- Email: [email protected]
Descending Aortopexy and Posterior Tracheopexy for Severe Tracheomalacia and Left Mainstem Bronchomalacia
Accepted Manuscript Title: Descending Aortopexy and Posterior Tracheopexy for Severe Tracheomalacia and Left Mainstem Bronchomalacia Author: Hester F. Shieh, C. Jason Smithers, Thomas E. Hamilton, David Zurakowski, Gary A. Visner, Michael A. Manfredi, Russell W. Jennings, Christopher W. Baird PII: DOI: Reference:
S1043-0679(18)30077-7 https://doi.org/10.1053/j.semtcvs.2018.02.031 YSTCS 1078
To appear in:
Seminars in Thoracic and Cardiovascular Surgery
Please cite this article as: Hester F. Shieh, C. Jason Smithers, Thomas E. Hamilton, David Zurakowski, Gary A. Visner, Michael A. Manfredi, Russell W. Jennings, Christopher W. Baird, Descending Aortopexy and Posterior Tracheopexy for Severe Tracheomalacia and Left Mainstem Bronchomalacia, Seminars in Thoracic and Cardiovascular Surgery (2018), https://doi.org/10.1053/j.semtcvs.2018.02.031. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39
AATS 2017 Original Manuscript: Congenital Heart Disease Scientific Session Descending aortopexy and posterior tracheopexy for severe tracheomalacia and left mainstem bronchomalacia Hester F. Shieh, MD a, C. Jason Smithers, MD a, Thomas E. Hamilton, MD a, David Zurakowski, PhD a, Gary A. Visner, DO b, Michael A. Manfredi, MD c, Russell W. Jennings, MD a*, Christopher W. Baird, MD d* a
Department of Surgery, Boston Children’s Hospital, Harvard Medical School, Boston, MA 02115 b Department of Pulmonology, Boston Children’s Hospital, Harvard Medical School, Boston, MA 02115 c Department of Gastroenterology, Boston Children’s Hospital, Harvard Medical School, Boston, MA 02115 d Department of Cardiac Surgery, Boston Children’s Hospital, Harvard Medical School, Boston, MA 02115 *Co-senior authors Conflict of Interest Statement and Sources of Funding: There are no conflicts of interest or funding. IRB Approval: IRB-P00021702 approved 11/10/2016 Corresponding author: Christopher W. Baird, MD Department of Cardiac Surgery Boston Children’s Hospital Harvard Medical School 300 Longwood Ave., 612 Farley Boston, MA 02115 Tel: (617) 355-5637 Fax: (617) 730-4698 Email: [email protected] Article word count: 2701 Keywords: aortopexy
1
Page 1 of 30
40
Glossary of Abbreviations
41
BRUE = brief resolved unexplained event
42
CHD = congenital heart disease
43
DLB = diagnostic laryngoscopy and bronchoscopy
44
EA = esophageal atresia
45
MDCT = multidetector computed tomography
46
TEF = tracheoesophageal fistula
2
Page 2 of 30
47
Central Picture
48 49
Central Picture Legend: Cardiomegaly and a left descending aorta compressing the left
50
mainstem bronchus.
51 52
Central Message: Descending aortopexy and posterior tracheopexy are clinically
53
effective in treating severe tracheobronchomalacia and left mainstem intrusion with
54
significant symptom and anatomic improvements.
55 56
Perspective Statement: Descending aortopexy and posterior tracheopexy are effective in
57
treating severe tracheobronchomalacia and left mainstem intrusion with significant
58
improvements in clinical symptoms and degree of airway collapse on bronchoscopy.
59
Complex cases warrant an individualized and flexible surgical approach guided by
60
preoperative imaging and intraoperative bronchoscopy in multidisciplinary specialized
61
centers.
3
Page 3 of 30
62
Abstract
63
Objective: Posterior descending aortopexy can relieve posterior intrusion of the left
64
mainstem bronchus that may limit the effectiveness of posterior tracheobronchopexy. We
65
review outcomes of patients undergoing both descending aortopexy and posterior
66
tracheopexy for severe tracheobronchomalacia with posterior intrusion and left mainstem
67
compression to determine if there were resolution of clinical symptoms and
68
bronchoscopic evidence of improvement in airway collapse. Methods: All patients who
69
underwent both descending aortopexy and posterior tracheopexy from October 2012 to
70
October 2016 were retrospectively reviewed. Clinical symptoms, tracheomalacia scores
71
based on standardized dynamic airway evaluation by anatomical region, and persistent
72
airway intrusion requiring reoperation were collected. Data were analyzed by Wald and
73
Wilcoxon signed-ranks tests. Results: 32 patients underwent descending aortopexy and
74
posterior tracheopexy at median age 18 months (IQR 6-40 months). Median follow up
75
was 3 months (IQR 1-7 months). There were statistically significant improvements in
76
clinical symptoms postoperatively, including cough, noisy breathing, prolonged and
77
recurrent respiratory infections, ventilator dependence, blue spells, and brief resolved
78
unexplained events (BRUEs) (all P<.001), as well as exercise intolerance (P=.033),
79
transient respiratory distress requiring positive pressure (P=.003), and oxygen
80
dependence (P=.007). Total tracheomalacia scores improved significantly (P<.001), with
81
significant segmental improvements in the middle (P=.003) and lower (P<.001) trachea,
82
and right (P=.011) and left (P<.001) mainstem bronchi. 2 patients (6%) had persistent
83
airway intrusion requiring reoperation with anterior aortopexy and/or tracheopexy.
84
Conclusions: Descending aortopexy and posterior tracheopexy are effective in treating
4
Page 4 of 30
85
severe tracheobronchomalacia and left mainstem intrusion with significant improvements
86
in clinical symptoms and degree of airway collapse on bronchoscopy.
87 88
Abstract Word Count: 248
5
Page 5 of 30
89
1. Introduction
90
Tracheobronchomalacia refers to a weakness or deformation of the airway such
91
that it is more susceptible to collapse with changes in pressure and compression by
92
adjacent thoracic structures [1]. It is often associated with esophageal atresia (EA),
93
tracheoesophageal fistula (TEF), and congenital heart disease (CHD) [2]. Severe
94
tracheobronchomalacia is characterized by dynamic airway collapse in spontaneously
95
breathing patients with anterior vascular compression and/or posterior membranous
96
tracheal intrusion [3]. Anterior ascending aortopexy addresses anterior vascular
97
compression by indirectly elevating the anterior wall of the trachea, but does not directly
98
address posterior membranous tracheal intrusion [4]. We recently reported a series of
99
patients who underwent posterior tracheopexy for severe symptomatic tracheomalacia
100
with posterior intrusion with promising short-term results [5]. The effectiveness of
101
posterior tracheopexy can be limited by left mainstem bronchomalacia in some patients.
102
Posterior descending aortopexy can be used to relieve left mainstem posterior intrusion
103
and compression between the descending aorta and pulmonary artery. We now review a
104
series of patients who underwent both descending aortopexy and posterior tracheopexy
105
for severe symptomatic tracheobronchomalacia with posterior intrusion and left
106
mainstem compression to determine if there were resolution of clinical symptoms and
107
bronchoscopic evidence of improvement in airway collapse.
108
2. Methods
109
The Esophageal and Airway Treatment (EAT) Center at Boston Children’s
110
Hospital is a multidisciplinary care team consisting of three pediatric surgeons, one
111
pediatric cardiothoracic surgeon, one pediatric pulmonologist, and two pediatric
6
Page 6 of 30
112
gastroenterologists. We retrospectively reviewed all patients who underwent both
113
descending aortopexy and posterior tracheopexy at Boston Children’s Hospital from
114
October 2012 to October 2016 under an approved institutional review board protocol
115
(IRB-P00021702).
116
Patient demographics, pre- and postoperative clinical symptoms and airway
117
evaluation, surgical techniques, and persistent airway intrusion requiring reoperation,
118
were collected. Clinical symptoms included cough, barking cough, noisy breathing,
119
prolonged pulmonary infection, recurrent pulmonary infections, exercise intolerance,
120
transient respiratory distress requiring positive pressure, oxygen dependence, ventilator
121
dependence, blue spells, and brief resolved unexplained events (BRUEs) (formerly
122
known as apparent life-threatening events (ALTEs)).
