Use of the Linear Stapler for Pharyngoesophageal Closure After Total Laryngectomy

In 1972 Paches et al. 1. published their experience in the Russian literature with the use of a stapling device during laryngectomy. However, this technique has not been used in the United States, although internal stapling methods have received recent popularity in the treatment of Zenker's diverticulum. 2.-4. Between October 1995 and October 1999 the senior author (r.a.s.) has performed 19 of 24 laryngectomies with the Ethicon (Somerville, NJ) linear stapler. Most closures were performed as “closed techniques” in which the larynx is not separated from the pharyngoesophagus until closure is completed. Once the larynx is skeletonized and definitive resection is imminent, the organ resection and stapled closure are accomplished in 5 to 10 minutes. This method is efficient, produces a reliable and secure suture line, and offers the theoretical advantage of the closed technique of avoiding salivary contamination from the oropharynx into the operative field. The total laryngectomy is performed in a standard fashion. If a neck dissection is required, the tissues are left in continuity with the larynx. After the inferior constrictor fibers are sectioned, the greater cornua of the hyoid bone and thyroid cartilages are sectioned with either bone cutters or heavy Mayo scissors (Fig. 1). The larynx is then pulled forward, putting the anterior wall of the pharyngoesophagus on a stretch. The trachea is transected, and the endotracheal tube converted to intubation of the distal tracheal segment. A 0° sinus endoscope is passed retrograde through the glottis to visualize the laryngeal surface of the epiglottis. The assistant produces external pressure on the vallecula, delivering the leading edge of the epiglottis into a more posterior location. A skin hook is advanced parallel to the endoscope until it is in position to engage the tip of the epiglottis and pull it into the glottic region. Under endoscopic view, the epiglottic tip is grasped with an Allis clamp and the skin hook disengaged and removed. At this point, the TL60 Ethicon linear stapler is prepared for use, and a second (reload) staple clip obtained. Usually the inferior staple line is performed first and the lower half of the larynx separated from the repaired pharynx (Fig. 2). The epiglottis is then pulled inferiorly into the glottis, placing the vallecula on a stretch in such a manner that the epiglottis will not be incorporated into the staple line (Fig. 3). The stapler is reloaded and inserted between the larynx and superior pharynx in a reversed direction. The open end of the device is inserted from the vallecular region in such a manner that the two staple lines meet or even cross at approximately the midpoint of the inferior constrictor fibers, although only mucosa is engaged in the repair. A few remaining sutures may be employed at the tongue base and juncture of the two staple lines if there is any question concerning tension or completeness of the closure, but this is optional. During application and eventual firing of the stapler, the device is positioned as close as possible to the thyroid ala to preserve as much pharyngeal mucosa as possible. This is the reason that the posteriorly projecting cornua of the thyroid cartilage and hyoid bone are removed before the stapler is positioned. The greater cornua of the thyroid cartilage and hyoid bone are removed to prepare the larynx for optimum placement of the staple device in closed laryngectomy. eso = esophagus; icm = inferior constrictor muscle. Inferior stapled pharyngeal repair has been completed, and the mucosal attachments to the larynx are divided. The epiglottis (epi) is pulled into the glottis–subglottis, opening the vallecula and allowing placement of the stapler in preparation for the superior repair and completion of the laryngectomy. In circumstances in which a tumor involves the epiglottis, vallecula, piriform sinus, or any location where the closed technique might risk or compromise adequate margins, the laryngeal resection is performed in the standard open fashion. After larynx removal, the pharyngeal mucosal edges are approximated in a midline linear fashion with the use of Allis clamps as close to the free edge as possible. The stapler is then used to create a superior and inferior stapled closure, and the mucosa incised and divided in such a way that only the mucosa previously grasped by the Allis clamps must be discarded. Once again, reinforcement sutures may be used at the juncture of the two staple lines. If primary puncture is preferred, special considerations are required for the closed staple technique (Fig. 4). A medium- to large-sized Hurst rubber dilator, which has been carefully evacuated of its mercury and has been deemed safe for clinical use, is a valuable asset for this procedure. A rectangular window 2 to 3 cm in length and 2 cm in width is created beginning 3 to 4 cm from the blunt tip. The empty dilator is inserted transorally through the region of the repair into the esophagus. The window is positioned before the dilator passage in such a manner that it can be palpated facing anteriorly through the esophagus. A puncture is created through the esophageal mucosa just above the cut edge of the posterior tracheal mucosa. A catheter of choice is inserted until it reaches the blunt, occluded end of the dilator. Then the dilator and catheter are advanced together a few centimeters into the thoracic esophagus. The dilator is advanced further, independent of the catheter, until the latter is disengaged from the window. The dilator is rotated 180° on axis to avoid catching the transesophageal puncture catheter in the window as the dilator is withdrawn. This puncture site is preferred to an intratracheal location, because it facilitates ease of changing valves in the future. With external housing and diaphragms, the transesophageal valve does not need to reside within the trachea proper. Tracheoesophageal puncture method. A. Puncture is placed just at the mucocutaneous junction with a hemostat into the bougie window. The catheter is then advanced into the closed end of the bougie. B. Bougie and catheter are advanced together into the mid thoracic esophagus. C. Bougie is further advanced while the catheter is held stationary, disengaging it from the window. D. Bougie is rotated 180°, preventing the window from catching the catheter as the bougie is withdrawn. E. Bougie is withdrawn and the catheter is advanced into the stomach. From October 1995 to October 1999 the senior author has performed 24 total laryngectomy