Ann Thorac Surg 2000;70:1197-1201
© 2000 The Society of Thoracic Surgeons
Original articles: general thoracic
Malignant laryngotracheal obstruction: a way to treat serial stenoses of the upper airways
Klaus Wassermann, MDa,
Frank Mathen, MDa,
Hans Edmund Eckel, MDb
a Third Medical Department, University of Cologne, Cologne, Germany
b Ear, Nose and Throat Department, University of Cologne, Cologne, Germany
Address reprint requests to Dr Klaus Wassermann, Klinik III für Innere Medizin der Universität zu Köln, Gebäude 0/D, Raum 409, Josef-Stelzmann-Strasse 9, 50924 Köln, Germany
 |
Abstract
|
|---|
Background. Six patients known to have inoperable esophageal carcinoma presented with stridor due to both malignant tracheal stenosis (n = 6) and bilateral vocal cord paralysis. Two patients also had respiratory-digestive fistula.
Methods. Patency was restored by endotracheal stenting plus unilateral cordectomy. Four patients had immediate relief. Two patients required enlargement of the cord incision. One of them declined reoperation and underwent tracheotomy.
Results. Stent function was uneventful. There was no dislodgement or mucous impaction. Fistula seal was complete. There was no aspiration through the new-shaped glottic orifice. Peak expiratory flow increased from 24.4% ± 9.7% predicted normal before to 40.5% ± 13.7% after the procedure, whereas the dyspnea score decreased from 74.2 ± 12.7 to 24.2 ± 14.0.
Conclusions. Restoration of airway continuity in serial laryngotracheal stenoses using a combined approach is a feasible technique in end-stage cancer patients. It effectively relieves respiratory distress and ensures voice preservation. In addition, it may avoid the risks of tracheotomy.
|
Introduction
|
|---|
Acute onset of respiratory distress or gradually worsening dyspnea in patients with known esophageal cancer is often the result of high-grade tracheal stenosis by invading malignancy. It may also reflect bilateral vocal cord paralysis (BVCP) [1]. Not infrequently, the vocal cord lesion precedes or succeeds the tracheal occlusion by a short delay. Every now and then, however, both the glottis and the trachea are severely obstructed at the same time. Because in this case the relative contribution of either site to the increased airflow resistance is difficult to assess [2], it is the two of them that need urgent treatment.
Airway stenting has become an established palliative endoscopic measure to restore sustained patency in malignant tracheobronchial obstruction. The current literature underscores prompt symptomatic and functional improvement brought about by all available endoprostheses [3]. Early and late complications are minimal and, if present, easy to overcome. There is virtually no procedure-related mortality and median survivals of up to 10.5 months have been documented [4].
Bilateral vocal cord paralysis in the paramedian position following thyroid gland surgery is now being treated successfully with the use of laser-assisted microlaryngoscopic interventions [5–10]. In particular, laser cordotomy seems to be favorable for the therapy of this complication [11]. Associated morbidity is usually low, and temporary tracheotomies are no longer required.
Hitherto, no data have been published on cordotomy in conjunction with central airway stenting for the man-agement of serial (glottic and traceobronchial) stenoses in oncologic patients. Therefore, this retrospective study seeks to evaluate the surgical technique, feasibility, complication rate, and efficiency of a combined laryngo-bronchologic approach for the symptomatic relief of serial central airway stenoses in 6 patients with locally advanced esophageal carcinoma. To ease the patients’ symptoms and to preserve their voice, we undertook laser-assisted unilateral cordotomy and endoscopic tumor coring followed by the insertion of stents in two subsequent sessions. We believe that in this particular group of tumor-stricken individuals, restoration of the normal airway passage is superior to bypassing it by a tracheostoma, which is a different and, as it may seem, more straightforward way to circumvent the obstacle [1]. But, given malignant tissue growth adjacent to larynx and trachea, tracheostomy is liable to attract complications by delayed wound closure and subsequent relapsing local infection. Also, permanent tracheal cannulation may be an additional cause of life-threatening mucus impaction and asphyxiation.
