Optimizing voice in conservation surgery for glottic cancer

Optimizing voice in conservation surgery for glottic cancer

OPTIMIZING VOICE IN CONSERVATION SURGERY FOR GLOTTIC CANCER GREG HARTIG, MD, STEVEN M. ZEITELS, MD Endoscopic and open resection techniques for glott...

8MB Sizes 17 Downloads 15 Views

OPTIMIZING VOICE IN CONSERVATION SURGERY FOR GLOTTIC CANCER GREG HARTIG, MD, STEVEN M. ZEITELS, MD

Endoscopic and open resection techniques for glottic carcinoma will be discussed. Technical aspects of both resection and reconstruction that allow for optimizing voice will be emphasized.

Conservation surgery for laryngeal carcinoma has origins predating total laryngectomy, and both horizontal and vertical techniques have proven oncologic efficacy spanning decades of use. H.B. Sands performed the first curative partial vertical laryngectomy for glottic carcinoma through a thyrotomy in 1863.1 Ten years later, Billroth performed the first total laryngectomy for carcinoma.2 From the 1940s through the 1960s, surgeons such as Alonso, Som, and Ogura developed and formalized standard partial vertical and horizontal laryngectomy procedures. In addition, much of the acceptance of conservation surgery is due to the work of Kirchner, who described growth characteristics of laryngeal carcinoma as they relate to laryngeal compartmentalization. This gave credence to the concept of oncologic safety with millimeter margins for partial laryngectomy surgery.3 When patients are carefully selected for conservation procedures, local control is similar to that achieved with total laryngectomy, and is often superior to that achieved with primary radiotherapy. Clear indications for the various conservation procedures are established by means of a thorough preoperative assessment. Careful endoscopy and radiographic imaging provide information that facilitates the design of consistently successful procedures by accurately estimating the location and volume of disease. Because there is a spectrum of vertical conservation procedures available, the surgeon can choose the procedure that will eradicate the disease with the least morbidit~ ensuring for maximal functional outcome at the same time. However, procedural selection is highly individualized, and influenced by both patient preference and surgeon bias. For example, some feel that T3 lesions of the glottis can be addressed with hemilaryngectomy, whereas others would limit indications to T1 and T2 tumors. 4,5 Although individual surgeons may successfully extend a given technique to include larger or more extensive tumors, this discussion will present conservative indications for the procedures discussed. For most otolaryngologists who practice conservation surgery, this will result in more consistent oncological and functional results with fewer patients requiring salvage total laryngectomy for recurrent tumor or chronic aspiration. From the Department of Otology and Laryngology, Harvard Medical School, and the Department of Otolaryngology, Massachusetts Eye and Ear Infirmary, Boston, MA. Address reprint requests to Steven M. Zeitels, MD, Department of Otolaryngology, Massachusetts Eye and Ear Infirmary, 243 Charles St, Boston, MA 02114. Copyright © 1998 by W.B. Saunders Company 1043-1810/98/0904-0006510.00/0 21 4

This article will concentrate on partial vertical resections moving from least to most aggressive. For the various procedures, indications relative to tumor characteristics and patient selection considerations will be discussed. Technical aspects of procedures will be limited to those factors involved in both the resection and reconstruction, which we feel are important for optimizing the postoperative vocal outcome.

T1 GLOTTIC CANCER: PHILOSOPHY OF PHONOMICROSURGICAL MANAGEMENT Phonomicrosurgical resection of Tla glottic cancer is an excellent treatment alternative to radiotherapy (XRT), which is the most commonly used treatment modality in the United States for the management of early glottic carcinoma. Ironicall~ many surgeons who support radical ablative resections for marginally resectable/curable disease are tentative about surgically treating very early, and often microinvasive, vocal-fold cancer. The development of phonomicrosurgical techniques for resection have improved the postoperative vocal outcome. 6-9 Furthermore, there are a number of new phonosurgical reconstructive techniques 1° that are enhancing the final voice result even further. The underlying philosophy is that the disease can be controlled equally well by endoscopic resection or XRT. However, the XRT cannot be confined to one vocal fold. In turn, glottal oscillation and the voice are better if a unilateral lesion is resected with ultranarrow margins with the vocal edge being reconstructed as necessar~ than if both vocal folds are irradiated. The surgical option will likely become relatively more important as medical resources become more tightly rationed. 11 Because any adequate treatment for T1 glottic cancer (endoscopic excision, transcervical excision, XRT) results in a cure rate of approximately 85% to 90%, 12 a primary parameter by which the clinician should judge the success of the treatment is the resulting quality of the voice. Recently, laryngologists have incorporated physiological principles of laryngeal sound production into the design of the oncological procedures. However, the surgeon must not lose sight of the variety of circumstances that might dictate a particular treatment approach. A primary goal of the endoscopic management of T1 glottic cancer is to narrow the cancer-free margin and thereby minimize patient morbidity without altering the cure rate. Phonomicrosurgical procedures 6-9 can precisely accommodate the narrow deep resection margin of T1 vocal-fold cancer to the depth of invasion (invisible third dimension) (Fig 1). A small amount of extra tissue that is

