If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
The most common indication for tracheal resection is stenosis secondary to cuff injury from either tracheostomy tube or endotracheal tube or tracheostomy stomal injury. These injuries occur in the upper, middle, or distal trachea and are amenable to resection and reconstruction with excellent results. Techniques have been devised to allow resection of up to one half of the adult trachea.
When postintubation stenosis and other inflammatory processes extend from the upper trachea into the subglottic larynx, the problem becomes more complex. If the lesion extends well above the lower border of the cricoid cartilage, circumferential resection is not possible because of the entry location of the recurrent laryngeal nerves into the larynx medial and posterior to the inferior cornua of the thyroid cartilage along the back of the posterior cricoid lamina (Figure I). The stenosis may involve only the anterolateral portion of the cricoid. When circumferential stenosis exists involving the posterior cricoid plate as well, the problem is much more formidable. The closer the stenosis comes to involving the vocal cords, the less suitable it is for resection, reconstruction, and preservation of laryngeal function.
in 1979 devised a radiological classification of upper airway stenosis based on the amount of involvement of the subglottic space (Figure II). Type A is a high tracheal stenosis not involving the cricoid. This lesion is easily treated by segmental resection and tracheotracheal anastomosis. In type B, stenosis reaches the lower border of the cricoid cartilage and involves anastomosis between the trachea and cricoid cartilage. Type C involves inflammation of the anterior portion of the cricoid cartilage. Correction requires resection of the anterior portion of the cricoid cartilage and may require resurfacing of the posterior portion of the cricoid surface if the mucosa is severely scarred. Type D involves inflammation and stenosis of the glottis and has insufficient subglottic space for reconstruction. There currently is no reliable single stage method of reconstruction for stenosis at this level.
There is no single factor responsible for subglottic stenosis. The vast majority of subglottic stenoses result from complications of endotracheal or tracheostomy tubes. In our experience, only endotracheal intubation was responsible for the largest number of subglottic stenoses. The point of greatest pressure from oral endotracheal tubes is the posterior cricoid plate (Figure III). The size of the larynx varies from individual to individual. It is recognized that women have a smaller larynx than men, and smaller adults have a proportionately smaller larynx. Larger endotracheal tubes will exert circumferential pressure at the glottic, subglottic, and cricoid levels and, if left in place long enough, may lead to circumferential necrosis at these levels. The aforementioned factors would seem to be responsible for the subglottic stenosis seen in patients that have been intubated with an oral endotracheal tube, as well as for posterior commissure stenosis. Tracheostomy tubes placed through the first tracheal ring or upward pressure by the tube in kyphotic patients may lead to retrograde erosion of the cricoid cartilage. Cricothyroidostomy through the cricothyroid membrane can also result in subglottic stenosis.
The next most common cause of subglottic stenosis results from idiopathic stenosis.
Failed attempts at repair or delayed recognition of injuries can result in subglottic stenosis.
There are a variety of other inflammatory or infectious problems that are extremely rare but have resulted in subglottic stenosis.
Selection of Patients
Patients determined to have subglottic stenosis should be carefully evaluated by an otolaryngologist to assess glottic function and the need for glottic procedures.
In many patients with subglottic laryngeal inflammatory processes, the inflammation extends to just below the vocal cords. If resection is to be carried into the subglottic area, it cannot extend all the way to the vocal cords with hope of uniformly good results. The operation we propose is reserved for patients who on radiological study and direct laryngoscopy have an adequate residual subglottic space (Figures IIA–C). It is particularly important to assess the involvement of mucosa overlying the posterior cricoid plate and the posterior portion of the subglottic larynx. The airway is evaluated radiologically by the usual techniques, and the tracheostomy tube is removed at the time of examination (Figures IVA–B).