|
Patients and methods
|
|---|
Case 1
This patient was a 50-year-old female in whom esophageal cancer had first been diagnosed in March 1997. The clinical workup had resulted in stage T3N0M0. After two courses of radiochemotherapy, transthoracic esophagectomy was performed in July 1997. A third course of radiochemotherapy followed in December 1997. In May 1998 she presented to the bronchology unit in respiratory distress and stridor. Bronchoscopy showed bilateral vocal cord paralysis in addition to high-grade rigid compression of the distal trachea (residual lumen: 3 to 5 mm) with substantial malignant infiltration of the carina. Treatment consisted of stent insertion plus right-sided laser cordotomy. A bifurcated silicone-hybrid stent that would have seemed most suitable [12], could not be passed through the narrowed tracheal segment. Even dilatation with a rigid bronchoscope was only partly successful. Further bougieing was felt to be quite risky due to imminent tracheal rupture. Therefore, two nested 12-mm Palmaz-stents [13] were chosen because of their maximum resistance to compressive forces and their elegant insertion technique using a thin catheter as a carrier. The next day posterior laser cordotomy was performed. The patient showed immediate symptomatic and functional improvement (Fig 1, Table 1). She died of cachexia and pneumonia secondary to metastatic lung disease 2 months later.
View larger version (131K):
[in this window]
[in a new window]
|
Fig 1. Patient 1. Roentgenogram of thoracic outlet. Two overlapping endotracheal Palmaz stents in situ.
|
|
Case 2
This 55-year-old male patient’s esophageal cancer was diagnosed in April 1998 and staged as inoperable T4N2M0. He presented with swallowing difficulties, dyspnea and stridor in June 1998. Bronchoscopy revealed BVCP and compressive stenosis of the distal trachea. The orifices of both main bronchi were involved in the stenosis. Left-sided posterior cordotomy was performed in July 1998 and bifurcated dynamic 13-mm stent placement followed 1 week later. The tumor-induced esophageal obstruction was efficiently treated by balloon dilatation. The patient improved but after 1 month was lost to follow-up. (Fig 2, Table 1).
View larger version (47K):
[in this window]
[in a new window]
|
Fig 2. Patient 2. (Left) Glottis 2 days postcordectomy, edema of resection wound. (Middle) Subglottic level. A trace of the resection wound is visible. In the background behind the cricoid, the proximal stent border. (Right) The proximal circumference of the dynamic stent.
|
|
Case 3
This 73-year-old male patient with a striking medical history of arterial hypertension, two ischemic cerebral infarcts, laryngeal carcinoma, and subsequent radiochemotherapy had a diagnosis of inoperable esophageal carcinoma T3N1M1 (brain) in October 1998. Radiochemotherapy led to partial remission. Relapsing local malignancy was treated with brachytherapy and the additional placement of an esophageal stent. Bronchoscopic control August 1999 demonstrated a midtracheal respiro-digestive fistula 2 cm long. The esophageal stent was visible within the orifice of the fistula. Tumor tissue overgrowth resulted in secondary stenosis of the esophagus and moderate impression of the tracheal lumen. There was near complete bilateral vocal cord paralysis with minor residual vocal cord motility. The initial presentation was with hoarseness but without significant dyspnea at rest. Polyflex-stent placement [14] for fistula seal was performed on August 18, 1999. Within the subsequent 2 weeks complete paralysis with paramedian position of the vocal cords developed. Acute stridor was the presenting symptom. Right-sided posterior cordotomy was then performed resulting in good respiratory function and fistula seal. The patient was still alive and in good respiratory condition 2 months after initial intervention (Fig 3, Table 1).
View larger version (48K):
[in this window]
[in a new window]
|
Fig 3. Patient 3. (Left) Glottic resection 2 days after microlaryngoscopic surgery (cordotomy). (Middle) Subglottic extension of wound with proximal stent border in the background. (Right) Polyflex stent with fistula underneath and distal intact carina.
|
|
|
Results
|
|---|
The majority of patients had prior treatment consisting of surgery or palliative radiochemotherapy, or both. They now presented with far advanced incurable local tumor growth, and all of them deliberately chose to undergo the combined laryngo-bronchoscopic approach instead of being tracheotomized. Their main reason was voice preservation.
