Introduction

For years, head and neck surgery was synonymous with radical “mutilating” procedures. Even with reconstruction using microvascular free-tissue transfer, functional organ preservation could not be guaranteed and, often, patients still required postoperative radiation with or without chemotherapy. While there were “conservation” or organ-preservation procedures, transcervical or transoral, operative indications were often too narrow to be of great clinical impact. As such, multidisciplinary care for head and neck cancers gradually shifted predominantly toward a radiation-based approach.

However, over the past 30 years, “endoscopic” head and neck surgery (eHNS) has evolved, incorporating transoral minimally invasive techniques, with transoral laser CO2 microsurgery (TLM) and now transoral robotic surgery (TORS). For cancers of the upper aerodigestive tract, eHNS has naturally developed around transoral exposure. For the neck, eHNS techniques have been developed that allow excision of parathyroid and thyroid tumors either via video-assisted small cervical incisions or through small inconspicuous incisions located some distance from the primary tumor site. Several surgeons have now proposed endoscopic lymphadenectomy using variations on these techniques.

Here we argue that eHNS offers great promise for patients, providing the surgeon with an oncologically sound approach with good functional outcomes, and thus a legitimate surgical option for “organ preservation.” Further, we propose that eHNS should play a larger role in the multidisciplinary care of patients with head and neck cancer (HNC). National Institutes of Health (NIH)–funded cooperative group trials must now be performed to study eHNS in a prospective fashion.

Endoscopic Laser Surgery and TLM

First devised by Strong and Jako in 1972 [1] and developed by Steiner [2] and others, endoscopic laser surgery (ELS) and TLM have emerged as a standard of care for early laryngeal cancer and also an attractive treatment alternative for select tumors of the oropharynx, supraglottis, and hypopharynx [3].

ELS has developed along traditional Halsteadian oncologic principles, with excision of the tumor in a circumferential, en bloc manner [4-6]. In contrast, during TLM, the surgeon may choose to divide the tumor under the operating microscope to assess the extent and depth of invasion and thus facilitate a more “logical” minimal resection [3]. By conserving the maximum amount of normal mucosa, crucial structures are preserved and function is optimized. Both methods have their proponents, but what is common to each is an approach that avoids the often debilitating sequelae that can follow, on occasion, radiation, chemoradiotherapy, or open surgery.

Laser Technology and Equipment

Fundamental to the continued benefit of eHNS are new technologies that allow exceptional visualization and permit precise tissue interaction through narrow or restrictive access. For certain patients and particular areas of the laryngopharynx, direct line-of-sight visibility may at times be impossible, and with it, line-of-sight cutting instruments such as the free-beam laser are of limited use. With its power and absorption characteristics, the carbon dioxide (CO2) laser remains among the most useful cutting devices for operation in the aerodigestive tract. Its long wavelength (10,600 nm) and physical properties have until recently mandated deploying the laser beam in a direct path from a microscope-mounted mirror manipulator to the end organ. This “line-of-site” restriction limits the “workspace” of the laser within which the surgeon can work and proves problematic as tumor size increases or patient anatomy restricts line-of-sight delivery. The Omni Guide System (Omni Guide Inc., Cambridge, MA) has been developed to deliver a CO2 laser via a flexible hollow core fiber [7]. The system allows delivery of CO2 laser energy to areas of the head and neck in which the direct visualization required for conventional linear CO2 laser systems cannot be acquired. The flexibility of the fiber delivery system permits handheld operation, or alternatively, the fiber can be mounted to a specialized robotic manipulator to facilitate cutting at increased angles, thus increasing the potential workspace at the tumor patient.

Other fiber-based laser technologies are beginning to find uses for the treatment of HNC. Thulium-ion based continuous-wave lasers have similar properties to CO2 lasers and demonstrate promise for benign laryngeal and tracheal disease, as well as for cancer [8]. The 2,013 nm wave length of this laser allows its energy to be delivered by a small caliber glass fiber, yet it retains water as a chromophore. This laser therefore offers a smooth vaporization pattern and excellent hemostasis, similar to that realized with the potassium titanyl phosphate laser, but with less risk of deep tissue penetration.

In addition, Remacle et al. [9••] recently introduced the scanning robotic micromanipulator for use with the carbon dioxide laser. This system provides the surgeon with three different cutting shapes besides the simple spot and also allows the laser surgeon to change the size (from 0.4 mm to 4 mm) as well as the depth of penetration. Using this “digital acublade” on varying pulsed modes, Remacle et al. [9••] showed significant reduction in thermal injury zone to 12.5 microns and increased operative precision, when compared to the traditional “dot-by-dot” micromanipulator approach.

