- Email: [email protected]
Primary tracheal anastomosis after resection of the cricoid cartilage with preservation of recurrent laryngeal nerves
Primary tracheal anastomosis after resection of the cricoid cartilage with preservation of recurrent laryngeal nerves Resections at the cricoid level pose the problems of damage to the recurrent laryngeal nerve and loss of circumferential cartilaginous support. Strictures within the cricoid ring have usually been managed with keels or stents, whereas neoplasms have been managed by laryngectomy. This paper reports on 6 patients with lesions involving the cricoid who were successfully treated by segmental tracheal resection and removal of all but a thin shell of posterior cricoid plate. The distal trachea was anastomosed at the subglottic level within 1 em. or less of the vocal cords. Two patients had traumatic transection at the cricotracheal level with disruption of cricoid cartilage and avulsion of both recurrent nerves. Of the other 4 patients with tracheal lesions involving the cricoid, 2 had postintubation strictures, another had chemical burns, and the fourth had adenoid cystic carcinoma. Primary healing and good clinical results were obtained in all 6 patients. In the 4 patients with intact recurrent nerves, nerve function was preserved. This technique provides a method for resection and reconstruction in one stage for selected lesions at the cricoid level.
F. G. Pearson, M.D.,* J. D. Cooper, M.D.* (by invitation), J.:M. Nelems, M.D.* (by invitation), and A. W. P. Van Nostrand, M.D. ** (by invitation), Toronto, Ontario, Canada
Resection of that part of the cricoid cartilage which lies below the level of the inferior border of the thyroid cartilage (Fig. 1) poses two problems. Complete transection of the airway at this level unavoidably divides both recurrent laryngeal nerves. The loss of circumferential cartilaginous support due to excision of the cricoid ring results in collapse of the airway at the subglottic From the University of Toronto and the Toronto General Hospital, Toronto, Ontario, Canada. Read at the Fifty-fifth Annual Meeting of The American Association for Thoracic Surgery, New York, N. Y., April 14, IS, and 16, 1975. Address for reprints: Dr. F. G. Pearson, Room 120 University Wing. Toronto General Hospital, Toronto, Ontario, Canada MSG IL7. * Department of Surgery, University of Toronto; Division of Thoracic Surgery, Toronto General Hospital. ··Department of Pathology, University of Toronto and Toronto General Hospital.
806
level. For these reasons, benign lesions which produce stenosis at the cricoid level are frequently managed by repetitive dilatation, staged plastic reconstruction, prolonged stenting, or permanent tracheostomy.>" Extension of malignant tracheal neoplasms into the cricoid area is usually considered an indication for laryngectomy. A few cases have been reported in which a successful primary anastomosis was accomplished following resection of the anterior cricoid arch and that part of the posterior cricoid plate, or rostrum, which lies inferior to the level of the cricothyroid articulation. r-s However, if a subglottic lesion requires transverse division of the airway at the level of the inferior border of the thyroid cartilage (Fig. 1), the prob-
Volume 70 Number 5 November, 1975
lems of recurrent laryngeal nerve preservation and airway support ensue. This paper reports on 6 patients with lesions involving the cricoid area who were managed by resection of segments of the trachea and excision of the cricoid arch and all but a thin shell of posterior cricoid plate. The distal trachea was anastomosed at the subglottic level within 1 em, of the vocal cords, and the intact recurrent laryngeal nerves were preserved . Operative technique
If the recurrent laryngeal nerves are intact, they must be identified on each side and followed upward to the posterolateral aspect of the cricoid cartilage, where they pass immediately behind the cricothyroid joints; beyond this point they can no longer be followed (Fig. 2). The nerves having been exposed in this way, the proximal resection line is begun by transection of the airway obliquely. The incision is started at the inferior border of the thyroid cartilage in front and is passed posteriorly and inferiorly to cross the lower margin of the cricoid plate below the level of the exposed nerves (Figs. 3A and 3B.) This results in complete resection of the anterior cricoid arch and a portion of its posterior plate. A segment of the posterior plate of the cricoid remains above and surrounds the posterolateral aspects of the mucous membrane in the subglottic region (Fig. 4, A and D) . In order to transect the airway at a higher level posteriorly, one must remove a rim of cricoid immediately subjacent to the posterior and lateral aspects of the submucosa. This procedure is best done with fine rongeurs (Fig. 4, B) and can be extended superiorly almost to the base of the vocal cords, if necessary. At this stage there will be some loss of external support to the mucous membrane of the airway, which tends to collapse (Fig. 4, C and E). The additional cricoid resection remains anterior to the cricothyroid joints. Otherwise, the recurrent laryngeal nerve, which is an immediate posterior relation of the joint (Fig. 5) , may be damaged. It is now possi-
Primary tracheal anastomosis
807
THYROID
CRICOID
TRACHEA
RECURRENT LARYNGEAL NERVE
Fig. 1. Diagram illustrating the level of resection at the inferior border of thyro id cartilage. Resection at this level will divide the recurrent laryngeal nerve s and remove the skeletal support of the cricoid ring.
ble to divide the mucous membrane posteriorly at the level of the inferior border of the thyroid cartilage within 1 em. of the vocal cords and above the level of the lesion to be resected. The distal resection line is made through either the cervical or mediastinal trachea, and the luminal diameter of the airway at this level is much larger than that at the proximal, subglottic resection line. The tips of the uppermost tracheal cartilage at the distal resection line are sutured together by plication of the membranous trachea (Fig. 6, A and B) . This maneuver decreases the luminal diameter of the trachea so that it will more readily fit the subglottic diameter (Fig. 6, C) . Furthermore, a complete cartilaginous ring has been created at the distal resection line. When anastomosed to the cuff of mucous membrane at the proximal resection line, this ring restores the skeletal support previously provided by the cricoid ring. A primary thyrotracheal anastomosis is achieved by advancement of the distal tracheal stump upward and anterior to the residual posterior shell of cricoid cartilage. (Fig. 7A and 7B). Interrupted sutures of No. 35 stainless steel wire are used for the anastomosis , with the knots tied inside the lumen on the posterior wall. The mucous membrane in the subglottic area is surpris-
808
Pearson et al.
The Journal of Thora cic and Card iovascular Surgery
Fig. 2. Operative photog raph showing mobilization of the right recurrent laryngeal nerve to the level of the inferior border of the plate of the cricoid cartilage. Above this point (black a rrow) the nerve passes behind the cricothyroid joint and enters the subglottis. T, Trachea. CT, Cricothyroid muscle.
