Diseases of the Esophagus (2014) ••, ••–•• DOI: 10.1111/dote.12179
Original article
Evaluation of aortopexy in the management of severe tracheomalacia after esophageal atresia repair E. Kay-Rivest, R. Baird, J.-M. Laberge, P. S. Puligandla The Montreal Children’s Hospital, Montreal, Quebec, Canada
SUMMARY. Severe tracheomalacia (TM) is a difficult problem in esophageal atresia (EA) patients. We reviewed our experience with aortopexy and other interventions for severe TM in this population. With review ethics board approval, a retrospective review of TM in postoperative EA patients was conducted (1989–2010). Demographics, perinatal, and surgical information regarding EA repair was collected. TM infants were analyzed for symptomatology, clinical severity, investigations, interventions, and outcomes. Data are presented as proportions or median(range). One hundred and thirty-two EA patients were reviewed. Most had type C atresia (87.3%), and 18 patients (13.6%) died. Twenty-five patients (18.9%) had TM of whom five (20%) died. Median symptom onset was 18 days (0–729) after EA repair, with stridor (64%) or retractions/distress (44%) being most frequent. Four and two patients had airway obstruction or cardiorespiratory arrest, respectively. Median time from symptom onset to investigations was 11 days; these were most commonly rigid bronchoscopy (56%) and fluoroscopy (36%). Ten patients (40%) had severe TM on bronchoscopy. Six underwent aortopexy, one fundoplication, and three were treated medically. Length of hospital stay (LOS) post-aortopexy was 13 days (5–60), and ventilation time was 2 days (0–9). LOS was 60.5 (1–69) days postdiagnosis in non-aortopexy patients. Readmission rates for respiratory issues were significantly less in the aortopexy (median 0 vs. 5; P = 0.048) group over 2-year follow up after discharge. Complications of aortopexy included transfusion (1) and temporary diaphragmatic paresis (1), and one mortality secondary to severe congenital cardiac anomalies. Our experience suggests that aortopexy is safe and effective for the treatment of severe TM. It is associated with reduced LOS compared with other treatment strategies and few complications or long-term sequelae. KEY WORDS: aortopexy, esophageal atresia, severe tracheomalacia.
INTRODUCTION Esophageal atresia (EA) is a congenital malformation of the esophagus that affects 1 in 3000 infants and is characterized by maldevelopment of the esophagus such that it does not connect to the stomach.1 A common finding among EA patients is tracheomalacia (TM), a condition in which the tracheal cartilage is ‘floppy’ or deficient.2–4 TM may be classified into mild, moderate, and severe types based on clinical, radiological, and investigational findings.5 In its most severe forms, TM can lead to life-threatening apneic spells and even death. Although most cases of mild and moderate TM resolve without intervention,2,6 the Address correspondence to: Dr Pramod S. Puligandla, MD, MSC, FRCSC, FACS, FAAP, Division of Pediatric General and Thoracic Surgery, Montreal Children’s Hospital, 2300 Tupper, Room C811, Montreal, QC, Canada H3H 1P3. Email: [email protected] © 2014 International Society for Diseases of the Esophagus
management of patients with severe TM is more controversial. Among the treatment options that have been described in the literature for severe TM, aortopexy seems to be the most successful.4 The aim of this study was to evaluate our single institution experience with aortopexy in infants with severe TM who had undergone prior EA repair, and compare them in terms of safety and effectiveness to patients treated using alternative strategies.
MATERIALS AND METHODS After obtaining ethics review board approval, we performed a retrospective chart review of all patients treated at the Montreal Children’s Hospital who underwent surgical correction of EA from 1989 to 2010 (n = 132). We excluded infants with EA that did not receive operative correction at our center. 1
2
Diseases of the Esophagus
Table 1 General demographics of patients after esophageal atresia (EA) repair Group
Number
Sex
Birth weight
Gestational age
Age at EA repair
Patients with Type C EA
Mortality
EA
132
74 : 58 M:F 6:4 M:F
2810 g (620–4560)
38 weeks (29.5–42.5)
1 day (1–180)
115 (87%)
18 (14%)
3083 g (1615–4050)
38 weeks (33.5–41.2)
1 day (1)
sTM
10
10 (100%)
1 (10%)
Data are presented as median (range) or absolute value and percentage. EA, entire esophageal atresia cohort; F, female; M, male; sTM, those with severe tracheomalacia based on investigations.
Patients with EA were co-managed primarily by neonatology and pediatric surgery, with subspecialty involvement based on patient evolution, and not in a systematic manner. All repairs were preferentially performed via an extrapleural thoracotomy, except in the rare circumstance of multiple interventions. In the vast majority of patients, a rigid bronchoscopy was performed during spontaneous respiration immediately prior to operative repair to assess fistula location and the presence/absence of other laryngotracheal anomalies. Postoperative care was protocolized since 2006, and structured follow up was organized via a multidisciplinary clinic. For the study cohort, we collected data on demographics, disease type, associated malformations, or syndromes as well as the operative details of EA repair and postoperative complications. Within this cohort, we identified patients diagnosed with TM as listed in the discharge summary (n = 25). These patients were then segregated based on their symptoms into mild (croup and bronchiolitis), moderate (wheezing, stridor, respiratory infection, and cyanosis), and severe (stridor with tidal breathing, retention of secretions, upper airway obstruction, reflex apnea, cardiac arrest, and dying spells) groups as defined by previous investigators.2,5,7,8 This classification system was used to inform the urgency of future investigations. Radiologic and/or bronchoscopic means were used to assess for coaptation of the tracheal walls during spontaneous respiration and confirmed the severity of TM. Among the patients diagnosed with TM, we focused on those labeled as ‘severe’ (n = 10) and analyzed their treatment course and outcomes. If they received an aortopexy, we reviewed the intervention itself, the complications, and the overall outcomes. No patient with mild or moderate TM confirmed with radiologic or bronchoscopic evaluation (non-coaptation of tracheal walls) underwent aortopexy. All aortopexies at our institution were performed via a limited left anterior ‘Chamberlain’ thoracotomy, followed by thymectomy and aortopexy using permanent trans-sternal sutures. Concomitant intraoperative bronchoscopy was used to confirm effectiveness (no further coaptation). In the patients treated by other means, we looked at the number of readmissions for respiratory issues and the length of stay after TM diagnosis as surrogate indicators of success of these
alternative strategies. The primary outcome variable in our study was mortality while secondary variables included length of stay, readmission rates for respiratory issues after discharge, and complications related to bronchoscopy or alternative treatments for TM.
