology and Nutrition, Cincinnati Children’s Hospital Medical Center, 3333 Burnet Avenue, Cincinnati, OH 45229–3026
a r t i c l e
i n f o
Article history: Received 14 January 2014 Accepted 27 January 2014 Key words: Pneumonia Esophageal atresia Periosteum Tracheomalacia Cardiopulmonary bypass Recurrent Larynx
a b s t r a c t Purpose: The purpose of this study is to determine the surgical outcome of slide tracheoplasty for the treatment of tracheoesophageal (TE) fistula in pediatric patients. Methods: After internal review board approval, the charts of pediatric patients (0–18 years old) who had undergone slide tracheoplasty for tracheoesophageal fistula were retrospectively reviewed. Patient information and surgical outcomes were reviewed. Results: Nine patients underwent slide tracheoplasty for correction of TE fistula. In five patients the original TE fistula was congenital. Other causes included battery ingestion, tracheostomy tube complications, foreign body erosion, and an iatrogenic injury. The average age at repair was 48 ± 64 months (range: 1–190). Seven patients had undergone previous TEF repair either open or endoscopically. There were no recurrences after repair. Two patients had sternal periosteum interposed between the esophagus and trachea. There were no TEF recurrences. A single patient had dehiscence of the tracheal anastomosis and underwent a second procedure. Conclusion: Slide tracheoplasty is an effective method to treat complex TE fistulas. The procedure was not associated with any recurrences. This is the first description of a novel, effective, and safe method to treat TE fistulas. © 2014 Elsevier Inc. All rights reserved.
Complex tracheoesophageal (TE) fistulas can be significantly challenging to surgically correct and carry significant morbidity and mortality if left unrepaired. These aberrant connections between the esophagus and trachea are often associated with esophageal atresia, dysmotility, tracheomalacia and various syndromes. Food may enter the airway and soil the lungs. The patients, often medically fragile, are then prone to repeat bouts of pneumonia and respiratory complications. Oral feeding becomes impossible for many. Despite refraining from eating, many continue to have saliva entering the airway, leading to persistent pulmonary injury. Repair of these aberrant connections therefore becomes essential to a healthy lower airway as well as reestablishing oral feeds. Surgical correction of TE fistulas preserves trachea and esophagus patency while closing the communication between these two structures. A number of procedures have been developed to address this problem. However, many of these procedures have had limited success when dealing with the most complex fistulas. Successful repair was first reported in the 1940’s using an extrapleural approach and primary anastomosis to address the fistula and esophageal atresia [1]. Since that time a number of different techniques have been employed [2–4]. These procedures have sought to decrease the ⁎ Corresponding author. Tel.: +1 513 636 4355; fax: +1 513 636 8133. E-mail address: [email protected] (M.J. Rutter). http://dx.doi.org/10.1016/j.jpedsurg.2014.01.022 0022-3468/© 2014 Elsevier Inc. All rights reserved.
morbidity associated with the approach while improving the closure rates. Various alterations in anastomosis have been proposed [1]. Some authors have used cervical esophagostomy and gastrostomy or replacement for treatment in patients with severe disease and atresia [3]. Endoscopic procedures are also employed; small fistulas can be cauterized from both the esophageal and tracheal sides. Fibrin glue can also be placed within the fistula [2]. Previous techniques closed the esophageal and tracheal components separately but continued to leave the two anastomosis lines juxtaposed. Breakdown of one closure made it likely that the opposite side would also fail. Some authors have used various materials, such as vein or pleura, between the two repair sites to help prevent breakdown of the closure [5]. Despite this interposed tissue, the two anastomotic lines continued to be continuity. Even with the success with these procedures, persistent and recurrent fistulas remain surgical challenges. Revision cases carry the additional difficulties of scaring, altered surgical anatomy and compromised blood supply to the tissue. Given these reasons, repair of recurrent fistulas is often difficult and associated with significant morbidity and mortality [3]. Slide tracheoplasty is a technique utilized to address various etiologies of tracheal stenosis [6,7]. It has proven a safe and highly effective treatment for patients with complete tracheal rings [6]. While this procedure has been successfully employed for tracheal abnormalities, the literature is absent of any reports using this
M.J. Provenzano et al. / Journal of Pediatric Surgery 49 (2014) 910–914
911
technique to address complex TE fistulas. We have found success with this technique when reserving it for complex cases or previously repaired fistulas. In this report we present the novel use of slide tracheoplasty for the treatment of TE fistulas and describe the surgical outcomes of this procedure. 1. Materials and methods 1.1. Data collection and analysis Data collection and patient selection were performed after approval by the Cincinnati Children’s Hospital Medical Center Institutional Review Board. Patients were identified who had a diagnosis of TE fistula and who had undergone repair by slide tracheoplasty. Pediatric patients (age 0–18 years old) were identified from 2000 to 2012. Demographic information, surgical data and outcomes were recorded and analyzed using Microsoft Excel (Redmond, WA). Patient gender, age at surgery and medical diagnosis were recorded. Previous surgical procedures were noted as well as intraoperative findings, surgical outcomes and complications. 