J.HYROID
"""tI' CRICOID
TRACHEA
-RECURRENT LARYNGEAL NERVE
Fig. 3A. · Diagram showing oblique resection line . The line begins anteriorly at the inferior border of thyroid cartilage and extends posteriorly through the lower border of the cricoid plate below the point of entry of the recurrent laryngeal nerve .
ingly sturdy and holds sutures well. Furthermore, the cuff of subglottic mucous membrane tends to stretch and is everted to accommodate the relatively larger airway diameter of the distal trachea. The result is a widely patent and well-supported airway at the anastomosis. It is essential that the completed anasto-
Fig. 38. Photograph of a cadaver specimen showing the same oblique line of transection.
Volume 70
Primary tracheal anastomosis
Number 5 November, 1975
• •
THYROID CRICOID RECURRENT N.
A
c
Fig. 4. A, Initial appearance of the upper resection line in cross section is illustrated diagramatically. A segment of the posterior cricoid plate surrounds the posterolateral aspect of the mucous membrane at the subglottic level. B, Technique of removal of cricoid cartilage subjacent to the posterolateral margins of the mucous membrane. C, Completed resection of cricoid cartilage results in some collapse of the soft tissues and subglottic lumen. D, Photograph of a cadaver specimen similar to the diagrammatic sketch shown in Fig. 4, A. E, Photograph of a cadaver specimen following submucosal resection of the posterior cricoid plate, similar to the diagrammatic sketch shown in Fig. 4, C.
Fig. S. Photomicrograph showing a cross section of the subglottic region through the lower half of the cricoid cartilage. C, Cricoid cartilage. T, Inferior cornu of thyroid cartilage. The recurrent laryngeal nerve lies within the black circle as an immediate posterior relation of the cricothyroid joint.
809
The Journal of
8 I0
Pearson et at.
Thoracic and Cardiovascular Surgery
A
I B
c
Fig. 6. A, Diagrammatic appearance of the distal tracheal resection line. The lumen is relatively larger than the subglottic lumen seen in Fig. 6, C. B, Technique of plication of the membranous trachea at the distal resection line. This procedure approximates the ends of the uppermost tracheal ring and produces a complete circle of cartilage to replace the resected cricoid ring. The luminal diameter at the distal line is also reduced. C, Diagrammatic appearance of the completed subglottic resection line seen in cross section. The vocal cords lie within 1 em. of the resection line.
Fig. 7A. Diagrammatic sketch of a thyrotracheal anastomosis. The distal trachea is "telescoped" in front of the shell of residual cricoid cartilage.
mosis be free of tension. Various techniques of tracheal and supralaryngeal mobilization may be necessary, the proper procedure depending on the length of trachea to be resected. Clinical experience Since January, 1973, 6 patients have undergone segmental tracheal resection, including an extensive removal of cricoid cartilage. In 4 cases the type of resection described under Operative technique was
Fig. 7B. Photograph of a completed anastomosis in a cadaver specimen. T, Thyroid cartilage. TR, Trachea. C, Cricoid cartilage.
done. In 2 patients, each of whom had bilateral recurrent laryngeal nerve palsy, the entire cricoid below the level of the inferior border of the thyroid cartilage was removed. Brief details of the pathology, operative procedure, and results are summarized in Table I. Cricoid resection in the presence of recurrent laryngeal nerve palsy. Two patients
Volume 70 Number 5
Primary tracheal anastomosis
November. 1975
811
Fig. 8. Postoperative tracheogram obtained in Case 1.
suffered blunt trauma to the neck, with transection of the airway at the cricotracheal junction and additional damage to the cricoid area (Cases 1 and 2). In each case, there was wide separation of the divided ends of the airway and complete disruption of both recurrent laryngeal nerves. In Case 1, there was a fracture of the anterior cricoid arch and circumferential loss of the mucosal lining within the cricoid ring. It is astonishing that this patient survived for 3 days without tracheostomy or intubation before this injury was recognized. Three days after the injury, a primary cricotracheal anastomosis was done, but the patient subsequently developed severe fibrous stenosis within the cricoid ring at the site of mucosal loss. The stricture was unimproved by repeated endoscopic dilatation and local injections of steroid (triamcinolone). Five months later all of the cricoid below the level of the thyroid cartilage was resected , and the airway was restored by thyrotracheal anastomosis. The anastomosis healed well, as illustrated in the postoperative tracheogram (Fig. 8). Patient 2 suffered a similar injury, with
Fig. 9A. Preoperative tracheogram (lateral projection) obtained in Case 2. The stricture begins at the level of the cricoid ring , and a tracheopharyngeal fistula has developed through the fracture in the posterior cricoid plate .
Fig. 9B. Postoperative tracheogram obtained in
Case 2.
The Journal of Thoracic and Cardiovascular
Pearson et al.
8 12
Surgery
Table I Patient (sex . age)
Pathology
M.F. (M,19)
Blunt trauma; transection of trachea at cricotracheal junction
Nil
Prior treatment Aug., 1972: Primary anastomosis ; healed withseven subcricoid stricture
R.R .
Blunt trauma; transection of trachea at cricotracheal junction, fracture cricoid arch and plate, and T. E. fistula Post-tracheostomy stricture at cricotracheallevel
Nil
June, 1973:Tracheostomy on day of trauma
Intact
Nov ., 1973: Exploration of subglottic stricture;no resection
Intact
Preop. radiat ion 3,000 rads in 3 weeks
(M,18)
R.O. (M,41)
Adenoid cystic carcinoma of trachea involving K.T. cricoid mucosa (M,54) L. J. (F, 19)
Caustic burn of larynx and upper half of trachea (formaldehyde and HCl)
E.B. (M,42)
Caustic burn of larynx and upper trachea (chlorine Intact inhalation, 1970)
. Intact
Feb.-Mar., 1974:Endoscopic dilatation and removal laryngeal granulomas; tracheostomy Feb. 7 to Mar. 5,1974 Tracheostomy since 1970;two operations with laryngofissure, mucosal grafts, and laryngotracheal stenting; one vocal cord resected
Legend: T.E., Tracheoesophageal. RLN, Recurrent larynge al nerves.
Fig. 10. A, Preoperative tracheogram obtained in Case 5 showing a long stricture due to inhalation of caustic material. B, Postoperative tracheogram following cricoid resection and resection of 7 em. of trachea.