Four patients had simultaneous complete BVCP and tracheal obstruction, and two more had tracheal stenosis/fistula and incomplete bilateral vocal cord paralysis at first presentation. However, complete paralysis associated with paramedian cord position developed within 2 weeks of endotracheal stenting. This was signaled by stridor reappearing after transient breath relief. Given the initial glottic aspect, which was one of immobile cords standing somewhat apart I think that complete paralysis was the result of progressing tumor growth and by no means an adverse effect of rigid bronchoscopy which had been necessary for stent placement.
Stent function proved satisfactory and uneventful. There was no dislodgement, mucous impaction, or major granulation. Fistula seal was efficient. Aspiration subsequent to cordotomy was not noticed in any of the patients. All could resume their normal diet on the first day after surgery. Two patients were thought to require surgical revision of the larynx because of secondary obstruction either by fibrin deposits upon the cordotomy wound or just by too small an incision. In 1 patient (patient 4, Table 1), the debris was removed by repeated bronchoscopies, and a more extended (subtotal) cordectomy was carried out thereafter. The further healing process was uncomplicated, and final respiratory and phonatory results were good. The other patient (patient 5, Table 1) declined recordectomy. He received a tracheostomy and died at home after a short period of relief.
Preoperative and postoperative spirometry were performed using standard equipment (MasterLab: Jäger, Würzburg, Germany). Normal values were derived from Quanjer [15]. Patients were asked to estimate shortness of breath according to a vertical visual analogue scale [16]. Mean best peak expiratory flow (PEF)—including tests of patient 5 before tracheotomy—increased from 24.4% (of the predicted normal) before to 40.5% after the procedure. Mean best peak inspiratory flow (PIF) did not show any substantial change from 1.11 L/s to 1.44 L/s, although 3 of 6 patients improved slightly (Fig 4, Table 1). Some mentioned minor expectoration difficulties and shortness of breath while climbing stairs, but except for patient 5 they experienced respiratory improvement as evidenced by a consistent score decrease from a mean 74.2 to 24.2 points (Fig 5, Table 1). They kept speaking clearly but in a whispering tone. Two patients died of fatal hemorrhage. The survival attained by the deceased patients was 1 to 2.5 months.
View larger version (15K):
[in this window]
[in a new window]
|
Fig 4. (A) Peak expiratory flow before and after the double procedure. (B) Peak inspiratory flow before and after the double procedure. Symbols denote individual patients.
|
|
View larger version (13K):
[in this window]
[in a new window]
|
Fig 5. Dyspnea score before and after the double procedure. Symbols denote individual patients. *Patient 5 (see Table 1) before tracheotomy.
|
|
|
Comment
|
|---|
Bilateral vocal cord paralysis is potentially life-threatening and requires surgical intervention to prevent acute asphyxiation. It sometimes develops during the natural course of esophageal and bronchogenic carcinoma and may then significantly contribute to these patients’ morbidity. If recognized in time, such glottic stenoses can now be treated with minimally invasive endoscopic procedures.
Patients in the present series underwent unilateral posterior cordotomy according to the method of Dennis and Kashima [6]. A surgical CO2 laser was used coupled to an operating microscope. All patients were intubated transorally for surgery. In manipulating the tube, care was taken not to dislodge previously inserted tracheal stents. The procedure included transection of the vocal process of the arytenoid cartilage, followed by the resection of a minor portion of the posterior part of the vocalis muscle and the lateral thyroarytenoid muscle. As described by Ossoff and coworkers [5, 7], the conus elasticus was carefully transected from the vocal process of the arytenoid cartilage down to the inner surface of the cricoid cartilage. The anterior portion of the vocal cord, however, was not included in the resection, because its removal would have contributed little to the desired airway enlargement. Extubation was performed 30 to 60 minutes after the end of the intervention. No patient had a tracheotomy at the time of surgery or during the postoperative recovery period. In the long run, unilateral posterior cordectomy was successful for 4 patients but, due to continuing breathlessness, required enlargement in two.
