Long-term Results of Innominate Artery Reimplantation for Tracheal Compression

Objective  To evaluate the safety and long-term efficacy of innominate artery reimplantation as treatment for innominate artery compression syndrome.

Design  Retrospective medical record review from January 1, 1992, to December 31, 2007, and telephone interview.

Setting  Children's hospital.

Patients  Twenty-two children with innominate artery compression of the trachea requiring surgical intervention.

Intervention  Innominate artery reimplantation for innominate artery compression syndrome.

Main Outcome Measures  Respiratory symptoms, rigid bronchoscopy results, operative time, estimated blood loss, duration of hospitalization, and complications.

Results  Twenty-two patients were seen with innominate artery compression of the trachea causing respiratory distress. The diagnosis was made based on chest magnetic resonance images, computed tomographic angiograms, and rigid bronchoscopy results demonstrating significant (>75%) tracheal stenosis. Following innominate artery reimplantation, 19 patients (86%) had complete resolution of symptoms and discontinuation of respiratory support. Two patients had partial resolution, and 1 patient had no improvement. The time course of resolution was immediate in 13 patients, after 3 months in 1 patient, and unknown in 5 patients. The mean operative time was 73 minutes, with a mean estimated blood loss of 18 mL. The mean postoperative duration of hospitalization was 6.5 days, with a mode and median stay of 3 days. One patient required prolonged intubation (5 days), and another patient developed postpericardiotomy syndrome. Telephone interview of 10 patients at a mean of 5 years after surgery revealed continued resolution of respiratory symptoms if previously resolved and no long-term complications.

Conclusions  Innominate artery reimplantation is a safe and effective treatment for innominate artery compression syndrome. Resolution of symptoms occurs quickly in most patients, with excellent long-term results.

Innominate artery compression of the trachea was first described by Gross and Neuhauser¹ in 1948, and it is now recognized as the most common cause of vascular airway compression.² The condition is thought to be due to more distal attachment of the innominate artery along the aortic arch, compressing the trachea as it crosses anteriorly. Mediastinal crowding has also been suggested as a plausible cause.³ This argument is supported by the observation that compression reduces with age and that the condition is rarely diagnosed beyond infancy.⁴

Regardless of the etiology, vascular compression of the trachea by the innominate artery is a significant cause of respiratory distress in the pediatric population. Patients typically are seen with apnea, dysphagia, expiratory stridor, deep tracheal cough, and increased work of breathing.⁵ In general, the condition is discovered in the first year of life; however, there are reports of patients whose conditions are diagnosed later in life.⁶

Successful treatment by aortopexy of the artery to the dorsal surface of the sternum has been reported by multiple authors achieving success rates in 56% to 87% of patients.^1,5,7 However, aortopexy has many downsides in that resolution of symptoms does not typically occur immediately. The procedure requires suturing the adventitia of the aorta, which may fail over time, and a large thymus may prevent adequate suspension. Hammocking the aorta with fascia and division of the thymus⁸ has been described to address some of these concerns.

Innominate artery reimplantation is an alternative surgical option with many advantages over aortopexy. This surgical approach has previously been described by one of us (J.A.H.) and colleagues⁹ and by others^7,10 and has been used predominantly at Primary Children's Medical Center, Salt Lake City, Utah, during the past 15 years. The main advantage of innominate artery reimplantation over aortopexy is the ability to immediately correct the cause of obstruction by isolating the innominate artery and moving it away from the trachea. We believe that in doing so the airway obstruction improves to a greater extent and symptoms resolve more quickly than with aortopexy. The objectives herein are to describe an additional 22 patients treated with innominate artery reimplantation and to assess the duration of hospitalization, estimated blood loss during surgery, complications, improvement of symptoms, and long-term follow-up.