123
Pre- and postoperative endoscopic airway evaluation was performed by the
124
primary surgeons involved. Diagnostic laryngoscopy and bronchoscopy (DLB) was done
125
under general anesthesia in spontaneously breathing patients. After assessing supraglottic
126
structures and vocal cord function, the vocal cords were anesthetized with topical
127
lidocaine, the larynx was assessed for presence of a laryngeal cleft, and a Hopkins II rod
128
lens was inserted through the cords to assess for TEF, tracheal diverticulum, and dynamic
129
motion in the tracheobronchial tree throughout the respiratory cycle. A standardized
130
tracheomalacia scoring system based on dynamic airway evaluation was used to
131
determine pre- and postoperative tracheomalacia scores (Table 1) [3,5-6]. The
132
tracheobronchial tree was scored by the percentage of open airway during cough or
133
Valsalva out of 100 at each anatomic region: upper (T1), middle (T2), and lower (T3)
134
trachea, and right and left mainstem bronchi, with a maximum score of 500. Dynamic
7
Page 7 of 30
135
airway multidetector computed tomography (MDCT) was performed preoperatively to
136
evaluate for aberrant vascular anatomy or associated lung parenchymal disease, and to
137
identify the artery of Adamkiewicz [6]. As it is difficult to justify postoperative radiation
138
and routine MDCT imaging in asymptomatic children, bronchoscopic airway evaluation
139
was used for postoperative follow up.
140
The operating surgeon determined the operative plan and approach based on
141
endoscopic evaluation and preoperative MDCT, as part of a multidisciplinary team.
142
Generally patients with associated esophageal disease underwent right posterior
143
thoracotomy, those with cardiac disease underwent median sternotomy, and those with
144
vascular rings underwent left thoracotomy. The esophagus, back wall of the trachea,
145
thoracic duct, and/or aorta were fully dissected and mobilized, taking care to protect the
146
left vagus nerve and left recurrent laryngeal nerve. In patients undergoing sternotomy, the
147
ductal ligament was often divided to extensively mobilize the ascending aorta, transverse
148
aortic arch, and descending aorta. A recurrent TEF or residual tracheal diverticulum from
149
a previously repaired TEF was corrected if present by resecting the TEF or diverticulum
150
flush with the tracheal wall under endoscopic visualization. Pexy procedures were all
151
done under direct flexible bronchoscopic guidance. Posterior descending aortopexy was
152
performed by passing autologous pleural or pericardial pledgeted polypropylene sutures
153
to secure the aorta to the side of the spine, and as posteriorly as necessary to relieve
154
posterior pressure off the left mainstem bronchus (Figure 1). This posterior movement of
155
the aorta may necessitate dividing one or more intercostals, and preoperative MDCT
156
helps to ensure that the artery of Adamkiewicz is not divided. Arm-leg blood pressure
157
measurements or arterial line tracing monitoring were used to confirm no descending
8
Page 8 of 30
158
aortic gradient. Posterior tracheopexy was performed by passing autologous pleural or
159
pericardial pledgeted sutures into but not through the posterior tracheal membrane, and
160
securing them to the anterior longitudinal spinal ligament, in such a fashion as to
161
optimize tracheal lumen and posterior tracheal membrane support. Suture placement is
162
guided by intraoperative bronchoscopic guidance to avoid full thickness bites and to
163
optimize placement of the sutures. Additional airway pexy procedures were similarly
164
performed by passing pledgeted sutures to secure the bronchial posterior membrane
165
posteriorly to the lateral edges of the anterior longitudinal ligament of the spine. In
166
patients undergoing sternotomy, anterior pexy sutures to support and/or displace the
167
vasculature and/or airway were then passed through the sternum and secured under direct
168
bronchoscopic visualization following sternal closure.
169
2.1 Statistical Analysis
170
To assess resolution of clinical symptoms, the percentage of patients with each
171
symptom pre- and postoperatively was compared by the Wald chi-square test using
172
logistic regression modeling with a generalized estimating equations (GEE) approach to
173
account for the binary paired data [7]. Changes in tracheomalacia scores for each airway
174
segment were determined by the Wilcoxon signed-ranks test [8]. Freedom from
175
reoperation was estimated using the Kaplan-Meier product-limit method [9]. Statistical
176
analysis was performed using IBM SPSS Statistics (version 23.0, IBM Corporation,
177
Armonk, NY). A two-tailed P<.05 was considered statistically significant.
178
3. Results
179
32 patients underwent descending aortopexy and posterior tracheopexy at median
180
age 18 months (interquartile range (IQR) 6-40 months). 63% (20 patients) were male.
9
Page 9 of 30
181
Median estimated gestational age (EGA) was 34 weeks (IQR 31-36 weeks). 66% (21
182
patients) were associated with EA, including 18 patients with type C EA, 2 patients with
183
type A EA, and 1 patient with type B EA. 19% (6 patients) had long gap EA. 69% (22
184
patients) had associated cardiac disease, and 19% (6 patients) had VACTERL syndrome.
185
63% (20 patients) had a prior EA repair, and 19% (6 patients) had a prior anterior
186
aortopexy. 19% (6 patients) had vascular rings, including 3 patients with double aortic
187
arch, 2 patients with right aortic arch with Kommerell diverticulum, and 1 patient with
188
circumflex aorta. 9% (3 patients) had a prior vascular ring division, 3% (1 patient) had a
189
prior coarctation repair, and 13% (4 patients) had a prior patent ductus arteriosus ligation.
190
3% (1 patient) had double superior vena cava and 3% (1 patient) had dysplastic
191
pulmonary valve stenosis. 6% (2 patients) had tetralogy of Fallot, one of which had
192
undergone prior repair. 22% (7 patients) had atrial and/or ventricular septal defects, 3 of
193
which had undergone prior septal defect closures.
194
Upper airway anomalies were common on preoperative and intraoperative
195
bronchoscopy. 3% (1 patient) had laryngomalacia. No patients had preoperative vocal
196
cord dysfunction or paralysis. 22% (7 patients) had laryngeal clefts, 6 patients with type 1
197
clefts and 1 patient with a type 3 cleft. 19% (6 patients) had some degree of subglottic
198
stenosis. The tracheobronchial tree was scored by anatomic region (Table 1).
199
Preoperatively, the middle (T2) trachea, lower (T3) trachea, and left mainstem bronchus
200
were the most severely affected, with median scores of 0 (IQR 0-45), 0 (IQR 0-30), and
201
20 (IQR 0-50), respectively.
202
Operative approach was by right thoracotomy in 56% (18 patients), median
203
sternotomy in 22% (7 patients), left thoracotomy in 19% (6 patients), and right neck
10
Page 10 of 30
204
dissection with right thoracotomy in 3% (1 patient). All patients underwent descending
205
aortopexy and posterior tracheopexy under intraoperative bronchoscopic guidance. 53%
206
(17 patients) underwent additional airway or vascular pexy procedures, including left
207
mainstem bronchopexy in 38% (12 patients), right mainstem bronchopexy in 16% (5
208
patients), innominate artery pexy in 9% (3 patients), anterior tracheopexy in 9% (3
209
patients), and anterior aortopexy in 9% (3 patients). 6% (2 patients) had an aberrant
210
subclavian artery behind the trachea, requiring mobilization of the artery in 1 patient and
211
division of the artery in 1 patient. 53% (17 patients) had an associated tracheal
212
diverticulum that was resected flush with the trachea. Concomitant procedures included
213
septal defect closure in 13% (4 patients), vascular ring division in 9% (3 patients),
214
tetralogy of Fallot repair in 3% (1 patient), and pulmonary valve replacement in 3% (1
215
patient).
216
Median days on the ventilator after surgery were 4 days (IQR 0-11 days). Median
217
total intensive care unit (ICU) stay was 10 days (IQR 2-23 days). Median total hospital
218
length of stay was 15 days (IQR 7-42 days). There were no significant early
219
complications including hemorrhage or infection. There were no complications with
220
erosion of the pledgetted sutures into the aorta or trachea, as the pledgets are autologous.
221
There were no mortalities.
222
Median clinical follow up was 3 months (IQR 1-7 months). There were
223
statistically significant improvements in clinical symptoms postoperatively, including
224
cough, barking cough, noisy breathing, prolonged and recurrent respiratory infections,
225
ventilator dependence, blue spells, and BRUEs (all P<.001), as well as exercise
226
intolerance (P=.033), transient respiratory distress requiring positive pressure (P=.003),
11
Page 11 of 30
227
and oxygen dependence (P=.007) (Figure 1). At latest follow up, no patients had
228
recurrence of a blue spell or BRUE.
229
47% (15 patients) underwent postoperative follow up evaluation with
230
bronchoscopy at median 2 months (IQR 1-6.5 months). Total tracheomalacia scores on
231
bronchoscopy improved significantly (P<.001), with significant segmental improvements
232
in the middle (T2) (P=.003) and lower (T3) (P<.001) trachea, and the right (P=.011) and
233
left (P<.001) mainstem bronchi (Table 1). The greatest areas of numerical improvement
234
were in the segments most affected preoperatively, namely the middle (T2) and lower
235
(T3) trachea, as well as the left mainstem bronchus.
