Ann Thorac Surg 2009;88:1970-1974. doi:10.1016/j.athoracsur.2009.08.039
© 2009 The Society of Thoracic Surgeons
Original Articles: Pediatric Cardiac
Influence of Tracheobronchomalacia on Outcome of Surgery in Children With Congenital Heart Disease and Its Management
Qiang Chen, MD, PhDa,
Simon Langton-Hewer, FRCPb,
Stephen Marriage, FRCPc,
Alison Hayes, FRCPd,
Massimo Caputo, MDa,
Ash Pawade, FRCSa,
Andrew J. Parry, FRCSa,*
a Department of Cardiac Surgery, Bristol Royal Hospital for Children, Bristol, United Kingdom
b Department of Respiratory Medicine, Bristol Royal Hospital for Children, Bristol, United Kingdom
c Department of Pediatric Intensive Care, Bristol Royal Hospital for Children, Bristol, United Kingdom
d Department of Pediatric Cardiology, Bristol Royal Hospital for Children, Bristol, United Kingdom
Accepted for publication August 17, 2009.
* Address correspondence to Dr Parry, Department of Congenital Cardiac Surgery, Bristol Royal Hospital for Children, Upper Maudlin St, Bristol, BS2 8HW, United Kingdom (Email: aj_parry{at}yahoo.co.uk).
 |
Abstract
|
|---|
Background: Patients with complex congenital heart disease associated with tracheobronchomalacia (TBM) remain difficult to manage after cardiac surgery. We studied the influence of TBM on the outcomes of pediatric patients after cardiac surgery for congenital heart disease to determine how to manage these patients better.
Methods: Twenty-two consecutive pediatric patients who had TBM diagnosed by bronchoscopy or dynamic contrast bronchography before or after cardiac surgery for congenital heart disease during a 5.5-year period were compared with an age- and procedure-matched control group operated on during the same period. Patients diagnosed postoperatively were investigated after a second failed extubation. Patients were managed by oxygen administration, endotracheal suctioning, and positive end-expiratory or continuous positive airway pressure through a nasotracheal tube or tracheostomy.
Results: There were 4 deaths within 1 year of surgery, all in the study group, with 2 early (neither of which appeared related to TBM) and 2 late. The estimated survival at 5 years was 82% (95% confidence interval, 59% to 93%) for the study group compared with 100% for control patients (p = 0.012). All deaths occurred in patients undergoing palliative procedures (p = 0.0004), and both children who underwent redo operations died (p = 0.02). Postoperatively, 50% of children with TBM required prolonged ventilation and tracheostomy. Compared with control patients the average postoperative ventilation time, pediatric intensive care unit stay, and hospital stay were 6.5, 11.5, and 20 days versus 1, 2, and 6.5 days, respectively (p < 0.001).
Conclusions: Although associated with longer postoperative ventilation time, pediatric intensive care unit stay, hospital stay, and mortality, outcomes after cardiac procedures in children with TBM are acceptable. Palliative and redo procedures in this group of patients are associated with significantly higher risk of death.
|
Introduction
|
|---|
Tracheobronchomalacia (TBM) is a rare condition characterized by weakness of the wall and supporting cartilage of the trachea and bronchi. It results in airway collapse during expiration, and may be a cause of respiratory distress in pediatric patients [1–3]. When it occurs in association with congenital heart disease, TBM has previously been considered to contribute to unfavorable outcome after cardiac surgical procedures owing to increased work of breathing, the requirement for prolonged ventilation, and long stay on the pediatric intensive care unit (PICU) in patients after cardiac surgery [3]. We reviewed our experience of TBM in pediatric patients undergoing cardiac surgery to determine whether we could identify factors that were associated with adverse outcomes and how to optimize their management.