Institutional review board approval was obtained. Twenty-two patients (age range, 3 days to 7 years) were treated for innominate artery compression syndrome by innominate artery reimplantation between January 1, 1992, and December 31, 2007, at Primary Children's Medical Center. The diagnosis of innominate artery compression syndrome was made based on rigid bronchoscopy results demonstrating pulsatile compression of the anterior distal trachea (>75% tracheal stenosis) (Figure). The diagnosis was confirmed by computed tomographic angiography or chest magnetic resonance imaging. Surgery was recommended for patients who displayed significant airway symptoms, including apnea, cyanosis, increased work of breathing, failure to thrive, oxygen dependence, or failure to wean from mechanical ventilatory support. Stridor alone was not an indication for surgery unless it was associated with other symptoms. The medical records of all patients who underwent innominate artery reimplantation were retrospectively reviewed to determine age at presentation, initial symptoms, duration of hospitalization, operative time, estimated blood loss, duration of intubation, resolution of symptoms, and complications. Attempts were made to contact all patients to conduct follow-up telephone interviews. Ten patients were contacted and completed the telephone interview.

The innominate artery was reimplanted as previously described by Hawkins et al.⁹ Briefly, a median sternotomy approach was used. The innominate artery was clamped after heparinization and detached from the aorta. The innominate artery was then reimplanted, approximately 1 cm more proximal on the greater curvature of the aorta, away from the trachea. Cardiopulmonary bypass was not used. Patients were admitted to the pediatric intensive care unit for postoperative recovery, typically for 24 hours, and were then transferred to the surgical floor for the remainder of their hospitalization.

Demographics, initial symptoms, and outcomes are given in Table 1. The study group consisted of 15 boys (68%) and 7 girls (32%). The mean age at the time of innominate artery reimplantation was 320 days, with a median age of 140 days (age range, 3 to 2861 days [7 years]). Patients often had multiple initial symptoms such as cough, stridor, cyanosis, and increased work of breathing. The most common initial symptom was stridor, which was seen in 19 of 22 patients (86%). Eleven patients (50%) required oxygen support or had cyanotic episodes. Apnea was reported in 9 patients (41%), as was cough. Endotracheal intubation for respiratory failure was observed in 2 patients (9%). Failure to thrive was seen in 2 patients (9%). One patient required continuous positive airway pressure, and 2 patients (9%) experienced dysphagia.

Nineteen patients (86%) demonstrated complete resolution of symptoms and discontinuation of respiratory support or oxygen following innominate artery reimplantation. Two patients had partial resolution, and 1 patient had no improvement. The percentage of compression on preoperative rigid bronchoscopy was documented in 10 patients, with a mean compression of 75% (range, 50%-90%). In the remaining 12 patients, preoperative compression was noted along with its location without giving a numeric value for the percentage of compression. On postoperative rigid bronchoscopy, no evidence of compression was seen in 21 of 22 patients (95%). One patient demonstrated mild pulsatile compression of the distal third of the trachea, which was improved from preoperative rigid bronchoscopy findings.

Of 19 patients who experienced complete resolution of symptoms, the time course of improvement was immediate in 13 patients (68%), after 3 months in 1 patient (5%), and unknown in 5 patients (26%). The 5 patients whose time courses of resolution of symptoms were not established could not be contacted by telephone or mail. The medical records of these 5 patients show postoperative medical histories and physical examinations between 1 and 7 years after surgery that indicate complete resolution of symptoms. However, the histories and physical examinations in these 5 patients were performed in association with other illnesses and give no indication of how long it took for symptoms to resolve and for respiratory support to be discontinued.

Follow-up telephone interviews were obtained in 10 patients at a mean of 5 years after surgery. Family members of 9 patients indicated that all symptoms had resolved immediately and that there were no residual symptoms. None of the 9 patients experienced limitation in activity because of airway symptoms. The only patient without improvement of symptoms was noted to have had 3 bronchoscopies at 11, 22, and 37 months after innominate artery reimplantation. While the preoperative rigid bronchoscopy results had demonstrated anterior compression of the trachea at the midtracheal level, all 3 postoperative bronchoscopies showed no evidence of compression of the trachea. Two years after undergoing surgery, the patient was diagnosed as having gastroesophageal reflux disease with significant esophagitis. The patient ultimately underwent Nissen fundoplication and had only temporary improvement. This patient was contacted by telephone, and it was confirmed that innominate artery reimplantation did not result in symptomatic improvement. The patient experiences attacks of severe stridor that frequently require treatment at the emergency department. Between attacks, the patient experiences no airway symptoms or limitation in activities.