1.2. Surgical procedure Our previous publication has described the slide tracheoplasty technique [6]. Modifications have been made to address the TE fistula. Microscopic direct laryngoscopy, bronchoscopy and esophagoscopy are initially performed to verify the location of the fistula. An esophageal bougie is placed. The technique varies depending upon the surgical approach. Patients undergoing a cervical incision will be intubated orally, or in the case of an existing tracheostomy, through the stoma site. Patients requiring a sternal incision and cardiopulmonary bypass will be intubated orally with the stoma sutured shut. In these patients, cardiopulmonary bypass is established prior to the beginning of tracheal work. In both approaches, the anterior wall of the trachea is freed from surrounding tissue as distal as possible. For patients with a cervical approach, mobilization often continues into the mediastinum and to the carina. In all patients, care is taken to preserve lateral tracheal attachments to maintain the blood supply and avoid damage to the recurrent laryngeal nerves. Retraction sutures using 2–0 Prolene (Ethicon, Blue Ash, OH) are placed through the distal tracheal rings to retract the trachea. Flexible bronchoscopy through the endotracheal tube or rigid bronchoscopy can be repeated to confirm the location of the fistula. A needle is placed through the anterior tracheal in the corresponding location of the fistula. The trachea is then divided both superior and inferior to the fistula tract, leaving a small portion of trachea attached to the tract (Fig. 1). The trachealis of the superior and inferior tracheal segments is then separated from the esophagus and mobilized. The trachea that remains attached to the fistula is then freed of it mucosa and the cartilage portion is removed (Fig. 2). The edges of the esophageal side of the fistula are freshened in preparation of closure. The tracheal mucosa is then inverted and folded into the esophageal portion of the fistula. Closure is performed with a series of interrupted vicryl sutures (Ethicon). In cases of large fistulas, the cartilage can be kept in continuity with the mucosa and used in the closure for added support. Periosteum is harvested from the sternum and placed on top of the esophageal closure (Fig. 3). Approximately 1 cm of the posterior wall of the inferior tracheal segment and the anterior wall of the superior segment are then divided vertically. The corners of the two segments are removed in order to achieve better approximation during closure. A running, polydioxanone (PDS) (Ethicon) is then used to close the anastomosis beginning with the posterior aspect of the trachea. The resulting, oblique anastomosis is longer than a corresponding end-to-end anastomosis, thereby distributing the tension across a longer area. Once all sutures are placed, fibrin glue is then applied across the
Fig. 1. The trachea is divided superior and inferior to the fistula site. A small portion of the trachea remains attached at the fistula and is used to reinforce the fistula repair.
tracheal closure. If occurring through a cervical incision, the previously placed Prolene retraction sutures can be placed around the hyoid as internal Grillo sutures. When performed through a sternotomy, the retraction sutures can be removed as the hyoid is not exposed. The patient remains intubated overnight in the intensive care unit and is generally extubated on postoperative day one. The endotracheal tube lumen should be large enough to accommodate a flexible bronchoscope containing a suction port, thereby allowing for the removal of secretions or blood. A repeat bronchoscopy can be performed in one to two weeks. Fig. 4 demonstrates preoperative and postoperative photos.
Fig. 2. The tracheal mucosa is removed from the cartilaginous rings and folded into the denuded fistula to reinforce the closure. In some instances the cartilage can also be used to reinforce the closure site.
912
M.J. Provenzano et al. / Journal of Pediatric Surgery 49 (2014) 910–914
Fig. 3. The sternal periosteum is placed between the tracheal and esophageal closure to reinforce the repair. Tracheal closure occurs through an oblique running anastomosis. The trachea mucosa that had been removed from the tracheal cartilage has been folded into the esophageal side of the fistula and incorporated into the repair.
2. Results Nine patients were identified who had a TE fistula repaired by slide tracheoplasty. The average age at surgery was 48 ± 64 months (range: 1–190 months). A variety of etiologies and associated anomalies were identified (Table 1). Seven of patients had undergone previous repairs including open and endoscopic procedures (Table 2). Two patients had a slide tracheoplasty as their initial procedure given the complexity of their case. One patient had suffered a foreign body erosion through a bronchi resulting in a large fistula which made repair via other techniques difficult. This patient had a slide on bypass. A second patient had a fistula in the presence of significant upper airway obstruction. In that case, the slide addressed both the fistula as well as some levels of stenosis. Three patients had additional airway procedures performed to address subglottic stenosis. Sternal periosteum was used in two procedures (Table 3). Slides were performed either through a sternal incision with the patient placed on cardiopulmonary bypass or through the neck with normal ventilator support. There were no recurrences after surgical repair. Five patients had a fistula greater than 1 cm in length. A single patient experienced a tracheal dehiscence that resulted in a second slide procedure. That patient required a temporary tracheostomy and was eventually decannulated. Five patients had a tracheostomy; four were present preoperative and one was placed postoperative during treatment of subglottic stenosis. Three patients who did not take oral intake prior to surgery were able to eat by mouth after fistula repair.