Volume 70 Number 5 November, 1975
Primary tracheal anastomosis
Operation Jan., 1973:Cricoid resection and superior laryngeal release; thyrotracheal anastomosis Sept., 1973: Cricoid resection and superiorlaryngeal release; closure ofT. E. fistula; thyrotracheal anastomosis Feb., 1974: Resection of cricoid and 2 em. trachea; superior laryngeal release; thyrotracheal anastomosis Mar., 1974: Resection of cricoid and 6 em. of trachea; superior laryngeal release; thyrotracheal anastomosis May, 1974:Cricoid and upper trachea resection (total length 7 em.); superior laryngeal release; right hilar release; thyrotracheal anastomosis Oct., 1974: Resection of cricoid and 5 em.of upper trachea; superior laryngeal release; thyrotracheal anastomosis
8 13
Follow-up
Complications None
Mar., 1975 (26 mo.): No clinical airway obstruction; weak voice
Granuloma at anastomosis; removed at bronchoscopy in Mar., 1974
April, 1975 (19 mo.): No clinical airway obstruction; weak voice
None
April, 1975 (14 mo.): Normal
Localized abscess and necrosis posterior aspect of anastomosis; spontaneous healing None
April, 1975 (13 rno.): Normal
April, 1975 (II rno.): Normal
Recurrent stenosis below vocal cords; long-term T-tube silicone (Montgomery) stent inserted
April, 1975: Final result pending; good airway with Montgomery stent in place; fair voice
complete transection of the airway at the cricotracheal level and fracture of both the anterior arch and posterior plate of the cricoid cartilage. He was initially treated by tracheostomy through the retracted tracheal stump, and he subsequently developed a tracheopharyngeal fistula at the site of fracture in the posterior cricoid plate (Fig. 9A). Three months after the injury, this lesion was managed by resection of all cricoid cartilage below the inferior border of the thyroid cartilage, closure of the esophageal fistula, and thyrotracheal anastomosis. The anastomosis healed well. It is illustrated in the postoperative tracheogram shown in Fig. 9B. The injury in both of these young men resulted in permanent loss of function of the recurrent laryngeal nerves. In spite of bilateral abductor paralysis, they are both managing remarkably well 26 and 19 months after thyrotracheal anastomosis. The airway at cord level is compatible with an active life. Hoarseness and weakness of the voice has not been so disabling as to warrant further laryngeal surgery. The thyrotracheal anastomosis is widely patent in both cases. Cricoid resection with preservation of
both recurrent laryngeal nerves. The underlying pathology in these 4 patients was cricotracheal stenosis following tracheostomy (Case 3), adenoid cystic carcinoma of the trachea with cricoid involvement (Case 4), and caustic cricotracheal strictures due to inhalation burns (Cases 5 and 6). Since the recurrent larygneal nerves were intact preoperatively in all 4 patients, a modified cricoid resection was done as previously described under Operative technique. Bilateral function of the recurrent laryngeal nerves was preserved in all 4 cases. In addition to the cricoid resection, a variable length of trachea was resected (2, 5, 6, and 7 em.). Next, thyrotracheal anastomosis was performed. In those patients undergoing resections of 5 em. or less, a transverse cervical incision was used (Cases 3 and 6). In 2 patients requiring resection of longer segments (6 and 7 em.), operative exposure was obtained through a transverse cervical incision with the addition of median sternotomy (Cases 4 and 5).
A superior laryngeal release- 7 was added in all 6 patients in order to reduce tension at the thyrotracheal anastomosis. This procedure, which divides the soft tissues be-
8 14
The Journal of Thoracic and Card iovascular Surgery
Pearson et al.
a result of edema or hematoma from the operative injury. Results
Fig. 11. Postoperative tracheogram obtained in Case 4. Resection of cricoid cartilage and 6 em. of trachea was done for adenoid cystic carcinoma.
tween the hyoid bone and thyroid cartilage, lowers the position of the thyroid cartilage and larynx in the neck and is the cause of transient swallowing problems during the postoperative period in most cases. In the patient requiring resection of a 7 em. segment (Case 5), anastomotic tension was further reduced by mobilization of the right pulmonary hilum.' Access for hilar mobilization was achieved by opening the right pleural space through the median sternotomy. A small, uncuffed metal tracheostomy tube was placed below the thyrotracheal anastomosis in all 6 patients for 1 to 2 weeks after operation. A temporary postoperative tracheostomy ensures an adequate airway at a time when airway obstruction is likely to occur in the subglottic area as
Excellent functional results were obtained in 5 of the 6 patients (Cases 1 to 5). In 4 patients the airway diameters at the level of thyrotracheal anastomosis appear normal in postoperative tracheograms. One patient has mild circumferential narrowing at the anastomosis but has no clinical evidence of upper airway obstruction. One patient developed significant stenosis at the thyrotracheal anastomosis 5 weeks after operation (Case 6). A silicone rubber, T-tube stent was introduced to splint the anastomosis and will be left in place for 6 to 9 months. The man was operated upon in October, 1974, and the final result is still pending. He is the only patient who did not have normal mucous membrane in the segment extending for 1 em. below the vocal cords. In this case, an inhalation bum produced severe damage to the larynx. Before admission to our hospital, he had undergone two major operations by laryngofissure, with excision of the left arytenoid cartilage and vocal cord and the application of free mucosal grafts supported by a laryngotracheal stent. The results obtained in 2 patients who underwent cricoid resection with preservation of the recurrent laryngeal nerves are shown in the preoperative and postoperative tracheograms illustrated in Fig. 10, A and B (Case 5) and Fig. 11 (Case 4). Discussion The operative technique which has been described affords the advantages of a onestage reconstruction and primary anastomosis with apposition of mucosa to mucosa. Stenting was necessary in only 1 of 6 patients. The operative procedures is technically difficult and requires a precise knowledge of the anatomy of the larynx and subglottic region . This anatomy is probably best learned from cadaver dissections . The operation is applicable in selected patients with subglottic pathology. Before
Volume 70 Number 5
Primary tracheal anastomosis
8 15
November, 1975
operation, the surgeon must ascertain that the mucous membrane is relatively normal throughout the first 1 em. of the subglottis below the vocal cords. Although the condition of the subglottic mucosa can be determined by preoperative bronchoscopy or laryngoscopy, the level of the pathology can be evaluated precisely only by contrast tracheograms obtained in anteroposterior, oblique, and lateral projections. REFERENCES
2
3
4
5
6
7
Ogura, J. H., and Powers, W. E.: Functional Restitution of Traumatic Stenosis of the Larynx and Pharynx, Laryngoscope 74: 1081, 1964. Ogura, J. H., and Biller, H. F.: Reconstruction of the Larynx Following Blunt Trauma, Ann. Otol. Rhinol. Laryngol. 80: 492, 1971. Harley, H. R. S.: Laryngotracheal Obstruction Complicating Tracheostomy or Endotracheal Intubation With Assisted Respiration: A critical Review, Thorax 26: 1971. Grillo, H. C.: The Management of Tracheal Stenosis Following Assisted Respiration, J. THoRAc. CARDIOVASC. SURG. 57: 52, 1969. Pearson, F. G., and Andrews, M. 1.: Detection and Management of Tracheal Stenosis Following Cuffed Tube Tracheostomy, Ann. Thorac. Surg, 12: 359, 1971. Gerwat, J., and Bryce, D. P.: The Management of Subglottic Laryngeal Stenosis by Resection and Direct Anastomosis, Laryngoscope 84: 940, 1974. Dedo, H. H., and Fishman, N. H.: Laryngeal Release and Sleeve Resection for Tracheal Stenosis, Ann. Otol. Rhinol. Laryngol. 78: 285, 1969.