OPERATIVETECHNIQUES IN OTOLARYNGOLOGY--HEAD AND NECK SURGERY, VOL 9, NO 4 (DEC), 1998: PP 214-223

432 6

al lamina propria

al ligament nuscle FIGURE 1. Schematic diagram displaying varied depths of neoplastic invasion despite, which is indeterminable based on surface appearance.

excised with the deep margin can have a profound effect on the vocal outcome, although not necessarily improving the cure rate. This is especially so for lesions that are confined to the epithelium or that minimally invade the superficial lamina propria. Reinke's space infusion has helped to precisely determine the depth of invasion of T1 cancers before committing to the depth of the excision. 6,13There are a number of patients with T1 cancers in whom the infusion has allowed for preservation of all of the vocal ligament and vocalis muscle, as well as part of the deep portion of the superficial lamina propria. The authors have not had any local failures in approximately 20 patients in whom this graded microsurgical approach was used. 7 The majority now have more than 3 years of follow-up. Endoscopic excision is associated with a very low complication rate, which consists mainly of minor postoperative bleeding u47 and granuloma formation. ~548 An en block excisional biopsy provides an accurate diagnosis as well as an effective treatment, ~4,~9 and does so with minimal morbidity. Whole-mount-section histological examination of resected specimens prevents overtreatment or undertreatment of small glottic lesions. All treatment options are preserved after endoscopic excision including further transoral resections. Unlike radiotherapy, endoscopic resection treats only the lesion without ablating the remaining normal glottal tissue. Postradiotherapy videostroboscopic exams suggest that both normal and cancerous tissues reflect fibrotic changes and impaired mucosal oscillation. 2° The use of radiotherapy for early cancers precludes its further use for tumor recurrence or for new primaries, and may even induce carcinogenesis in atypical epitheliumY ,22 The cost of endoscopic excision is significantly less than open laryngeal surgery or radiation, because either of these latter procedures are typically preceded by a staging endoscopy, lu2 Also, the patient sustains the increased burden of time and travel commitments required by radiotherapy.

DISEASE IN THE ANTERIOR COMMISSURE The anterior commissure tendon, or Broyle's ligament, is a confluence of the vocal ligament, the thyroepiglottic ligament, the conus elasticus, and the internal perichondrium of the thyroid ala. There is a misconception that T1

HARTIGAND ZEITELS

cancers at the anterior commissure have a great predilection for understaging and that many of these lesions have occult invasion of the thyroid cartilage (T4 stage). This is based upon the misunderstanding that the anatomy of the dense anterior-commissure ligament is a less resilient tumor barrier than the adjacent thin thyroid perichondrium (Fig 2). Kirchner 23,24has made clear that Tla and Tlb carcinomas rarely transgress Broyle's ligament to invade thyroid cartilage. Anterior commissure tumors that have thyroid cartilage invasion typically display cephalad surface invasion of the infrapetiole region of the supraglottis, or caudal surface invasion of the subglottis (both T2 by surface staging criteria) (Fig 3). There has been divergent opinion as to whether cancer can be endoscopically eradicated from the anterior commissure. The proscriptions imposed by some are based primarily on the difficulty in obtaining adequate surgical exposure in this area3 s The rate-limiting factors for resection of early cancer in the anterior commissure are the true extent of the disease (is it invading cartilage?), and the endoscopic exposure required to encompass the lesion. Davis m and the Boston University group, and later Koufman et al, is demonstrated that cancer could be removed from the anterior commissure. However, it required great skill to excise the lesion without vaporizing the specimen. Vaporizing cancer without clear en block resection margins is inadequate surgical oncological technique. This factor, as well as underestimating the extent of disease, probably led to the reported failures by a number of investigators. 25-28 If one accepts Kirchner's histopathological data regarding the invasion pattern of T1 glottic cancer, there is no reason to believe that an adequate soft tissue resection of small volume soft tissue disease is not adequate treatment. This premise is further substantiated by a number of reports that did not find a correlation between anterior commissure involvement with T1 glottic cancer and failure of radiotherapy as a curative modality. 29-3° The extensive European experience reported by Steiner, 3t Eckel and Thumfart, 16 and Rudert 32 substantiates that glottal cancer in the anterior commissure can be removed transorally. The problem with any surgical approach to the anterior commissure (for true Tla, Tlb, and T2 lesions) is that these procedures disturb the structural integrity of the anterior commissure and result in a poor voice. Lesions that are invading cartilage are T4 by stage, require open partial laryngectomy, and are not suitable for endoscopic ,:::~ Macula : ,~ flava Inner perichondrium

i ~: ~

~ '~%

//

~ \:

~-.