236
6% (2 patients) had persistent airway intrusion requiring reoperation, using
237
anterior aortopexy and/or anterior tracheopexy. The 2 reoperations were within 2 months
238
following the index surgery. Kaplan-Meier analysis estimated 92% of patients to be free
239
from reoperation at 3 months follow up (95% confidence interval 85-99%).
240
6% (2 patients) had tracheostomies preoperatively, all for severe tracheomalacia.
241
Overall, 9% (3 patients) had tracheostomies postoperatively, with no significant
242
difference when compared to preoperatively (P=.31). The 2 patients with preoperative
243
tracheostomies on the ventilator were able to wean to tracheostomy collar
244
postoperatively. One additional patient with associated severe cardiac disease, transferred
245
intubated on mechanical ventilation, had a tracheostomy placed for prolonged intubation
246
and ventilator weaning.
247
4. Discussion
248
Tracheobronchomalacia is an underestimated disease, given the wide spectrum of
249
disease with nonspecific chronic respiratory symptoms that are commonly misdiagnosed
12
Page 12 of 30
250
[1-2,10-11]. It is a common respiratory problem in patients with EA and CHD [12-15].
251
Among at least CHD patients, tracheobronchomalacia has been associated with increased
252
ventilator days, length of stay, and mortality [15-17]. Excessive airway collapse leads to
253
ineffective ventilation and poor clearance of secretions, resulting in frequent respiratory
254
infections, respiratory failure, and apneic events, all of which are less well tolerated in
255
patients with EA and CHD, underscoring the importance of early diagnosis in these
256
populations [1-2,10,18]. Our multidisciplinary care team routinely uses a standardized
257
reporting and scoring system based on anatomic region for endoscopic evaluation [3,5-6].
258
The greater the severity of airway collapse, indicated by a lower tracheomalacia score,
259
combined with the presence of clinical symptoms, may indicate the need for surgical
260
intervention, possibly concomitant with EA or CHD repair.
261
Preoperative dynamic airway MDCT is used in conjunction with bronchoscopy to
262
inform the operative plan. MDCT is particularly useful in evaluating complex cases with
263
aberrant vascular anatomy and associated cardiac or esophageal anomalies. Several
264
different types of disease resulting in complex tracheobronchomalacia include a right
265
aortic arch with large Kommerell diverticulum (Figure 2A), a right to left circumflex
266
aorta (Figure 2B), a double aortic arch compressing the trachea (Figure 2C), a left aortic
267
arch compressing the trachea (Figure 2D), and a left descending aorta compressing the
268
left mainstem bronchus (Figure 2E).
269
The management of severe tracheobronchomalacia remains difficult with little
270
consensus on treatment and surgical approach [1-2,15,19]. Surgical options include pexy
271
procedures (ascending and/or descending aortopexy, anterior and/or posterior
272
tracheopexy), tracheal resection, internal stents, and external stabilization [1-5,15,20-24].
13
Page 13 of 30
273
Anterior ascending aortopexy indirectly supports the anterior tracheal wall and is the
274
most commonly used technique, but has a reported failure rate of 10-25% in the literature
275
[15,22-24]. Direct anterior and/or posterior tracheopexy, as first reported by our group,
276
improve airway patency by directly addressing anterior malformed tracheal cartilage and
277
posterior membranous tracheal intrusion [2-3,5].
278
Left mainstem bronchomalacia is a challenging entity that may limit airflow,
279
mucus clearance, and the effectiveness of direct tracheopexy in some cases, which is the
280
focus of this paper [25]. The descending aorta may be anteriorly displaced, intruding into
281
the back wall of the left mainstem bronchus, while the pulmonary artery may be
282
posteriorly displaced, compressing the airway from the front, resulting in left mainstem
283
vascular compression. Surgical options include posterior descending aortopexy,
284
descending aortic translocation, pulmonary artery anterior fixation, internal stents, and
285
external splints [26-32].
286
In this series, we show that descending aortopexy and posterior tracheopexy are
287
clinically effective in treating severe tracheobronchomalacia with posterior intrusion and
288
left mainstem compression. Postoperatively, there were significant improvements in
289
clinical symptoms, as well as anatomic tracheomalacia scores. Anterior and posterior
290
tracheopexy or bronchopexy provide direct support to the airway, whereas anterior
291
ascending and posterior descending aortopexy indirectly support the airway by directly
292
addressing
293
tracheobronchomalacia, especially those with altered anatomy associated with EA or
294
CHD, warrant an individualized and flexible surgical approach guided by intraoperative
vascular
compression
of
the
airway.
Complex
cases
of
severe
14
Page 14 of 30
295
bronchoscopy. In fact, in our series, 53% of patients underwent concomitant airway
296
and/or vascular pexy procedures to optimize each individual airway.
297
Posterior descending aortopexy is a technically challenging procedure. Through a
298
right posterior thoracotomy approach, sutures are placed in the anterior aspect of the
299
descending aorta, and then blindly placed on the left side of the spine close to the
300
costovertebral junction by passing the needle out the anterior longitudinal ligament of the
301
spine. Through a median sternotomy approach, it can be even more challenging to
302
mobilize and expose the descending aorta, and place sutures to posteriorly displace it. We
303
find it best to monitor upper and lower extremity blood pressure or arterial line tracings
304
to avoid aortic stenosis.
305
There are a number of limitations to this study. Retrospective chart review was
306
used to collect the data and follow up, including pre- and postoperative clinical symptoms
307
and bronchoscopy findings. Although patients are followed closely by our
308
multidisciplinary clinic, further studies could utilize a prospective structured clinical
309
symptom questionnaire to further standardize reporting. Bronchoscopy can be subjective
310
and was performed by three primary operating surgeons. One study in adults showed
311
appropriate inter- and intraobserver reliability in flexible bronchoscopy, however less is
312
known in the pediatric population [33]. Future work can include bronchoscopic analysis
313
by independent observers to make a more statistically valid comparison. Postoperative
314
endoscopic evaluation was not available for all patients, however we used the
315
standardized scoring system to demonstrate resolution of tracheomalacia postoperatively
316
in those evaluated. Our standard protocol for endoscopic postoperative evaluation is at 1
317
year for longitudinal airway assessment if clinically asymptomatic, unless the patient is
15
Page 15 of 30
318
undergoing another scheduled procedure. Our study cohort included a heterogeneous
319
group of complex patients often requiring concomitant airway/vascular pexy procedures
320
or adjunct therapies that may have contributed to outcomes and confounded the influence
321
of surgical treatment alone. As our preference is to correct all anomalies that may affect
322
airway or cardiac function at the initial operation to try to prevent multiple reoperations,
323
it would be nearly impossible to isolate patients undergoing only one procedure. Follow
324
up intervals were relatively short-term and variable.
325
In conclusion, descending aortopexy and posterior tracheopexy are effective in
326
treating severe tracheobronchomalacia and left mainstem compression with significant
327
improvement or resolution of clinical symptoms and degree of airway collapse on
328
bronchoscopy. Further studies to follow long-term outcomes of this technique are
329
certainly warranted and ongoing. Given the heterogeneity and complexity of this patient
330
population with significant morbidity, treatment and long-term follow up is best done in
331
multidisciplinary specialized centers for individualized patient care.
332 333 334 335 336
Acknowledgment H.F.S. was supported by the Joshua Ryan Rappaport Fellowship of the Department of Surgery at Boston Children’s Hospital.
337 338
16
Page 16 of 30
339
References
340
1. Fraga JC, Jennings RW, Kim PC. Pediatric tracheomalacia. Semin Pediatr Surg 2016;
341 342 343 344
25:156–64. 2. Bairdain S, Zurakowski D, Baird CW, Jennings RW. Surgical treatment of tracheobronchomalacia: a novel approach. Paediatr Respir Rev 2016; 19:16–20. 3. Bairdain S, Smithers CJ, Hamilton TE, Zurakowski D, Rhein L, Foker JE, et al.
345
Direct
346
tracheobronchomalacia: short-term outcomes in a series of 20 patients. J Pediatr Surg
347
2015; 50:972–7.
tracheobronchopexy
to
correct
airway
collapse
due
to
severe
348
4. Jennings RW, Hamilton TE, Smithers CJ, Ngerncham M, Feins N, Foker JE. Surgical
349
approaches to aortopexy for severe tracheomalacia. J Pediatr Surg 2014; 49:66-70,
350
discussion 70-1.
351 352 353
5. Shieh HF, Smithers CJ, Hamilton TE, Zurakowski D, Rhein LM, Manfredi MA, et al. Posterior tracheopexy for severe tracheomalacia. J Pediatr Surg 2017; in press. 6. Ngerncham
M,
Lee
EY,
Zurakowski
D,
Tracy
DA,
Jennings
R.