Discussion DR. HERMES C. GRILLO Boston, Mass.
This is an important contribution for the solution of a problem which, while not very common, presents enormous difficulties when it occurs. Until now, only prolonged stenting procedures have been available for such problems. There is a spectrum of inflammatory lesions within the first 3 em. below the vocal cords, and this lesion is a very special part of that spectrum. I emphasize this because I do find there is a tendency to label several lesions as "high tracheal stenosis." Stenosis involving the upper trachea is one problem. Even if one half of the first ring is intact, the problem is simple. If the stenosis reaches the lower border of the cricoid, the problem becomes a little more
difficult technically and the results a little less good. When the stenosis involves the lower part of the cricoid-low subglottic stricture-and part of that ring can be saved by being beveled, the problem gets a little more complicated. When the level of stenosis creeps up to a point above the top of the cricoid, a true intralaryngeal subglottic stenosis is present. It is to this last group of patients that Dr. Pearson and his colleagues have directed this superb and very elegant technique. We have had single-stage solution for those problems in the past. Dr. Pearson very properly cautioned us that this will be a difficult technique unless we take the time to explore the anatomy in intricate detail and be very certain of what we are doing. Even then it is going to require masterful technique to get the kind of results reported. In passing, Dr. Pearson presented cases of bilateral cord transection in which this technique was applied. It is important to point out that people with permanent palsies of both cords can have a functional larynx. However, in some patients with permanent palsies from trauma, the cords remain in mid position. Thus an otolaryngologist should pin one of the cords or do an arytenoidectomy first to obtain a glottic airway before an attempt is made to repair the subglottic airway. DR. WILLIAM E. NEVILLE Newark, N. J.
As Dr. Pearson has so aptly described, a tracheal resection at the cricoid level poses enumerable technical problems. There is no question that a primary anastomosis should be performed if at all possible. However, this may not be possible. I would like to submit an alternate method which we have used to advantage on occasion when long segments of the anterior portion of the trachea have been involved, but where the membranous portion can be left intact. I like to refer to this incision as a slash tracheostomy done by an eager general surgical resident. The trachea was incised anteriorly from the cricoid cartilage down to the manubrium because of severe respiratory insufficiency. The patient was discharged after a stormy hospital course, but he was readmitted with complete tracheal obstruction. It was necessary to dilate the trachea before insertion of an endotracheal tube. Invagination of the anterior cartilages into the lumen created a web inside of the trachea from the cricoid cartilage to the manubrium. We resected the anterior and lateral margins of the trachea but left the membranous portion. The silicone rubber tracheal prosthesis was invaginated into the upper and lower tracheal open-
The Journal of
8 16
Pearson et at.
Thoracic and Cardiovascular Surgery
ings so that the airway was re-established. Then we simply reapproximated the thyroid and the strap muscles of the neck over the prosthesis. From our results in a few cases, I think this is a worthwhile operation when primary tracheal anastomosis cannot be done. DR. JOHN R. BENFIELD Torrance, Calif.
I wonder if Dr. Pearson could help some of us who operate in this area less frequently than he does by commenting on two points. First, exactly how do you handle the thyroid gland? It seems to me there are two alternatives: (1) to split it down the middle and reflect both sides laterally or (2) to divide the superior thyroid vessel and retract the whole gland toward the opposite side of the predominant operation. I would like to know which approach you usually use. The second question, and perhaps the more important point, has to do with the function of the laryngeal nerve in the postoperative period. Is there a need for tracheostomy or prolonged endotracheal intubation in the management of these patients? In the late postoperative period, how many of your patients had complete return of function? DR. PEARSON (Closing) I wish to thank Dr. Grillo for his lucid and supportive discussion. He made reference to the problem of bilateral recurrent nerve palsy, which results in an unpredictable airway at the level of the glottis. In our 2 patients with blunt trauma and complete disruption of both recurrent laryngeal nerves, the vocal cords remained in an intermediate or partially abducted position. This was most fortunate, since the cords did not seriously impair their airway, and yet they retained sufficient voice to be understandable. Vocal cord function has not returned in either case, the one having been followed for 18 months and the other for about 2 years. These 2 patients were seen by our otolaryngologists who recommended that nothing further should be done. Any effort to improve their voice would potentially compromise the airway. Dr. Ogura, an otolaryngologist in St. Louis, Missouri, has written extensively concerning surgical reconstruction after laryngeal and laryngotracheal injury. He routinely excises one arytenoid cartilage and secures one vocal cord to the side of the larynx in patients who have sustained blunt trauma with division of both recurrent laryngeal nerves. Although his experience is much more extensive than ours, it would appear that routine excision and lateralization of the vocal
cord is not necessary. In our 2 patients, nothing was done to the vocal cords, and the outcome has been perfectly satisfactory. However, I understand that there may be late changes at cord level in patients with recurrent nerve palsy because of loss of muscle bulk and scar replacement. Thus these young men may still require corrective surgery at a later date. A question was asked regarding tracheostomy in patients undergoing cricoid resection. In all 6 of our patients a tracheostomy was placed below the anastomosis. The operative field is immediately below the vocal cords in the narrowest part of the airway, and we have assumed that minimal edema or hematoma would obstruct the airway at the level of the glottis in the early postoperative period. For this reason we have placed a small, temporary tracheostomy tube below the anastomosis. It has not been left for longer than 2 weeks in any patient, other than the single individual who still has aT-tube stent and in whom the result is pending. Dr. Benfield asked about cord function in those 4 patients in whom we were able to preserve the recurrent nerves. All of these patients had normal cord function with a return of normal voice within a week or 10 days of operation-as soon as the tracheostomy tube was removed. I think we have been fortunate in that none of the 4 patients has had even a transient palsy. Dr. Benfield also asked about the management of the thyroid gland in the process of identifying the recurrent nerves. If the nerves are to be identified, the dissection is tedious since one must free the nerves completely from their posterolateral thyroid relations. In these patients, this has been done in two ways. The lobe has been reflected forward wtih the dissection begun behind. We have also split the isthmus and freed the lobe from the lateral aspect of the trachea to identify and free the nerve posteriorly. In the patient with adenoid cystic carcinoma, it was necessary to remove the right lobe of the thyroid with the tumor and tracheal segment. However, it is nearly always possible to leave at least one lobe of thyroid. I agree with Dr. Neville that there are some patients in whom a primary anastomosis cannot be achieved. I am concerned, however, about the use of a prosthesis in this location. The subglottic region is the narrowest part of the airway, lying immediately below the vocal cords. I believe that any prosthesis, even a silicone prosthesis, is likely to cause some local tissue reaction in this critical area. I would anticipate a significant incidence of obstructive problems due to granulation tissue or edema at the upper end of any permanent prosthetic replacement.