-4

|

!' ....

Superficial lamina propria FIGURE 2. Ligamentous composition of the anterior commissure tendon.

215

FIGURE 4. The glottiscope has been slightly withdrawn to enhance exposure of the vestibular folds and infrapetiole region to facilitate laser excision of those tissues.

sive cancer; 2. Dissection within the superficial lamina propria for microinvasive cancer that is not attached to the vocal ligament; 3. Dissection between the deep lamina propria (vocal ligament) and the vocalis muscle for lesions that are attached to the ligament but not through it; 4. Dissection within the thyroarytenoid muscle for lesions

FIGURE 3. Pattern of caudal and cephalad spread of cancer into thyroid cartilage at the anterior commissure.

excision. An endoscopic exploration of the infrapetiole region (Figs 4-6) of the supraglottis allows for definitive determination of whether a presumed T1 lesion has cartilage invasion (T4). 33 This procedure facilitates precise staging (without disarticulating Broyle's ligament), commensurate management, and optimal post-treatment voice quality. ENDOSCOPIC TECHNIQUE

The goal of endoscopic treatment of an isolated T1 lesion of the musculo-membranous vocal fold is eradication of the disease with maximal preservation of the normal layered microstructure. This approach results in the optimal postoperative voice without compromising oncological cure. There are four basic procedures that are based on the depth of excision (Fig 7)7: 1. Dissection just deep of the epithelial basement membrane and superficial to the superficial lamina propria for atypical epithelium and microinva21 6

FIGURE 5. The lower supraglottal tissues have been resected to enhance exposure of the tumor and the superior surface of the glottis. OPTIMIZING VOICE IN GLOTTIC CANCER SURGERY

tion between the vocal ligament and the vocalis muscle can be performed equally well with cold instruments alone, or with assistance by the laser. Dissection within the muscle is performed most precisely with the CO2 laser, which allows for improved visualization because of its hemostatic cutting properties. The subepithelial saline-epinephrine infusion into Reinke's space (Figs 8,9) improves preexcisional assessment of lesion depth. If the tumor has invaded the vocal ligament, the SLP at the perimeter of the lesion will distend, creating a contour depression in the region of the cancer. The subepithelial infusion assists with the surgeon's technical execution of the surgery in a number of other ways: 1. The infusion facilitates mucosal incisions by improving visualization of the lateral border of the lesion, and by distending the SLP so that the overlying epithelium is under tension; 2. The infusion also increases the depth of the superficial lamina propria, which facilitates less traumatic dissection in this layer and leads to regenerated epithelium that is more flexible; 3. The epinephrine and hydrostatic pressure of the infusion vasoconstricts microvasculature in the SLP, and this improves visualization and precise dissection; 4. If the laser is used, the saline acts as a heat sink, which decreases thermal trauma to the normal vocal fold tissue. VOCAL OUTCOME AFTER ENDOSCOPIC RESECTION OF THE MUSCULO-MEMBRANOUS REGION

FIGURE 6. The anterior commissure tendon can be partially dissociated from the thyroid cartilage to determine if there is invasion into the cartilage [T4 lesion].

penetrating the vocal ligament and invading the vocalis muscle. This approach can be further fine tuned by performing partial resections of any of the layered microstructure. The specimen is always oriented for wholemount histological analysis, and frozen-section margin assessment is employed selectively to verify a complete excision. 6,7 If dissection is performed in the superficial lamina propria (SLP), cold instruments facilitate precise tangential dissection around the curving vocal fold. This allows for maximal preservation of the superficial lamina propria, and for pliability of the regenerating epithelium. 6,34Dissec-

/ I ]]

II

FIGURE 7. Schematic diagram displaying varied options for ultranarrow margin resection based on the depth of tumor invasion. HARTIG AND ZEITELS