RESULTS The demographic data for the total EA cohort (n = 132) is presented in Table 1. The median gestational age and birth weight for the cohort were 38 weeks and 2810 g, respectively. Almost 90% of infants were diagnosed with a proximal EA and distal tracheoesophageal fistula (type C). The overall mortality in the cohort was 13.6%. Sixteen infants had an anastomotic leak (12%), and 19 (14%) required more than one esophageal dilatation within the first year postoperatively. Twenty-five patients (19%) within the EA cohort were diagnosed with TM. Of these with any form of TM, whether mild, moderate, or severe, five expired, including three who died soon after the diagnosis of TM but before any intervention. These three patients were classified as either mild or moderate TM based on initial investigations using bronchoscopy and/ or fluoroscopy. While the exact cause of death could not be directly attributed to TM, the final cause of death was reported as ‘cardiorespiratory arrest’. The remaining two deceased patients had complex congenital heart disease. One of these cardiac patients had investigations that indicated only mild to moderate TM (i.e. no coaptation of tracheal walls). The fifth patient had severe TM and underwent aortopexy (see below). The most common presenting signs and symptoms of TM, irrespective of severity, were stridor when irritated (16/25; 64%), chest wall retractions (11/25; 44%), and stridor with feeding (10/25; 40%). The most frequent investigations used to diagnose TM were rigid bronchoscopy (14/25; 56%) and fluoroscopy (9/25; 36%), with one child undergoing both investigations. These investigations were performed after the patients developed the first symptoms of TM. The median delay between the first symptoms of TM and investigation was 11 days (range of 0–1497 days). The patient for whom the interval was very © 2014 International Society for Diseases of the Esophagus
Aortopexy in severe tracheomalacia
long presented with one episode of perioral cyanosis early in life, but no further action was taken until years later when a bronchoscopy was performed because of persistent respiratory difficulties. The median delay between EA repair and first symptoms of TM was 18 days (range of 0–729 days), with one premature patient with a long-gap type C EA developing symptoms of TM before definitive repair of EA at three months of age. Ten (40%) patients within the TM cohort were categorized as ‘severe’ based on their symptomatology and confirmatory investigations (Table 2). Two of these 10 infants had an anastomotic leak that resolved without surgery, both of which ultimately proceeded to aortopexy. Two different infants with severe TM also required more than one esophageal dilatation in the first year postoperatively. Of the 10 patients with severe TM, six were treated by aortopexy (see Fig. 1). Four of the six patients that underwent aortopexy occurred after the establishment of our multidisciplinary EA clinic in 2006; three fourths of patients managed without aortopexy were treated prior to this time. Half of the patients (3/6) treated with aortopexy had VACTERL association. Furthermore, 4/6 patients treated with aortopexy had at least one episode of perioral cyanosis prior to the intervention. Complications after aortopexy included one patient with bleeding that required a single transfusion and one patient with temporary diaphragmatic paresis that resolved over several months. One patient died as a result of surgical complications from complex congenital heart disease (single ventricle and coarctation of the aorta) 3 months after aortopexy. All patients who underwent aortopexy had complete resolution of their symptoms postoperatively. The short-term outcomes after aortopexy were favorable. Patients were ventilated for a median 2 days (range of 0–9 days) and oral feeding was resumed similarly (median 2 days; range of 1–13 days). The median length of stay in hospital for aortopexy patients was 13 days (range of 5–60 days). The remaining four patients with severe TM who did not undergo aortopexy had a median length of hospital stay of 60.5 days (range of 1–69 days). One of these patients was treated by continuous positive airway pressure, two were treated medically for their gastroesophageal reflux disease (GERD), and the last patient was treated by Nissen fundoplication for GERD. Importantly, significantly reduced readmission rates for respiratory symptoms were noted in the aortopexy group (median 0 [0–3] vs. 5 [0–10]; P = 0.0481) in the 2-year follow-up period after aortopexy.