Vocal cord function is assessed by fluoroscopy of the larynx and by direct examination. Such evaluation is essential not only as a baseline for postoperative conparison of function but also to avoid performing subglotic airway reconstruction in a patient who, in addition has obstruction at the glottic level because of preexising bilateral palsy. Endoscopy with magnifying telescopes are used to evaluate the anatomy in detail (Figure V). It is most important to determine if there is adequate space beneath the glottis for repair because it is often difficult to determine this beforehand. The degree of inflammation in the area where the anastomosis would be carried out is carefully assessed. Inflammation is one of the important factors determining timing of resection and reconstruction. If inflammation does exits, it is best to dilate the stenosis and delay reconstruction. Careful endoscopic measurements should be taken to determine the exact length of involvement and the amount of normal trachea available for reconstruction.
Airway management is critical to a successful outcome. It requires close cooperation and patience between the anesthesiologist and surgeon.
Previous radiologica evaluation is invaluable in understanding the degree and extent of involvement of the airway. The most important aspect to successful management of these difficult airways is the attainment of a deep level of anesthesia with spontaneous ventilation by the patient. A satisfactory level of anesthesia may take 15 to 20 minutes for induction. When this level of anesthesia is reached, the upper airway is evaluated with a laryngoscope, or rigid bronchoscope, and a straight magnifying telescope. The glottis should be inspected very carefully before dilation to properly evaluate the degree of mucosal involvement, especially of the posterior commissure. Active inflammation with friable mucosa may preclude reconstruction at this time. Dilation is initiated by small woven bougies through a large rigid bronchoscope. Subsequent dilations are performed with small pediatric rigid bronchoscopes used as dilators under direct vision. Dilation is performed with gentle forward pressure in a corkscrew fashion under direct vision at all times. If resection is planned, it is only necessary to dilate to a diameter that will allow placement of a no. 6 endotracheal tube. If reconstruction is to be delayed, it should be possible to dilate to the size of a 7 or 8 rigid bronchoscope. This should be satisfactory for days to weeks depending on the origin of the stenosis. Patients should be made aware of this and know to seek medical attention before critical stenosis develops. Because of the complexities involved in reconstruction, tracheostomy is best avoided because it will only complicate subsequent reconstruction. Primary reconstruction or repeat dilation are the best ways to manage patients with subglottic stenosis.
A collar incision is usually adequate for exploration. An existing tracheotomy may be included in the incision or separately excised. It is preferable to do a laryngeal release through a short transverse incision over the hyoid bone if exposure is inadequate through the collar incision. The anterior surface of the airway is expose from the thyroid notch to the carina. Dissection is kept close to the airway to avoid injury to the recurrent laryngeal nerves. The nerves are not identified but injury is avoided by staying away from their course.
Postoperative Management of Airway
Although these anastomoses have proved to be surprisingly competent initially, they may achieve only a percentage of a normal cross-sectional airway area because of the amount of disease involvement that is present submucosally even at the immediate subglottic level. In some patients, it is judicious to use a small tracheostomy temporarily as an alternative airway. It may also be necessary to leave this airway in place for some time until the edema subsides sufficiently to permit extubation. For some patients, it is impossible to place such a tracheostomy tube without endangering either the anastomosis or the innominate artery. In these patients, an area should be walled off and marked as noted. Such patients should be extubated in the operating room. If the airway is adequate, they are allowed to breathe on their own and are watched carefully for the next few days. If they do not breathe adequately or if they develop trouble in the immediate postoperative period, a small endotracheal tube is gently inserted between the vocal cords into the trachea. Ventilation is not usually required because there is no insult to the pulmonary parenchyma. If required, a cuff may be placed well below the anastomotic area with safety. The endotracheal tube is left in place for a number of days. It is usually withdrawn in the operating room. If the patient does not breathe adequately, it is replaced, the wound is reopened, and a tracheostomy tube is placed at the premarked position. By this time the anastomosis and the innominate artery are walled off. In some patients where there is no room for a safe tracheostomy, the endotracheal tube can again be placed until the patient can be extubated. If a tracheostomy tube is placed too close to the anastomosis, erosion may lead to recurrent subglottic stenosis, which probably will not be reparable. We have not found it necessary to splint the anastomosis with an inlaying T-tube or other type of tube. When a tracheostomy tube is placed at the original operation, the anastomosis and innominate artery are walled off by local strap muscles. Figure VI shows the radiographic appearance after surgical correction of the subglottic stenosis seen in Figure IVA.