The laryngologist can precisely accommodate a microcontrolled excision to the three dimensional characteristics of the vocal cord cancer. Unlike radiation, 2° this approach avoids trauma to uninvolved tissue, such as the contralateral fold, during treatment of a unilateral lesion. It is to be expected that increasing the magnitude of the resection, especially its depth, leads to deteriorating vocal function. ~,35,36 Therefore, the graded resection approach that was suggested by Hirano 33 and Koufman 15 was further developed. 6,7 The mucosal microflap technique, which is useful for resection of epithelial atypia and microinvasive cancer, can sometimes be confined to the superior surface of the vocal fold. After reepithelialization, typically, the patient can comfortably increase subglottal driving pressures to achieve perceptually and objectively measured (acoustical analysis) normal voice at conversational levels. Maximal range tasks reveal some mild limitation of frequency variation (dynamic range) and loudness capability. These vocal findings show mild progression in a similar pattern with extension of the resection to the medial vocalizing surface of the vocal fold. 6 When the surgical dissection is confined to the SLP in one vocal fold, the perceptual assessment of the resulting voice is usually normal. Stroboscopy reveals mild impairment of mucosal wave propagation, in regard both to amplitude and magnitude, wherever there has been dissection in the SLP. As the deep resection margin extends to include the vocal ligament, there is further acoustic, aerodynamic, and stroboscopic impairment. When the SLP is excised as a component of the cancer resection, the regenerated epithelial surface is adherent to the underlying body of the vocal cord. 6,37This results in unavoidable stiffness in the regenerated epithelium. We have not seen evidence that the pliable superficial lamina-propria regenerates after removal in the way that epithelium does. However, after a resection of the vocal fold epithelium and its underlying laminae propria, the healed neocord is usually smooth and straight, and 217

FIGURE 8. The subepithelial infusion reveals tumor invasion into the vocal ligament.

glottal closure is usually normal. Vocal limitations may still be imperceptible at conversational levels in some patients, if glottal closure is complete. Obvious hoarseness is usually observed when complete glottal closure does not occur, because of tissue loss and a resulting vocal-edge excavation.6, 37 When a significant amount of vocalis muscle is removed as part of the deep margin, the healed neocord is usually concave. The excavated neocord results in incomplete glottal closure during phonation. This leads to a stiff, leaking glottal valve and a hoarse, disordered voice. As the depth of an excision leads to valvular incompetence of the glottis, there is increased subglottal pressures and flows; acoustical instability (jitter and shimmer); and supraglottal muscle strain patterns. Stroboscopy reveals no substantive mucosal wave oscillation, but rather a mass vibration of the stiff neocord body. 6 Once the neocord is healed, phonosurgical reconstruction may ensue if the vocal edge is excavated. Initiall]6 this consists of microlaryngoscopic fat injection. This may take

one or two injections based on h o w much tissue was initially resected. As necessary, a medialization laryngoplasty can be performed once there is adequate glottal tissue medial to the thyroid lamina. The advantage of the medialization is that it is done under local anesthesia, with intravenous sedation, which allows for phonatory feedback during the procedureJ ° OPEN

CORDECTOMY

VIA LARYNGOFISSURE

Open cordectomy can be used in patients who have T1 lesions of the true vocal fold and who are not candidates for laser resection. This includes patients with larynges not well visualized on suspension laryngoscopy, those who have failed primary radiotherapy, those who have lesions approaching the anterior commissure, and those with lesions extending onto the anterior vocal process. The latter two indications can now often be approached transorally due to improvements in endoscopic instrumentation (laryngoscopes, hand instruments, microspot lasers).

FIGURE 9. The subepithelial infusion lifts the lesion, which has microinvasion through the epithelial basement membrane. This enhances the precision of the procedure and facilitates maximal preservation of the vocal folds layered microstructure.

218

OPTIMIZING VOICE IN GLOTTIC CANCER SURGERY

FIGURE 10. A cross-sectional view of resected tissue during open cordectomy.

In this procedure, one should plan on a temporary tracheotomy in most patients. However, in younger and healthier patients, the procedure can be accomplished without tracheotomy. Generally, these patients will have a delayed recovery secondary to the laryngofissure and, for many, tracheotomy as compared to an endoscopic resection. TECHNIQUE

After accomplishing adequate exposure of the larynx, the thyroid cartilage perichondrium is divided in the