F. G. Pearson, M.D.,* J. D. Cooper, M.D.* (by invitation), J.:M. Nelems, M.D.* (by invitation), and A. W. P. Van Nostrand, M.D. ** (by invitation), Toronto, Ontario, Canada
Resection of that part of the cricoid cartilage which lies below the level of the inferior border of the thyroid cartilage (Fig. 1) poses two problems. Complete transection of the airway at this level unavoidably divides both recurrent laryngeal nerves. The loss of circumferential cartilaginous support due to excision of the cricoid ring results in collapse of the airway at the subglottic From the University of Toronto and the Toronto General Hospital, Toronto, Ontario, Canada. Read at the Fifty-fifth Annual Meeting of The American Association for Thoracic Surgery, New York, N. Y., April 14, IS, and 16, 1975. Address for reprints: Dr. F. G. Pearson, Room 120 University Wing. Toronto General Hospital, Toronto, Ontario, Canada MSG IL7. * Department of Surgery, University of Toronto; Division of Thoracic Surgery, Toronto General Hospital. ··Department of Pathology, University of Toronto and Toronto General Hospital.
806
level. For these reasons, benign lesions which produce stenosis at the cricoid level are frequently managed by repetitive dilatation, staged plastic reconstruction, prolonged stenting, or permanent tracheostomy.>" Extension of malignant tracheal neoplasms into the cricoid area is usually considered an indication for laryngectomy. A few cases have been reported in which a successful primary anastomosis was accomplished following resection of the anterior cricoid arch and that part of the posterior cricoid plate, or rostrum, which lies inferior to the level of the cricothyroid articulation. r-s However, if a subglottic lesion requires transverse division of the airway at the level of the inferior border of the thyroid cartilage (Fig. 1), the prob-
Volume 70 Number 5 November, 1975
lems of recurrent laryngeal nerve preservation and airway support ensue. This paper reports on 6 patients with lesions involving the cricoid area who were managed by resection of segments of the trachea and excision of the cricoid arch and all but a thin shell of posterior cricoid plate. The distal trachea was anastomosed at the subglottic level within 1 em, of the vocal cords, and the intact recurrent laryngeal nerves were preserved . Operative technique
If the recurrent laryngeal nerves are intact, they must be identified on each side and followed upward to the posterolateral aspect of the cricoid cartilage, where they pass immediately behind the cricothyroid joints; beyond this point they can no longer be followed (Fig. 2). The nerves having been exposed in this way, the proximal resection line is begun by transection of the airway obliquely. The incision is started at the inferior border of the thyroid cartilage in front and is passed posteriorly and inferiorly to cross the lower margin of the cricoid plate below the level of the exposed nerves (Figs. 3A and 3B.) This results in complete resection of the anterior cricoid arch and a portion of its posterior plate. A segment of the posterior plate of the cricoid remains above and surrounds the posterolateral aspects of the mucous membrane in the subglottic region (Fig. 4, A and D) . In order to transect the airway at a higher level posteriorly, one must remove a rim of cricoid immediately subjacent to the posterior and lateral aspects of the submucosa. This procedure is best done with fine rongeurs (Fig. 4, B) and can be extended superiorly almost to the base of the vocal cords, if necessary. At this stage there will be some loss of external support to the mucous membrane of the airway, which tends to collapse (Fig. 4, C and E). The additional cricoid resection remains anterior to the cricothyroid joints. Otherwise, the recurrent laryngeal nerve, which is an immediate posterior relation of the joint (Fig. 5) , may be damaged. It is now possi-
Primary tracheal anastomosis
807
THYROID
CRICOID
TRACHEA
RECURRENT LARYNGEAL NERVE
Fig. 1. Diagram illustrating the level of resection at the inferior border of thyro id cartilage. Resection at this level will divide the recurrent laryngeal nerve s and remove the skeletal support of the cricoid ring.
ble to divide the mucous membrane posteriorly at the level of the inferior border of the thyroid cartilage within 1 em. of the vocal cords and above the level of the lesion to be resected. The distal resection line is made through either the cervical or mediastinal trachea, and the luminal diameter of the airway at this level is much larger than that at the proximal, subglottic resection line. The tips of the uppermost tracheal cartilage at the distal resection line are sutured together by plication of the membranous trachea (Fig. 6, A and B) . This maneuver decreases the luminal diameter of the trachea so that it will more readily fit the subglottic diameter (Fig. 6, C) . Furthermore, a complete cartilaginous ring has been created at the distal resection line. When anastomosed to the cuff of mucous membrane at the proximal resection line, this ring restores the skeletal support previously provided by the cricoid ring. A primary thyrotracheal anastomosis is achieved by advancement of the distal tracheal stump upward and anterior to the residual posterior shell of cricoid cartilage. (Fig. 7A and 7B). Interrupted sutures of No. 35 stainless steel wire are used for the anastomosis , with the knots tied inside the lumen on the posterior wall. The mucous membrane in the subglottic area is surpris-
808
Pearson et al.
The Journal of Thora cic and Card iovascular Surgery
Fig. 2. Operative photog raph showing mobilization of the right recurrent laryngeal nerve to the level of the inferior border of the plate of the cricoid cartilage. Above this point (black a rrow) the nerve passes behind the cricothyroid joint and enters the subglottis. T, Trachea. CT, Cricothyroid muscle.
J.HYROID
"""tI' CRICOID
TRACHEA
-RECURRENT LARYNGEAL NERVE
Fig. 3A. · Diagram showing oblique resection line . The line begins anteriorly at the inferior border of thyroid cartilage and extends posteriorly through the lower border of the cricoid plate below the point of entry of the recurrent laryngeal nerve .
ingly sturdy and holds sutures well. Furthermore, the cuff of subglottic mucous membrane tends to stretch and is everted to accommodate the relatively larger airway diameter of the distal trachea. The result is a widely patent and well-supported airway at the anastomosis. It is essential that the completed anasto-
Fig. 38. Photograph of a cadaver specimen showing the same oblique line of transection.
Volume 70
Primary tracheal anastomosis
Number 5 November, 1975
• •
THYROID CRICOID RECURRENT N.