Table 1 Demographic information for 9 patients.
Fig. 4. Preoperative photos of the trachea show the small fistula tract with a suction catheter inserted. The trachea itself shows moderate tracheomalacia with the fistula distal near the carina. Postoperative photos show the fistula tract closed with a small, closed dimple in the mucosa.
Age at surgery: Female: Esophageal atresia: H-type fistulas: Etiology: Congenital Battery ingestion Iatrogenic Tracheostomy tube erosion Foreign body erosion Associated anomalies Esophageal stricture VACTERL DiGeorge GERD Tracheomalacia Cardiac anomalies GERD: Gastroesophageal reflux disease.
48 ± 64 months (range: 1–190) 4 (44%) 3 (33%) 3 (33%) 5 (55%) 1 (11%) 1 (11%) 1 (11%) 1 (11%) 3 (66%) 1 (11%) 1 (11%) 4 (44%) 5 (56%) 2 (22%)
M.J. Provenzano et al. / Journal of Pediatric Surgery 49 (2014) 910–914 Table 2 Associated procedures. Previous procedures No. of patients Total no. previous procedures Endoscopic (no. pts) Open (no. pts) Additional airway procedures performed after slide tracheoplasty ssLTP ACCG CTR dsLTP APCCG, arytenoidectomy, petiole repositioning
6 (67%) 19 5 (56%) 2 (22%) 2 (22%) 1 (11%) 1 (11%)
ssLTP: single stage laryngotracheoplasty; ACCH: anterior costal cartilage graft; CTR: cricotracheal resection; dsLTP: double stage laryngotracheoplasty; APCCG: anterior posterior costal cartilage graft.
3. Discussion While other surgical techniques have successfully treated TE fistulas, a subgroup of patients have recurrences, persistence or complex fistulas. Tsai and colleagues reported a 10% recurrence rate [8], consistent with the experience of other authors [5]. Some patients experience multiple recurrences, making subsequent repairs increasingly difficult [3]. Certain procedures, such as fistula ligation, are associated with higher rates of recurrence [4]. The technique presented here was not associated with any recurrences or persistent fistulas. This despite the difficulties associated with revision procedures. The difference in recurrence rate between our series and previously published work likely results from the slide tracheoplasty technique. Unlike other procedures, this technique offsets the tracheal and esophageal repairs. By advancing the trachealis the esophageal and tracheal anastomosis lines are not in continuity. Breakdown of one closure line should not affect the other closure. In contrast, other procedures keep the esophageal and tracheal closures juxtaposed. In that scenario, breakdown of one closure jeopardizes the other. The use of sternal periosteum in two patients highlights an additional method to prevent recurrences. The periosteum was placed between the tracheal and esophageal walls to support the fistula closures. We employ this technique for laryngeal cleft repair and have found that even when one side of the fistula breaks down, the periosteum remains intact and the cleft does not recur. Others have reported the use of pericardium, pleura or vein [5]. We choose to use periosteum given the ease of harvesting it from a cervical incision and the large amount of available material. To our knowledge, this is the first instance of its use in repairing TE fistulas. Although not a vascularized material, its durability likely exceeds that of other free tissue grafts such as vein or free muscle. The slide tracheoplasty technique also addresses the difficulty with tension along the anastomosis, a finding noted to effect surgical outcomes [3,5]. The long, oblique anastomosis line in this technique distributes the tension across a greater surface area, preventing break
Table 3 Surgical procedure and results. Cardiopulmonary bypass Sternal periosteum Fistula size b1 cm N1 cm Tracheostomy Yes No Surgical outcomes Recurrence Infection VC paralysis Dehiscence Mild tracheal stenosis
3 (33%) 2 (22%) 4 (44%) 5 (56%) 5 (56%) 4 (44%) 0 (0%) 0 (0%) 1 (11%) 1 (11%) 2 (22%)
913
down of the anastomosis. When the slide is done through a cervical incision, tension along the anastomosis can be decreased through the hyoid release and use of internal retraction sutures. While this technique carries significant benefits for TE fistula closure, it also presents challenges and risks. Significant mobilization of the trachea can be difficult especially in the setting of previous surgical repairs. Despite this, previous reports have demonstrated successful slide tracheoplasty after a previous fistula repair [9,10]. A majority of the patients in our series had undergone a previous procedure. Despite these past operations, scarring and distorted anatomy, slide tracheoplasty was performed in all patients. We have found it difficult to completely free the trachea distally. Mobilization in this area is facilitated by sternotomy and cardiopulmonary bypass but can be difficult when performing the slide through a cervical incision. One patient had previously underwent a tracheal resection but still had an uneventful slide procedure. Despite this success, caution should be taken in those patients with a shortened trachea as mobilization and the anastomosis may be difficult. Freeing of the trachea also requires attention to recurrent laryngeal nerve position. The