|
Material and Methods
|
|---|
Clinical records of two groups of pediatric patients who underwent cardiac surgery for congenital heart disease between June 1998 and December 2003 in our institution were reviewed. After consultation with the local ethics committee we were advised that as the study was an anonymous, retrospective service evaluation, formal ethical committee approval was not required. Of a cohort of 1,578 patients undergoing surgery for congenital heart disease in our unit during this time (1,094 open and 484 closed cases), 22 patients were identified who had TBM diagnosed by bronchoscopy or dynamic contrast bronchography. The control group consisted of age- and procedure-matched patients in whom TBM was not demonstrated. This was created by using 2 control patients for each study patient, one immediately before and one immediately after the study patient. One patient in the study group underwent repair of a left pulmonary artery sling for which there were no control patients. This made the total number of patients in the control group 42. Data collected included primary and associated diagnoses, cardiac surgical procedures performed, findings at bronchoscopy and bronchography, ventilation details, intubation times, length of PICU and hospital stays after surgery, and morbidity and mortality during hospital admission and at outpatient follow-up clinics. The demographics and the cardiac procedures performed in the two groups of patients are summarized in Table 1.
In the study group, the diagnosis of TBM was made either preoperatively from a history of respiratory distress requiring ventilation, gastroesophageal reflux, stridor, or dyspnea or postoperatively after failed extubation, or as a result of gastroesophageal reflux or stridor. Patients in whom TBM had not been diagnosed preoperatively had ventilation weaned as dictated by clinical condition and then proceeded to extubation. If extubation failed, the patients were reintubated, and all factors—respiratory, cardiac and nutritional—were optimized for the subsequent 48 to 72 hours, after which they were again extubated. If this second attempt failed, and in the absence of any other clinical causes for failure to extubate (eg, paralyzed hemidiaphragm, lobar collapse, and so forth), we proceeded to investigate the child for TBM. Initially this was by flexible bronchoscopy and bronchography.
Bronchoscopy was performed using a 2.2-, 2.8-, or 3.6-mm Olympus flexible fiberoptic bronchoscope passed through a face mask or endotracheal tube while using inhalational anesthesia with the child breathing spontaneously. Assessment was made of the intrabronchial distending pressure required to maintain airway patency using a pressure manometer placed within the airway circuit.
Bronchography was performed under general inhalational anesthesia, again with the patient breathing spontaneously with an endotracheal tube placed just below the subglottis. Soluble contrast was injected into the trachea through a feeding tube passed through the endotracheal tube and placed distal to the tip. Videofluoroscopy of the tracheobronchial tree was captured in two dimensions for 10 to 15 seconds from the moment of contrast injection. This modality is increasingly being used as it demonstrates well the distal bronchial tree, which cannot be assessed by bronchoscopy. Again, assessment was made of the intrabronchial pressure required to maintain airway patency [3].
In those patients in whom initial investigation suggested airway compromise as a result of external compression, computed tomography scanning was performed.
The severity of TBM was defined as follows: mild, less than 50%; moderate, 50% to 80%; and severe, greater than 80% narrowing of the airway lumen.
Preoperatively, patients diagnosed with TBM were thoroughly investigated to ensure there was no evidence of respiratory tract infection. For those who were ventilated, aggressive physiotherapy and endotracheal suctioning was used to optimize the respiratory tract. However, care was taken to ensure there was not undue delay in addressing the cardiac abnormality.
Postoperatively, all patients with TBM were managed by oxygen administration, endotracheal suctioning, and positive end-expiratory or continuous positive airway pressure. If the patient had not been extubated after a further 10 days and if the initial investigations had demonstrated that the pressure required to distend the airways was greater than 10 cm H2O, a tracheostomy was performed. If the distending pressure was less than 10 cm H2O the child was progressed onto noninvasive ventilation using prong or mask continuous positive airway pressure.
Nonparametric Mann-Whitney U tests and Kruskal-Wallis tests were used to make intergroup comparisons with respect to ventilation time, PICU time, and hospital stay. Kaplan-Meier survival estimates were plotted for the two groups and for subgroups within the TBM cohort and compared using a log-rank test. A two-tailed Fisher's exact test was used to compare the demographic characters and survival between groups when the number was too small for a log-rank test.
|
Results
|
|---|
In 12 patients the diagnosis of TBM was made preoperatively owing to respiratory distress requiring ventilation (n = 6), gastroesophageal reflux (n = 2), stridor (n = 3), or dyspnea (n = 1). The diagnosis was made postoperatively in 10 patients after failed extubation (n = 8), or because of gastroesophageal reflux (n = 1) or stridor (n = 1).