The acquisition of a variety of specialized endoscopes and endoscopic instruments is an important consideration in developing an ELS or TLM practice; these include a selection of graspers, manipulators, diathermy, and suction devices and endoscopic haemostatic clip applicators (Fig. 1, top).

Fig. 1
figure 1

Top, The equipment required for transoral laser microsurgery (TLM). Bottom, The postoperative result following TLM for the treatment of a T1 verrucous carcinoma of the right glottic larynx at 18 months. No radiation therapy was given

Full size image

eHNS for Early Laryngeal Cancer

In 2002, a review by the Cochrane Collaboration was performed examining the evidence base for the treatment of laryngeal cancer with conventional open laryngeal surgery versus radiotherapy versus endoscopic surgery (with or without laser) in T1 and T2 laryngeal cancer [10]. Only one randomized controlled trial was found investigating the relationship of surgery versus radiotherapy for early glottic cancer [11]. The multicenter randomized controlled trial, published in 1990 from Eastern Europe, recruited patients to open surgery, radiotherapy, or a combination of radiotherapy and chemotherapy but crucially did not include endolaryngeal surgery. The Cochrane review concluded that the methodology of the study was insufficient to draw conclusions.

So, to answer the question of optimal treatment for early laryngeal cancer, two large multicenter randomized controlled trials comparing radiotherapy to ELS and TLM for early glottic cancer were proposed by Coman in Brisbane, Australia and by Birchall in Bristol, United Kingdom. Unfortunately, as a result of difficulty in patient recruitment and questions regarding treatment equipoise, both trials were not completed [12, 13].

Without level I randomized data, one must turn to the available evidence-based data reported in the literature. A search of Medline or PubMed will reveal several hundred case series studies and a handful of nonrandomized comparative cohort studies reporting the oncologic effectiveness of TLM or endoscopic laser surgery in the treatment of laryngeal cancer. Local control rates for TLM/ELS in the treatment of early laryngeal cancer have been reported at between 85% and 96% for T1 tumors, and 66% and 100% for T2 lesions (Fig. 1, bottom) [14••, 15-18].

eHNS for Intermediate-Stage to Advanced-Stage Laryngeal Cancer

Given the collective inability of nonsurgical therapeutic modalities to make a significant impact on overall survival in advanced HNC, a strong emphasis on functional results as an indicator of treatment success has evolved. In some centers in Europe and North America, transoral laser microsurgery has been advocated as an effective organ-sparing approach for select advanced laryngeal cancer. A combined multicenter study of 117 patients undergoing TLM with or without adjuvant radiotherapy for stage III and IV laryngeal cancer demonstrated 2-year and 5-year local disease control rates of 82% and 74%, respectively [19]. Only 15/117 patients (13%) in the cohort required adjuvant radiotherapy to the primary tumor site. In the subgroup of patients treated in North America, the average length of stay in hospital after surgery was only 5.7 days (median 5 days, range 1–16 days). In terms of functional outcomes, 68/117 patients were alive with no evidence of disease recurrence and a preserved larynx at their last follow-up. As a primary surgical approach, TLM achieved a 2-year actual laryngeal preservation rate of 92% (83/90).Two of 68 patients (3%) were tracheotomy dependent, and five of 68 patients (7%) were feeding tube dependant. It was concluded that TLM offers acceptable oncologic and functional results, with most patients receiving single-modality therapy with the added benefit of shortened periods of hospitalization.

eHNS and Oropharyngeal Cancer

The role of eHNS techniques in the management of oropharyngeal squamous cell carcinoma (OPSCC) is perhaps more controversial. Since many institutions rely on primary radiotherapy with or without chemotherapy for OPSCC and achieve acceptable locoregional control rates, critics suggest that the rationale for using surgery for this disease is unclear. In a study of 69 patients with select T1-T3, N0-N2 OPSCC treated with TLM surgery alone without adjuvant radiotherapy to the primary tumor or neck, local control was achieved in 66/69 patients [20]. Furthermore, locoregional control for all patients was 84%, and all patients who developed regional recurrence were successfully salvaged with surgery or radiotherapy. At the Washington University, Haughey and his team, using both surgery and radiation, demonstrated an overall survival rate of 88% for 84 patients with stage III or IV disease. Surgical complications were not common, and most patients (81%) had acceptable swallowing function, with only 3.4% of patients with gastrostomy [21].

These TLM experiences suggest a greater role for surgery first, not only to render local and regional disease control but to more accurately identify the requirement for and extent of any adjuvant therapy. TLM has considerable advantages over radiotherapy or concurrent chemoradiotherapy, including lower morbidity, short duration of treatment, and patient acceptability. TLM may be preferable to patients not only because of its low morbidity and mortality, but also because they can return quickly to their normal routine. Perhaps for properly selected patients, radiotherapy or eHNS are equally effective. Yet, without a national cooperative group to study surgery, a randomized prospective trial seems unlikely. Nonetheless, as head and neck surgeons, we urge that such a prospective trial be performed through the existing NIH-sponsored intergroup framework.