nerves’ normal location within the tracheoesophageal groove can be altered from previous surgery, scarring or aberrant development. We have found that adhering closely to the tracheal wall protects the nerves. One patient in our series did have unilateral vocal cord paralysis. This patient had a postoperative dehiscence requiring reoperation, wound exploration and a repeat slide. It is unknown if the recurrent laryngeal nerve injury occurred at the first or second procedure or whether it was a preexisting consequence of the original button battery injury. A single patient (1/9, 11%) had tracheal dehiscence postoperatively. This is a major, life threatening complication. Close postoperative monitoring is essential for early detection. Wound crepitus or respiratory compromise must be promptly and thoroughly invested. There must be a low tolerance for returning to the operating room. While our patient was successfully treated with a repeat slide, it has been our experience that other interventions are occasionally required to secure the airway in the event of tracheal dehiscence. These can include placement of a tracheostomy, T-tube, tracheal stenting or reconstruction with cartilage grafts. Repeat slide tracheoplasties in the setting of dehiscence often require additional tracheal mobilization as the tissue is more friable and tension at the anastomosis needs to be significantly reduced. The single dehiscence reported here is higher than our previously published work concerning slide tracheoplasty for treatment of complete tracheal rings [6]. The esophageal component in this series may be responsible for the higher dehiscence rate. Esophageal involvement may reduce blood supply to the trachealis while also exposing the anastomosis to esophageal contents in the event of a partial breakdown. This patient suffered significant intimal damage from the battery ingestion that caused the fistula. This likely contributed to the poor esophageal and tracheal blood supply that hindered healing and predisposed this patient to dehiscence. Associated esophageal and tracheal anomalies did not affect surgical outcome. The two patients with tracheal pouches had successful closure of their fistulas without surgical difficulties. Those patients with tracheomalacia may sometimes benefit from the altered configuration of the trachea that results from the slide. However, this is not the primary purpose of this procedure and we did not assess for changes in tracheomalacia symptoms after the slide. Any respiratory improvements were likely secondary to closure of the fistula and prevention of pneumonias. Two patients had continued tracheal stenosis even after the slides, a finding seen in our slide tracheoplasty experience [6]. This series demonstrated that slide tracheoplasty can successfully repair complex, recurrent or persistent TE fistulas. It is a procedure reserved for those instances and should not be employed as initial treatment for simple fistulas that could best be addressed via other
914
M.J. Provenzano et al. / Journal of Pediatric Surgery 49 (2014) 910–914
means. In our series no patients had a recurrence. Sternal periosteum is easily harvested and provides an additional level of repair reinforcement. Repair in this manner does not prevent future airway procedures and was associated with relatively few complications. This technique offers a novel approach to treating these fistulas. References [1] Manning PB, Morgan RA, Coran AG, et al. Fifty years' experience with esophageal atresia and tracheoesophageal fistula. Beginning with Cameron Haight's first operation in 1935. Ann Surg 1986;204:446–53. [2] Richter GT, Ryckman F, Brown RL, et al. Endoscopic management of recurrent tracheoesophageal fistula. J Pediatr Surg 2008;43:238–45. [3] Ghandour KE, Spitz L, Brereton RJ, et al. Recurrent tracheo-oesophageal fistula: experience with 24 patients. J Paediatr Child Health 1990;26:89–91.
[4] Touloukian RJ. Long-term results following repair of esophageal atresia by end-toside anastomosis and ligation of the tracheoesophageal fistula. J Pediatr Surg 1981;16:983–8. [5] Kovesi T, Rubin S. Long-term complications of congenital esophageal atresia and/ or tracheoesophageal fistula. Chest 2004;126:915–25. [6] Rutter MJ, Cotton RT, Azizkhan RG, et al. Slide tracheoplasty for the management of complete tracheal rings. J Pediatr Surg 2003;38:928–34. [7] Elliott MJ, Speggiorin S, Vida VL, et al. Slide tracheoplasty as a rescue technique after unsuccessful patch tracheoplasty. Ann Thorac Surg 2009;88:1029–31. [8] Tsai JY, Berkery L, Wesson DE, et al. Esophageal atresia and tracheoesophageal fistula: surgical experience over two decades. Ann Thorac Surg 1997;64:778–83 [discussion 783–774]. [9] Le Bret E, Roubertie F, Roger G. Slide tracheoplasty in infant with congenital tracheal stenosis and tracheomalacia after esophageal atresia with tracheoesophageal fistula repair. J Pediatr Surg 2009;44:e15–7. [10] Anton-Pacheco JL, Kalicinski P, Kansy A, et al. Slide tracheoplasty in an infant with congenital tracheal stenosis and oesophageal atresia with tracheoesophageal fistula. Eur J Cardiothorac Surg 2012;42:892–3.