Bronchoscopy was performed in 19 patients (86%), and dynamic contrast tracheobronchography in 9 patients (41%). The diagnosis of TBM was confirmed solely by bronchoscopy in 12 patients (55%) or bronchography in 2 patients (9%), and by a combination of these techniques in 7 patients (32 %). One patient (4.6%) underwent a magnetic resonance imaging scan to make the diagnosis. Tracheobronchomalacia was mild in 5 patients, moderate in 8 patients, and severe in 9 patients. The site of TBM was confined to the trachea in 6 patients, and there was isolated bronchial involvement in 5 patients and combined tracheal and bronchial malacia in 11 patients.
Follow-up was 100% complete for a mean period of 75.5 ± 12.8 months (range, 54 to 104 months) for the study group and 80 ± 10.8 months (range, 33 to 108 months) for the control group.
Survival
During the study period 4 patients died within 1 year of surgery, all in the TBM group. There were two early deaths (within 1 month of surgery). The first child with a relatively small left ventricle died of intestinal infection and severe dehydration after repair of aortic coarctation. The second child, with known TBM, had undergone an uneventful Blalock-Taussig shunt. He was being weaned to extubation, and the continuous positive airway pressure support had been reduced. He had a bradycardic arrest in the PICU from which he could not be resuscitated. There were two late deaths. Both patients experienced sudden death at home, 4 and 8 months after initial discharge from the hospital after surgery; postmortem examinations revealed no specific causes. There was no difference in mortality between those diagnosed with TBM preoperatively or postoperatively (p = 0.63). In addition, no difference in mortality could be demonstrated between those with mild, moderate, or severe TBM (p = 0.63). The site of TBM, which was confined to trachea, isolated bronchial involvement, or combined tracheal and bronchial malacia, could not be shown to affect mortality (p = 0.77).
No deaths occurred after 1 year. Estimated survival of patients with TBM at 5 years was 82% (95% confidence interval, 59% to 93%), which was significantly lower than for those without TBM (log-rank test, p = 0.004; Fig 1).
View larger version (16K):
[in this window]
[in a new window]
|
Fig 1. Kaplan-Meier survival estimates for patients with tracheobronchial malacia at 5 years was 82% (95% confidence interval, 59% to 93%) which was significantly lower than for those without tracheobronchial malacia (p = 0.004).
|
|
Univariate analysis of subgroups within the TBM group showed redo operations and palliative procedures to be risk factors for death. Two patients (9%) underwent redo operations; both these children died. This was significantly worse than that in the control group (2 of 2 versus 2 of 22; two-tailed Fisher's exact test p = 0.026). Six patients (27%) underwent operations that were palliative, and all deaths occurred in the palliation group. Comparison of survival between these two groups using univariate analysis showed a statistically significant difference (p = 0.0004; Fig 2). Prematurity, urgency, and prolonged postoperative ventilation were not risk factors for death (p = 0.15, p = 0.25, p = 0.51, respectively).
View larger version (15K):
[in this window]
[in a new window]
|
Fig 2. Kaplan-Meier survival estimates for patients underwent corrective or palliative cardiac procedures. Comparison of survival between these two groups using univariate analysis showed a statistically significant difference (p = 0.0004).
|
|
Multivariate analysis could not be performed as there were too few events per variable for this to be done [4].
Ventilation Time, Length of Pediatric Intensive Care Unit Stay, and Length of Hospital Stay
The median length of postoperative PICU stay was 11.5 days (2 to 191 days) in the study group compared with 2 days (1 to 17 days) for control patients (p < 0.001; Table 2). Similarly the median length of postoperative hospital stay was 20 days (6 to 201 days) in the study group compared with 6.5 days (5 to 23 days) for control patients (p < 0.001).