We have treated 80 patients in whom the subglottic larynx was partially resected with trachea for inflammatory stenotic processes and in whom primary reconstruction was performed.
The anterior cricoid arch was resected in all patients. This series excludes patients with primary and secondary tumors because this is a much different problem than inflammatory stenosis. In inflammatory disease, the process frequently extends above the stenosis in the subglottic larynx, nearly to the vocal cords. Resection, therefore, cannot be carried proximally to include all of the inflammation. For tumors, once the excision of tumor has been accomplished, the line of anastomosis lies in normal tissue.
Fifty of the 80 patients had lesions that resulted from intubation done for ventilatory support. (Table 1). Thirty-one of these had endotracheal tubes only; in 16, the lesion was believed to result from stomal erosion of the cricoid cartilage, and in three the originating trauma was an elective cricothyroidostomy for ventilatory support. In five patients, the stenosis was of traumatic origin. In four of these it was because of blunt trauma, usually with separation of larynx and trachea and with injury to the lower larynx, and in one it was because of gouging of the anterior laryngotracheal wall by a flying object. A patient suffering from an inhalation burn was the first treated in this series. One patient was referred from another institution with a postoperative stenosis after an attempt at a complex laryngotracheal repair with a hyoid graft, performed to correct deformity because of a goiter.
TABLE 1Results of Surgical Treatment of Subglottic Stenosis
Of the 23 remaining patients, 19 had idiopathic laryngotracheal stenosis. We have now seen over 60 patients with idiopathic subglottic stenosis and operated on over 50 (unpublished data, April 1998).
Five patients had tracheoesohageal fistulas when first seen. In two, the fistula had resulted from blunt trauma and in three, from intubation and ventilation with tracheoesophageal erosion. In 16 patients, paralysis or paresis of one or both vocal cords was identified preoperatively. In numerous others there was malfunction of one or both vocal cords.
In those patients in whom the stenotic process involved only the anterior portion of the subglottic larynx (49 out of 80), resection of this portion of the cricoid arch in an arcuate line extending up to a point just short of the midline of the thyroid cartilage anteriorly sufficed. The posterior margin of resection was along the lower border of the cricoid cartilage. In patients in whom the subglottic process was circumferential, extending in front of the posterior plate of the cricoid, the line of posterior mucosal resection was incised above the level of stenosis, approaching the arytenoid cartilages. All involved mucosa and scar tissue was excised from the front of the posterior cricoid plate, leaving the cartilage intact posteriorly, to be surfaced by a broad-based flap of membranous trachea advanced from below. The margin of the anterior defect in the subglottic larynx was sutured in both types of resection to a prow-shaped segment of one distal tracheal ring, shaped to fit. The posterior membranous flap of tracheal wall was required for resurfacing of the bared cricoid in 31 of 80 patients. In 49 patients, it was possible to perform an anastomosis without resecting the tissues overlying the posterior cricoid plate.
In the five patients with tracheoesophageal fistula, including one with three adjacent fistulas, the esophagus was closed in two layers with inverting sutures, and a pedicled flap of strap muscle was interposed between the esophageal closure and the laryngotracheal anastomosis anteriorly.
In the first 20 patients, tracheostomy was used in 14 at the time of the original operation and an endotracheal tube in one. In the subsequent 60 patients, tracheostomy was used only nine times. This represented a drop in use of protective airway devices from 75% to 15%.
There was only a single postoperative mortality in 80 patients. This was related to a fatal myocardial infarction. Two patients suffered early and long-term failure of treatment. One was the only man who was retrospectively classified as having idiopathic laryngotracheal stenosis. The other patient had had a tracheostomy in infancy and had undergone multiple reconstructions and T-tube splinting before attempted repair here.
In classifying the results obtained (Table 1), we have deemed as excellent the patient with a normal voice and without any limitation of respiration at rest or on exercise. Patients were considered to have a good result if they suffered only slight lessening of maximum volume of voice, slight hoarseness that did not impede vocal use, slight weakness of voice after prolonged use, diminished ability to sing, and if their breathing was adequate for all normal activities. Patients labeled as having a satisfactory results were those with a hoarse voice, and either slight wheezing or shortness of breath on exercise not sufficient to impair usual activities. Eighteen patients achieved an excellent result, 48 a good result, and eight were classified as satisfactory. As noted, there were two failures and one immediate postoperative death. Three are listed as uncertain despite initial good results, because their follow-up was for less than 6 months.