DISCUSSION TM is a relatively common but potentially lifethreatening problem that has been reported to occur © 2014 International Society for Diseases of the Esophagus
3
in association with EA since the 1970s.9 The pathophysiological implications of TM relate to the ‘softness’ of the trachea and its potential for collapse, thereby leading to dynamic airway obstruction. While the majority of patients with mild or moderate TM can be managed expectantly, those infants with severe TM represent a special population of patients for whom surgical intervention is likely indicated. We defined severe TM in our study as stridor with tidal breathing and significant upper airway obstruction leading to reflex apnea, perioral cyanosis, ‘dying spells’, and/or cardiac arrest. Our preferred approach for these patients is to perform an aortopexy, which we have demonstrated in this study to have acceptable risks and the complete resolution of symptoms. The decision to intervene operatively for pediatric patients with TM of any etiology has been debated in the literature. A recent Cochrane review failed to identify any Level I evidence supporting the routine use of surgery for severe TM.10 However, the authors did support surgery as part of a tailored, patientspecific approach to the management of severe TM based on several case series where aortopexy was used. In our review of 132 infants with EA repaired at our institution, approximately one fifth presented with some form of TM, a frequency that compares favorably with other published reports.4,11,12 While many of these patients did not require any specific treatment, 10 infants with TM were classified as severe, representing 7.6% of the total EA cohort, and did receive additional therapeutic interventions. All of these patients had persistent symptoms such as episodes of perioral cyanosis, and many had other associated anomalies (i.e. VACTERL), or both. Six of these infants underwent aortopexy after confirming coaptation (i.e. ‘kissing’) of the tracheal walls with spontaneous respiration during rigid bronchoscopy. While the small number of patients in this study precluded robust statistical analyses, the overall effects of aortopexy were very positive with weaning from ventilation and the resumption of enteral feeding within 2 days of the procedure in the majority of patients. Most importantly, all patients had complete resolution of their respiratory symptoms, including episodes of cyanosis or apnea. Moreover, those infants with severe TM who did not receive aortopexy had longer lengths of stay and, more significantly, more visits/admissions to hospital for respiratory issues in the 2 years after discharge from hospital. Thus, our results suggest that while aortopexy may be considered an ‘invasive’ means of treatment, it is associated with acceptable risks13 and long-term positive benefits for these patients and their families. The reason why TM occurs is unclear because the symptoms of TM are rarely identified prior to EA repair. Earlier publications have suggested that
M
F M
M M
M
F
F
F
2
3 4
5 6
7
8
9
10
3560
4050
2085
2340
3200 1615
3135 3740
2820
3030
BW (g)
EATEF
EATEF Fryn’s omphalocele EATEF laryngeal cleft
EATEF
EATEF VACTERL EATEF single ventricle CoA EATEF VACTERL EATEF
EATEF VACTERL laryngeal cleft
EATEF
Associated diagnoses
Perioral cyanosis, airway obstruction, stridor Stridor, retractions
Perioral cyanosis, stridor with feeding Stridor, barking cough, recurrent airway infection Perioral cyanosis Perioral cyanosis, barking cough, stridor Perioral cyanosis, stridor Airway obstruction, stridor wheezing Wheezing, retractions, stridor, desaturations Respiratory arrest
Presenting symptoms of severe
Characteristics of infants with severe tracheomalacia
25
81
10
9
11 6
6 13
26
1
Interval between symptoms and investigation (d)
CPAP
Antireflux meds
Nissen
Antireflux Meds
Aortopexy Aortopexy
Aortopexy Aortopexy
Aortopexy
Aortopexy
Treatment
–
–
–
–
28 68
19 85
83
92
Age at aortopexy (d)
–
–
–
Diaphragmatic paresis Died from cardiac complications None Bleeding requiring transfusion –
None
None
Complications of aortopexy
34
–
–
–
1 2
9 2
2
0
Ventilation post-aortopexy or other intervention (d)
60
1†
69
61
5 23
60 Until death
13
8
LOS (d)
All patients had repair of a proximal esophageal atresia and distal tracheoesophageal fistula. †This patient was evaluated as an outpatient. Despite being diagnosed with severe tracheomalacia, the treating team decided against any further treatment, accounting for the LOS of 1 day. However, this patient required eight readmissions for respiratory difficulties. BW, birth weight; CoA, coarctation of aorta; CPAP, continuous positive airway pressure; d, days; EATEF, esophageal atresia with tracheoesophageal atresia; F, female; LOS, length of stay after aortopexy in days; M, male; VACTERL, VACTERL association.
M
1
Sex
Table 2
4 Diseases of the Esophagus
© 2014 International Society for Diseases of the Esophagus
Aortopexy in severe tracheomalacia
5
Fig. 1 Proposed algorithm for patients after esophageal repair who experience respiratory difficulties (also applicable to patients experiencing feeding difficulties). TM, tracheomalacia.