The median postoperative ventilation time was 6.5 days (1 to 180 days) in the study group compared with 1 day (0.42 to 12 days) for the control group (p < 0.001). The median ventilation time was significantly longer in patients diagnosed to have TBM postoperatively than preoperatively (18.5 days versus 2.5 days; p = 0.024). However, there was no significant difference between these groups as regards PICU or hospital stay (p = 0.086, p = 0.20, respectively). There was a significant difference in duration of ventilation and PICU stay relative to the severity of TBM (p = 0.048, p = 0.049, respectively; the more severe the disease, the longer the ventilation time and PICU stay), although the duration of hospital stay was similar between the subgroups (p = 0.18). There was no difference in ventilation time (p = 0.68), PICU stay (p = 0.67), or hospital stay (p = 0.39) relative to the site of TBM.
Eleven patients (50%) required tracheostomy for prolonged ventilation for more than 10 days. All patients with severe TBM and 2 with moderate TBM required tracheostomy. No patient required home ventilation after discharge.
|
Comment
|
|---|
Tracheobronchomalacia, although rare, is a common and important cause of persistent ventilatory requirement in infancy with a reported incidence of up to 50% [5–8]. Children may present with clinical signs of airway obstruction, such as stridor, wheeze, cyanotic spells, and reflex apnea, or recurrent respiratory tract infections [3]. However, it may remain silent until the postoperative period when it becomes manifest by difficulty in ventilation or unexpected requirement for prolonged ventilation [2, 3]. Rigid or flexible bronchoscopy and bronchography are the standard methods of diagnosis of TBM [3, 9]. In the present series, 55% of patients were diagnosed before surgery, of which half required preoperative ventilation for respiratory distress. Forty-five percent of patients had the diagnosis made after cardiac surgery, of whom 70% failed extubation. This suggests that in patients with unexpected postoperative ventilatory difficulty, one should consider the possibility of TBM. Bronchoscopy and bronchography can be performed easily and safely in postoperative intubated patients and can provide accurate information about the tracheobronchial tree. It is therefore now our policy to investigate patients when a second attempted extubation fails without known cause.
Tracheobronchomalacia can usually be managed with positive end-expiratory pressure through a nasotracheal tube or a tracheostomy [3]. Positive airway pressure results in an increase in dynamic compliance and a decrease in total respiratory system resistance [10]. The optimal positive end-expiratory pressure required during bronchoscopy or bronchography to open the airways can be measured (up to 18 cm H2O may be required [11]), and this can facilitate subsequent ventilatory management and decision making. Medical therapy such as bronchodilators and steroids may also reduce the dynamic compression of the large airways by improving peripheral airways disease and reducing edema [11]. Surgery such as tracheopexy, airway stenting or implants, and lesion resection is considered only when the above conservative therapies have failed, and underlying causes such as congenital heart disease have been corrected [10, 11]. In our series, more than half of the patients were known to have TBM before cardiac surgery; in this subgroup of patients, prolonged ventilation may be expected. We were able to manage these patients using conventional ventilation with positive end-expiratory pressure, and eventually continuous positive airway pressure either through a nasotracheal tube or tracheostomy (Fig 3A). In this subgroup of 12 patients, 4 required tracheostomy and the median postoperative ventilation time was 2.5 days.
View larger version (14K):
[in this window]
[in a new window]
|
Fig 3. (A) Recommendation for perioperative management of pediatric cardiac surgical patients with known tracheobronchial malacia. (B) Recommendation for perioperative management of pediatric cardiac surgical patients with preoperatively unrecognized tracheobronchial malacia. (CPAP = continuous positive airway pressure.)
|
|
In the patients in whom the diagnosis of TBM was made postoperatively, suspicion was raised after the patient failed to wean from mechanical ventilation in 70%. Once TBM was identified by bronchoscopy or bronchography, conservative management was performed as described above (Fig 3B). The median ventilation time was significantly longer in this group than for those in whom the diagnosis had been made preoperatively (17.5 versus 2.5 days; p = 0.02), partly owing to the delay in making the diagnosis. In addition, 70% of patients in this group required tracheostomy. There was one early death in each subgroup, although neither death appeared to be directly related to respiratory problems. There was no survival difference between these two subgroups.