A
c
Fig. 4. A, Initial appearance of the upper resection line in cross section is illustrated diagramatically. A segment of the posterior cricoid plate surrounds the posterolateral aspect of the mucous membrane at the subglottic level. B, Technique of removal of cricoid cartilage subjacent to the posterolateral margins of the mucous membrane. C, Completed resection of cricoid cartilage results in some collapse of the soft tissues and subglottic lumen. D, Photograph of a cadaver specimen similar to the diagrammatic sketch shown in Fig. 4, A. E, Photograph of a cadaver specimen following submucosal resection of the posterior cricoid plate, similar to the diagrammatic sketch shown in Fig. 4, C.
Fig. S. Photomicrograph showing a cross section of the subglottic region through the lower half of the cricoid cartilage. C, Cricoid cartilage. T, Inferior cornu of thyroid cartilage. The recurrent laryngeal nerve lies within the black circle as an immediate posterior relation of the cricothyroid joint.
809
The Journal of
8 I0
Pearson et at.
Thoracic and Cardiovascular Surgery
A
I B
c
Fig. 6. A, Diagrammatic appearance of the distal tracheal resection line. The lumen is relatively larger than the subglottic lumen seen in Fig. 6, C. B, Technique of plication of the membranous trachea at the distal resection line. This procedure approximates the ends of the uppermost tracheal ring and produces a complete circle of cartilage to replace the resected cricoid ring. The luminal diameter at the distal line is also reduced. C, Diagrammatic appearance of the completed subglottic resection line seen in cross section. The vocal cords lie within 1 em. of the resection line.
Fig. 7A. Diagrammatic sketch of a thyrotracheal anastomosis. The distal trachea is "telescoped" in front of the shell of residual cricoid cartilage.
mosis be free of tension. Various techniques of tracheal and supralaryngeal mobilization may be necessary, the proper procedure depending on the length of trachea to be resected. Clinical experience Since January, 1973, 6 patients have undergone segmental tracheal resection, including an extensive removal of cricoid cartilage. In 4 cases the type of resection described under Operative technique was
Fig. 7B. Photograph of a completed anastomosis in a cadaver specimen. T, Thyroid cartilage. TR, Trachea. C, Cricoid cartilage.
done. In 2 patients, each of whom had bilateral recurrent laryngeal nerve palsy, the entire cricoid below the level of the inferior border of the thyroid cartilage was removed. Brief details of the pathology, operative procedure, and results are summarized in Table I. Cricoid resection in the presence of recurrent laryngeal nerve palsy. Two patients
Volume 70 Number 5
Primary tracheal anastomosis
November. 1975
811
Fig. 8. Postoperative tracheogram obtained in Case 1.
suffered blunt trauma to the neck, with transection of the airway at the cricotracheal junction and additional damage to the cricoid area (Cases 1 and 2). In each case, there was wide separation of the divided ends of the airway and complete disruption of both recurrent laryngeal nerves. In Case 1, there was a fracture of the anterior cricoid arch and circumferential loss of the mucosal lining within the cricoid ring. It is astonishing that this patient survived for 3 days without tracheostomy or intubation before this injury was recognized. Three days after the injury, a primary cricotracheal anastomosis was done, but the patient subsequently developed severe fibrous stenosis within the cricoid ring at the site of mucosal loss. The stricture was unimproved by repeated endoscopic dilatation and local injections of steroid (triamcinolone). Five months later all of the cricoid below the level of the thyroid cartilage was resected , and the airway was restored by thyrotracheal anastomosis. The anastomosis healed well, as illustrated in the postoperative tracheogram (Fig. 8). Patient 2 suffered a similar injury, with
Fig. 9A. Preoperative tracheogram (lateral projection) obtained in Case 2. The stricture begins at the level of the cricoid ring , and a tracheopharyngeal fistula has developed through the fracture in the posterior cricoid plate .
Fig. 9B. Postoperative tracheogram obtained in
Case 2.
The Journal of Thoracic and Cardiovascular
Pearson et al.
8 12
Surgery
Table I Patient (sex . age)
Pathology
M.F. (M,19)
Blunt trauma; transection of trachea at cricotracheal junction
Nil
Prior treatment Aug., 1972: Primary anastomosis ; healed withseven subcricoid stricture
R.R .
Blunt trauma; transection of trachea at cricotracheal junction, fracture cricoid arch and plate, and T. E. fistula Post-tracheostomy stricture at cricotracheallevel
Nil
June, 1973:Tracheostomy on day of trauma
Intact
Nov ., 1973: Exploration of subglottic stricture;no resection
Intact
Preop. radiat ion 3,000 rads in 3 weeks
(M,18)
R.O. (M,41)
Adenoid cystic carcinoma of trachea involving K.T. cricoid mucosa (M,54) L. J. (F, 19)
Caustic burn of larynx and upper half of trachea (formaldehyde and HCl)
E.B. (M,42)
Caustic burn of larynx and upper trachea (chlorine Intact inhalation, 1970)
. Intact
Feb.-Mar., 1974:Endoscopic dilatation and removal laryngeal granulomas; tracheostomy Feb. 7 to Mar. 5,1974 Tracheostomy since 1970;two operations with laryngofissure, mucosal grafts, and laryngotracheal stenting; one vocal cord resected
Legend: T.E., Tracheoesophageal. RLN, Recurrent larynge al nerves.
Fig. 10. A, Preoperative tracheogram obtained in Case 5 showing a long stricture due to inhalation of caustic material. B, Postoperative tracheogram following cricoid resection and resection of 7 em. of trachea.
