Keywords

Introduction

Endoscopic endonasal, transoral, and transcervical approaches have recently been developed as promising alternatives to traditional transoral microsurgery for the craniovertebral junction (CVJ), and these approaches may become more mainstream as experience with them increases (drawbacks of these newer approaches are a steeper learning curve and loss of three-dimensional visualization) [40, 41].

The transoral-transpharyngeal approach historically remains the “gold standard” for anterior approaches to the upper cervical spine when indicated according to the Menezes algorithm [15]. However, there are still technical difficulties with the operating microscope, such as the need to see and work through a narrow opening in a deep cavity and the need to improve visualization; soft palate splitting and even hard palate resection along with extended maxillotomy are occasionally required. To overcome such complications, endoscopic-assisted procedures have been developed for CVJ decompression. The endoscopic approaches to the CVJ include the endoscopic endonasal approach, the endoscopic transoral approach, the robot-assisted endoscopic transoral approach, combined endoscopic transnasal and transoral approaches, and the endoscopic transcervical approach [18, 22]. The aim of the present review is to provide an update on the anatomical fundamentals of endoscopic-assisted surgery for the CVJ and to report on the available clinical results.

Anatomical Studies of Endoscopic Craniovertebral Approaches

At present, the most commonly used endoscopically assisted approaches to the CVJ include the transnasal, transoral, and transcervical routes (Table 1).

Table 1 Major findings in anatomical studies of endoscopic-assisted approaches to the craniovertebral junction (CVJ)
Full size table

Endoscopic Transoral Approach

In 2004, de Divitiis et al. studied an endoscopic transoral-transclival intradural approach on 15 cadavers, without maxillotomy or mandibulotomy, and estimated a safe entry zone achieved endoscopically through the clivus [10].

In 2006, Balasingam et al. conducted a cadaveric anatomical study to assess the area of surgical exposure and the available liberty of action for instrument manipulation by four different surgical approaches to the extracranial periclival region: the traditional transoral route, transoral with a palate split, LeFort I osteotomy, and median labioglossomandibulotomy [4].

In 2009, Pillai et al. performed an odontoidectomy in nine specimens by a direct transoral approach; endoscope-assisted (five cases) or a combined endoscopic-microscopic procedure, evaluating the surgical working area and the surgical freedom; the authors concluded that the endoscope and image guidance allowed them to approach the ventral CVJ transorally with minimal tissue dissection, no palatal splitting, and no compromise of surgical freedom [31].

Endoscopic Endonasal Approach

The main advantages of the endoscopic endonasal approach to the ventral CVJ are minimal invasiveness, unlimited surgical access to the rostral midline CVJ, avoidance of palatal split, and less operative morbidity overall compared with the transoral approach. Thanks to a relatively inclined surgical trajectory, in a rostral-to-caudal direction, the compressive pathology of basilar invagination, including the lower clivus and odontoid tip, may be removable without removing the C1 anterior arch, thus maintaining the stability of C1–C2 [2]. In 2009, Kassam’s team published the concept of the “nasopalatine line” (NPL) [12]. The NPL is a reliable predictor of the maximal length of inferior dissection, and odontoid surgery can be performed safely according to a preoperative radiological study of the potential anatomical limitations of the endonasal approach. In 2012 Aldana et al. proved that a line in the midsagittal plane, the nasoaxial line (NAxL), connecting the midpoint of the distance from the rhinion to the anterior nasal spine of the maxillary bone and the C2 vertebra, tangential to the posterior nasal spine of the palatine bone, accurately predicted the lowest limit of this approach on the cervical spine [1].

Endoscopic Transcervical Approach

In 2011, Russo et al. [35] described the microsurgical anatomy and limits of exposure of the endoscopically assisted high anterior cervical, submandibular approach to the clivus and foramen magnum; the optimal route to access pathologies located ventral to the pontomedullary region. Two extensions of the approach were studied and described: an extended anterior far-lateral clivectomy and an inferior petrosectomy, thus extending the exposure to the anterior foramen magnum and the anterior cerebellopontine region.