Contents lists available at ScienceDirect
Journal of Pediatric Surgery journal homepage: www.elsevier.com/locate/jpedsurg
Slide tracheoplasty for the treatment of tracheoesophogeal fistulas Matthew J. Provenzano a, Michael J. Rutter a,⁎, Daniel von Allmen b, Peter B. Manning c, R. Paul Boesch d, Philip E. Putnam e, Angela P. Black a, Alessandro de Alarcon a a b c d e
Division of Pediatric Otolaryngology – Head and Neck Surgery, Cincinnati Children’s Hospital Medical Center, 3333 Burnet Avenue, Cincinnati, OH 45229–3026 Division of General and Thoracic Surgery, Division of Pediatric Surgery, Cincinnati Children’s Hospital Medical Center, 3333 Burnet Avenue, Cincinnati, OH 45229–3026 The Heart Center: Cardiology Cardiothoracic Surgery, St. Louis Children’s Hospital, One Children's Place, Suite 5S50, Saint Louis, MO 63110 Pediatric and Adolescent Medicine, Mayo Clinic, 201W Center St, Rochester, MN 55902 Division of Gastroenterology, Hepatology and Nutrition, Cincinnati Children’s Hospital Medical Center, 3333 Burnet Avenue, Cincinnati, OH 45229–3026
a r t i c l e
i n f o
Article history: Received 14 January 2014 Accepted 27 January 2014 Key words: Pneumonia Esophageal atresia Periosteum Tracheomalacia Cardiopulmonary bypass Recurrent Larynx
a b s t r a c t Purpose: The purpose of this study is to determine the surgical outcome of slide tracheoplasty for the treatment of tracheoesophageal (TE) fistula in pediatric patients. Methods: After internal review board approval, the charts of pediatric patients (0–18 years old) who had undergone slide tracheoplasty for tracheoesophageal fistula were retrospectively reviewed. Patient information and surgical outcomes were reviewed. Results: Nine patients underwent slide tracheoplasty for correction of TE fistula. In five patients the original TE fistula was congenital. Other causes included battery ingestion, tracheostomy tube complications, foreign body erosion, and an iatrogenic injury. The average age at repair was 48 ± 64 months (range: 1–190). Seven patients had undergone previous TEF repair either open or endoscopically. There were no recurrences after repair. Two patients had sternal periosteum interposed between the esophagus and trachea. There were no TEF recurrences. A single patient had dehiscence of the tracheal anastomosis and underwent a second procedure. Conclusion: Slide tracheoplasty is an effective method to treat complex TE fistulas. The procedure was not associated with any recurrences. This is the first description of a novel, effective, and safe method to treat TE fistulas. © 2014 Elsevier Inc. All rights reserved.
Complex tracheoesophageal (TE) fistulas can be significantly challenging to surgically correct and carry significant morbidity and mortality if left unrepaired. These aberrant connections between the esophagus and trachea are often associated with esophageal atresia, dysmotility, tracheomalacia and various syndromes. Food may enter the airway and soil the lungs. The patients, often medically fragile, are then prone to repeat bouts of pneumonia and respiratory complications. Oral feeding becomes impossible for many. Despite refraining from eating, many continue to have saliva entering the airway, leading to persistent pulmonary injury. Repair of these aberrant connections therefore becomes essential to a healthy lower airway as well as reestablishing oral feeds. Surgical correction of TE fistulas preserves trachea and esophagus patency while closing the communication between these two structures. A number of procedures have been developed to address this problem. However, many of these procedures have had limited success when dealing with the most complex fistulas. Successful repair was first reported in the 1940’s using an extrapleural approach and primary anastomosis to address the fistula and esophageal atresia [1]. Since that time a number of different techniques have been employed [2–4]. These procedures have sought to decrease the ⁎ Corresponding author. Tel.: +1 513 636 4355; fax: +1 513 636 8133. E-mail address: [email protected] (M.J. Rutter). http://dx.doi.org/10.1016/j.jpedsurg.2014.01.022 0022-3468/© 2014 Elsevier Inc. All rights reserved.
morbidity associated with the approach while improving the closure rates. Various alterations in anastomosis have been proposed [1]. Some authors have used cervical esophagostomy and gastrostomy or replacement for treatment in patients with severe disease and atresia [3]. Endoscopic procedures are also employed; small fistulas can be cauterized from both the esophageal and tracheal sides. Fibrin glue can also be placed within the fistula [2]. Previous techniques closed the esophageal and tracheal components separately but continued to leave the two anastomosis lines juxtaposed. Breakdown of one closure made it likely that the opposite side would also fail. Some authors have used various materials, such as vein or pleura, between the two repair sites to help prevent breakdown of the closure [5]. Despite this interposed tissue, the two anastomotic lines continued to be continuity. Even with the success with these procedures, persistent and recurrent fistulas remain surgical challenges. Revision cases carry the additional difficulties of scaring, altered surgical anatomy and compromised blood supply to the tissue. Given these reasons, repair of recurrent fistulas is often difficult and associated with significant morbidity and mortality [3]. Slide tracheoplasty is a technique utilized to address various etiologies of tracheal stenosis [6,7]. It has proven a safe and highly effective treatment for patients with complete tracheal rings [6]. While this procedure has been successfully employed for tracheal abnormalities, the literature is absent of any reports using this
M.J. Provenzano et al. / Journal of Pediatric Surgery 49 (2014) 910–914
911