Volume 70 Number 5 November, 1975
Primary tracheal anastomosis
Operation Jan., 1973:Cricoid resection and superior laryngeal release; thyrotracheal anastomosis Sept., 1973: Cricoid resection and superiorlaryngeal release; closure ofT. E. fistula; thyrotracheal anastomosis Feb., 1974: Resection of cricoid and 2 em. trachea; superior laryngeal release; thyrotracheal anastomosis Mar., 1974: Resection of cricoid and 6 em. of trachea; superior laryngeal release; thyrotracheal anastomosis May, 1974:Cricoid and upper trachea resection (total length 7 em.); superior laryngeal release; right hilar release; thyrotracheal anastomosis Oct., 1974: Resection of cricoid and 5 em.of upper trachea; superior laryngeal release; thyrotracheal anastomosis
8 13
Follow-up
Complications None
Mar., 1975 (26 mo.): No clinical airway obstruction; weak voice
Granuloma at anastomosis; removed at bronchoscopy in Mar., 1974
April, 1975 (19 mo.): No clinical airway obstruction; weak voice
None
April, 1975 (14 mo.): Normal
Localized abscess and necrosis posterior aspect of anastomosis; spontaneous healing None
April, 1975 (13 rno.): Normal
April, 1975 (II rno.): Normal
Recurrent stenosis below vocal cords; long-term T-tube silicone (Montgomery) stent inserted
April, 1975: Final result pending; good airway with Montgomery stent in place; fair voice
complete transection of the airway at the cricotracheal level and fracture of both the anterior arch and posterior plate of the cricoid cartilage. He was initially treated by tracheostomy through the retracted tracheal stump, and he subsequently developed a tracheopharyngeal fistula at the site of fracture in the posterior cricoid plate (Fig. 9A). Three months after the injury, this lesion was managed by resection of all cricoid cartilage below the inferior border of the thyroid cartilage, closure of the esophageal fistula, and thyrotracheal anastomosis. The anastomosis healed well. It is illustrated in the postoperative tracheogram shown in Fig. 9B. The injury in both of these young men resulted in permanent loss of function of the recurrent laryngeal nerves. In spite of bilateral abductor paralysis, they are both managing remarkably well 26 and 19 months after thyrotracheal anastomosis. The airway at cord level is compatible with an active life. Hoarseness and weakness of the voice has not been so disabling as to warrant further laryngeal surgery. The thyrotracheal anastomosis is widely patent in both cases. Cricoid resection with preservation of
both recurrent laryngeal nerves. The underlying pathology in these 4 patients was cricotracheal stenosis following tracheostomy (Case 3), adenoid cystic carcinoma of the trachea with cricoid involvement (Case 4), and caustic cricotracheal strictures due to inhalation burns (Cases 5 and 6). Since the recurrent larygneal nerves were intact preoperatively in all 4 patients, a modified cricoid resection was done as previously described under Operative technique. Bilateral function of the recurrent laryngeal nerves was preserved in all 4 cases. In addition to the cricoid resection, a variable length of trachea was resected (2, 5, 6, and 7 em.). Next, thyrotracheal anastomosis was performed. In those patients undergoing resections of 5 em. or less, a transverse cervical incision was used (Cases 3 and 6). In 2 patients requiring resection of longer segments (6 and 7 em.), operative exposure was obtained through a transverse cervical incision with the addition of median sternotomy (Cases 4 and 5).
A superior laryngeal release- 7 was added in all 6 patients in order to reduce tension at the thyrotracheal anastomosis. This procedure, which divides the soft tissues be-
8 14
The Journal of Thoracic and Card iovascular Surgery
Pearson et al.
a result of edema or hematoma from the operative injury. Results
Fig. 11. Postoperative tracheogram obtained in Case 4. Resection of cricoid cartilage and 6 em. of trachea was done for adenoid cystic carcinoma.
tween the hyoid bone and thyroid cartilage, lowers the position of the thyroid cartilage and larynx in the neck and is the cause of transient swallowing problems during the postoperative period in most cases. In the patient requiring resection of a 7 em. segment (Case 5), anastomotic tension was further reduced by mobilization of the right pulmonary hilum.' Access for hilar mobilization was achieved by opening the right pleural space through the median sternotomy. A small, uncuffed metal tracheostomy tube was placed below the thyrotracheal anastomosis in all 6 patients for 1 to 2 weeks after operation. A temporary postoperative tracheostomy ensures an adequate airway at a time when airway obstruction is likely to occur in the subglottic area as
Excellent functional results were obtained in 5 of the 6 patients (Cases 1 to 5). In 4 patients the airway diameters at the level of thyrotracheal anastomosis appear normal in postoperative tracheograms. One patient has mild circumferential narrowing at the anastomosis but has no clinical evidence of upper airway obstruction. One patient developed significant stenosis at the thyrotracheal anastomosis 5 weeks after operation (Case 6). A silicone rubber, T-tube stent was introduced to splint the anastomosis and will be left in place for 6 to 9 months. The man was operated upon in October, 1974, and the final result is still pending. He is the only patient who did not have normal mucous membrane in the segment extending for 1 em. below the vocal cords. In this case, an inhalation bum produced severe damage to the larynx. Before admission to our hospital, he had undergone two major operations by laryngofissure, with excision of the left arytenoid cartilage and vocal cord and the application of free mucosal grafts supported by a laryngotracheal stent. The results obtained in 2 patients who underwent cricoid resection with preservation of the recurrent laryngeal nerves are shown in the preoperative and postoperative tracheograms illustrated in Fig. 10, A and B (Case 5) and Fig. 11 (Case 4). Discussion The operative technique which has been described affords the advantages of a onestage reconstruction and primary anastomosis with apposition of mucosa to mucosa. Stenting was necessary in only 1 of 6 patients. The operative procedures is technically difficult and requires a precise knowledge of the anatomy of the larynx and subglottic region . This anatomy is probably best learned from cadaver dissections . The operation is applicable in selected patients with subglottic pathology. Before
Volume 70 Number 5
Primary tracheal anastomosis
8 15
November, 1975
operation, the surgeon must ascertain that the mucous membrane is relatively normal throughout the first 1 em. of the subglottis below the vocal cords. Although the condition of the subglottic mucosa can be determined by preoperative bronchoscopy or laryngoscopy, the level of the pathology can be evaluated precisely only by contrast tracheograms obtained in anteroposterior, oblique, and lateral projections. REFERENCES
2
3
4
5
6
7
Ogura, J. H., and Powers, W. E.: Functional Restitution of Traumatic Stenosis of the Larynx and Pharynx, Laryngoscope 74: 1081, 1964. Ogura, J. H., and Biller, H. F.: Reconstruction of the Larynx Following Blunt Trauma, Ann. Otol. Rhinol. Laryngol. 80: 492, 1971. Harley, H. R. S.: Laryngotracheal Obstruction Complicating Tracheostomy or Endotracheal Intubation With Assisted Respiration: A critical Review, Thorax 26: 1971. Grillo, H. C.: The Management of Tracheal Stenosis Following Assisted Respiration, J. THoRAc. CARDIOVASC. SURG. 57: 52, 1969. Pearson, F. G., and Andrews, M. 1.: Detection and Management of Tracheal Stenosis Following Cuffed Tube Tracheostomy, Ann. Thorac. Surg, 12: 359, 1971. Gerwat, J., and Bryce, D. P.: The Management of Subglottic Laryngeal Stenosis by Resection and Direct Anastomosis, Laryngoscope 84: 940, 1974. Dedo, H. H., and Fishman, N. H.: Laryngeal Release and Sleeve Resection for Tracheal Stenosis, Ann. Otol. Rhinol. Laryngol. 78: 285, 1969.
Discussion DR. HERMES C. GRILLO Boston, Mass.