midline, and the cartilage is likewise divided without entering the airway. Next, the cricothyroid membrane is entered from below and a straight cut is made through the anterior commissure, slightly to the side of the lesion. If the procedure is being used for tumors of the middle third of the fold, this will still allow an adequate anterior margin and better preserve the anterior attachment of the contralateral cord (Fig 10). Detachment of the contralateral cord is one reason for a poor mucosal wave being found in the normal cord postoperatively. Next, cuts are made at the level of the subglottis, and at the apex of the laryngeal ventricle. The deep extent of the resection can preserve or include the thyroid perichondrium. In lesions that are more anterior, it is advisable to take this layer to provide a more substantial margin. The posterior aspect of the resection is dictated by the extent of the tumor. When performing an open cordectomy without a tracheotomy one can simply retract the endotracheal tube to the noninvolved side of the larynx. Reconstruction of the cordectomy defect is best accomplished by mobilizing the false fold mucosa, and suturing over the cordectomy defect with absorbable (chromic gut or polyglycolic acid) suture (Fig 11). Typically, one mobilizes the false fold by elevating the underlying thyroid perichondrium, allowing a bloodless plane of mobilization without damaging paraglottic soft tissue. During closure, careful reapproximation of the anterior commissure will be an important component in optimizing voice. When the remaining vocal folds meet at different vertical levels, voice is usually breathy and weak. This can result in vibration occurring between the normal vocal fold and the false fold of the operated side. Brasnu et al reported their voice results of 151 patients with and without false fold flap reconstruction. The incidence of moderate and severe voice alteration was significantly higher in the nonreconstructed group. 38Without reconstruction, the patient typically has a persistent glottal gap during phonatory tasks. Although other methods of glottic reconstruction are available, 39 we feel that this technique is simple and results in consistently good voice.

B

i ¸¸

j~/f

FIGURE 11. False fold mobilization and closure of cordectomy defect. HARTIG AND ZEITELS

219

PARTIAL

VERTICAL

LARYNGECTOMY

Using the same laryngofissure approach described for cordectom~ one can address selected T1 and T2 lesions of the glottic larynx. Partial vertical laryngectomy is contraindicated if there is fixation of the vocal fold, as this implies significant involvement of paraglottic musculature or the cricoarytenoid joint. Impaired mobility due to tumor bulk or localized anterior paraglottic involvement is not contraindicated, but should be viewed with caution to be assured that the tumor is not more extensive than suspected. Preoperative radiographic imaging can be very helpful in this regard. This procedure is also contraindicated in tumors which invade thyroid cartilage, extend across the ventricle onto the false fold, or involve the interarytenoid mucosa. TECHNIQUE

In regard to thyroid cartilage resection, the classic procedure of removing all but a thin posterior strut of thyroid lamina cartilage is rarely necessary. In many resections, cartilage resection can be limited to removal of a rectangular block of cartilage anteriorly (Fig 12). Because the vocal fold is closer to the cartilage anteriorly, cartilage resection is more important in this location. If a patient had a tumor which came close to the cartilage posteriorly, h e / s h e would have deep paraglottic involvement and would not be a candidate for a partial vertical laryngectomy procedure. Soft tissue cuts may also be variable. The superior extent of the resection is usually above the false vocal fold and the inferior cut at the subglottis. For tumors which involve the infraglottic surface of the fold, one must be cautious of submucosal involvement down to the level of the cricoid cartilage along the conus elasticus. If this is suspected intraoperatively, a rim of cricoid cartilage should also be taken to maintain oncological safety. The posterior extent of the resection is the variable that has the most significant impact on the person's functional outcome. In patients who lose the entire arytenoid, the likelihood of aspiration is significantly higher. Often those who have a portion of the arytenoid removed do better (Fig 13). Even though the residual arytenoid may become fixed, the bulk provided to approximate the contralateral arytenoid helps complete the protective sphincteric action occurring during swallowing. In regard to voice after partial vertical laryngectomy procedures, the goal should be to provide adequate long term bulk to approximate the contralateral normal cord. A number of reconstructive techniques have been described. Those that fill the void with healthy viable tissue are most successful and are discussed below. Leaving as much thyroid cartilage as safely possible is also important in reconstruction, particularly with muscle flap

FIGURE 12. (A) shows classic cartilage resection while (B) shows a necessary cartilage resection. 220

FIGURE 13. A cross-sectional view of resected tissue during partial vertical laryngectomy.