354
Tracheobronchomalacia in pediatric patients with esophageal atresia: comparison of
355
diagnostic laryngoscopy/bronchoscopy and dynamic airway multidetector computed
356
tomography. J Pediatr Surg 2015; 50:402–7.
357 358 359 360 361
7. McCullagh P, Nelder JA. Generalized Linear Models, 2nd ed. London: Chapman & Hall; 1989:98-148. 8. Altman DG. Practical Statistics for Medical Research. London: Chapman & Hall; 1991:189-228. 9. Cox DR, Oakes D. Analysis of Survival Data. London, Chapman & Hall; 1984:48-61.
17
Page 17 of 30
362
10. Boogaard R, Huijsmans SH, Pijnenburg MW, Tiddens HA, de Jongste JC, Merkus PJ.
363
Tracheomalacia
364
characteristics. Chest 2005; 128:3391–7.
and
bronchomalacia
in
children:
incidence
and
patient
365
11. Fischer AJ, Singh SB, Adam RJ, Stoltz DA, Baranano CF, Kao S, et al.
366
Tracheomalacia is associated with lower FEV1 and Pseudomonas acquisition in
367
children with CF. Pediatr Pulmonol 2014; 49:960–70.
368 369
12. Spitz L, Kiely E, Brereton RJ. Esophageal atresia: five year experience with 148 cases. J Pediatr Surg 1987; 22:103–8.
370
13. Filler RM, Messineo A, Vinograd I. Severe tracheomalacia associated with
371
esophageal atresia: results of surgical treatment. J Pediatr Surg 1992; 27:1136-40,
372
discussion 1140-1.
373
14. Cartabuke RH, Lopez R, Thota PN. Long-term esophageal and respiratory outcomes
374
in children with esophageal atresia and tracheoesophageal fistula. Gastroenterol Rep
375
2016; 4:310-4.
376 377
15. Ragalie WS, Mitchell ME. Advances in surgical treatment of congenital airway disease. Semin Thorac Cardiovasc Surg 2016; 28(1):62-8.
378
16. Chen Q, Langton-Hewer S, Marriage S, Hayes A, Caputo M, Pawade A, et al.
379
Influence of tracheobronchomalacia on outcome of surgery in children with
380
congenital heart disease and its management. Ann Thorac Surg 2009; 88(6):1970-4.
381
17. Pfammatter JP, Casaulta C, Pavlovic M, Berdat PA, Frey U, Carrel T. Important
382
excess morbidity due to upper airway anomalies in the perioperative course in infant
383
cardiac surgery. Ann Thorac Surg 2006; 81(3):1008-12.
384
18. Carden
KA,
Boiselle
PM,
Waltz
DA,
Ernst
A.
Tracheomalacia
and
18
Page 18 of 30
385
tracheobronchomalacia in children and adults: an in-depth review. Chest 2005;
386
127:984–1005.
387 388 389 390 391 392 393 394
19. Goyal V, Masters IB, Chang AB. Interventions for primary (intrinsic) tracheomalacia in children. Cochrane Database Syst Rev 2012; 10:CD005304. 20. Torre M, Carlucci M, Speggiorin S, Elliott MJ. Aortopexy for the treatment of tracheomalacia in children: review of the literature. Ital J Pediatr 2012; 38:62. 21. Dave S, Currie BG. The role of aortopexy in severe tracheomalacia. J Pediatr Surg 2006; 41:533–7. 22. Weber TR, Keller MS, Fiore A. Aortic suspension (aortopexy) for severe tracheomalacia in infants and children. Am J Surg 2002; 184:573–7, discussion 577.
395
23. Mitchell ME, Rumman N, Chun RH, Rao A, Martin T, Beste DJ, et al. Anterior
396
tracheal suspension for tracheobronchomalacia in infants and children. Ann Thorac
397
Surg 2014; 98:1246-53.
398 399
24. Lee SY, Kim SJ, Baek JS, Kwak JG, Lee C, Lee CH, et al. Outcomes of aortopexy for patients with congenital heart disease. Pediatr Cardiol 2013; 34(6):1469-75.
400
25. Hungate RG, Newman B, Meza MP. Left mainstem bronchial narrowing: a vascular
401
compression syndrome? Evaluation by magnetic resonance imaging. Pediatr Radiol
402
1998; 28(7):527-32.
403
26. Sacco O, Santoro F, Ribera E, Magnano GM, Rossi GA. Short-length ligamentum
404
arteriosum as a cause of congenital narrowing of the left main stem bronchus. Pediatr
405
Pulmonol 2016; 51(12):1356-61.
406
27. Arcieri L, Serio P, Nenna R, Di Maurizio M, Baggi R, Assanta N, et al. The role of
407
posterior aortopexy in the treatment of left mainstem bronchus compression. Interact
19
Page 19 of 30
408
Cardiovasc Thorac Surg 2016; 23(5):699-704.
409
28. Valerie EP, Durrant AC, Forte V, Wales P, Chait P, Kim PC. A decade of using
410
intraluminal tracheal/bronchial stents in the management of tracheomalacia and/or
411
bronchomalacia: is it better than aortopexy? J Pediatr Surg 2005; 40:904–7.
412
29. Morrison RJ, Hollister SJ, Niedner MF, Mahani MG, Park AH, Mehta DK, et al.
413
Mitigation of tracheobronchomalacia with 3D-printed personalized medical devices
414
in pediatric patients. Sci Transl Med 2015; 7:285ra64.
415 416
30. Kosloske AM. Left mainstem bronchopexy for severe bronchomalacia. J Pediatr Surg 1991; 26(3):260-2.
417
31. Baird CW, Prabhu S, Buchmiller TL, Smithers C, Jennings R. Direct
418
tracheobronchopexy and posterior descending aortopexy for severe left mainstem
419
bronchomalacia associated with congenital pulmonary airway malformation and left
420
circumflex aortic arch. Ann Thorac Surg 2016; 102(1):e1-4.
421
32. McKenzie ED, Roeser ME, Thompson JL, De Leon LE, Adachi I, Heinle JS, et al.
422
Descending aortic translocation for relief of distal tracheal and proximal bronchial
423
compression. Ann Thorac Surg 2016; 102(3):859-62.
424
33. Majid A, Gaurav K, Sanchez JM, Berger RL, Folch E, Fernandez-Bussy S, et al.
425
Evaluation of tracheobronchomalacia by dynamic flexible bronchoscopy. A pilot
426
study. Ann Am Thorac Soc 2014; 11:951–5.
427 428
20
Page 20 of 30
429
Figure 1. Posterior Descending Aortopexy. A. Illustrates anatomic relationship of
430
aorta, esophagus, and spine. B. Cross-sectional view, in which the esophagus is rotated to
431
the right and the descending aorta is moved to the left and secured to the side of the spine
432
as posteriorly as necessary to relieve posterior pressure off the left mainstem bronchus. C.
433
Descending aortopexy sutures are tied, relieving left mainstem posterior intrusion and
434
compression between the descending aorta and pulmonary artery.
435 436
Figure 2: Pre- and Postoperative Clinical Symptoms.
437 438
Figure 3: Preoperative Dynamic Airway Multidetector Computed Tomography
439
(MDCT) of Complex Cases of Tracheobronchomalacia. A. Right aortic arch with
440
large Kommerell diverticulum. B. Right to left circumflex aorta. C. Double aortic arch
441
compressing the trachea. D. Left aortic arch compressing the trachea. E. Left descending
442
aorta compressing the left mainstem bronchus.
443 444
21
Page 21 of 30
445
Table 1: Tracheomalacia Scores. Pre- and postoperative tracheomalacia scores based
446
on standardized bronchoscopic evaluation. Scores are percentage of open airway out of
447
100 for each anatomical region. Data are median (IQR). Location
448 449 450 451 452 453 454 455
Postoperative (n=15) 100 (80-100) 75 (60-100) 100 (70-100) 100 (100-100) 70 (50-100)
P value
T1 T2 T3 Right bronchus Left bronchus
Preoperative (n=32) 80 (70-95) 0 (0-45) 0 (0-30) 78 (12-100) 20 (0-50)
Total
215 (145-268)
450 (360-475)
<.001*
.182 .003* <.001* .011* <.001*
Video 1: Descending Aortopexy and Posterior Tracheopexy. Operative procedure and its relevance as discussed by Dr. Russell Jennings.