Comparison Studies

In a study on nine cadaver heads, in 2009, Baird et al. assessed surgical access to the craniovertebral junction using three endoscopic approaches: endonasal, transoral, and transcervical. Data suggested that the surgical goals of lower clival and odontoid decompression were achieved using the endonasal and transoral approaches, and the distance to the target area was shorter in the first approach. The transcervical approach was unable to achieve more than 1 cm of lower clival resection, thus not allowing complete odontoid resection [5]. In 2010, Seker et al. reported that the transnasal endoscopic approach provided a shorter route to the CVJ, while the transoral approach achieved a wider opening [36].

However, the two approaches should be considered as complementary rather than as alternatives. When removing large lesions that extend from the upper clivus to below C2, the transnasal and transoral routes may be successfully combined. The transcervical approach has the clear clinical advantage of reducing the risk of meningitis and of cerebrospinal fluid leak; its advantages also include maintaining a sterile surgical field, a familiar approach, and an optimal surgical trajectory for pathological findings lower than C2.

In 2012, Dallan et al. [11] investigated a new robotic surgical setting, the DaVinci system, in two cadavers, comparing the traditional transoral and the combined transoral-transnasal approaches to the CVJ. They concluded that the lower the placement of the robotic arms, the easier was the dissection of the rhinopharynx, basisphenoid, and upper clivus.

Visocchi et al. [42] compared the surgical exposition angle and the working channel volume of both the transnasal and transoral approaches in a cadaver, by means of a comparative neuroradiological “real-time” study. They concluded that the transnasal approach, as widely discussed, is a viable strategy for reaching the CVJ, but that this approach has limited angular (nostrils, choanae) and linear (NPL) surgical exposure, which, in our view, makes it suitable only for certain types of diseases and prevents its systematic applicability in other conditions, such as lateral tumors and pathologies caudal to C2. However, an obvious advantage of this approach is that there is no need to cut the soft palate; this minimizes potential postoperative morbidities, such as swallowing disturbances and hypernasal speech, which have a major negative impact on the quality of life (if there is a palatine veil dysfunction). The transoral approach provides a better exposure of the CVJ, both on the sagittal plane and on the transverse plane. Finally, the combination of the two approaches must be considered as an option for accomplishing a particular surgical goal. From a purely anatomical point of view, the results of Visocchi et al. seem to suggest that, in normal anatomical conditions, the transnasal approach to the CVJ is an oblique approach, which allows only the piecemeal removal of CVJ pathology and is not recommended for large tumors and low and far laterally sited CVJ pathologies. The transnasal approach is limited in the caudal direction down to the NAxL, whereas the transoral approach is limited in the rostral direction in an anatomically normal specimen [42]. In a further study, Visocchi and colleagues have confirmed the NAxL to be a reliable preoperative predictor of the maximal extent of inferior dissection for the transnasal approach. Moreover, these authors identified the corresponding palatal line for evaluating the upper limit of the transoral approach (from the inferior dental arch up to the hard palate), which represents the maximal extent of superior dissection; they called it the surgical palate inferior arcade (SPIA), and, interestingly, it can be found by a simple lateral head X-ray examination with open mouth. The NAxL appears to vary more than the SPIA. Finally, the pros and cons of each approach have to be taken into account, as well as the choice of a combined transoral and transnasal approach [47].

Surgical Studies (Table 2)

Table 2 Surgical results of endoscopic-assisted surgery for the craniovertebral junction
Full size table

In regard to complications associated with the endoscopic endonasal approach, Valero et al., in 2015 [29], in a comprehensive literature search of several databases indexing the English-language literature published from 1990 to November 13, 2014, reported cerebrospinal fluid (CSF) leakage in 18 % with this approach. One patient developed meningitis that was complicated by sepsis and death, resulting in a procedure-related mortality of 1.4 %. Transient velopharingeal insufficiency was seen in three patients (4.2 %) and two patients had respiratory failure in the perioperative period.