The 2 patients who demonstrated partial resolution of symptoms showed no airway compression on postoperative rigid bronchoscopy. One patient required respiratory support before surgery that was discontinued immediately after innominate artery reimplantation. However, mild noisy breathing and dysphagia continued in this patient, despite the absence of compression seen on rigid bronchoscopy. The other patient has continued noisy breathing and, although postoperative rigid bronchoscopy results showed no significant airway obstruction, mild pulsatile compression of the distal third of the trachea was noted. Neither of these 2 patients could be contacted for long-term follow-up.

The mean operative time was 73 minutes, with a median operative time of 72 minutes (range, 57-123 minutes). The mean estimated blood loss during surgery was 18 mL (range, 5-80 mL). The mean postoperative duration of hospitalization was 156 hours (6.5 days), with a mode and median stay of 72 hours (3 days) (range, 48- 720 hours [2-30 days]).

Surgical complications were infrequent. One patient required prolonged intubation (5 days), and another patient developed postpericardiotomy syndrome. Other complications included respiratory acidosis that was seen in 1 patient in the first 24 hours after innominate artery reimplantation and resolved on its own, as well as pulmonary edema in another patient that required diuresis.

Long-term follow-up revealed no neurologic deficits related to innominate artery reimplantation. One patient reported headaches occurring a couple of times each month. Another patient who had surgery in 2005 reported 4 seizures since the operation; however, the cause was not attributed to innominate artery reimplantation.

Long-term follow-up in patients who underwent aortopexy for vascular compression of the trachea has demonstrated variable success rates. Schier et al¹¹ described 24 patients with a mean follow-up of 9 years and demonstrated a 71% aortopexy success rate (17 patients). At a mean of 4.75 years after surgery, Anand et al¹² reviewed 8 patients who had undergone aortopexy and noted a 75% success rate (6 patients). Erwin et al⁵ demonstrated complete resolution of symptoms following aortopexy in 39 of 45 patients (87%), although 5 patients (11%) required multiple surgical procedures. Greenholz et al¹³ reported that 5 of 9 patients (56%) became symptom free after aortopexy. In a 2007 study, Abdel-Rahman et al¹⁴ conducted long-term follow-up on 20 patients for a mean of 7.8 years after aortopexy. They reported immediate and permanent improvement in 80% (16 of 20) of patients. Table 2 summarizes the published success rates of aortopexy and innominate artery reimplantation.

In comparison, our study demonstrated complete resolution of symptoms after innominate artery reimplantation in 19 of 22 patients (86%) and anatomical improvement based on rigid bronchoscopy results in 21 of 22 patients (96%). None of our patients required reoperation. Our success rate after innominate artery reimplantation is consistent with previously published series. Hawkins et al⁹ reported complete resolution of symptoms in 27 of 29 patients (93%) after innominate artery reimplantation. Partial resolution of symptoms was seen in the remaining 2 patients (7%). Combining the present study and the study by Hawkins et al, complete resolution of symptoms was seen in 46 of 51 patients (90%) after innominate artery reimplantation. Only 1 patient (2%) in this combined series had no resolution of symptoms. Myer et al⁶ reported similar success with innominate artery reimplantation in a group of 16 patients.