technique to address complex TE fistulas. We have found success with this technique when reserving it for complex cases or previously repaired fistulas. In this report we present the novel use of slide tracheoplasty for the treatment of TE fistulas and describe the surgical outcomes of this procedure. 1. Materials and methods 1.1. Data collection and analysis Data collection and patient selection were performed after approval by the Cincinnati Children’s Hospital Medical Center Institutional Review Board. Patients were identified who had a diagnosis of TE fistula and who had undergone repair by slide tracheoplasty. Pediatric patients (age 0–18 years old) were identified from 2000 to 2012. Demographic information, surgical data and outcomes were recorded and analyzed using Microsoft Excel (Redmond, WA). Patient gender, age at surgery and medical diagnosis were recorded. Previous surgical procedures were noted as well as intraoperative findings, surgical outcomes and complications. 1.2. Surgical procedure Our previous publication has described the slide tracheoplasty technique [6]. Modifications have been made to address the TE fistula. Microscopic direct laryngoscopy, bronchoscopy and esophagoscopy are initially performed to verify the location of the fistula. An esophageal bougie is placed. The technique varies depending upon the surgical approach. Patients undergoing a cervical incision will be intubated orally, or in the case of an existing tracheostomy, through the stoma site. Patients requiring a sternal incision and cardiopulmonary bypass will be intubated orally with the stoma sutured shut. In these patients, cardiopulmonary bypass is established prior to the beginning of tracheal work. In both approaches, the anterior wall of the trachea is freed from surrounding tissue as distal as possible. For patients with a cervical approach, mobilization often continues into the mediastinum and to the carina. In all patients, care is taken to preserve lateral tracheal attachments to maintain the blood supply and avoid damage to the recurrent laryngeal nerves. Retraction sutures using 2–0 Prolene (Ethicon, Blue Ash, OH) are placed through the distal tracheal rings to retract the trachea. Flexible bronchoscopy through the endotracheal tube or rigid bronchoscopy can be repeated to confirm the location of the fistula. A needle is placed through the anterior tracheal in the corresponding location of the fistula. The trachea is then divided both superior and inferior to the fistula tract, leaving a small portion of trachea attached to the tract (Fig. 1). The trachealis of the superior and inferior tracheal segments is then separated from the esophagus and mobilized. The trachea that remains attached to the fistula is then freed of it mucosa and the cartilage portion is removed (Fig. 2). The edges of the esophageal side of the fistula are freshened in preparation of closure. The tracheal mucosa is then inverted and folded into the esophageal portion of the fistula. Closure is performed with a series of interrupted vicryl sutures (Ethicon). In cases of large fistulas, the cartilage can be kept in continuity with the mucosa and used in the closure for added support. Periosteum is harvested from the sternum and placed on top of the esophageal closure (Fig. 3). Approximately 1 cm of the posterior wall of the inferior tracheal segment and the anterior wall of the superior segment are then divided vertically. The corners of the two segments are removed in order to achieve better approximation during closure. A running, polydioxanone (PDS) (Ethicon) is then used to close the anastomosis beginning with the posterior aspect of the trachea. The resulting, oblique anastomosis is longer than a corresponding end-to-end anastomosis, thereby distributing the tension across a longer area. Once all sutures are placed, fibrin glue is then applied across the
Fig. 1. The trachea is divided superior and inferior to the fistula site. A small portion of the trachea remains attached at the fistula and is used to reinforce the fistula repair.
tracheal closure. If occurring through a cervical incision, the previously placed Prolene retraction sutures can be placed around the hyoid as internal Grillo sutures. When performed through a sternotomy, the retraction sutures can be removed as the hyoid is not exposed. The patient remains intubated overnight in the intensive care unit and is generally extubated on postoperative day one. The endotracheal tube lumen should be large enough to accommodate a flexible bronchoscope containing a suction port, thereby allowing for the removal of secretions or blood. A repeat bronchoscopy can be performed in one to two weeks. Fig. 4 demonstrates preoperative and postoperative photos.
Fig. 2. The tracheal mucosa is removed from the cartilaginous rings and folded into the denuded fistula to reinforce the closure. In some instances the cartilage can also be used to reinforce the closure site.
912
M.J. Provenzano et al. / Journal of Pediatric Surgery 49 (2014) 910–914
Fig. 3. The sternal periosteum is placed between the tracheal and esophageal closure to reinforce the repair. Tracheal closure occurs through an oblique running anastomosis. The trachea mucosa that had been removed from the tracheal cartilage has been folded into the esophageal side of the fistula and incorporated into the repair.
2. Results Nine patients were identified who had a TE fistula repaired by slide tracheoplasty. The average age at surgery was 48 ± 64 months (range: 1–190 months). A variety of etiologies and associated anomalies were identified (Table 1). Seven of patients had undergone previous repairs including open and endoscopic procedures (Table 2). Two patients had a slide tracheoplasty as their initial procedure given the complexity of their case. One patient had suffered a foreign body erosion through a bronchi resulting in a large fistula which made repair via other techniques difficult. This patient had a slide on bypass. A second patient had a fistula in the presence of significant upper airway obstruction. In that case, the slide addressed both the fistula as well as some levels of stenosis. Three patients had additional airway procedures performed to address subglottic stenosis. Sternal periosteum was used in two procedures (Table 3). Slides were performed either through a sternal incision with the patient placed on cardiopulmonary bypass or through the neck with normal ventilator support. There were no recurrences after surgical repair. Five patients had a fistula greater than 1 cm in length. A single patient experienced a tracheal dehiscence that resulted in a second slide procedure. That patient required a temporary tracheostomy and was eventually decannulated. Five patients had a tracheostomy; four were present preoperative and one was placed postoperative during treatment of subglottic stenosis. Three patients who did not take oral intake prior to surgery were able to eat by mouth after fistula repair.
