Robotic Conservation Surgery of the Larynx
Gregory S. Weinstein
INTRODUCTION
The role of transoral robotic surgery (TORS) for laryngeal cancer, at the University of Pennsylvania, is essentially limited to TORS-supraglottic partial laryngectomy (SGPL). The rationale for rarely using TORS for glottic cancer is that the approaches that existed for glottis cancer prior to the introduction of TORS were effective, efficient, and safe from the functional and oncologic perspective. In general, when a new medical technology or treatment approach is introduced, adoption by clinicians requires that the new approach solves the problems with the existing therapies.
In my opinion, the standard transoral and open organ preservation surgery techniques as well as the nonsurgical approaches for glottic cancer, in selected patients, work well. At our institution, a few cases of glottic carcinoma have been operated upon using TORS. In general, this has not been found to be superior to our standard approaches. In addition, although there have been reports of robotic total laryngectomy elsewhere, these have not been yet performed at the University of Pennsylvania and therefore I am not in a position to evaluate this transoral approach in comparison to the standard open approach.
SGPL was originally described as an open procedure, which was associated with tracheostomy and a prolonged period of rehabilitation. In my opinion, although transoral laser microsurgery (TLM)-SGPL was a major improvement over open SGPL in terms of morbidity, the difficulties due to line of site exposure as well as challenges associated with surgeon’s acceptance of multiple transtumoral cuts through cancer have resulted in limited adoption of this technique. In addition, the decrease in incidence of patients with supraglottic cancer who are candidates for SGPL has also contributed to fewer surgeons being trained in a technique that has a particularly long learning curve.
Given these issues, our team developed TORS-SGPL in the canine model in 2005 and in 2007 was the first to report our new procedure in humans. I have found a significant decrease in operative time and improvement in laryngeal exposure using TORS-SGPL when compared to TLM-SGPL, which I attribute to the use of the robotic technology. I have also found that the most significant benefit of TORS over TLM-SGPL is the ability to have the two hands of the assistant help the surgeon during the case. This chapter is limited to the technique of TORS-SGPL. Extended versions of the standard SGPL procedure have been described previously, and in the case of the SGPL, these would include (1) resection of the base of the tongue, (2) unilateral arytenoid resection, (3) unilateral pyriform sinus resection, and (4) unilateral or bilateral vocal cord resection. These extended procedures are not commonly performed at the University of Pennsylvania and are not included in this discussion.
HISTORY
The most common presenting symptoms in patients with supraglottic carcinoma are hoarseness, a muffled voice, otalgia, pain in the throat, or a mass in the neck. With rare exception, the more extensive lesions, which have severe obstructive symptoms such as shortness of breath, are already so large that they usually have spread to anatomic areas which preclude performing any type of SGPL.
PHYSICAL EXAMINATION
The physical examination of the primary site in the outpatient setting includes indirect or direct laryngoscopy as well as a general examination of the head and neck. Examination of the neck is important since unresectable metastasis to the neck precludes operating on the primary and bilateral extensive metastasis to the necks could result in resecting both internal jugular veins which would both also be a contraindication to TORS-SPGL.
When evaluating the larynx, the surgeon should examine the surface extent of the primary cancer and the mobility of both the vocal cords and arytenoids, as this will reflect the surface as well as the three-dimensional extent of the cancer. There are specific areas of surface extent that are important to evaluate since resection of these areas may impact on short- and long-term functional outcomes and thus impact on resectability using SGPL (see Contraindications section below).
The surfaces to evaluate include the (1) vallecula, (2) pyriform sinus, (3) mucosa of the base of the tongue, (4) arytenoid(s), (5) ventricle, (6) anterior commissure, and (7) true vocal cords. Mobility of the arytenoids and vocal cords has been correlated with depth of invasion of certain laryngeal structures.
Accurate assessment of the mobility of the vocal cords and arytenoids provides the clinician the following information: (1) a fixed vocal cord or one with impaired mobility is an indication of malignant involvement at the glottic level, (2) fixation of the arytenoid indicates lateral cricoarytenoid muscle or cricoarytenoid joint involvement, and (3) pseudofixation of the arytenoid occurs when the bulk of the tumor involves the portion of the arytenoid adjacent to the aryepiglottic fold or false cord. The clinical finding of pseudofixation of the arytenoid is seen when a cursory look from above reveals that the arytenoid is not moving but when the scope is positioned to look past the tumor at the vocal cord itself the examiner perceives mobility. Pseudofixation indicates that neither the thyroarytenoid muscle nor the cricoarytenoid joint is involved by the cancer and, barring other contraindications, does not preclude SGPL.