In the meantime, we argue that it is the changing epidemiology of oropharyngeal cancer that will lead to a greater role for eHNS in the management of this disease. Several reports from the United States and abroad have documented the precipitous rise in the incidence of OPSCC in individuals younger than 40 years [22••, 23••]. Molecular epidemiologic evidence suggests an association between human papillomavirus (HPV) and OPSCC [22••]. Different reports have demonstrated the presence of HPV DNA in at least 50% of OPSCCs. Moreover, patients who are seropositive for HPV subtype 16 have a sevenfold increased risk of developing a squamous cell cancer positive for HPV 16 DNA as compared to their seronegative counterparts [24]. Chaturvedi et al. [23••] and others have attributed the increase in incidence of HPV-related oral and OPSCCs in the United States between 1973 and 2004 to changing sexual habits. Thus, HPV infection, transmitted sexually, likely accounts for the rising incidence of oropharyngeal cancers in younger patients. A genetic susceptibility to a latent chronic HPV infection is probably necessary for the development of HPV+ HNC—though no genetic marker (such as a single nucleotide polymorphism) has yet been identified; the International Head and Neck Cancer Epidemiology consortium has identified this as an important research objective.

The good news is that HPV-positive tumors may respond better to radiation with or without chemotherapy [25]. While tobacco use remains an important covariate, an HPV-associated oropharyngeal cancer may have less genomic instability and be more homogenous at the molecular level. As these issues are addressed in the laboratory and in cooperative group trials, we believe that now is the time to revisit our basic assumptions about a radiation-based approach. If HPV-associated oropharyngeal cancers have a more favorable outcome than non-HPV associated disease, shouldn’t new options in eHNS also be evaluated prospectively?

In fact, one compelling argument is that, given the potential long-term sequelae [26] of radiation for a younger population [27•, 28], eHNS must play a greater role in the management of patients with oropharyngeal cancer.

Robotic Surgery: An Emerging New Aspect of eHNS

For endoscopic neck surgery, natural access through both the mouth and axilla provides ample opportunity for innovation and improvement of current endoscopic and minimally invasive techniques. There are several exciting and novel applications that have emerged for eHNS over the last decade—many of them relying upon transoral and transaxillary robotics.

Czech writer Josef Čapek introduced the term “robot” to his brother Karel Čapek, a fellow playwright who used the expression to describe several characters in his stage play R.U.R. (Rossum’s Universal Robots) [29]. While examination of the origins of the term “robot” is relatively straightforward, a precise technical definition of what a robot actually is remains somewhat more difficult to describe. In the field of medicine, the Society of American Gastrointestinal and Endoscopic Surgeons and the Minimally Invasive Robotic Association produced a consensus document in 2007 on robotic surgery. The authors defined “robotic surgery” as “a surgical procedure or technology that adds a computer technology-enhanced device to the interaction between a surgeon and a patient during a surgical operation and assumes some degree of control heretofore completely reserved for the surgeon” [30].

eHNS and Robotic Technology

The daVinci Surgical System (Intuitive Surgical Inc., Sunnyvale, CA) is currently the most popular commercial platform for robotic surgery. Widely used for minimally invasive surgery in cardiac, abdominopelvic, and urological procedures, daVinci is, in fact, a robotic system in three parts. At the operating room table, surgical instruments are deployed from a “patient-side” cart and are safely placed by the surgeon within the patient’s body under direct and/or endoscopic guidance. A binocular 12-mm endoscope with dual zero or thirty degree optics is then introduced to visualize the target anatomy. The three-dimensional surgical anatomy is then recreated using the “vision” cart, where computer processing links the image and the spatial relationships of the instruments in a virtual environment. The surgeon, sitting at a remote console, can then operate within this virtual three-dimensional environment using “master controllers” that control and direct movements of the robotic instruments. The surgeon has an unprecedented perspective on the surgical anatomy but also the ability to operate on it with 540 degrees of wristed instrumentation and at either zero-degree or thirty-degree angles. Furthermore, motion scaling increases precision and reduces the larger hand movements required by humans, while eliminating tremor and fatigue.