Table 1 Demographic information for 9 patients.
Fig. 4. Preoperative photos of the trachea show the small fistula tract with a suction catheter inserted. The trachea itself shows moderate tracheomalacia with the fistula distal near the carina. Postoperative photos show the fistula tract closed with a small, closed dimple in the mucosa.
Age at surgery: Female: Esophageal atresia: H-type fistulas: Etiology: Congenital Battery ingestion Iatrogenic Tracheostomy tube erosion Foreign body erosion Associated anomalies Esophageal stricture VACTERL DiGeorge GERD Tracheomalacia Cardiac anomalies GERD: Gastroesophageal reflux disease.
48 ± 64 months (range: 1–190) 4 (44%) 3 (33%) 3 (33%) 5 (55%) 1 (11%) 1 (11%) 1 (11%) 1 (11%) 3 (66%) 1 (11%) 1 (11%) 4 (44%) 5 (56%) 2 (22%)
M.J. Provenzano et al. / Journal of Pediatric Surgery 49 (2014) 910–914 Table 2 Associated procedures. Previous procedures No. of patients Total no. previous procedures Endoscopic (no. pts) Open (no. pts) Additional airway procedures performed after slide tracheoplasty ssLTP ACCG CTR dsLTP APCCG, arytenoidectomy, petiole repositioning
6 (67%) 19 5 (56%) 2 (22%) 2 (22%) 1 (11%) 1 (11%)
ssLTP: single stage laryngotracheoplasty; ACCH: anterior costal cartilage graft; CTR: cricotracheal resection; dsLTP: double stage laryngotracheoplasty; APCCG: anterior posterior costal cartilage graft.
3. Discussion While other surgical techniques have successfully treated TE fistulas, a subgroup of patients have recurrences, persistence or complex fistulas. Tsai and colleagues reported a 10% recurrence rate [8], consistent with the experience of other authors [5]. Some patients experience multiple recurrences, making subsequent repairs increasingly difficult [3]. Certain procedures, such as fistula ligation, are associated with higher rates of recurrence [4]. The technique presented here was not associated with any recurrences or persistent fistulas. This despite the difficulties associated with revision procedures. The difference in recurrence rate between our series and previously published work likely results from the slide tracheoplasty technique. Unlike other procedures, this technique offsets the tracheal and esophageal repairs. By advancing the trachealis the esophageal and tracheal anastomosis lines are not in continuity. Breakdown of one closure line should not affect the other closure. In contrast, other procedures keep the esophageal and tracheal closures juxtaposed. In that scenario, breakdown of one closure jeopardizes the other. The use of sternal periosteum in two patients highlights an additional method to prevent recurrences. The periosteum was placed between the tracheal and esophageal walls to support the fistula closures. We employ this technique for laryngeal cleft repair and have found that even when one side of the fistula breaks down, the periosteum remains intact and the cleft does not recur. Others have reported the use of pericardium, pleura or vein [5]. We choose to use periosteum given the ease of harvesting it from a cervical incision and the large amount of available material. To our knowledge, this is the first instance of its use in repairing TE fistulas. Although not a vascularized material, its durability likely exceeds that of other free tissue grafts such as vein or free muscle. The slide tracheoplasty technique also addresses the difficulty with tension along the anastomosis, a finding noted to effect surgical outcomes [3,5]. The long, oblique anastomosis line in this technique distributes the tension across a greater surface area, preventing break
Table 3 Surgical procedure and results. Cardiopulmonary bypass Sternal periosteum Fistula size b1 cm N1 cm Tracheostomy Yes No Surgical outcomes Recurrence Infection VC paralysis Dehiscence Mild tracheal stenosis
3 (33%) 2 (22%) 4 (44%) 5 (56%) 5 (56%) 4 (44%) 0 (0%) 0 (0%) 1 (11%) 1 (11%) 2 (22%)
913
down of the anastomosis. When the slide is done through a cervical incision, tension along the anastomosis can be decreased through the hyoid release and use of internal retraction sutures. While this technique carries significant benefits for TE fistula closure, it also presents challenges and risks. Significant mobilization of the trachea can be difficult especially in the setting of previous surgical repairs. Despite this, previous reports have demonstrated successful slide tracheoplasty after a previous fistula repair [9,10]. A majority of the patients in our series had undergone a previous procedure. Despite these past operations, scarring and distorted anatomy, slide tracheoplasty was performed in all patients. We have found it difficult to completely free the trachea distally. Mobilization in this area is facilitated by sternotomy and cardiopulmonary bypass but can be difficult when performing the slide through a cervical incision. One patient had previously underwent a tracheal resection but still had an uneventful slide procedure. Despite this success, caution should be taken in those patients with a shortened trachea as mobilization and the anastomosis may be difficult. Freeing of the trachea also requires attention to recurrent laryngeal nerve position. The