22
Page 22 of 30
456
Figure 1
457 458 459 460 461
Figure 2
462 463 464
23
Page 23 of 30
465
466
Figure 3
A
B
C
D
467
468 469
E
24
Page 24 of 30
470
25
Page 25 of 30
471 472 473
Central PictureR2.jpg
26
Page 26 of 30
474 475 476
Figure 1R2.jpg
27
Page 27 of 30
477 478 479
FIgure 2R2.jpg
28
Page 28 of 30
480 481 482
jtcvs-17-798R1_Figure 2R2.tif
29
Page 29 of 30
483 484
video thumb.tif
30
Page 30 of 30
S1043-0679(18)30077-7 https://doi.org/10.1053/j.semtcvs.2018.02.031 YSTCS 1078
To appear in:
Seminars in Thoracic and Cardiovascular Surgery
Please cite this article as: Hester F. Shieh, C. Jason Smithers, Thomas E. Hamilton, David Zurakowski, Gary A. Visner, Michael A. Manfredi, Russell W. Jennings, Christopher W. Baird, Descending Aortopexy and Posterior Tracheopexy for Severe Tracheomalacia and Left Mainstem Bronchomalacia, Seminars in Thoracic and Cardiovascular Surgery (2018), https://doi.org/10.1053/j.semtcvs.2018.02.031. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39
AATS 2017 Original Manuscript: Congenital Heart Disease Scientific Session Descending aortopexy and posterior tracheopexy for severe tracheomalacia and left mainstem bronchomalacia Hester F. Shieh, MD a, C. Jason Smithers, MD a, Thomas E. Hamilton, MD a, David Zurakowski, PhD a, Gary A. Visner, DO b, Michael A. Manfredi, MD c, Russell W. Jennings, MD a*, Christopher W. Baird, MD d* a
Department of Surgery, Boston Children’s Hospital, Harvard Medical School, Boston, MA 02115 b Department of Pulmonology, Boston Children’s Hospital, Harvard Medical School, Boston, MA 02115 c Department of Gastroenterology, Boston Children’s Hospital, Harvard Medical School, Boston, MA 02115 d Department of Cardiac Surgery, Boston Children’s Hospital, Harvard Medical School, Boston, MA 02115 *Co-senior authors Conflict of Interest Statement and Sources of Funding: There are no conflicts of interest or funding. IRB Approval: IRB-P00021702 approved 11/10/2016 Corresponding author: Christopher W. Baird, MD Department of Cardiac Surgery Boston Children’s Hospital Harvard Medical School 300 Longwood Ave., 612 Farley Boston, MA 02115 Tel: (617) 355-5637 Fax: (617) 730-4698 Email: [email protected] Article word count: 2701 Keywords: aortopexy
1
Page 1 of 30
40
Glossary of Abbreviations
41
BRUE = brief resolved unexplained event
42
CHD = congenital heart disease
43
DLB = diagnostic laryngoscopy and bronchoscopy
44
EA = esophageal atresia
45
MDCT = multidetector computed tomography
46
TEF = tracheoesophageal fistula
2
Page 2 of 30
47
Central Picture
48 49
Central Picture Legend: Cardiomegaly and a left descending aorta compressing the left
50
mainstem bronchus.
51 52
Central Message: Descending aortopexy and posterior tracheopexy are clinically
53
effective in treating severe tracheobronchomalacia and left mainstem intrusion with
54
significant symptom and anatomic improvements.
55 56
Perspective Statement: Descending aortopexy and posterior tracheopexy are effective in
57
treating severe tracheobronchomalacia and left mainstem intrusion with significant
58
improvements in clinical symptoms and degree of airway collapse on bronchoscopy.
59
Complex cases warrant an individualized and flexible surgical approach guided by
60
preoperative imaging and intraoperative bronchoscopy in multidisciplinary specialized
61
centers.
3
Page 3 of 30
62
Abstract
63
Objective: Posterior descending aortopexy can relieve posterior intrusion of the left
64
mainstem bronchus that may limit the effectiveness of posterior tracheobronchopexy. We
65
review outcomes of patients undergoing both descending aortopexy and posterior
66
tracheopexy for severe tracheobronchomalacia with posterior intrusion and left mainstem
67
compression to determine if there were resolution of clinical symptoms and
68
bronchoscopic evidence of improvement in airway collapse. Methods: All patients who
69
underwent both descending aortopexy and posterior tracheopexy from October 2012 to
70
October 2016 were retrospectively reviewed. Clinical symptoms, tracheomalacia scores
71
based on standardized dynamic airway evaluation by anatomical region, and persistent
72
airway intrusion requiring reoperation were collected. Data were analyzed by Wald and
73
Wilcoxon signed-ranks tests. Results: 32 patients underwent descending aortopexy and
74
posterior tracheopexy at median age 18 months (IQR 6-40 months). Median follow up
75
was 3 months (IQR 1-7 months). There were statistically significant improvements in
76
clinical symptoms postoperatively, including cough, noisy breathing, prolonged and
77
recurrent respiratory infections, ventilator dependence, blue spells, and brief resolved
78
unexplained events (BRUEs) (all P<.001), as well as exercise intolerance (P=.033),
79
transient respiratory distress requiring positive pressure (P=.003), and oxygen
80
dependence (P=.007). Total tracheomalacia scores improved significantly (P<.001), with
81
significant segmental improvements in the middle (P=.003) and lower (P<.001) trachea,
82
and right (P=.011) and left (P<.001) mainstem bronchi. 2 patients (6%) had persistent
83
airway intrusion requiring reoperation with anterior aortopexy and/or tracheopexy.
84
Conclusions: Descending aortopexy and posterior tracheopexy are effective in treating
4
Page 4 of 30
85
severe tracheobronchomalacia and left mainstem intrusion with significant improvements
86
in clinical symptoms and degree of airway collapse on bronchoscopy.
87 88
Abstract Word Count: 248
5
Page 5 of 30
89
1. Introduction
90
Tracheobronchomalacia refers to a weakness or deformation of the airway such
91
that it is more susceptible to collapse with changes in pressure and compression by
92
adjacent thoracic structures [1]. It is often associated with esophageal atresia (EA),
93
tracheoesophageal fistula (TEF), and congenital heart disease (CHD) [2]. Severe
94
tracheobronchomalacia is characterized by dynamic airway collapse in spontaneously
95
breathing patients with anterior vascular compression and/or posterior membranous
96
tracheal intrusion [3]. Anterior ascending aortopexy addresses anterior vascular
97
compression by indirectly elevating the anterior wall of the trachea, but does not directly
98
address posterior membranous tracheal intrusion [4]. We recently reported a series of
99
patients who underwent posterior tracheopexy for severe symptomatic tracheomalacia
100
with posterior intrusion with promising short-term results [5]. The effectiveness of
101
posterior tracheopexy can be limited by left mainstem bronchomalacia in some patients.
102
Posterior descending aortopexy can be used to relieve left mainstem posterior intrusion
103
and compression between the descending aorta and pulmonary artery. We now review a
104
series of patients who underwent both descending aortopexy and posterior tracheopexy
105
for severe symptomatic tracheobronchomalacia with posterior intrusion and left
106
mainstem compression to determine if there were resolution of clinical symptoms and
107
bronchoscopic evidence of improvement in airway collapse.
108
2. Methods
109
The Esophageal and Airway Treatment (EAT) Center at Boston Children’s
110
Hospital is a multidisciplinary care team consisting of three pediatric surgeons, one
111
pediatric cardiothoracic surgeon, one pediatric pulmonologist, and two pediatric
6
Page 6 of 30
112
gastroenterologists. We retrospectively reviewed all patients who underwent both
113
descending aortopexy and posterior tracheopexy at Boston Children’s Hospital from
114
October 2012 to October 2016 under an approved institutional review board protocol
115
(IRB-P00021702).
116
Patient demographics, pre- and postoperative clinical symptoms and airway
117
evaluation, surgical techniques, and persistent airway intrusion requiring reoperation,
118
were collected. Clinical symptoms included cough, barking cough, noisy breathing,
119
prolonged pulmonary infection, recurrent pulmonary infections, exercise intolerance,
120
transient respiratory distress requiring positive pressure, oxygen dependence, ventilator
121
dependence, blue spells, and brief resolved unexplained events (BRUEs) (formerly
122
known as apparent life-threatening events (ALTEs)).
123
Pre- and postoperative endoscopic airway evaluation was performed by the
124
primary surgeons involved. Diagnostic laryngoscopy and bronchoscopy (DLB) was done
125
under general anesthesia in spontaneously breathing patients. After assessing supraglottic
126
structures and vocal cord function, the vocal cords were anesthetized with topical
127
lidocaine, the larynx was assessed for presence of a laryngeal cleft, and a Hopkins II rod
128
lens was inserted through the cords to assess for TEF, tracheal diverticulum, and dynamic
129
motion in the tracheobronchial tree throughout the respiratory cycle. A standardized
130
tracheomalacia scoring system based on dynamic airway evaluation was used to
131
determine pre- and postoperative tracheomalacia scores (Table 1) [3,5-6]. The
132
tracheobronchial tree was scored by the percentage of open airway during cough or
133
Valsalva out of 100 at each anatomic region: upper (T1), middle (T2), and lower (T3)
134
trachea, and right and left mainstem bronchi, with a maximum score of 500. Dynamic
7
Page 7 of 30
135
airway multidetector computed tomography (MDCT) was performed preoperatively to
136
evaluate for aberrant vascular anatomy or associated lung parenchymal disease, and to
137
identify the artery of Adamkiewicz [6]. As it is difficult to justify postoperative radiation
138
and routine MDCT imaging in asymptomatic children, bronchoscopic airway evaluation
139
was used for postoperative follow up.
