Tracheal stenosis involving >5 cm of trachea is debilitating for patients, impedes their quality of life and may even become life-threatening. Large, open defects often result in these patients because of multiple failed tracheal resections and endoscopic dilatations. Tracheal transplantation has been proposed as a curative solution, but revascularization of the trachea as an allograft has generally been considered prohibitively challenging due to a tenuous blood supply. We report a multidisciplinary effort resulting in the first vascularized, single-stage, deceased donor tracheal allograft transplantation in a woman with long-segment tracheal stenosis.
Long-segment tracheal stenosis (ie, stenotic lesions >5 cm in length) is debilitating, and generally leads to tracheostomy-dependence, an inability to speak and a need for repeated surgeries. Failed tracheal resections, endoscopic dilations or laser procedures in these patients may result in large tracheal stomata and these procedures do not guarantee reliable resolution of their stenosis1. Extensive defects may become life-threatening due to airway compromise, tracheoinnominate fistulae formation, or pulmonary infections1. Deceased donor tracheal allograft transplantation (DDTAT) can theoretically restore native airway patency for such patients, but single-stage transplantation has been avoided, driven largely by concerns regarding blood supply of the tracheoesophageal complex2.
We report the successful anesthetic management of a patient who underwent DDTAT for long-standing tracheal stenosis and tracheostomy dependence. Our group has described the surgical approach to this case elsewhere3. Written consent/Health Insurance Portability and Accountability Act authorization was obtained.
A 56-year-old woman with asthma, hypertension, noninsulin dependent (type-2) diabetes, coronary artery disease, hyperlipidemia, sarcoidosis and long-segment tracheal stenosis (~8 cm) presented for DDTAT. The patient had a prolonged intubation of ~3 weeks following an asthma attack in 2014. She subsequently developed severe tracheal stenosis (ie, >70%) and was left tracheostomy-dependent for the ensuing 7 years. In that time period, she underwent six tracheal procedures in order to address her stenosis, including endoscopic dilatations, laser and open resections. Approximately 2–3 years before DDTAT, several planning meetings were held involving otolaryngology, transplant medicine and surgery, thoracic surgery, intensive care, interventional pulmonology, and anesthesiology. These meetings were used to detail the clinical and scientific rationale, ensure necessary local and national approvals (ie, from the United Network for Organ Sharing), and to do “tabletop” exercises, tracing the chronology of the patient’s planned perioperative journey and considering potential challenges.
The tracheal donor was a 37-year-old man who sustained a subarachnoid hemorrhage and brain death. Following surgical exposure of the donor trachea and verification of usable vascular anatomy by the otolaryngology team (ie, adequate length, blood flow via doppler and no evidence of calcification), the recipient was brought to an adjacent operating room. Initial vital signs were all normal. She received nebulized albuterol through her tracheostomy (7.0 cuffless Shiley tracheostomy tube, Medtronic, Minneapolis, MN) before the induction of general anesthesia with propofol (100 mg), rocuronium (100 mg), and fentanyl (250 mcg). The patient’s tracheostomy device was then removed and replaced with a flexible 6.5 mm reinforced endotracheal tube. A pulse oximeter was placed on the patient’s left index finger and a 20-G intra-arterial catheter was placed in the patient’s right radial artery in anticipation of an intrathoracic approach with potential for great vessel compression. A tracheobronchial blocker was available for lung isolation. A processed electroencephalogram (Bispectral index monitor, Medtronic, Minneapolis, MN).
The maintenance anesthetic consisted of propofol, remifentanil, and rocuronium infusions. Methadone (10 mg) and ketamine (100 mg) were also administered. In spite of potential concerns for postoperative respiratory depression with methadone, this analgesic is a routine part of our head and neck enhanced recovery protocol and was deemed safe in this patient given the open access to the airway via tracheal stoma. A goal mean arterial pressure between 75 and 85 mm Hg was maintained with a phenylephrine infusion as needed. Close communication was maintained with surgical team during periods of endotracheal tube manipulation/removal, which were numerous. To decrease fire risk, the fraction of inspired oxygen was maintained at 0.3 during periods of tracheal surgery involving electrocautery. Immunosuppression with methylprednisolone 500 mg and antithymocyte globulin 200 mg were given over 240 minutes after adequate surgical exposure. Thoracic surgery involvement and one-lung ventilation were ultimately deemed unnecessary given satisfactory exposure of the surgical field. Venovenous extracorporeal membrane oxygenation was considered before the surgical exposure but was also deemed unnecessary. The cardiothoracic surgery team remained in the operating room on standby throughout the case, however.
The surgical team in the donor operating room then performed local vascular perfusion with University of Wisconsin solution and the trachea was taken en bloc to the recipient operating room. After 78 minutes of cold ischemia time, the allograft was taken out of ice and placed in the surgical field. Ventilation was maintained via the most caudad (lowest) portion of the tracheal stoma as the graft was placed in the recipient tracheal bed. The posterior membranous trachea was reconstructed first, and an operating microscope was then brought in to perform 2 arterial and 2 venous microvascular anastomoses, which included (in sequence):
- The allograft left lingual artery to the recipient left facial artery.