procedures; in 19 of these, repair was performed with the Ethicon linear TL60 stapler. In five patients the extent of tumor suggested the need for conventional repair rather than stapled closure. In three of these cases, tumor extent required significant resection of esophageal mucosa, raising questions about the adequacy of esophageal patency after repair. The staple technique was abandoned in these patients because of concern for sacrificing an additional 2 mm of mucosa beneath the stapler jaws. In the two other cases, significant tongue base extension required removal of so much tissue that use of the stapler was judged unsafe for supporting a tension-free closure. These were the only contraindicated circumstances for stapled closure, and every other patient in this series requiring laryngectomy was deemed suitable for this technique. Twelve of the 19 mucosal repairs were performed with the closed technique, and in the remaining 7 with conventional open methodology. Of these 12 patients, 2 demonstrated a pinpoint leak on barium swallow test before alimentation was initiated. In these two patients tube feedings were continued for another 4 days and subsequent radiograph images were negative for fistula. Neither of these two patients demonstrated signs of clinical infection. No patient developed stenosis at the stapled repair. The patient who underwent conventional suture closure demonstrated a large tumor extent leaving a smaller than usual remnant. A stapled closure in this patient might have compromised the already reduced pharyngoesophageal remnant. Four of the 19 patients (21%) had a previous full course of radiation therapy and none developed fistula after stapled closure. In patients with nonirradiated larynges, the mean time from laryngectomy to initiation of tube feedings was 5.7 days and the median time was 5 days. The patients with irradiated larynges were fed, on average, at 7 days (range, 5–8 d). Of the 19 patients who had stapled repair, 10 were treated with total laryngectomy alone and the other 9 required either unilateral or bilateral neck dissection. Nine patients who had stapled repair underwent concurrent primary transesophageal puncture. The approximate time for completion of the stapled repair and laryngeal resection in the closed technique was 5 minutes, and 8 minutes, on average, was required for the open techniques. In contrast, most laryngectomies performed by a standard two-layer suture closure require nearly 45 minutes for completion of the repair. Stapled closure of defects of the alimentary tract is the routine rather than the exception for various conditions of the gastrointestinal tract and tracheobronchial tree. These closures are efficiently performed and very reliable. The only principal difference between stapled and sutured closure is the inability to achieve inversion of mucosal edges with staples. This is more of a theoretical than an actual disadvantage. None of the patients in this series developed clinical fistulization. Twenty-one percent of surgically treated patients with stapled closure had a previous full course of irradiation, and none of these individuals developed wound complications. Although this series of patients received their stapled closures with the Ethicon TL-60 linear stapler, other units of similar design by other manufacturers are available (Table I). These stapling devices are disposable, and those specifically adapted to laryngectomy create a staple line of up to 60 mm in length. This seems to be the ideal length of closure, because longer staple lines would not adapt properly to the requirements of staying as close as possible to the laryngeal framework and vallecula if one longer linear closure alone is created. Each staple is 4.8 mm long and 4 mm wide. The staples are manufactured from titanium; they are not ferromagnetic and thus are compatible with magnetic resonance imaging. The devices all create a double-layered closure with the staples offset from those in the adjacent row, effectively leaving no gaps between staples along the repair (Fig. 5). In patients who have had irradiation or who have thicker tissues at the tongue base, longer staples may be better suited to an effective closure. These Ethicon staplers (TLH-60) are 5.5 mm in length and create a larger staple loop. Appearance of the dual staple line closure. A. Single closed staple after firing of the device. B. Staggered staple line closure as viewed from the top. C. Staggered staple line closure as viewed from the side. Before the use of staplers, our usual feeding routine began at 7 days after surgery for patients with nonirradiated larynges and 10 days for patients with irradiated larynges. Those who did not demonstrate cellulitis or fever were believed to be at low risk for fistula and thus begun on clear liquids at the appointed time. The authors have been comfortable with obtaining esophagograms in the patients who had stapled repair at the fourth or fifth postoperative day in patients with nonirradiated larynges and initiating a full liquid diet. Patients with irradiated larynges were similarly treated at 7 days after surgery. The clinical experience to date has revealed that patients with stapled closure can begin oral feedings at an early interval after surgery without increased risk of fistula and wound breakdown. With the advent of organ preservation protocols, it is possible that a subpopulation of patients with heavily irradiated larynges will come to salvage surgery. It appears that stapled closures are very secure and may offer advantages over routine suture techniques, because there are no gaps between the staples. Only a large prospective comparative series will clarify this hypothesis. The technology is easy to master, but practice with the device before clinical application is mandatory. Colleagues who perform intestinal anastomoses are adept at using the equipment and are an excellent resource for practice and initial ventures in the operating room setting. The linear stapler has long been approved for use in the alimentary tract, although pharyngeal closure after laryngectomy is a relatively new and unique application. The larynx can be separated from the pharyngoesophagus using staple techniques that can allow direct internal inspection of the alimentary lumen or resection without this requirement. These techniques are simple to learn, efficient in application, and associated with low complication rates. These repairs may in fact be more secure than traditional closure techniques, but only larger experience with stapled closure will support or challenge these early experiences.