140
The operating surgeon determined the operative plan and approach based on
141
endoscopic evaluation and preoperative MDCT, as part of a multidisciplinary team.
142
Generally patients with associated esophageal disease underwent right posterior
143
thoracotomy, those with cardiac disease underwent median sternotomy, and those with
144
vascular rings underwent left thoracotomy. The esophagus, back wall of the trachea,
145
thoracic duct, and/or aorta were fully dissected and mobilized, taking care to protect the
146
left vagus nerve and left recurrent laryngeal nerve. In patients undergoing sternotomy, the
147
ductal ligament was often divided to extensively mobilize the ascending aorta, transverse
148
aortic arch, and descending aorta. A recurrent TEF or residual tracheal diverticulum from
149
a previously repaired TEF was corrected if present by resecting the TEF or diverticulum
150
flush with the tracheal wall under endoscopic visualization. Pexy procedures were all
151
done under direct flexible bronchoscopic guidance. Posterior descending aortopexy was
152
performed by passing autologous pleural or pericardial pledgeted polypropylene sutures
153
to secure the aorta to the side of the spine, and as posteriorly as necessary to relieve
154
posterior pressure off the left mainstem bronchus (Figure 1). This posterior movement of
155
the aorta may necessitate dividing one or more intercostals, and preoperative MDCT
156
helps to ensure that the artery of Adamkiewicz is not divided. Arm-leg blood pressure
157
measurements or arterial line tracing monitoring were used to confirm no descending
8
Page 8 of 30
158
aortic gradient. Posterior tracheopexy was performed by passing autologous pleural or
159
pericardial pledgeted sutures into but not through the posterior tracheal membrane, and
160
securing them to the anterior longitudinal spinal ligament, in such a fashion as to
161
optimize tracheal lumen and posterior tracheal membrane support. Suture placement is
162
guided by intraoperative bronchoscopic guidance to avoid full thickness bites and to
163
optimize placement of the sutures. Additional airway pexy procedures were similarly
164
performed by passing pledgeted sutures to secure the bronchial posterior membrane
165
posteriorly to the lateral edges of the anterior longitudinal ligament of the spine. In
166
patients undergoing sternotomy, anterior pexy sutures to support and/or displace the
167
vasculature and/or airway were then passed through the sternum and secured under direct
168
bronchoscopic visualization following sternal closure.
169
2.1 Statistical Analysis
170
To assess resolution of clinical symptoms, the percentage of patients with each
171
symptom pre- and postoperatively was compared by the Wald chi-square test using
172
logistic regression modeling with a generalized estimating equations (GEE) approach to
173
account for the binary paired data [7]. Changes in tracheomalacia scores for each airway
174
segment were determined by the Wilcoxon signed-ranks test [8]. Freedom from
175
reoperation was estimated using the Kaplan-Meier product-limit method [9]. Statistical
176
analysis was performed using IBM SPSS Statistics (version 23.0, IBM Corporation,
177
Armonk, NY). A two-tailed P<.05 was considered statistically significant.
178
3. Results
179
32 patients underwent descending aortopexy and posterior tracheopexy at median
180
age 18 months (interquartile range (IQR) 6-40 months). 63% (20 patients) were male.
9
Page 9 of 30
181
Median estimated gestational age (EGA) was 34 weeks (IQR 31-36 weeks). 66% (21
182
patients) were associated with EA, including 18 patients with type C EA, 2 patients with
183
type A EA, and 1 patient with type B EA. 19% (6 patients) had long gap EA. 69% (22
184
patients) had associated cardiac disease, and 19% (6 patients) had VACTERL syndrome.
185
63% (20 patients) had a prior EA repair, and 19% (6 patients) had a prior anterior
186
aortopexy. 19% (6 patients) had vascular rings, including 3 patients with double aortic
187
arch, 2 patients with right aortic arch with Kommerell diverticulum, and 1 patient with
188
circumflex aorta. 9% (3 patients) had a prior vascular ring division, 3% (1 patient) had a
189
prior coarctation repair, and 13% (4 patients) had a prior patent ductus arteriosus ligation.
190
3% (1 patient) had double superior vena cava and 3% (1 patient) had dysplastic
191
pulmonary valve stenosis. 6% (2 patients) had tetralogy of Fallot, one of which had
192
undergone prior repair. 22% (7 patients) had atrial and/or ventricular septal defects, 3 of
193
which had undergone prior septal defect closures.
194
Upper airway anomalies were common on preoperative and intraoperative
195
bronchoscopy. 3% (1 patient) had laryngomalacia. No patients had preoperative vocal
196
cord dysfunction or paralysis. 22% (7 patients) had laryngeal clefts, 6 patients with type 1
197
clefts and 1 patient with a type 3 cleft. 19% (6 patients) had some degree of subglottic
198
stenosis. The tracheobronchial tree was scored by anatomic region (Table 1).
199
Preoperatively, the middle (T2) trachea, lower (T3) trachea, and left mainstem bronchus
200
were the most severely affected, with median scores of 0 (IQR 0-45), 0 (IQR 0-30), and
201
20 (IQR 0-50), respectively.
202
Operative approach was by right thoracotomy in 56% (18 patients), median
203
sternotomy in 22% (7 patients), left thoracotomy in 19% (6 patients), and right neck
10
Page 10 of 30
204
dissection with right thoracotomy in 3% (1 patient). All patients underwent descending
205
aortopexy and posterior tracheopexy under intraoperative bronchoscopic guidance. 53%
206
(17 patients) underwent additional airway or vascular pexy procedures, including left
207
mainstem bronchopexy in 38% (12 patients), right mainstem bronchopexy in 16% (5
208
patients), innominate artery pexy in 9% (3 patients), anterior tracheopexy in 9% (3
209
patients), and anterior aortopexy in 9% (3 patients). 6% (2 patients) had an aberrant
210
subclavian artery behind the trachea, requiring mobilization of the artery in 1 patient and
211
division of the artery in 1 patient. 53% (17 patients) had an associated tracheal
212
diverticulum that was resected flush with the trachea. Concomitant procedures included
213
septal defect closure in 13% (4 patients), vascular ring division in 9% (3 patients),
214
tetralogy of Fallot repair in 3% (1 patient), and pulmonary valve replacement in 3% (1
215
patient).
216
Median days on the ventilator after surgery were 4 days (IQR 0-11 days). Median
217
total intensive care unit (ICU) stay was 10 days (IQR 2-23 days). Median total hospital
218
length of stay was 15 days (IQR 7-42 days). There were no significant early
219
complications including hemorrhage or infection. There were no complications with
220
erosion of the pledgetted sutures into the aorta or trachea, as the pledgets are autologous.
221
There were no mortalities.
222
Median clinical follow up was 3 months (IQR 1-7 months). There were
223
statistically significant improvements in clinical symptoms postoperatively, including
224
cough, barking cough, noisy breathing, prolonged and recurrent respiratory infections,
225
ventilator dependence, blue spells, and BRUEs (all P<.001), as well as exercise
226
intolerance (P=.033), transient respiratory distress requiring positive pressure (P=.003),
11
Page 11 of 30
227
and oxygen dependence (P=.007) (Figure 1). At latest follow up, no patients had
228
recurrence of a blue spell or BRUE.
229
47% (15 patients) underwent postoperative follow up evaluation with
230
bronchoscopy at median 2 months (IQR 1-6.5 months). Total tracheomalacia scores on
231
bronchoscopy improved significantly (P<.001), with significant segmental improvements
232
in the middle (T2) (P=.003) and lower (T3) (P<.001) trachea, and the right (P=.011) and
233
left (P<.001) mainstem bronchi (Table 1). The greatest areas of numerical improvement
234
were in the segments most affected preoperatively, namely the middle (T2) and lower
235
(T3) trachea, as well as the left mainstem bronchus.
236
6% (2 patients) had persistent airway intrusion requiring reoperation, using
237
anterior aortopexy and/or anterior tracheopexy. The 2 reoperations were within 2 months
238
following the index surgery. Kaplan-Meier analysis estimated 92% of patients to be free
239
from reoperation at 3 months follow up (95% confidence interval 85-99%).