patients needing aortopexy experienced more complications after EA repair (anastomotic leaks and strictures) or had longer gaps between the two esophageal ends.9,13,14 Our results do not support these findings as the incidence of esophageal leaks and strictures requiring more than one dilation in the first year postdischarge were similar between the severe TM group and the rest of the EA cohort evaluated in this study, part of which was published in detail previously.15 While TM in EA has traditionally been thought to be the result of a pre-existing anatomical problem affecting the trachea,6 it is unclear if operative technique and extensive dissection may contribute to its development. Moreover, increased inspiratory efforts during oral feeding, combined © 2014 International Society for Diseases of the Esophagus
with distension of the upper esophageal pouch, could also contribute to collapse of an already weakened tracheal wall. In the postoperative period after EA repair, physicians are often faced with an infant that has difficulty feeding that may or may not be associated with coughing, oxygen desaturation, regurgitation, respiratory distress, or cyanosis. Indeed, it is often very difficult to discriminate feeding difficulties from GERD or other significant postoperative problems commonly experienced by infants with EA (Fig. 1). Based on our experience, we suggest that these infants have a contrast esophagogram as an initial study to rule out postoperative stricture formation and/or refistulization. On the lateral views, attention is also
6
Diseases of the Esophagus
paid to any fluctuation in the diameter of the tracheal air column during breathing or swallowing. If this study is negative or equivocal, we suggest that flexible or rigid bronchoscopy be performed expeditiously as the next investigation. This is of particular importance if episodes of perioral cyanosis or acute lifethreatening events are reported despite ongoing medical therapy for GERD, which we suggest be started in the immediate postoperative period.16,17 Ideally, bronchoscopy should be performed with the infant spontaneously breathing and can be accomplished by using an intravenous induction technique under the guidance of an experienced pediatric anesthesiologist. Bronchoscopy will help rule out a recurrent fistula and also identify severe TM. While there is no consensus definition of the bronchoscopic findings and severity of TM, we define severe TM as near (>80%) to complete reduction of the anteroposterior tracheal diameter during expiration. Bronchoscopy can also be useful in identifying other laryngotracheal anomalies such as laryngeal clefts that may have been missed at the time of EA repair. In our center, if severe TM is confirmed at the time of bronchoscopy, we generally proceed with immediate aortopexy under the same general anesthesia. While a standard anterior approach provides excellent exposure, there is increasing experience with a thoracoscopic technique that appears to provide similar results and excellent visualization.13 In the event that there is no significant TM or evidence of refistulization, we suggest maximizing the medical management of GERD and considering fundoplication for those infants with ongoing symptoms. We acknowledge that this study is limited by its retrospective nature and small sample size. We also recognize that there has been a change in our management practices over time, particularly over the last 7 years during which we have developed a fully multidisciplinary EA clinic that includes pediatric surgeons, gastroenterologists, pediatricians, and other subspecialty services such as respirology, cardiology, and otolaryngology. The resultant harmonization of practice has led to aggressive investigation for TM in symptomatic patients and early treatment once identified. Indeed, the management of these patients early in our series was subject to individual physician/ service biases where later practice involved consensus building. The consequent evolution in management philosophy favoring aortopexy, as seen by the increasing trend in our dataset over time, appears to be
associated with superior patient outcomes. Larger multicenter studies may aid in further clarifying this issue. In conclusion, aortopexy is a safe and effective means of treating severe TM. While stricture, refistulization, and GERD are problems that can potentially affect all patients after EA repair, our algorithm provides a straightforward approach to the infant with feeding and respiratory difficulties in the postoperative period. We believe that aortopexy is likely indicated in all patients with severe TM and can be performed with minimal complications and positive lasting effects. References 1 Laberge J M, Guttman F M. Esophageal atresia and related tracheoesophageal anomalies. In: Donnellan W, (ed.). Abdominal Surgery of Infancy and Childhood. Luxembourg: Harwood Academic Publishers GmbH, 1996; 11/1–34. 2 McNamara V M, Crabbe D C G. Tracheomalacia. Paediatr Respir Rev 2004; 5: 147–54. 3 Weber T R, Keller M S, Fiore A. Aortic suspension (aortopexy) for severe tracheomalacia in infants and children. Am J Surg 2002; 184: 573–7. 4 Chetcuti P, Phelan P D. Respiratory morbidity after repair of oesophageal atresia and tracheo-oesophageal fistula. Arch Dis Child 1993; 68: 167–70. 5 Benjamin B, Cohen D, Glaason M. Tracheomalacia in association with tracheoesophgeal fistula. Surgery 1976; 79: 504–8. 6 Wailoo M P, Emery J L. The trachea in children with tracheoesophageal fistula. Histopathology 1979; 3: 329–38. 7 Malone P S, Kiely E M. Role of aortopexy in the management of primary tracheomalacia and tracheobronchomalacia. Arch Dis Child 1990; 65: 238–440. 8 Fayoux P, Sfeir R. Management of severe tracheomalacia. J Pediatr Gastroenterol Nutr 2011; 52: 533–4. 9 Filler R M, Messineo A, Vinograd I. Severe tracheomalacia associated with esophageal atresia: results of surgical treatment. J Pediatr Surg 1992; 27: 1136–40. 10 Masters I B, Chang A B. Interventions for primary (intrinsic) tracheomalacia in children. Cochrane Database Syst Rev 2005; 4: CD005304. 11 Nasr A, Ein S H, Gerstle J T. Infants with repaired esophageal atresia and distal tracheoesophageal fistula with severe respiratory distress: is it tracheomalacia, reflux, or both? J Pediatr Surg 2005; 40: 901–3. 12 Spitz L, Kiely E, Brereton R J. Esophageal atresia: five year experience with 148 cases. J Pediatr Surg 1987; 22: 103–8. 13 Dave S, Currie B G. The role of aortopexy in severe tracheomalacia. J Pediatr Surg 2006; 41: 533–7. 14 Corbally M T, Spitz L, Kiely E, Brereton R J, Drake D P. Aortopexy for tracheomalacia in oesophageal anomalies. Eur J Pediatr Surg 1993; 3: 264–6. 15 AlShehri A, Lo A, Baird R. An analysis of early non-mortality outcome prediction in esophageal atresia. J Pediatr Surg 2012; 47: 881–4. 16 Tovar J A, Fragoso A C. Gastroesophageal reflux after repair of esophageal atresia. Eur J Pediatr Surg 2013; 23: 175–81. 17 Baird R, Laberge J M, Lévesque D. Anastomotic stricture after esophageal atresia repair: a critical review of recent literature. Eur J Pediatr Surg 2013; 23: 204–13.