INDICATIONS
The T staging system provides useful standardized descriptions of malignancies, which allow the clinician to prognosticate for an individual patient or compare the potential outcomes between treatments. In the case of da Vinci transoral surgery, the Food and Drug Administration has cleared all T1 and T2 lesions. T stage indications published in the surgical literature for TORS-SGPL include selected T1, T2, and T3 cancers of the supraglottis. However, selection criteria based on the T staging system are not adequate for “en bloc” surgical triage because the T staging system divides the cancer into four categories (T1-T4), most often based on the superficial and deep extent of the cancer although sometimes not, which are too broad and general to constitute specific criteria for a particular “en bloc” operation. I emphasize the words “en bloc” here because there are treatments such as radiation therapy (whole organ treatment) or TLM (which uses piecemeal resection), in which the T staging system is used for treatment triage. In the treatments in which “en bloc” resection is not used typically, any lesion below a maximal T stage is considered “treatable,” and the radiation or surgical margins are defined by the extent of the cancer, not by predefined standardized resection margins. Although there are likely as many ways to perform the precise steps of an “en bloc” surgical resection as there are qualified surgeons, there is a general consensus among practicing surgeons as to what constitutes a standardized resection for a particular “en bloc” procedure. This is true of “en bloc” procedures such as total laryngectomy, supracricoid partial laryngectomy, thyroid lobectomy, radical maxillectomy, parotidectomy, and TORS-SGPL. Simply put, specific indications for a particular “en bloc” resection occurs when the surgeon predicts that the malignancy in question, in both its’ superficial and deep extent, fits within the standardized margins of resection of a particular procedure. The standardized margins of resection of TORS-SGPL include the entire epiglottis and preepiglottic space as well as both false cords.
In my opinion, the surgeon must be able to visualize in the mind’s eye the three-dimensional defect of the particular operation and then, using clinical assessment in the outpatient setting, imaging studies, and intraoperative endoscopy, needs to predict that negative margins will be achieved if a particular cancer of the supraglottis is resected with the standard TORS-SGPL. Hence, although a T2 supraglottic cancer might be a candidate for TORS-SGPL, if the lesion involves the glottis, medial wall of the pyriform, or the posterior aspect of the arytenoid, it would not fit within the standardized margins of resection of the TORS-SGPL and hence the lesion would not be amenable to TORS-SGPL. On the other hand, if a T3 supraglottic cancer involved the anterior false cord and the superior petiole of the epiglottis with radiologic evidence of minimal preepiglottic space invasion, the surgeon may expect that when the standard TORS-SGPL is performed, the full extent of the cancer will fall within the three-dimensional extent of the resection and therefore it is reasonable to proceed. Since, of course, the number of combinations of both superficial and deep extent of supraglottic cancers that might be amenable to TORS-SGPL is innumerable, it would be impossible to list them all here. (This is true of specific indications for all en bloc resections, not just TORS-SGPL.)
A skeptic might argue that nonobjective selection criteria such as described herein are as much art as science and indeed are largely dependent on the knowledge and skill set of the individual surgeon. While I would in large part agree with this statement, I would counter that while nonsurgical triage may appear more objective because it is dependent on T staging, if clinicians lack anatomical knowledge of the larynx or the
clinical acumen to precisely define the superficial and deep extent of the cancer, they would likely not be able to adequately T stage the lesion. In my opinion, the very different approach to patient selection for “en bloc” surgical resection versus nonsurgical approaches such as radiation or chemoradiation is one of the major impediments to setting up a randomized trial for comparison of treatments.
clinical acumen to precisely define the superficial and deep extent of the cancer, they would likely not be able to adequately T stage the lesion. In my opinion, the very different approach to patient selection for “en bloc” surgical resection versus nonsurgical approaches such as radiation or chemoradiation is one of the major impediments to setting up a randomized trial for comparison of treatments.