240
6% (2 patients) had tracheostomies preoperatively, all for severe tracheomalacia.
241
Overall, 9% (3 patients) had tracheostomies postoperatively, with no significant
242
difference when compared to preoperatively (P=.31). The 2 patients with preoperative
243
tracheostomies on the ventilator were able to wean to tracheostomy collar
244
postoperatively. One additional patient with associated severe cardiac disease, transferred
245
intubated on mechanical ventilation, had a tracheostomy placed for prolonged intubation
246
and ventilator weaning.
247
4. Discussion
248
Tracheobronchomalacia is an underestimated disease, given the wide spectrum of
249
disease with nonspecific chronic respiratory symptoms that are commonly misdiagnosed
12
Page 12 of 30
250
[1-2,10-11]. It is a common respiratory problem in patients with EA and CHD [12-15].
251
Among at least CHD patients, tracheobronchomalacia has been associated with increased
252
ventilator days, length of stay, and mortality [15-17]. Excessive airway collapse leads to
253
ineffective ventilation and poor clearance of secretions, resulting in frequent respiratory
254
infections, respiratory failure, and apneic events, all of which are less well tolerated in
255
patients with EA and CHD, underscoring the importance of early diagnosis in these
256
populations [1-2,10,18]. Our multidisciplinary care team routinely uses a standardized
257
reporting and scoring system based on anatomic region for endoscopic evaluation [3,5-6].
258
The greater the severity of airway collapse, indicated by a lower tracheomalacia score,
259
combined with the presence of clinical symptoms, may indicate the need for surgical
260
intervention, possibly concomitant with EA or CHD repair.
261
Preoperative dynamic airway MDCT is used in conjunction with bronchoscopy to
262
inform the operative plan. MDCT is particularly useful in evaluating complex cases with
263
aberrant vascular anatomy and associated cardiac or esophageal anomalies. Several
264
different types of disease resulting in complex tracheobronchomalacia include a right
265
aortic arch with large Kommerell diverticulum (Figure 2A), a right to left circumflex
266
aorta (Figure 2B), a double aortic arch compressing the trachea (Figure 2C), a left aortic
267
arch compressing the trachea (Figure 2D), and a left descending aorta compressing the
268
left mainstem bronchus (Figure 2E).
269
The management of severe tracheobronchomalacia remains difficult with little
270
consensus on treatment and surgical approach [1-2,15,19]. Surgical options include pexy
271
procedures (ascending and/or descending aortopexy, anterior and/or posterior
272
tracheopexy), tracheal resection, internal stents, and external stabilization [1-5,15,20-24].
13
Page 13 of 30
273
Anterior ascending aortopexy indirectly supports the anterior tracheal wall and is the
274
most commonly used technique, but has a reported failure rate of 10-25% in the literature
275
[15,22-24]. Direct anterior and/or posterior tracheopexy, as first reported by our group,
276
improve airway patency by directly addressing anterior malformed tracheal cartilage and
277
posterior membranous tracheal intrusion [2-3,5].
278
Left mainstem bronchomalacia is a challenging entity that may limit airflow,
279
mucus clearance, and the effectiveness of direct tracheopexy in some cases, which is the
280
focus of this paper [25]. The descending aorta may be anteriorly displaced, intruding into
281
the back wall of the left mainstem bronchus, while the pulmonary artery may be
282
posteriorly displaced, compressing the airway from the front, resulting in left mainstem
283
vascular compression. Surgical options include posterior descending aortopexy,
284
descending aortic translocation, pulmonary artery anterior fixation, internal stents, and
285
external splints [26-32].
286
In this series, we show that descending aortopexy and posterior tracheopexy are
287
clinically effective in treating severe tracheobronchomalacia with posterior intrusion and
288
left mainstem compression. Postoperatively, there were significant improvements in
289
clinical symptoms, as well as anatomic tracheomalacia scores. Anterior and posterior
290
tracheopexy or bronchopexy provide direct support to the airway, whereas anterior
291
ascending and posterior descending aortopexy indirectly support the airway by directly
292
addressing
293
tracheobronchomalacia, especially those with altered anatomy associated with EA or
294
CHD, warrant an individualized and flexible surgical approach guided by intraoperative
vascular
compression
of
the
airway.
Complex
cases
of
severe
14
Page 14 of 30
295
bronchoscopy. In fact, in our series, 53% of patients underwent concomitant airway
296
and/or vascular pexy procedures to optimize each individual airway.
297
Posterior descending aortopexy is a technically challenging procedure. Through a
298
right posterior thoracotomy approach, sutures are placed in the anterior aspect of the
299
descending aorta, and then blindly placed on the left side of the spine close to the
300
costovertebral junction by passing the needle out the anterior longitudinal ligament of the
301
spine. Through a median sternotomy approach, it can be even more challenging to
302
mobilize and expose the descending aorta, and place sutures to posteriorly displace it. We
303
find it best to monitor upper and lower extremity blood pressure or arterial line tracings
304
to avoid aortic stenosis.
305
There are a number of limitations to this study. Retrospective chart review was
306
used to collect the data and follow up, including pre- and postoperative clinical symptoms
307
and bronchoscopy findings. Although patients are followed closely by our
308
multidisciplinary clinic, further studies could utilize a prospective structured clinical
309
symptom questionnaire to further standardize reporting. Bronchoscopy can be subjective
310
and was performed by three primary operating surgeons. One study in adults showed
311
appropriate inter- and intraobserver reliability in flexible bronchoscopy, however less is
312
known in the pediatric population [33]. Future work can include bronchoscopic analysis
313
by independent observers to make a more statistically valid comparison. Postoperative
314
endoscopic evaluation was not available for all patients, however we used the
315
standardized scoring system to demonstrate resolution of tracheomalacia postoperatively
316
in those evaluated. Our standard protocol for endoscopic postoperative evaluation is at 1
317
year for longitudinal airway assessment if clinically asymptomatic, unless the patient is
15
Page 15 of 30
318
undergoing another scheduled procedure. Our study cohort included a heterogeneous
319
group of complex patients often requiring concomitant airway/vascular pexy procedures
320
or adjunct therapies that may have contributed to outcomes and confounded the influence
321
of surgical treatment alone. As our preference is to correct all anomalies that may affect
322
airway or cardiac function at the initial operation to try to prevent multiple reoperations,
323
it would be nearly impossible to isolate patients undergoing only one procedure. Follow
324
up intervals were relatively short-term and variable.
325
In conclusion, descending aortopexy and posterior tracheopexy are effective in
326
treating severe tracheobronchomalacia and left mainstem compression with significant
327
improvement or resolution of clinical symptoms and degree of airway collapse on
328
bronchoscopy. Further studies to follow long-term outcomes of this technique are
329
certainly warranted and ongoing. Given the heterogeneity and complexity of this patient
330
population with significant morbidity, treatment and long-term follow up is best done in
331
multidisciplinary specialized centers for individualized patient care.
332 333 334 335 336
Acknowledgment H.F.S. was supported by the Joshua Ryan Rappaport Fellowship of the Department of Surgery at Boston Children’s Hospital.
337 338
16
Page 16 of 30
339
References
340
1. Fraga JC, Jennings RW, Kim PC. Pediatric tracheomalacia. Semin Pediatr Surg 2016;
341 342 343 344
25:156–64. 2. Bairdain S, Zurakowski D, Baird CW, Jennings RW. Surgical treatment of tracheobronchomalacia: a novel approach. Paediatr Respir Rev 2016; 19:16–20. 3. Bairdain S, Smithers CJ, Hamilton TE, Zurakowski D, Rhein L, Foker JE, et al.
345
Direct
346
tracheobronchomalacia: short-term outcomes in a series of 20 patients. J Pediatr Surg
347
2015; 50:972–7.
tracheobronchopexy
to
correct
airway
collapse
due
to
severe
348
4. Jennings RW, Hamilton TE, Smithers CJ, Ngerncham M, Feins N, Foker JE. Surgical
349
approaches to aortopexy for severe tracheomalacia. J Pediatr Surg 2014; 49:66-70,
350
discussion 70-1.
351 352 353
5. Shieh HF, Smithers CJ, Hamilton TE, Zurakowski D, Rhein LM, Manfredi MA, et al. Posterior tracheopexy for severe tracheomalacia. J Pediatr Surg 2017; in press. 6. Ngerncham
M,
Lee
EY,
Zurakowski
D,
Tracy
DA,
Jennings
R.
354
Tracheobronchomalacia in pediatric patients with esophageal atresia: comparison of
355
diagnostic laryngoscopy/bronchoscopy and dynamic airway multidetector computed
356
tomography. J Pediatr Surg 2015; 50:402–7.