Forty-nine patients were contacted in late 1990 in follow-up from 6 months to 12 years. In 13 additional patients, follow-up was available between 2 to 10 years after operation, and in four between 1 to 2 years. In three, follow-up data was available at between 6 months and 1 year, and in three less than 6 months. The fact that no patient who achieved an excellent or good status was found to deteriorate in subsequent months or years should be noted. In contrast, an apparent failure improved over 3 years, as noted earlier, and most who required tracheostomies eventually were successfully decannulated. It may be concluded that, if anything, the long-term results represent a minimum statement.
Early complications, in addition to the need for prolonged intubation because of glottic edema or anastomotic edema, included superficial wound infection in one patient that required drainage, suture granuloma in the incision of a second patient that required removal of the suture, and re-exploration for air leak in one patient. A pinhole leak was closed with a local muscle flap and healed promptly. Eight patients had difficulty with deglutition or with aspiration postoperatively. All had undergone extensive resection, and four of them had suprahyoid laryngeal release performed. One required a gastrostomy tube for nutrition and two others had such tubes placed at the time of surgery in anticipation of difficulty, especially because one of these two had also undergone reclosure of a recurrent tracheoesophageal fistula. With time and retraining in swallowing, it was possible to remove all gastrostomy tubes. Tracheal granulations were removed bronchoscopically in two patients who had anastomoses performed with Tevdek (Dernatel, Inc, Fall River, MA) and in five who had anastomoses performed with Vicryl (Ethicon, Inc). One of these patients restenosed. A polyp of the vocal cord was lasered in one patient. The patient with leukemia required a prolonged period for healing of the tracheal stoma, which had been made at the time of reconstruction. Twenty-two patients had a persistent hoarse voice in varying degree and 16 had weakness of the voice when attempting to project their speech. In four cases, these were expected consequences of bilateral vocal cord paralyses because of trauma and, in some others, of unilateral paralyses that was preexisting. Eight complained of an alteration of singing voice.
Otolaryngological literature is filled with descriptions of multiple modes of treatment of non-neoplastic stenosis involving subglottic larynx and upper trachea. Conservative measures include dilatation, intubation, stenting, steroid injection, cryotherapy, electrocoagulation and laser therapy. Operative procedures, often complex, include scar excision, incision of cricoid anteriorly or posteriorly, grafts of buccal mucosa or skin, free cartilage inserts, pedicled hyoid grafts, and creation of cutaneous gutters variously supported; all are often stented for a prolonged period and are usually accompanied by tracheostomy. More procedures than encouraging results have been recorded.
described a single-stage repair using partial subperichondrial resection of the posterior cricoid to remove posterior stenotic scar, narrowing of the trachea by suturing the ends of cartilage together and intusussception of the trachea into the groove created by the subperichondrial resection. A follow-up report
in 1986 of 28 patients with non-neoplastic stenosis (postintubation, traumatic, burn, and idiopathic) and seven with neoplasm so treated, produced a good airway in 26 and a limitation in two. A T-tube was used postoperatively in 10, and 13 had laryngeal release. Couraud et al
updated their experience between 1978 and 1988 with laryngotracheal resection for stenosis resulting from intubation or trauma, including 27 patients who had a Pearson-type procedure (F.G. Pearson, MD, Toronto, Canada)
and seven who had total or subtotal cricoid plate resection with stenting. All had good results.
The similar and generally good results obtained in this difficult group, of patients by units employing similar single-stage operations suggest that many complex procedures still used might well be abandoned. However, diagnostic precision is essential, operative timing must be carefully judged, operative technique is exacting, and postoperative care early and late must be meticulous. These are not operations to be done occasionally.
in: Pearson FG Deslauriers J Ginsburg RJ Thoracic Surgery. Churchill Livingstone,
New York, NY1995: 322