The overall mortality in this series was 18%, which is in contrast to the previous report of 60% mortality for children with TBM [3]. In the latter series, all 15 children who needed ventilation for malacia that involved either main bronchus for longer than 14 days died, and all 3 children who needed ventilation for malacia of any severity for longer than 21 days died. This is again in contrast to our findings that although 75% required prolonged ventilation, the length of ventilation did not affect mortality.
Furthermore, comparing subgroups of patients with TBM we found that patients who belonged to subgroups that were predicted to be high risk (such as those undergoing urgent operations and premature infants) did not have a significantly higher mortality. However, those undergoing palliative procedures or repeat operations had significantly poorer outcomes. During follow-up most patients did well particularly after the first year, which demonstrates the self-limiting nature of TBM and the ability of the child to grow out of it.
In conclusion, TBM may be unsuspected in preoperative workup. If uncovered in children who become symptomatic after having undergone cardiac surgery, a longer duration of ventilation and longer PICU and hospital stays should be anticipated. However, overall outcomes are satisfactory, and surgery should not be withheld from this group of patients. Complete correction of cardiac defects should be strived for in patients with TBM. Nevertheless, the excess mortality seen in this group may be unrelated to their TBM, and this deserves further study.
|
Acknowledgments
|
|---|
We would like to thank Linda Hunt, PhD, for assistance with the statistical analysis for this paper.
|
References
|
|---|
- Holinger PH, Johnston KC, Parchet VN, Zimmermann AA. Congenital malformations of the trachea, bronchi and lung Ann Otol Rhinol Laryngol 1952;61:1159-1180.[Medline]
- Doull IJ, Mok Q, Tasker RC. Tracheobronchomalacia in preterm infants with chronic lung disease Arch Dis Child Fetal Neonatal Ed 1997;76:F203-F205.[Abstract/Free Full Text]
- Burden RJ, Shann F, Butt W, Ditchfield M. Tracheobronchial malacia and stenosis in children in intensive care: bronchograms help to predict outcome Thorax 1999;54:511-517.[Abstract/Free Full Text]
- Feinstein A. Multivariable analysis; an introductionNew Haven: Yale University Press; 1996. pp. 423.
- Miller RW, Woo P, Kellman RK, Slagle TS. Tracheobronchial abnormalities in infants with bronchopulmonary dysplasia J Pediatr 1987;111:779-782.[Medline]
- Greenholz SK, Hall RJ, Lilly JR, Shikes RH. Surgical implications of bronchopulmonary dysplasia J Pediatr Surg 1987;22:1132-1136.[Medline]
- Downing GJ, Kilbride HW. Evaluation of airway complications in high-risk preterm infants: application of flexible fiberoptic airway endoscopy Pediatrics 1995;95:567-572.[Abstract/Free Full Text]
- Lindahl H, Rintala R, Malinen L, Leijala M, Sairanen H. Bronchoscopy during the first month of life J Pediatr Surg 1992;27:548-550.[Medline]
- Rozycki HJ, Van Houten ML, Elliott GR. Quantitative assessment of intrathoracic airway collapse in infants and children with tracheobronchomalacia Pediatr Pulmonol 1996;21:241-245.[Medline]
- Zinman R. Tracheal stenting improves airway mechanics in infants with tracheobronchomalacia Pediatr Pulmonol 1995;19:275-281.[Medline]
- Sotomayor JL, Godinez RI, Borden S, Wilmott RW. Large-airway collapse due to acquired tracheobronchomalacia in infancy Am J Dis Child 1986;140:367-371.[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
T. Cotts, J. Hirsch, M. Thorne, and R. Gajarski
Tracheostomy after pediatric cardiac surgery: Frequency, indications, and outcomes
J. Thorac. Cardiovasc. Surg.,
February 1, 2011;
141(2):
413 - 418.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