techniques of reconstruction, as the cartilage keeps the flap in a medial position. The following reconstruction techniques are preferred. Use of epiglottic cartilage was first described by Kambic4° but popularized by TuckerY This technique has the advantage of using local mucosal-covered tissue, and is felt to be best with more extensive vertical partial and fronto-lateral partial laryngectomy procedures. After resection, the base of the epiglottis is pulled inferiorly, and tenotomy scissors are used to dissect preepiglottic fat away from the undersurface of the cartilage. Next, the hyoepiglottic ligament is divided. This is performed without entry into the vallecula. Once mobilized, the epiglottis can be rotated and sewn to the subglottic margin of resection, which will cover the entire defect with healthy mucosalized tissue. The entire epiglottis need not be used; instead, the epiglottis can be divided at the midline and the ipsilateral half alone utilized. Use of 2.0 and 3.0 polyglycolic acid sutures is preferred as the epiglottis is often under tension (Fig 14). Closure of the laryngofissure is then performed. In patients who have had an arytenoid removed, recovery is more protracted and swallowing rehabilitation to avoid aspiration becomes a greater concern. The contralateral mobile arytenoid must extend further medially, and consequently the remaining normal half of the larynx will become hypertrophic in an effort to produce an adequate sphincter. When this happens, the likelihood of supraglottic voicing is increased. When vibration and phonation occur in supraglottic rather than glottic tissues, increased roughness and a lower fundamental frequency result. A second option for reconstruction is the use of a bipedicled strap muscle flap (Fig 15). This option is preferred for lesser anterior resections as it is less disruptive to the supraglottic soft tissues than epiglottic mobilization. This flap consists of the outer thyroid cartilage perichondrium along with the attached strap muscles. The muscles are carefully mobilized along their lateral border to preserve the blood supply to the muscle. If the muscles are devascularized, the amount of atrophy and scar contracture that occurs will increase. Typicall~ one can still plan on some atrophy even with careful dissection, and thereOPTIMIZING VOICE IN GLOTTIC CANCER SURGERY

I

FIGURE 14. Epiglottic advancement/reconstruction.

fore one should overcompensate by placing more muscle than seems necessary within the defect (30-40% extra). The perichondrium now lines the laryngeal defect. Mucosa from the pharynx, lower lip, buccal area, nasal septum, and epiglottis can optionally be used to cover the perichon-

FIGURE 16. Delineation of true fold cut on microlaryngoscopy prior to open fronto-lateral resection.

drium, but can be difficult to secure. When used, they should be placed immediately opposite the contralateral anterior true vocal fold in order to offer the best phonatory advantage. Skin grafts are not recommended.

FRONTO-LATERAL HEMILARYNGECTOMY

I II I II I I I I L I I

L

I

FIGURE 15. Bipedicle strap muscle reconstruction. HARTIG AND ZEITELS

Patients with lesions that cross the anterior commissure to involve one third or less of the contralateral vocal fold are candidates for fronto-lateral procedures (Tlb and T2b glottic lesions). Vocal fold fixation or thyroid cartilage involvement are, again, contraindications. With these procedures, one must take thyroid cartilage as a component of the resection. Broyles ligament 42 is an amalgamation of the vocal ligaments, thyroepiglottic ligament, and the inner perichondrium of the thyroid laminae. Early disease seldomly invades the cartilage unless there is surface exten-

FIGURE 17. Cross-sectional view of resected tissue during fronto-lateral partial laryngectomy.

221

FIGURE 18. Reattachment of true folds anteriorly with T-shaped

keel placement.

sion subglottically or to the infrapetiole region of the supraglottis. 23,24This, in turn, makes a site where no clean plane exists between the anterior commissure and the thyroid cartilage. However, it is not usually necessary to take the thyroid cartilage more than I cm to either side of the midline. Although tumor may extend along the true fold posterior to this point, it will be far enough from the cartilage to allow an adequate margin without taking cartilage. The amount of remaining cartilage becomes important in reconstruction and in the avoidance of laryngeal foreshortening. TECHNIQUE

Exposure of the thyroid cartilage is the same as that in other open vertical procedures. The difficult portion of the resection is in making the first cut from the cricothyroid membrane up through the vocal fold on the less involved side of the larynx. In order to make the cut at the proper position, one can make a marker cut in the leading edge of the vocal fold at a point 5 m m behind the posterior aspect

of the tumor at the beginning of the operation. This can be performed on direct laryngoscop~ creating the incision with a microscissors (Fig 16). This marker can then be visualized from below and followed from the subglottis up through the vocal fold. In this way, the cut on the less involved fold is in the correct location. The contralateral cut will be made under direct visualization, and is easy to make accurately (Fig 17). Reconstruction may be of two types, primary closure or flap closure. For resections in which only a very small amount of vocal fold and anterior thyroid cartilage is removed, primary reapproximation of the vocal ligaments anteriorl~ and reapproximation of the thyroid cartilages to one another, will produce a functional but slightly foreshortened larynx. Depending on the magnitude of cartilage resection, the thyroid lamina can be sutured together primarily (wire or proline), or the larynx is stented with a T shaped keel. This allows the patient to be decannulated and resume an oral diet. The T-shaped keel will prevent webbing at the new anterior commissure (Fig 18). These patients are discharged after several days, and the keel is removed as a separate open outpatient procedure 6-8 weeks later. In situations where a greater amount of the anterior larynx has been removed, primary closure is not possible. The goal becomes the creation of a neoglottic airway which has adequate dimensions for respiration and preserves the sphincteric action of mobile arytenoids. Phonation occurs through approximation of the arytenoids to one another or remaining supraglottic soft tissue. This will produce a hoarse but serviceable voice without significant range of pitch. Reconstruction in patients who have lost significant amounts of anterior soft tissue and cartilage is accomplished with a full epiglottic mobilization, bilateral strap muscle/perichondrial flaps, or both. These tissues are brought over an intralaryngeal stent which is maintained for 6-8 weeks before being removed endoscopically. The stent used must function as an adequate lumen keeper without producing necrosis of the flap or unoperated laryngeal tissues. Molded Silastic (Dow-Corning Co, Midland, MI) laryngeal stents work well (Fig 19).