© 2014 International Society for Diseases of the Esophagus
Original article
Evaluation of aortopexy in the management of severe tracheomalacia after esophageal atresia repair E. Kay-Rivest, R. Baird, J.-M. Laberge, P. S. Puligandla The Montreal Children’s Hospital, Montreal, Quebec, Canada
SUMMARY. Severe tracheomalacia (TM) is a difficult problem in esophageal atresia (EA) patients. We reviewed our experience with aortopexy and other interventions for severe TM in this population. With review ethics board approval, a retrospective review of TM in postoperative EA patients was conducted (1989–2010). Demographics, perinatal, and surgical information regarding EA repair was collected. TM infants were analyzed for symptomatology, clinical severity, investigations, interventions, and outcomes. Data are presented as proportions or median(range). One hundred and thirty-two EA patients were reviewed. Most had type C atresia (87.3%), and 18 patients (13.6%) died. Twenty-five patients (18.9%) had TM of whom five (20%) died. Median symptom onset was 18 days (0–729) after EA repair, with stridor (64%) or retractions/distress (44%) being most frequent. Four and two patients had airway obstruction or cardiorespiratory arrest, respectively. Median time from symptom onset to investigations was 11 days; these were most commonly rigid bronchoscopy (56%) and fluoroscopy (36%). Ten patients (40%) had severe TM on bronchoscopy. Six underwent aortopexy, one fundoplication, and three were treated medically. Length of hospital stay (LOS) post-aortopexy was 13 days (5–60), and ventilation time was 2 days (0–9). LOS was 60.5 (1–69) days postdiagnosis in non-aortopexy patients. Readmission rates for respiratory issues were significantly less in the aortopexy (median 0 vs. 5; P = 0.048) group over 2-year follow up after discharge. Complications of aortopexy included transfusion (1) and temporary diaphragmatic paresis (1), and one mortality secondary to severe congenital cardiac anomalies. Our experience suggests that aortopexy is safe and effective for the treatment of severe TM. It is associated with reduced LOS compared with other treatment strategies and few complications or long-term sequelae. KEY WORDS: aortopexy, esophageal atresia, severe tracheomalacia.
INTRODUCTION Esophageal atresia (EA) is a congenital malformation of the esophagus that affects 1 in 3000 infants and is characterized by maldevelopment of the esophagus such that it does not connect to the stomach.1 A common finding among EA patients is tracheomalacia (TM), a condition in which the tracheal cartilage is ‘floppy’ or deficient.2–4 TM may be classified into mild, moderate, and severe types based on clinical, radiological, and investigational findings.5 In its most severe forms, TM can lead to life-threatening apneic spells and even death. Although most cases of mild and moderate TM resolve without intervention,2,6 the Address correspondence to: Dr Pramod S. Puligandla, MD, MSC, FRCSC, FACS, FAAP, Division of Pediatric General and Thoracic Surgery, Montreal Children’s Hospital, 2300 Tupper, Room C811, Montreal, QC, Canada H3H 1P3. Email: [email protected] © 2014 International Society for Diseases of the Esophagus
management of patients with severe TM is more controversial. Among the treatment options that have been described in the literature for severe TM, aortopexy seems to be the most successful.4 The aim of this study was to evaluate our single institution experience with aortopexy in infants with severe TM who had undergone prior EA repair, and compare them in terms of safety and effectiveness to patients treated using alternative strategies.
MATERIALS AND METHODS After obtaining ethics review board approval, we performed a retrospective chart review of all patients treated at the Montreal Children’s Hospital who underwent surgical correction of EA from 1989 to 2010 (n = 132). We excluded infants with EA that did not receive operative correction at our center. 1
2
Diseases of the Esophagus
Table 1 General demographics of patients after esophageal atresia (EA) repair Group
Number
Sex
Birth weight
Gestational age
Age at EA repair
Patients with Type C EA
Mortality
EA
132
74 : 58 M:F 6:4 M:F
2810 g (620–4560)
38 weeks (29.5–42.5)
1 day (1–180)
115 (87%)
18 (14%)
3083 g (1615–4050)
38 weeks (33.5–41.2)
1 day (1)
sTM
10
10 (100%)
1 (10%)
Data are presented as median (range) or absolute value and percentage. EA, entire esophageal atresia cohort; F, female; M, male; sTM, those with severe tracheomalacia based on investigations.
Patients with EA were co-managed primarily by neonatology and pediatric surgery, with subspecialty involvement based on patient evolution, and not in a systematic manner. All repairs were preferentially performed via an extrapleural thoracotomy, except in the rare circumstance of multiple interventions. In the vast majority of patients, a rigid bronchoscopy was performed during spontaneous respiration immediately prior to operative repair to assess fistula location and the presence/absence of other laryngotracheal anomalies. Postoperative care was protocolized since 2006, and structured follow up was organized via a multidisciplinary clinic. For the study cohort, we collected data on demographics, disease type, associated malformations, or syndromes as well as the operative details of EA repair and postoperative complications. Within this cohort, we identified patients diagnosed with TM as listed in the discharge summary (n = 25). These patients were then segregated based on their symptoms into mild (croup and bronchiolitis), moderate (wheezing, stridor, respiratory infection, and cyanosis), and severe (stridor with tidal breathing, retention of secretions, upper airway obstruction, reflex apnea, cardiac arrest, and dying spells) groups as defined by previous investigators.2,5,7,8 This classification system was used to inform the urgency of future investigations. Radiologic and/or bronchoscopic means were used to assess for coaptation of the tracheal walls during spontaneous respiration and confirmed the severity of TM. Among the patients diagnosed with TM, we focused on those labeled as ‘severe’ (n = 10) and analyzed their treatment course and outcomes. If they received an aortopexy, we reviewed the intervention itself, the complications, and the overall outcomes. No patient with mild or moderate TM confirmed with radiologic or bronchoscopic evaluation (non-coaptation of tracheal walls) underwent aortopexy. All aortopexies at our institution were performed via a limited left anterior ‘Chamberlain’ thoracotomy, followed by thymectomy and aortopexy using permanent trans-sternal sutures. Concomitant intraoperative bronchoscopy was used to confirm effectiveness (no further coaptation). In the patients treated by other means, we looked at the number of readmissions for respiratory issues and the length of stay after TM diagnosis as surrogate indicators of success of these
alternative strategies. The primary outcome variable in our study was mortality while secondary variables included length of stay, readmission rates for respiratory issues after discharge, and complications related to bronchoscopy or alternative treatments for TM.