Beginning in 2005, Hockstein et al. [31, 32], Weinstein et al. [33••, 34], and O’Malley et al. [35] presented a landmark series of papers, methodically defining the field of TORS. Whereas McLeod and Melder [36] placed the robotic arms through a slotted laryngoscope, in TORS three of the four daVinci robotic arms are placed through the mouth using standard retractors such as the Fehr-Kastenbauer, Crowe-Davis, or Dingman—even for supraglottic and rare selected laryngeal tumors. A12-mm binocular camera is introduced into the oral cavity/pharynx followed by two other arms carrying interchangeable 5-mm wide working instruments. Interestingly, the TORS setup still relies heavily on the presence of a human assistant to sit at the patient’s side and provide assistance with precisely targeted suction and countertraction. A variety of manipulators and dissectors can be utilized during surgery. The cutting device is usually monopolar electrocautery. The recent adaptation of robot manipulators for the deployment of laser cutting technology has proved a popular addition to the eHNS armamentarium. While monopolar diathermy is an effective instrument, some eHNS operators prefer the precision, nominal thermal damage, and reduced tissues necrosis offered by fiber laser technologies such as the carbon dioxide and thulium laser [37, 38].

The body of literature supporting the evidence base for robotic head and neck surgery remains in its infancy. The first reported experiences of adapting the head and neck applications were proof-of-principle studies in mannequin, animal, and cadaveric models using the daVinci surgical robot [31, 32, 35, 39, 40]. The first published robotic procedure on a human was excision of a simple vallecular cyst in a 46-year-old woman in 2003 [36]. This has been followed by translation of the TORS techniques practiced in the laboratory setting to experience in human subjects. Investigators from the University of Pennsylvania have led the way in the initial development of TORS methods and reported success in the treatment of supraglottic and tongue-base tumors [33••, 34, 35]. In December 2009, TORS was approved by the US Food and Drug Administration (FDA) for use in benign and selected malignant tumors of the head and neck [27•].

eHNS for the Neck? Thyroid, Parathyroid, and Cervical Lymphatics

Minimally invasive transcervical surgery can trace its origins to the first endoscopic parathyroidectomy performed in 1996 [41]. This innovation saw the use of 5-mm laparoscopic instruments placed between the platysma and strap muscles in a working space maintained between constant C02 insufflation. Following these initial experiences, a variety of minimally invasive video-assisted thyroidectomy (MIVAT) [42, 43••] approaches have evolved. With these techniques the thyroid gland is removed with or without gas insufflation via cervical, anterior chest wall, and transaxillary approaches. Several studies have emerged from Asia reporting endoscopic transaxillary approaches to the thyroid [44•, 45, 46]. Chang et al. [47] recently reported on their combined endoscopic/robotic transaxillary approach as well. More recently, Kang et al. [48••] from Seoul, Korea, reported on 200 patients that have undergone transaxillary robotic thyroidectomy (Fig. 2) incorporating the daVinci Surgical System.

Fig. 2
figure 2

Top, The transaxillary approach for robotic thyroidectomy. Bottom, Options for the placement of incisions and robotic instruments. (Used with permission from Elsevier, from The Atlas of Head and Neck Surgery, edited by Cohen and Clayman, 2010)

Full size image

This convergence of techniques in minimally invasive thyroid surgery and the robotic platform of the daVinci Surgical System may provide several distinct advantages and unique opportunities for the endoscopic head and neck surgeon. The three-dimensional environment created with 30-degree capable endoscopy improves operative field visualization, and 540 degrees enabled wristed instrumentation facilitates operative dexterity. The conventional open thyroidectomy approach, however, remains safe, effective, and time-honored, and for this reason, many remain skeptical of robot-assisted techniques.

As the radical neck dissection has progressively been replaced by functional and selective neck dissection without compromising oncologic outcomes, the extent and morbidity of neck dissection has steadily diminished. Authors have proposed minimally invasive endo-robotic and/or endoscopic neck surgery in human cadaver models [49]. But Werner et al. [50] were the first to demonstrate successful management of cervical lymphadenopathy via endoscopic approach for squamous carcinoma in the setting of sentinel lymph node biopsy. Chung and colleagues have already demonstrated the feasibility of robotic neck dissection for well-differentiated thyroid cancer (Chung, Personal communication).

Conclusions

The emergence of eHNS represents the natural evolution for the discipline of head and neck surgery. The burgeoning field of endoscopic head and neck surgery now includes TLM, TORS, as well as video-assisted and robotic surgery of the neck and thyroid. For the glottic larynx, transoral laser microsurgery remains the gold standard. For supraglottic laryngeal, oropharyngeal, and hypopharyngeal tumors, as well as selected other lesions, both TLM and TORS can be used. In the neck, minimally invasive video-assisted techniques are now an important part of thyroid and parathyroid surgery. More recently, robotic thyroidectomy has emerged and represents not only a revolutionary new approach to the thyroid, but also a stepping stone to more complex operations of the neck—neck dissection or even tracheoesophageal and/or laryngeal framework surgery. In the next 5 years, with robotic surgery and laser technology as a common platform, we foresee the development and widespread use of several new and innovative procedures for head and neck surgery, via transoral and transaxillary approaches.