FIGURE 19. Bilateral strap muscle flaps with intralumenal stent after fronto-lateral resection.

222

OPTIMIZING VOICE IN GLOTTIC CANCER SURGERY

SUMMARY Successful conservation surgery for glottic cancer requires careful interdependent selection of patients, lesions, and procedures. The technical goal is to minimize trauma to uninvolved tissue and to wisely utilize local tissues for reconstruction, while insuring for an oncologically sound procedure. This approach will facilitate maximal functional preservation of the airway, deglutition, and voice in appropriately selected patients. Recent advances in phonosurgical reconstruction should continue to influence the classic glottal cancer procedures over the next decade.

REFERENCES 1. Sands HB: Case of cancer of the larynx, successfully removed by laryngotomy: With an analysis of 50 cases of tumors of the larynx. Treated by operation. N Y Med J 1:110,1865 2. Billroth T: Quoted by C. Gussenbauer: Ueber die erste durch T. Billroth an Menschen ausgeh.thrte kehlkopf-exstirpation, und die anwendung eines kunstlichen kehlopfes. Verh Otsch Ges Chir, 1874 3. Kirshner JA: Growth and spread of laryngeal cancer as related to partial laryngectomy, in Alberti PW, Bryce DP, (eds): Workshops from the Centennial Conference on Laryngeal Cancer. New York, AppletonCentury-Crofts, 1976, p 54 4. Dandiilidis J, et al: Vertical partial laryngectomy our results after treating 81 cases of T2 and T3 laryngeal carcinomas. J Laryngol Oto] 106:349, 1992 5. Olsen KD, Desanto LW: Partial vertical laryngectomy: Indications and surgical technique. Am J Otolaryngo111:153, 1990 6. Zeitels SM: Premalignant epithelium and microinvasive cancer of the vocal fold: The evolution of phonomicrosurgical management. Laryngoscope 67:1-51, 1995 (suppl) 7. Zeitels SM: Phonomicrosurgical treatment of early glottic cancer and carcinoma in situ. Am J Surg 172:704-709, 1996 8. Zeitels SM: Endoscopic management of the larynx for benign and malignant disease, in Myers E, Bluestone, CD (eds): Advances in Otolaryngology--Head and Neck Surgery. Chicago, Mosby, 1998, pp 1-40 9. Zeitels SM: Microflap excisional biopsy for atypia and microinvasive glottic cancer. Op Tech Otolaryngol Head Neck Surg 4:218-222, 1993 10. Isshiki N: Phonosurgery: Theory and Practice. Tokyo, Japan, SpringerVerlag, 1989 11. Myers EN, Wagner RL, Johnson JT: Microlaryngoscopic surgery for T1 glottic lesions: A cost effective option. Ann Otol Rhinol Laryngol 103:28-30, 1993 12. Cragle SP, Brandenburg JH: Laser cordectomy or radiotherapy: Cure rates, communication, and cost. Otolaryngol Head Neck Surg 108:648653, 1993 13. Zeitels SM, Vaughan CW: A submucosal vocal fold infusion needle. Otolaryngol Head Neck Surg 105:478-479, 1991 14. Vaughan CW, Strong MS, Jako GJ: Laryngeal carcinoma: Transoral treatment using the CO2 laser. Am J Surg 136:490-493, 1978 15. Koufman JA: The endoscopic management of early squamous carcinoma of the vocal cord with the carbon dioxide surgical laser: Clinical experience and a proposed subclassification. Otolaryngo! Head Neck Surg 95:531-537, 1986 16. Eckel H, Thumfart WF: Laser surgery for the treatment of larynx carcinomas: Indications, techniques, and preliminary results. Ann Otol Rhinol Laryngo1101:113-118, 1992 17. Thomas J, Olsen KD, Neel HB, et al: Recurrences after endoscopic management of early (T1) glottic cancer. Laryngoscope 104:1099-1104, 1994 18. Olsen K, Thomas JV, DeSanto LW, Suman VJ: Indications and results of cordectomy for early glottic carcinoma. Otolaryngol Head Neck Surg 108:277-282, 1993