RESULTS The demographic data for the total EA cohort (n = 132) is presented in Table 1. The median gestational age and birth weight for the cohort were 38 weeks and 2810 g, respectively. Almost 90% of infants were diagnosed with a proximal EA and distal tracheoesophageal fistula (type C). The overall mortality in the cohort was 13.6%. Sixteen infants had an anastomotic leak (12%), and 19 (14%) required more than one esophageal dilatation within the first year postoperatively. Twenty-five patients (19%) within the EA cohort were diagnosed with TM. Of these with any form of TM, whether mild, moderate, or severe, five expired, including three who died soon after the diagnosis of TM but before any intervention. These three patients were classified as either mild or moderate TM based on initial investigations using bronchoscopy and/ or fluoroscopy. While the exact cause of death could not be directly attributed to TM, the final cause of death was reported as ‘cardiorespiratory arrest’. The remaining two deceased patients had complex congenital heart disease. One of these cardiac patients had investigations that indicated only mild to moderate TM (i.e. no coaptation of tracheal walls). The fifth patient had severe TM and underwent aortopexy (see below). The most common presenting signs and symptoms of TM, irrespective of severity, were stridor when irritated (16/25; 64%), chest wall retractions (11/25; 44%), and stridor with feeding (10/25; 40%). The most frequent investigations used to diagnose TM were rigid bronchoscopy (14/25; 56%) and fluoroscopy (9/25; 36%), with one child undergoing both investigations. These investigations were performed after the patients developed the first symptoms of TM. The median delay between the first symptoms of TM and investigation was 11 days (range of 0–1497 days). The patient for whom the interval was very © 2014 International Society for Diseases of the Esophagus
Aortopexy in severe tracheomalacia
long presented with one episode of perioral cyanosis early in life, but no further action was taken until years later when a bronchoscopy was performed because of persistent respiratory difficulties. The median delay between EA repair and first symptoms of TM was 18 days (range of 0–729 days), with one premature patient with a long-gap type C EA developing symptoms of TM before definitive repair of EA at three months of age. Ten (40%) patients within the TM cohort were categorized as ‘severe’ based on their symptomatology and confirmatory investigations (Table 2). Two of these 10 infants had an anastomotic leak that resolved without surgery, both of which ultimately proceeded to aortopexy. Two different infants with severe TM also required more than one esophageal dilatation in the first year postoperatively. Of the 10 patients with severe TM, six were treated by aortopexy (see Fig. 1). Four of the six patients that underwent aortopexy occurred after the establishment of our multidisciplinary EA clinic in 2006; three fourths of patients managed without aortopexy were treated prior to this time. Half of the patients (3/6) treated with aortopexy had VACTERL association. Furthermore, 4/6 patients treated with aortopexy had at least one episode of perioral cyanosis prior to the intervention. Complications after aortopexy included one patient with bleeding that required a single transfusion and one patient with temporary diaphragmatic paresis that resolved over several months. One patient died as a result of surgical complications from complex congenital heart disease (single ventricle and coarctation of the aorta) 3 months after aortopexy. All patients who underwent aortopexy had complete resolution of their symptoms postoperatively. The short-term outcomes after aortopexy were favorable. Patients were ventilated for a median 2 days (range of 0–9 days) and oral feeding was resumed similarly (median 2 days; range of 1–13 days). The median length of stay in hospital for aortopexy patients was 13 days (range of 5–60 days). The remaining four patients with severe TM who did not undergo aortopexy had a median length of hospital stay of 60.5 days (range of 1–69 days). One of these patients was treated by continuous positive airway pressure, two were treated medically for their gastroesophageal reflux disease (GERD), and the last patient was treated by Nissen fundoplication for GERD. Importantly, significantly reduced readmission rates for respiratory symptoms were noted in the aortopexy group (median 0 [0–3] vs. 5 [0–10]; P = 0.0481) in the 2-year follow-up period after aortopexy.