Liu et al. 2015 [26] reported the operative technique and technical nuances used in their institution. In particular they use two surgeons (neurosurgeon and otolaryngologist) with a three- to four-hand approach via binostril access. They start with a 30° angled HD (Hight Definition) 4-mm endoscope. A zero degree endoscope is preferred in cases of cranial settling in which the odontoid is located very high, above the hard palate. A pedicled nasoseptal flap is prepared on both sides. In some cases of platybasia, it may be necessary to perform a sphenoidotomy and extend the midline incision from the floor of the sphenoid sinus down to the inferior clivus, especially if the odontoid process is located in a retroclival position.

The Menezes group (Dlouhy et al. [15]) emphasize the importance of intraoperative reduction strategies. If reduction cannot be achieved, a 540° procedure may be necessary in some cases (depending on the pathology), whereby the posterior approach and incision is temporarily closed and the patient is repositioned supine for a ventral decompression, followed by reopening of the posterior incision and posterior fixation. All patients undergo neck flexion/extension magnetic resonance imaging (MRI) of the CVJ. The patient is positioned supine with halo crown traction; an intraoperative three-dimensional computed tomography (3D CT) scan is obtained in traction. The patient is then placed prone and another 3D CT scan is obtained. The updated algorithm is shown in [13].

Conclusions

The progressive worldwide blooming of transoral procedures, thanks to the intensive care and the improvements in intraoperative neurophysiological monitoring techniques (once considered pioneering and very selective), is spreading the expertise in this field of surgery to a new population of surgeons. These techniques are performed alone or in conjunction with posterior procedures [43].

The pure endonasal and cervical endoscopic approach deserves consideration, but it still has three disadvantages, the first two being: (1) the steep learning curve and (2) the lack of 3D perception of the surgical field, which could be an operationally limiting factor. Image clarity would be diminished when endoscopes smaller than 2.7 mm are used. Standard 4-mm endoscopes give a good image quality, but 2.7-mm scopes provide better maneuverability. The third disadvantage is that there is a limited working channel, depending on the variability of the NAxL, which can make it difficult to remove huge tumors.

In our opinion, endoscopically assisted transoral surgery with 30° endoscopes represents an emerging alternative to standard microsurgical techniques for transoral approaches to the anterior CVJ. Used in conjunction with traditional microsurgery and intraoperative fluoroscopy, this endoscopically assisted transoral approach provides a safe and improved method for anterior decompression, with or without a reduced need for extensive soft palate splitting, hard palate resection, or extended maxillotomy. Virtually no surgical limitations exist for this approach, compared with the pure endonasal and transcervical approaches to the CVJ, in normal anatomical conditions.

Of note, the endoscope has an interesting role as “support” for the standard transoral microsurgical approach, since 30° angulated endoscopy strongly improves the visual but not the working channel and volume, even though soft palate splitting is often still required. In our opinion, the transoral (microsurgical or video-assisted) approach with sparing of the soft palate still remains the gold standard compared with the “pure” transnasal and transcervical approaches, due to the wider working channel provided by the former technique. The transnasal endoscopic approach alone appears to be superior when the CVJ lesion exceeds the upper limit of the inferior third of the clivus. Furthermore, the combined transnasal and transoral procedures can be tailored according to the specific pathological and radiological findings.

According to a recent anatomical study, the lower incidence of postoperative dysphagia with the endonasal approach is likely related to the lower density of neuronal elements from the pharyngeal plexus above the palatal plane [38].

However, the time to extubation and oral feeding was significantly shorter in the endonasal group in that study. Similarly, Ponce-Gómez and colleagues reported their own series of patients treated using both approaches and found comparable rates of neurological improvement after odontoidectomy, with less time to extubation and oral feeding, as well as shorter hospital stay, in the endonasal group [39].

Finally, to further validate all the endoscopic techniques, experience is required with greater numbers of patients and long-term follow-up. In our opinion, and in agreement with other authors, the endoscopic endonasal approach should be considered a complementary approach, rather than an alternative, to the standard transoral-transpharyngeal route [29].