nerves’ normal location within the tracheoesophageal groove can be altered from previous surgery, scarring or aberrant development. We have found that adhering closely to the tracheal wall protects the nerves. One patient in our series did have unilateral vocal cord paralysis. This patient had a postoperative dehiscence requiring reoperation, wound exploration and a repeat slide. It is unknown if the recurrent laryngeal nerve injury occurred at the first or second procedure or whether it was a preexisting consequence of the original button battery injury. A single patient (1/9, 11%) had tracheal dehiscence postoperatively. This is a major, life threatening complication. Close postoperative monitoring is essential for early detection. Wound crepitus or respiratory compromise must be promptly and thoroughly invested. There must be a low tolerance for returning to the operating room. While our patient was successfully treated with a repeat slide, it has been our experience that other interventions are occasionally required to secure the airway in the event of tracheal dehiscence. These can include placement of a tracheostomy, T-tube, tracheal stenting or reconstruction with cartilage grafts. Repeat slide tracheoplasties in the setting of dehiscence often require additional tracheal mobilization as the tissue is more friable and tension at the anastomosis needs to be significantly reduced. The single dehiscence reported here is higher than our previously published work concerning slide tracheoplasty for treatment of complete tracheal rings [6]. The esophageal component in this series may be responsible for the higher dehiscence rate. Esophageal involvement may reduce blood supply to the trachealis while also exposing the anastomosis to esophageal contents in the event of a partial breakdown. This patient suffered significant intimal damage from the battery ingestion that caused the fistula. This likely contributed to the poor esophageal and tracheal blood supply that hindered healing and predisposed this patient to dehiscence. Associated esophageal and tracheal anomalies did not affect surgical outcome. The two patients with tracheal pouches had successful closure of their fistulas without surgical difficulties. Those patients with tracheomalacia may sometimes benefit from the altered configuration of the trachea that results from the slide. However, this is not the primary purpose of this procedure and we did not assess for changes in tracheomalacia symptoms after the slide. Any respiratory improvements were likely secondary to closure of the fistula and prevention of pneumonias. Two patients had continued tracheal stenosis even after the slides, a finding seen in our slide tracheoplasty experience [6]. This series demonstrated that slide tracheoplasty can successfully repair complex, recurrent or persistent TE fistulas. It is a procedure reserved for those instances and should not be employed as initial treatment for simple fistulas that could best be addressed via other
914
M.J. Provenzano et al. / Journal of Pediatric Surgery 49 (2014) 910–914
means. In our series no patients had a recurrence. Sternal periosteum is easily harvested and provides an additional level of repair reinforcement. Repair in this manner does not prevent future airway procedures and was associated with relatively few complications. This technique offers a novel approach to treating these fistulas. References [1] Manning PB, Morgan RA, Coran AG, et al. Fifty years' experience with esophageal atresia and tracheoesophageal fistula. Beginning with Cameron Haight's first operation in 1935. Ann Surg 1986;204:446–53. [2] Richter GT, Ryckman F, Brown RL, et al. Endoscopic management of recurrent tracheoesophageal fistula. J Pediatr Surg 2008;43:238–45. [3] Ghandour KE, Spitz L, Brereton RJ, et al. Recurrent tracheo-oesophageal fistula: experience with 24 patients. J Paediatr Child Health 1990;26:89–91.
[4] Touloukian RJ. Long-term results following repair of esophageal atresia by end-toside anastomosis and ligation of the tracheoesophageal fistula. J Pediatr Surg 1981;16:983–8. [5] Kovesi T, Rubin S. Long-term complications of congenital esophageal atresia and/ or tracheoesophageal fistula. Chest 2004;126:915–25. [6] Rutter MJ, Cotton RT, Azizkhan RG, et al. Slide tracheoplasty for the management of complete tracheal rings. J Pediatr Surg 2003;38:928–34. [7] Elliott MJ, Speggiorin S, Vida VL, et al. Slide tracheoplasty as a rescue technique after unsuccessful patch tracheoplasty. Ann Thorac Surg 2009;88:1029–31. [8] Tsai JY, Berkery L, Wesson DE, et al. Esophageal atresia and tracheoesophageal fistula: surgical experience over two decades. Ann Thorac Surg 1997;64:778–83 [discussion 783–774]. [9] Le Bret E, Roubertie F, Roger G. Slide tracheoplasty in infant with congenital tracheal stenosis and tracheomalacia after esophageal atresia with tracheoesophageal fistula repair. J Pediatr Surg 2009;44:e15–7. [10] Anton-Pacheco JL, Kalicinski P, Kansy A, et al. Slide tracheoplasty in an infant with congenital tracheal stenosis and oesophageal atresia with tracheoesophageal fistula. Eur J Cardiothorac Surg 2012;42:892–3.