DISCUSSION TM is a relatively common but potentially lifethreatening problem that has been reported to occur © 2014 International Society for Diseases of the Esophagus
3
in association with EA since the 1970s.9 The pathophysiological implications of TM relate to the ‘softness’ of the trachea and its potential for collapse, thereby leading to dynamic airway obstruction. While the majority of patients with mild or moderate TM can be managed expectantly, those infants with severe TM represent a special population of patients for whom surgical intervention is likely indicated. We defined severe TM in our study as stridor with tidal breathing and significant upper airway obstruction leading to reflex apnea, perioral cyanosis, ‘dying spells’, and/or cardiac arrest. Our preferred approach for these patients is to perform an aortopexy, which we have demonstrated in this study to have acceptable risks and the complete resolution of symptoms. The decision to intervene operatively for pediatric patients with TM of any etiology has been debated in the literature. A recent Cochrane review failed to identify any Level I evidence supporting the routine use of surgery for severe TM.10 However, the authors did support surgery as part of a tailored, patientspecific approach to the management of severe TM based on several case series where aortopexy was used. In our review of 132 infants with EA repaired at our institution, approximately one fifth presented with some form of TM, a frequency that compares favorably with other published reports.4,11,12 While many of these patients did not require any specific treatment, 10 infants with TM were classified as severe, representing 7.6% of the total EA cohort, and did receive additional therapeutic interventions. All of these patients had persistent symptoms such as episodes of perioral cyanosis, and many had other associated anomalies (i.e. VACTERL), or both. Six of these infants underwent aortopexy after confirming coaptation (i.e. ‘kissing’) of the tracheal walls with spontaneous respiration during rigid bronchoscopy. While the small number of patients in this study precluded robust statistical analyses, the overall effects of aortopexy were very positive with weaning from ventilation and the resumption of enteral feeding within 2 days of the procedure in the majority of patients. Most importantly, all patients had complete resolution of their respiratory symptoms, including episodes of cyanosis or apnea. Moreover, those infants with severe TM who did not receive aortopexy had longer lengths of stay and, more significantly, more visits/admissions to hospital for respiratory issues in the 2 years after discharge from hospital. Thus, our results suggest that while aortopexy may be considered an ‘invasive’ means of treatment, it is associated with acceptable risks13 and long-term positive benefits for these patients and their families. The reason why TM occurs is unclear because the symptoms of TM are rarely identified prior to EA repair. Earlier publications have suggested that
M
F M
M M
M
F
F
F
2
3 4
5 6
7
8
9
10
3560
4050
2085
2340
3200 1615
3135 3740
2820
3030
BW (g)
EATEF
EATEF Fryn’s omphalocele EATEF laryngeal cleft
EATEF
EATEF VACTERL EATEF single ventricle CoA EATEF VACTERL EATEF
EATEF VACTERL laryngeal cleft
EATEF
Associated diagnoses
Perioral cyanosis, airway obstruction, stridor Stridor, retractions
Perioral cyanosis, stridor with feeding Stridor, barking cough, recurrent airway infection Perioral cyanosis Perioral cyanosis, barking cough, stridor Perioral cyanosis, stridor Airway obstruction, stridor wheezing Wheezing, retractions, stridor, desaturations Respiratory arrest
Presenting symptoms of severe
Characteristics of infants with severe tracheomalacia
25
81
10
9
11 6
6 13
26
1
Interval between symptoms and investigation (d)
CPAP
Antireflux meds
Nissen
Antireflux Meds
Aortopexy Aortopexy
Aortopexy Aortopexy
Aortopexy
Aortopexy
Treatment
–
–
–
–
28 68
19 85
83
92
Age at aortopexy (d)
–
–
–
Diaphragmatic paresis Died from cardiac complications None Bleeding requiring transfusion –
None
None
Complications of aortopexy
34
–
–
–
1 2
9 2
2
0
Ventilation post-aortopexy or other intervention (d)
60
1†
69
61
5 23
60 Until death
13
8
LOS (d)
All patients had repair of a proximal esophageal atresia and distal tracheoesophageal fistula. †This patient was evaluated as an outpatient. Despite being diagnosed with severe tracheomalacia, the treating team decided against any further treatment, accounting for the LOS of 1 day. However, this patient required eight readmissions for respiratory difficulties. BW, birth weight; CoA, coarctation of aorta; CPAP, continuous positive airway pressure; d, days; EATEF, esophageal atresia with tracheoesophageal atresia; F, female; LOS, length of stay after aortopexy in days; M, male; VACTERL, VACTERL association.
M
1
Sex
Table 2
4 Diseases of the Esophagus
© 2014 International Society for Diseases of the Esophagus
Aortopexy in severe tracheomalacia
5
Fig. 1 Proposed algorithm for patients after esophageal repair who experience respiratory difficulties (also applicable to patients experiencing feeding difficulties). TM, tracheomalacia.