This is an important contribution for the solution of a problem which, while not very common, presents enormous difficulties when it occurs. Until now, only prolonged stenting procedures have been available for such problems. There is a spectrum of inflammatory lesions within the first 3 em. below the vocal cords, and this lesion is a very special part of that spectrum. I emphasize this because I do find there is a tendency to label several lesions as "high tracheal stenosis." Stenosis involving the upper trachea is one problem. Even if one half of the first ring is intact, the problem is simple. If the stenosis reaches the lower border of the cricoid, the problem becomes a little more
difficult technically and the results a little less good. When the stenosis involves the lower part of the cricoid-low subglottic stricture-and part of that ring can be saved by being beveled, the problem gets a little more complicated. When the level of stenosis creeps up to a point above the top of the cricoid, a true intralaryngeal subglottic stenosis is present. It is to this last group of patients that Dr. Pearson and his colleagues have directed this superb and very elegant technique. We have had single-stage solution for those problems in the past. Dr. Pearson very properly cautioned us that this will be a difficult technique unless we take the time to explore the anatomy in intricate detail and be very certain of what we are doing. Even then it is going to require masterful technique to get the kind of results reported. In passing, Dr. Pearson presented cases of bilateral cord transection in which this technique was applied. It is important to point out that people with permanent palsies of both cords can have a functional larynx. However, in some patients with permanent palsies from trauma, the cords remain in mid position. Thus an otolaryngologist should pin one of the cords or do an arytenoidectomy first to obtain a glottic airway before an attempt is made to repair the subglottic airway. DR. WILLIAM E. NEVILLE Newark, N. J.
As Dr. Pearson has so aptly described, a tracheal resection at the cricoid level poses enumerable technical problems. There is no question that a primary anastomosis should be performed if at all possible. However, this may not be possible. I would like to submit an alternate method which we have used to advantage on occasion when long segments of the anterior portion of the trachea have been involved, but where the membranous portion can be left intact. I like to refer to this incision as a slash tracheostomy done by an eager general surgical resident. The trachea was incised anteriorly from the cricoid cartilage down to the manubrium because of severe respiratory insufficiency. The patient was discharged after a stormy hospital course, but he was readmitted with complete tracheal obstruction. It was necessary to dilate the trachea before insertion of an endotracheal tube. Invagination of the anterior cartilages into the lumen created a web inside of the trachea from the cricoid cartilage to the manubrium. We resected the anterior and lateral margins of the trachea but left the membranous portion. The silicone rubber tracheal prosthesis was invaginated into the upper and lower tracheal open-
The Journal of
8 16
Pearson et at.
Thoracic and Cardiovascular Surgery
ings so that the airway was re-established. Then we simply reapproximated the thyroid and the strap muscles of the neck over the prosthesis. From our results in a few cases, I think this is a worthwhile operation when primary tracheal anastomosis cannot be done. DR. JOHN R. BENFIELD Torrance, Calif.
I wonder if Dr. Pearson could help some of us who operate in this area less frequently than he does by commenting on two points. First, exactly how do you handle the thyroid gland? It seems to me there are two alternatives: (1) to split it down the middle and reflect both sides laterally or (2) to divide the superior thyroid vessel and retract the whole gland toward the opposite side of the predominant operation. I would like to know which approach you usually use. The second question, and perhaps the more important point, has to do with the function of the laryngeal nerve in the postoperative period. Is there a need for tracheostomy or prolonged endotracheal intubation in the management of these patients? In the late postoperative period, how many of your patients had complete return of function? DR. PEARSON (Closing) I wish to thank Dr. Grillo for his lucid and supportive discussion. He made reference to the problem of bilateral recurrent nerve palsy, which results in an unpredictable airway at the level of the glottis. In our 2 patients with blunt trauma and complete disruption of both recurrent laryngeal nerves, the vocal cords remained in an intermediate or partially abducted position. This was most fortunate, since the cords did not seriously impair their airway, and yet they retained sufficient voice to be understandable. Vocal cord function has not returned in either case, the one having been followed for 18 months and the other for about 2 years. These 2 patients were seen by our otolaryngologists who recommended that nothing further should be done. Any effort to improve their voice would potentially compromise the airway. Dr. Ogura, an otolaryngologist in St. Louis, Missouri, has written extensively concerning surgical reconstruction after laryngeal and laryngotracheal injury. He routinely excises one arytenoid cartilage and secures one vocal cord to the side of the larynx in patients who have sustained blunt trauma with division of both recurrent laryngeal nerves. Although his experience is much more extensive than ours, it would appear that routine excision and lateralization of the vocal
cord is not necessary. In our 2 patients, nothing was done to the vocal cords, and the outcome has been perfectly satisfactory. However, I understand that there may be late changes at cord level in patients with recurrent nerve palsy because of loss of muscle bulk and scar replacement. Thus these young men may still require corrective surgery at a later date. A question was asked regarding tracheostomy in patients undergoing cricoid resection. In all 6 of our patients a tracheostomy was placed below the anastomosis. The operative field is immediately below the vocal cords in the narrowest part of the airway, and we have assumed that minimal edema or hematoma would obstruct the airway at the level of the glottis in the early postoperative period. For this reason we have placed a small, temporary tracheostomy tube below the anastomosis. It has not been left for longer than 2 weeks in any patient, other than the single individual who still has aT-tube stent and in whom the result is pending. Dr. Benfield asked about cord function in those 4 patients in whom we were able to preserve the recurrent nerves. All of these patients had normal cord function with a return of normal voice within a week or 10 days of operation-as soon as the tracheostomy tube was removed. I think we have been fortunate in that none of the 4 patients has had even a transient palsy. Dr. Benfield also asked about the management of the thyroid gland in the process of identifying the recurrent nerves. If the nerves are to be identified, the dissection is tedious since one must free the nerves completely from their posterolateral thyroid relations. In these patients, this has been done in two ways. The lobe has been reflected forward wtih the dissection begun behind. We have also split the isthmus and freed the lobe from the lateral aspect of the trachea to identify and free the nerve posteriorly. In the patient with adenoid cystic carcinoma, it was necessary to remove the right lobe of the thyroid with the tumor and tracheal segment. However, it is nearly always possible to leave at least one lobe of thyroid. I agree with Dr. Neville that there are some patients in whom a primary anastomosis cannot be achieved. I am concerned, however, about the use of a prosthesis in this location. The subglottic region is the narrowest part of the airway, lying immediately below the vocal cords. I believe that any prosthesis, even a silicone prosthesis, is likely to cause some local tissue reaction in this critical area. I would anticipate a significant incidence of obstructive problems due to granulation tissue or edema at the upper end of any permanent prosthetic replacement.