Operative time, estimated blood loss, duration of hospitalization, and complications are also important considerations in choosing a treatment modality. A 2005 study by Valerie et al¹⁵ retrospectively reviewed 11 patients undergoing aortopexy. Their study found that 4 of 11 patients (36%) developed pericardial effusion as a complication of surgery. The authors reported a mean operative time of 190 minutes, with a mean estimated blood loss of 26 mL. Schier et al¹¹ described surgical complications in 4 of 24 patients (17%) undergoing aortopexy in their study. Clevenger et al¹⁶ reported that 83% of patients experienced major postoperative complications after aortopexy. In their series, the mean postoperative duration of hospitalization was 25 days.

By comparison, our study found a mean estimated blood loss of 18 mL and a mean operative time of 73 minutes associated with innominate artery reimplantation. The estimated blood loss is comparable to the mean 26-mL estimated blood loss associated with aortopexy as reported by Valerie et al.¹⁵ Our mean operative time of 73 minutes is shorter than the mean operative time of 190 minutes reported by Valerie et al for aortopexy. This is an important point because an objection to innominate artery reimplantation has been the additional complexity and operative time. Our study recorded a mean duration of hospitalization after innominate artery reimplantation of 6.5 days, with a mode and mean stay of 3 days, which is much shorter than the 25 days recorded in the only published series (to our knowledge) reporting on duration of hospitalization after aortopexy.¹⁶ Two of 22 patients (9%) in our study had surgical complications after innominate artery reimplantation. One patient in our study required prolonged intubation of 5 days following surgery. Another patient experienced postpericardiotomy syndrome. This compares favorably with the 17% to 83% complication rates reported in aortopexy studies.^5,7

Our results indicate that the long-term success rates in treating innominate artery compression syndrome with innominate artery reimplantation are as good as or better than the long-term success rates in patients undergoing aortopexy (Table 2). Aortopexy for vascular compression of the trachea has had reported success rates ranging from 56% to 87%.^5,11-14 Innominate artery reimplantation success rates have ranged from 86% to 93%.^6,9 Success rates may vary based on reporting methods. Our rates are based on medical records, follow-up interviews, and postoperative rigid bronchoscopy results. Although 3 of our 22 patients reported lingering symptoms, postoperative rigid bronchoscopy results indicated that only 1 of 22 patients had residual vasculature compression of the trachea (a 95% success rate). The only postoperative rigid bronchoscopy finding that demonstrated any remaining compression of the trachea by the innominate artery showed mild pulsatile compression of the distal third of the trachea.

Innominate artery compression syndrome is associated with esophageal atresia,¹⁷ which may cause persistent symptoms, although we did not observe this in our study. One patient in our study had postoperative dysphagia that did not resolve, but it was not associated with esophageal atresia.

In conclusion, innominate artery reimplantation is a safe and effective treatment for innominate artery compression syndrome. The operative time, estimated blood loss, and duration of hospitalization are comparable or lower than those associated with aortopexy in published series. Resolution of symptoms occurs quickly in most patients, with excellent long-term results.

Correspondence: J. Fredrik Grimmer, MD, Division of Otolaryngology, Department of Surgery, School of Medicine, Primary Children's Medical Center, University of Utah, 100 N Medical Dr, Ste 4500, Salt Lake City, UT 84113 (fred.grimmer@imail.org).

Submitted for Publication: January 29, 2008; final revision received April 14, 2008; accepted May 5, 2008.

Author Contributions: Drs Grimmer, Hawkins, Park, and Kouretas had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Grimmer, Hawkins, Park, and Kouretas. Acquisition of data: Grimmer, Herway, and Hawkins. Analysis and interpretation of data: Grimmer, Herway, Hawkins, Park, and Kouretas. Drafting of the manuscript: Grimmer and Herway. Critical revision of the manuscript for important intellectual content: Grimmer, Hawkins, Park, and Kouretas. Administrative, technical, and material support: Grimmer, Park, and Kouretas. Study supervision: Grimmer, Hawkins, and Park.

Financial Disclosure: None reported.

Previous Presentation: This study was presented at the 2008 American Society of Pediatric Otolaryngology Scientific Program; May 4, 2008; Orlando, Florida.