patients needing aortopexy experienced more complications after EA repair (anastomotic leaks and strictures) or had longer gaps between the two esophageal ends.9,13,14 Our results do not support these findings as the incidence of esophageal leaks and strictures requiring more than one dilation in the first year postdischarge were similar between the severe TM group and the rest of the EA cohort evaluated in this study, part of which was published in detail previously.15 While TM in EA has traditionally been thought to be the result of a pre-existing anatomical problem affecting the trachea,6 it is unclear if operative technique and extensive dissection may contribute to its development. Moreover, increased inspiratory efforts during oral feeding, combined © 2014 International Society for Diseases of the Esophagus
with distension of the upper esophageal pouch, could also contribute to collapse of an already weakened tracheal wall. In the postoperative period after EA repair, physicians are often faced with an infant that has difficulty feeding that may or may not be associated with coughing, oxygen desaturation, regurgitation, respiratory distress, or cyanosis. Indeed, it is often very difficult to discriminate feeding difficulties from GERD or other significant postoperative problems commonly experienced by infants with EA (Fig. 1). Based on our experience, we suggest that these infants have a contrast esophagogram as an initial study to rule out postoperative stricture formation and/or refistulization. On the lateral views, attention is also
6
Diseases of the Esophagus
paid to any fluctuation in the diameter of the tracheal air column during breathing or swallowing. If this study is negative or equivocal, we suggest that flexible or rigid bronchoscopy be performed expeditiously as the next investigation. This is of particular importance if episodes of perioral cyanosis or acute lifethreatening events are reported despite ongoing medical therapy for GERD, which we suggest be started in the immediate postoperative period.16,17 Ideally, bronchoscopy should be performed with the infant spontaneously breathing and can be accomplished by using an intravenous induction technique under the guidance of an experienced pediatric anesthesiologist. Bronchoscopy will help rule out a recurrent fistula and also identify severe TM. While there is no consensus definition of the bronchoscopic findings and severity of TM, we define severe TM as near (>80%) to complete reduction of the anteroposterior tracheal diameter during expiration. Bronchoscopy can also be useful in identifying other laryngotracheal anomalies such as laryngeal clefts that may have been missed at the time of EA repair. In our center, if severe TM is confirmed at the time of bronchoscopy, we generally proceed with immediate aortopexy under the same general anesthesia. While a standard anterior approach provides excellent exposure, there is increasing experience with a thoracoscopic technique that appears to provide similar results and excellent visualization.13 In the event that there is no significant TM or evidence of refistulization, we suggest maximizing the medical management of GERD and considering fundoplication for those infants with ongoing symptoms. We acknowledge that this study is limited by its retrospective nature and small sample size. We also recognize that there has been a change in our management practices over time, particularly over the last 7 years during which we have developed a fully multidisciplinary EA clinic that includes pediatric surgeons, gastroenterologists, pediatricians, and other subspecialty services such as respirology, cardiology, and otolaryngology. The resultant harmonization of practice has led to aggressive investigation for TM in symptomatic patients and early treatment once identified. Indeed, the management of these patients early in our series was subject to individual physician/ service biases where later practice involved consensus building. The consequent evolution in management philosophy favoring aortopexy, as seen by the increasing trend in our dataset over time, appears to be
associated with superior patient outcomes. Larger multicenter studies may aid in further clarifying this issue. In conclusion, aortopexy is a safe and effective means of treating severe TM. While stricture, refistulization, and GERD are problems that can potentially affect all patients after EA repair, our algorithm provides a straightforward approach to the infant with feeding and respiratory difficulties in the postoperative period. We believe that aortopexy is likely indicated in all patients with severe TM and can be performed with minimal complications and positive lasting effects. References 1 Laberge J M, Guttman F M. Esophageal atresia and related tracheoesophageal anomalies. In: Donnellan W, (ed.). Abdominal Surgery of Infancy and Childhood. Luxembourg: Harwood Academic Publishers GmbH, 1996; 11/1–34. 2 McNamara V M, Crabbe D C G. Tracheomalacia. Paediatr Respir Rev 2004; 5: 147–54. 3 Weber T R, Keller M S, Fiore A. Aortic suspension (aortopexy) for severe tracheomalacia in infants and children. Am J Surg 2002; 184: 573–7. 4 Chetcuti P, Phelan P D. Respiratory morbidity after repair of oesophageal atresia and tracheo-oesophageal fistula. Arch Dis Child 1993; 68: 167–70. 5 Benjamin B, Cohen D, Glaason M. Tracheomalacia in association with tracheoesophgeal fistula. Surgery 1976; 79: 504–8. 6 Wailoo M P, Emery J L. The trachea in children with tracheoesophageal fistula. Histopathology 1979; 3: 329–38. 7 Malone P S, Kiely E M. Role of aortopexy in the management of primary tracheomalacia and tracheobronchomalacia. Arch Dis Child 1990; 65: 238–440. 8 Fayoux P, Sfeir R. Management of severe tracheomalacia. J Pediatr Gastroenterol Nutr 2011; 52: 533–4. 9 Filler R M, Messineo A, Vinograd I. Severe tracheomalacia associated with esophageal atresia: results of surgical treatment. J Pediatr Surg 1992; 27: 1136–40. 10 Masters I B, Chang A B. Interventions for primary (intrinsic) tracheomalacia in children. Cochrane Database Syst Rev 2005; 4: CD005304. 11 Nasr A, Ein S H, Gerstle J T. Infants with repaired esophageal atresia and distal tracheoesophageal fistula with severe respiratory distress: is it tracheomalacia, reflux, or both? J Pediatr Surg 2005; 40: 901–3. 12 Spitz L, Kiely E, Brereton R J. Esophageal atresia: five year experience with 148 cases. J Pediatr Surg 1987; 22: 103–8. 13 Dave S, Currie B G. The role of aortopexy in severe tracheomalacia. J Pediatr Surg 2006; 41: 533–7. 14 Corbally M T, Spitz L, Kiely E, Brereton R J, Drake D P. Aortopexy for tracheomalacia in oesophageal anomalies. Eur J Pediatr Surg 1993; 3: 264–6. 15 AlShehri A, Lo A, Baird R. An analysis of early non-mortality outcome prediction in esophageal atresia. J Pediatr Surg 2012; 47: 881–4. 16 Tovar J A, Fragoso A C. Gastroesophageal reflux after repair of esophageal atresia. Eur J Pediatr Surg 2013; 23: 175–81. 17 Baird R, Laberge J M, Lévesque D. Anastomotic stricture after esophageal atresia repair: a critical review of recent literature. Eur J Pediatr Surg 2013; 23: 204–13.
© 2014 International Society for Diseases of the Esophagus