- The allograft left internal jugular vein to the recipient left internal jugular vein.
- The allograft right inferior thyroid artery to the recipient right transverse cervical artery.
- The allograft right facial vein to the recipient right common facial vein.
After completion of the first 2 anastomoses and 26 minutes of warm ischemia time, reperfusion occurred. The graft was supplied with blood via the patient’s left lingual and right inferior thyroid arteries. There was brisk perfusion of the distal tracheal cartilage and membranous portions of the trachea to the level of the carina. There were no hemodynamic changes during reperfusion, but preprepared syringes of calcium chloride, sodium bicarbonate and vasoactive agents were available. The remaining vascular anastomoses were completed, and then a period of preoxygenation occurred in preparation for an extended apneic period while the lower tracheal defect was closed. After closure of the lower stoma, the patient was intubated from the upper stoma by the anesthesia team using a 4.0 bronchoscope with a 6.0 reinforced endotracheal tube.
At the conclusion of the case, the patient was breathing spontaneously with pressure support of 5 cm H2O to limit positive pressure on the graft and suture sites. For graft surveillance purposes and to limit potential postoperative emergency airway manipulation, the decision was made to keep the patient intubated. A T-shaped laryngeal tube was utilized to maintain airway patency cephalad to the graft, and next to it a 6.0 mm reinforced endotracheal tube was left indwelling for bronchoscopy and the option of positioning the tip lower along the tracheal graft as needed (eg, to lessen compression by the cuff on a given area). Both devices were in the same stoma, above the graft. The anesthesia team performed intraoperative bronchoscopy to position the endotracheal tube tip just above the distal anastomosis.
The patient remained hemodynamically stable throughout the 11-hour case. In total, she received 3 L of Plasmalyte, 500 mL of 5% human albumin solution, and 4 units packed red blood cells owing to baseline anemia and surgical blood loss of 500 mL with a nadir hematocrit of 22. The patient’s total urine output was 1700 mL.
On postoperative day 6, the patient presented from the intensive care for the first of several planned bronchoscopies. At this point, the patient required minimal pressure support, and upon visualization of vital mucosa and healing suture lines, both the T-shaped laryngeal tube and endotracheal tube were removed and replaced with a Provox LaryTube, to allow for increased comfort and tolerance by the patient. Bedside bronchoscopy with biopsy was performed on the allograft on days 3, 6, 17, and 32, 45, and 72. A healthy, well-appearing graft was evident with serial bronchoscopy (Fig. 1). Pathology examinations and electron micrographs revealed neovascularization and cilia generation over the first 2–3 postoperative weeks. She was discharged home on postoperative day 37 and remains well over 9 months postoperatively and is no longer tracheostomy tube dependent.
Approaches to tracheal stenosis management vary depending on etiology and severity. Our patient’s comorbidities and multiple failed tracheal procedures left her tracheostomy-dependent with a large open defect, and without surgical options. While stenosis can be managed endoscopically or with laser treatments, not every patient will have successful treatment4. Published studies have revealed variable success rates, regardless of treatment modality, and large defects may result4–7. Patient selection and close collaboration with transplant staff is crucial to creating a DDTAT program, finding an appropriate donor and successfully carrying out the surgery. Indeed, the planning and approval required to undertake this case took several years, and the team felt encouraged to pursue this treatment based on their prior work with animal models of tracheal transplant8–11.
Our anesthetic approach borrowed from routinely employed management techniques in microvascular free flap reconstruction, including judicious intravenous fluid administration12 and a focus on early extubation, which is not only important in head and neck free flap surgery13, but also important in lung transplantation where it decreases anastomotic injury and postoperative infections14,15. A total intravenous anesthetic was utilized to limit waste gases during endotracheal tube manipulation and to facilitate smooth emergence. Ultimately a joint decision was made to keep the patient intubated to limit coughing and allow for surveillance bronchoscopies and biopsies. It is likely, however, that we will extubate future DDTAT patients immediately following surgery. Subsequent bronchoscopies, it should be noted, can be done with minimal sedation and airway topicalization given the insensate nature of the graft.
Adequate blood pressure was maintained via intermittent phenylephrine infusion, which has been proved safe in microvascular surgery15. A conservative transfusion threshold of Hgb >8 g/dL was also used. We were prepared for an intrathoracic component of the surgery, but ultimately did not need to deploy lung isolation or manage great vessel compression. However, future cases may require such maneuvers, as well as extracorporeal membranous oxygenation, to complete the surgery.
While true end-stage tracheal pathologies are rare, they are not isolated. In addition, our team is encountering patients with tracheal damage following COVID-19 (Coronavirus Disease 2019)-related airway procedures. In the aftermath of the COVID-19 pandemic, patients may present with severe tracheal stenosis. Indeed, innumerable percutaneous tracheostomies16–18, and prolonged intubations occurred during the pandemic16. This fact, along with potential direct tracheal damage from the infection itself19 may increase the need for DDTAT in post-COVID-19 patients. This is the first DDTAT case being presented and future work depends upon further investigation and data collection and reporting. As such, we believe anesthesiologists should consider the anesthetic management of DDTAT, and report further cases as they occur.