357 358 359 360 361
7. McCullagh P, Nelder JA. Generalized Linear Models, 2nd ed. London: Chapman & Hall; 1989:98-148. 8. Altman DG. Practical Statistics for Medical Research. London: Chapman & Hall; 1991:189-228. 9. Cox DR, Oakes D. Analysis of Survival Data. London, Chapman & Hall; 1984:48-61.
17
Page 17 of 30
362
10. Boogaard R, Huijsmans SH, Pijnenburg MW, Tiddens HA, de Jongste JC, Merkus PJ.
363
Tracheomalacia
364
characteristics. Chest 2005; 128:3391–7.
and
bronchomalacia
in
children:
incidence
and
patient
365
11. Fischer AJ, Singh SB, Adam RJ, Stoltz DA, Baranano CF, Kao S, et al.
366
Tracheomalacia is associated with lower FEV1 and Pseudomonas acquisition in
367
children with CF. Pediatr Pulmonol 2014; 49:960–70.
368 369
12. Spitz L, Kiely E, Brereton RJ. Esophageal atresia: five year experience with 148 cases. J Pediatr Surg 1987; 22:103–8.
370
13. Filler RM, Messineo A, Vinograd I. Severe tracheomalacia associated with
371
esophageal atresia: results of surgical treatment. J Pediatr Surg 1992; 27:1136-40,
372
discussion 1140-1.
373
14. Cartabuke RH, Lopez R, Thota PN. Long-term esophageal and respiratory outcomes
374
in children with esophageal atresia and tracheoesophageal fistula. Gastroenterol Rep
375
2016; 4:310-4.
376 377
15. Ragalie WS, Mitchell ME. Advances in surgical treatment of congenital airway disease. Semin Thorac Cardiovasc Surg 2016; 28(1):62-8.
378
16. Chen Q, Langton-Hewer S, Marriage S, Hayes A, Caputo M, Pawade A, et al.
379
Influence of tracheobronchomalacia on outcome of surgery in children with
380
congenital heart disease and its management. Ann Thorac Surg 2009; 88(6):1970-4.
381
17. Pfammatter JP, Casaulta C, Pavlovic M, Berdat PA, Frey U, Carrel T. Important
382
excess morbidity due to upper airway anomalies in the perioperative course in infant
383
cardiac surgery. Ann Thorac Surg 2006; 81(3):1008-12.
384
18. Carden
KA,
Boiselle
PM,
Waltz
DA,
Ernst
A.
Tracheomalacia
and
18
Page 18 of 30
385
tracheobronchomalacia in children and adults: an in-depth review. Chest 2005;
386
127:984–1005.
387 388 389 390 391 392 393 394
19. Goyal V, Masters IB, Chang AB. Interventions for primary (intrinsic) tracheomalacia in children. Cochrane Database Syst Rev 2012; 10:CD005304. 20. Torre M, Carlucci M, Speggiorin S, Elliott MJ. Aortopexy for the treatment of tracheomalacia in children: review of the literature. Ital J Pediatr 2012; 38:62. 21. Dave S, Currie BG. The role of aortopexy in severe tracheomalacia. J Pediatr Surg 2006; 41:533–7. 22. Weber TR, Keller MS, Fiore A. Aortic suspension (aortopexy) for severe tracheomalacia in infants and children. Am J Surg 2002; 184:573–7, discussion 577.
395
23. Mitchell ME, Rumman N, Chun RH, Rao A, Martin T, Beste DJ, et al. Anterior
396
tracheal suspension for tracheobronchomalacia in infants and children. Ann Thorac
397
Surg 2014; 98:1246-53.
398 399
24. Lee SY, Kim SJ, Baek JS, Kwak JG, Lee C, Lee CH, et al. Outcomes of aortopexy for patients with congenital heart disease. Pediatr Cardiol 2013; 34(6):1469-75.
400
25. Hungate RG, Newman B, Meza MP. Left mainstem bronchial narrowing: a vascular
401
compression syndrome? Evaluation by magnetic resonance imaging. Pediatr Radiol
402
1998; 28(7):527-32.
403
26. Sacco O, Santoro F, Ribera E, Magnano GM, Rossi GA. Short-length ligamentum
404
arteriosum as a cause of congenital narrowing of the left main stem bronchus. Pediatr
405
Pulmonol 2016; 51(12):1356-61.
406
27. Arcieri L, Serio P, Nenna R, Di Maurizio M, Baggi R, Assanta N, et al. The role of
407
posterior aortopexy in the treatment of left mainstem bronchus compression. Interact
19
Page 19 of 30
408
Cardiovasc Thorac Surg 2016; 23(5):699-704.
409
28. Valerie EP, Durrant AC, Forte V, Wales P, Chait P, Kim PC. A decade of using
410
intraluminal tracheal/bronchial stents in the management of tracheomalacia and/or
411
bronchomalacia: is it better than aortopexy? J Pediatr Surg 2005; 40:904–7.
412
29. Morrison RJ, Hollister SJ, Niedner MF, Mahani MG, Park AH, Mehta DK, et al.
413
Mitigation of tracheobronchomalacia with 3D-printed personalized medical devices
414
in pediatric patients. Sci Transl Med 2015; 7:285ra64.
415 416
30. Kosloske AM. Left mainstem bronchopexy for severe bronchomalacia. J Pediatr Surg 1991; 26(3):260-2.
417
31. Baird CW, Prabhu S, Buchmiller TL, Smithers C, Jennings R. Direct
418
tracheobronchopexy and posterior descending aortopexy for severe left mainstem
419
bronchomalacia associated with congenital pulmonary airway malformation and left
420
circumflex aortic arch. Ann Thorac Surg 2016; 102(1):e1-4.
421
32. McKenzie ED, Roeser ME, Thompson JL, De Leon LE, Adachi I, Heinle JS, et al.
422
Descending aortic translocation for relief of distal tracheal and proximal bronchial
423
compression. Ann Thorac Surg 2016; 102(3):859-62.
424
33. Majid A, Gaurav K, Sanchez JM, Berger RL, Folch E, Fernandez-Bussy S, et al.
425
Evaluation of tracheobronchomalacia by dynamic flexible bronchoscopy. A pilot
426
study. Ann Am Thorac Soc 2014; 11:951–5.
427 428
20
Page 20 of 30
429
Figure 1. Posterior Descending Aortopexy. A. Illustrates anatomic relationship of
430
aorta, esophagus, and spine. B. Cross-sectional view, in which the esophagus is rotated to
431
the right and the descending aorta is moved to the left and secured to the side of the spine
432
as posteriorly as necessary to relieve posterior pressure off the left mainstem bronchus. C.
433
Descending aortopexy sutures are tied, relieving left mainstem posterior intrusion and
434
compression between the descending aorta and pulmonary artery.
435 436
Figure 2: Pre- and Postoperative Clinical Symptoms.
437 438
Figure 3: Preoperative Dynamic Airway Multidetector Computed Tomography
439
(MDCT) of Complex Cases of Tracheobronchomalacia. A. Right aortic arch with
440
large Kommerell diverticulum. B. Right to left circumflex aorta. C. Double aortic arch
441
compressing the trachea. D. Left aortic arch compressing the trachea. E. Left descending
442
aorta compressing the left mainstem bronchus.
443 444
21
Page 21 of 30
445
Table 1: Tracheomalacia Scores. Pre- and postoperative tracheomalacia scores based
446
on standardized bronchoscopic evaluation. Scores are percentage of open airway out of
447
100 for each anatomical region. Data are median (IQR). Location
448 449 450 451 452 453 454 455
Postoperative (n=15) 100 (80-100) 75 (60-100) 100 (70-100) 100 (100-100) 70 (50-100)
P value
T1 T2 T3 Right bronchus Left bronchus
Preoperative (n=32) 80 (70-95) 0 (0-45) 0 (0-30) 78 (12-100) 20 (0-50)
Total
215 (145-268)
450 (360-475)
<.001*
.182 .003* <.001* .011* <.001*
Video 1: Descending Aortopexy and Posterior Tracheopexy. Operative procedure and its relevance as discussed by Dr. Russell Jennings.
22
Page 22 of 30
456
Figure 1
457 458 459 460 461
Figure 2
462 463 464
23
Page 23 of 30
465
466
Figure 3
A
B
C
D
467
468 469
E
24
Page 24 of 30
470
25
Page 25 of 30
471 472 473
Central PictureR2.jpg
26
Page 26 of 30
474 475 476
Figure 1R2.jpg
27
Page 27 of 30
477 478 479
FIgure 2R2.jpg
28
Page 28 of 30
480 481 482
jtcvs-17-798R1_Figure 2R2.tif
29
Page 29 of 30
483 484
video thumb.tif
30
Page 30 of 30