As might have been expected, lung function tests following the interventions were far from normalized. Whereas PEF results were within the realm of postcordectomy values reported in a previous paper [11], PIF was essentially unchanged. This may be best explained by the patients’ miserable physical condition and not as reflecting insufficient inspiratory glottic patency, because PIF depends on respiratory muscle strength and motivation more than PEF does. Eventually, the leading central resistance now was the "new" vocal cords since voice preservation requires a certain degree of glottic closure. The stented trachea no longer contributed to residual obstruction. It is true that any surgical widening of the glottis for the treatment of bilateral vocal cord abduction paralysis trades voice for airway. A related problem is the potential risk of aspiration after surgical procedures on the larynx for glottic airway rehabilitation. This risk is of particular concern in patients with coexisting tracheal, pulmonary, or cardiac disease and may lead to severe aspiration pneumonia. Therefore, a compromise must be found between retaining voice quality and restoring an adequate translaryngeal airway. It is tissue-sparing resections like the one described in this report that generate superior phoniatric results associated with only minor airway stenosis [11]. Furthermore, with the arytenoid cartilages left intact no aspiration will come from such a procedure.
Placement of stents used in this study has been described elsewhere [4, 13, 14]. In short, orolaryngeal intubation was performed with an operating laryngoscope attached to a chest support (Karl Storz, Tuttlingen, Germany). It was positioned to expose the entire circumference of the vocal cords. Ventilation was sustained by a low-frequency jet through a nozzle fixed to the outer orifice of the instrument. A rigid bronchoscope was used to dilate the narrowed trachea or to peel off the congested or infiltrated mucosa. If necessary, the main stem bronchi were pretreated in a similar way.
Stent insertion seems to be of immediate and sustained benefit whenever the tracheobronchial tree is compromised by extrinsic tumor compression or respiratory–digestive fistula. Most patients (77% to 100%) experience prompt symptomatic and functional improvement. According to published data [3] the main serious side effects are stent dislodgement, mucus impaction within the devices, and granulation at both ends. Complication rates vary between 5 and 25% irrespective of the particular material or shape. Stent location within the central airways (trachea and bifurcation) is associated with median survivals of 4 to 10.5 months. Whether survival is prolonged in comparison to nonstented patients with similar tumor growth remains a matter of debate [4]. Compared to the pertinent literature, our patients’ survival was distinctly shorter. One reason may have been a more advanced cancer stage as evidenced by the additional damage of the laryngeal nerves. Second, esophageal carcinoma, once infiltrating the trachea and adjacent structures, may grow more aggressively than lung cancer with concomitant tracheal involvement. Two of 6 patients died of fatal hemorrhage, which is a larger percentage than found in other study groups where lung cancer is a frequent cause of tracheal stenosis. The relatively short observation period may have contributed to the total lack of noted stent complications which tend to increase with indwelling time. Even tumor growth through the open meshes of the Palmaz stent, as has occasionally been reported, was missing in the present study.
Conclusions
Palliative measures in end-stage tumor patients should provide efficient long-standing relief from distressing conditions such as pain, dyspnea, and recurrent pneumonias without being too invasive, painful, or risky by themselves. In no case should they be associated with any substantial morbidity, let alone mortality. In our experience, cordotomy plus tracheal stenting reliably provides prompt and adequate palliation for cancer patients with serial tracheal occlusion and bilateral vocal cord paralysis. However, there were some drawbacks that necessitated reintervention: occasional massive fibrin deposits upon the resection wound and a cord incision that proved to be too narrow to be of real benefit. There happened to be no major stent problems such as dislocation or mucus impaction, as have been described elsewhere.