HARTIG AND ZEITELS

19. Davis RK, Jako GJ, Hyams VJ, Shapshay SM: The anatomic limitations of CO2 laser cordectomy. Laryngoscope 92:980-984, 1982 20. Lehman JJ, Bless DM, Brandenburg JH: An objective assessment of voice production after radiation therapy for stage I squamous cell carcinoma of the glottis. Otolaryngol Head Neck Surg 98:121-129, 1988 21. Hellquist H, Lundgren J, Olofsson J: Hyperplasia, keratosis, dysplasia and carcinoma in situ of the vocal cords--A follow-up study. Clin Otolaryngol 7:11-27, 1982 22. DeSanto LW: Selection of treatment for in situ and early invasive carcinoma of the glottis, in PW Alberti and DB Bryce (eds): Workshops from the Centennial Conference on Laryngeal Cancer. New York, Appleton-Century-Crofts, 1976, pp 146-150 23. Kirchner JA, Carter D: Intralaryngeal barriers to the spread of cancer. Acta Otolaryngologica (Stockholm) 103:503-513, 1987 24. Kirchner JA: What have whole organ sections contributed to the treatment of laryngeal cancer? Ann Otol Rhinol Laryngol 98:661-667, 1989 25. Wolfensberger M, Dort JC: Endoscopic laser surgery for early glottic carcinoma: A clinical and experimental study. Laryngoscope 100:11001105, 1990 26. Wetmore SJ, Key M, Suen JY: Laser therapy for T1 glottic carcinoma of the larynx. Arch Otolaryngol Head Neck Surg 112:853-855, 1986 27. Krespi Y, Meltzer CJ: Laser surgery for vocal cord carcinoma involving the anterior commissure. Ann Otol Rhinol Laryngol 98:105-109, 1989 28. Casiano R, Cooper JD, Lundy DS, Chandler JR: Laser cordectomy for T1 glottic carcinoma: A 10-year experience and videostroboscopic findings. Otolaryngol Head Neck Surg 104:831-837, 1991 29. Benninger M, Gillen J, Thieme P, et al: Factors associated with recurrence and voice quality following radiation therapy for T1 and T2 glottic carcinomas. Laryngoscope 104:294-298, 1994 30. Mendenhall W, Parsons JT, Stringer SP, Cassissi NJ: Management of Tis, T1, T2 squamous cell carcinoma of the glottic larynx. Am J Otolaryngo115:250-257, 1994 31. Steiner W: Results of curative laser microsurgery of laryngeal carcinomas. Am J Otolaryngo| 14:116-121, 1993 32. Rudert HH: Transoral CO2-/aser surgery of early glottic cancer (CIS-T2), in Smee R, Bridger P (eds): Laryngeal Cancer: Proceedings of the 2nd World Congress on Laryngeal Cancer. Amsterdam, Netherlands, Elsevier, 1994, pp 389-392 33. Zeitels SM: lnfrapetiole exploration of the supraglottis for exposure of the anterior glottal commissure. J Voice 12:117-122, 1998 34. Zeitels SM: Laser versus cold instruments for microlaryngoscopic surgery. Laryngoscope 106:545-552, 1996 35. Hirano M, Hirade Y, Kawasaki H: Vocal function following carbon dioxide laser surgery for glottic carcinoma. Ann Otol Rhinol Laryngol 94:232-235, 1985 36. McGuirt WF, BlaIock D, Koufman JA, Feebs RS: Voice analysis of patients with endoscopically treated early laryngeal carcinoma. Ann Otol Rhinol Laryngol 101:142-146, 1992 37. Zeitels SM: Transoral treatment of early glottic cancer, in Smee R, Bridger P (eds): Laryngeal Cancer: Proceedings of the 2nd World Congress on Laryngeal Cancer. Amsterdam, Elsevier, 1994, pp 373383 38. Brasnu D, Laccourreye O, Weinstein G, et al: False vocal cord reconstruction of the glottis following vertical partial laryngectomy: A preliminary analysis. Laryngoscope 102:717-719, 1992 39. Milutinovic Z: Composite myomucosal reconstruction of the vocal fold. Eur Arch Otorhinolaryngo1252:119-122, 1995 40. Kambic V: Epiglottoplast--New technique for laryngeal reconstruction. Radiologia Yugoslavia (suppl II):33-43, 1977 41. Tucker HM, Wood GB, Levine H, et al: Glottic reconstruction after near total laryngectomy. Laryngoscope 89:609-618, 1979 42. Broyles EN: The anterior commissure tendon. Ann Otol Rhinol Laryngo152:342-345, 1943

223