Whether the presented approach is a true advantage over usual tracheostomy in terms of survival, reduced respiratory emergencies, lower infection rates, and the like will be elucidated further. It seems unquestionable, however, that quality-of-life issues are better accounted for in end-stage cancer patients with central airway stenosis if endoscopic techniques are applied to restore the airway and tracheotomies can be avoided. These are preliminary data. Clearly, more research will be needed to elaborate criteria for individual treatment selection, with a choice of different endolaryngeal procedures readily available.
|
References
|
|---|
-
Strauss A., Pinder M., Lipman J., Conidaris M. Acute stridor as a presentation of bilateral abductor vocal cord paralysis. Anaesthesia 1996;51:1046-1048.[Medline]
-
Wassermann K., Koch A., Warschkow A., Mathen F., Müller-Ehmsen J., Eckel H.E. Measuring. "in-situ" central airway resistance in patients with laryngotracheal stenosis. Laryngoscope 1999;109:1516-1520.[Medline]
-
Wassermann K. Atemwegsstents. Ein skeptisches Plädoyer für die Palliativmedizin. Deutsch Med Wschr 2000;725:429-435.
-
Wassermann K., Eckel H.E., Michel O., Müller R.P. Emergency stenting of malignant obstruction of the upper airways. Ann Otol Rhinol Laryngol 1998;107:149-154.[Medline]
-
Ossoff R.H., Sisson G.A., Duncavage J.A., Moselle H.I., Andrews P.E., McMillan W.G. Endoscopic laser arytenoidectomy for the treatment of bilateral vocal cord paralysis. Laryngoscope 1984;94:1293-1297.[Medline]
-
Dennis D.P., Kashima H. Carbon dioxide laser posterior cordectomy for treatment of bilateral vocal cord paralysis. Ann Otol Rhinol Laryngol 1989;98:930-934.[Medline]
-
Ossoff R.H., Duncavage J.A., Shapshay S.M., Krespi Y.P., Sisson G.A. Endoscopic laser arytenoidectomy revisited. Ann Otol Rhinol Laryngol 1990;99:764-771.[Medline]
-
Crumley R.L. Endoscopic laser medical arytenoidectomy for airway management in bilateral laryngeal paralysis. Ann Otol Rhinol Laryngol 1993;102:81-84.[Medline]
-
Remacle M., Lawson G., Mayne A., Jamart J. Subtotal carbon dioxide laser arytenoidectomy by endoscopic approach for treatment of bilateral cord immobility in adduction. Ann Otol Rhinol Laryngol 1996;105:438-445.[Medline]
-
Bigenzahn W., Hoefler H. Minimally invasive laser surgery for the treatment of bilateral vocal cord paralysis. Laryngoscope 1996;106:791-793.[Medline]
-
Eckel H.E., Thumfart M., Vössing M., Wassermann K., Thumfart W.F. Cordectomy versus arytenoidectomy in the management of bilateral vocal cord paralysis. Ann Otol Rhinol Laryngol 1994;103:852-857.[Medline]
-
Freitag L., Eicker R., Linz B., Greschuchna D. Theoretical and experimental basis for the development of a dynamic airway stent. Eur Respir J 1994;7:2038-2045.[Abstract]
-
Müller C., Dienemann H., Hoffmann H., et al. Expandierbare Metallmaschenstents zur Behandlung von Trachealstenosen und Tracheomalazie. Zentralbl Chir 1993;118:543-548.[Medline]
-
Wassermann K., Koch A., Müller-Ehmsen J., Reuter M., Michel O., Eckel H.E. Clinical and laboratory evaluation of a new thin-walled self-expanding tracheobronchial silicone stent. J Thorac Cardiovasc Surg 1997;114:527-534.[Abstract/Free Full Text]
-
Quanjer P.H. Clinical respiratory physiology. Bull Eur Physiopathol Respir 1983;19(Suppl 5):49-50.
-
Gift A.G. Validation of a vertical visual analogue scale as a measure of clinical dyspnea. Rehab Nurs 1989;14:323-325.[Medline]
Accepted for publication March 29, 2000.
Top
Abstract
Introduction
