Herman Schloffer, according to Liu et al, performed the first successful modern transsphenoidal approach for a pituitary tumor in March of 1907. It was a tedious three-stage procedure, with wide exposure involving removal of the nasal turbinates and the majority of the bones of the sphenoid rostrum and vomer. It was not until 1909 that Theodor Kocher provided direct resection of the nasal septum (submucosal) to allow better visualization of the sella, preventing major lateral sinus wall invasion although still using a midline nose incision.
Hirsch, in 1910, provided the first true endonasal transseptal approach based on a procedure used by Hajek, his mentor, for sinusitis surgery. Hirsch noted gradual improvement in the patient’s visual symptoms over a 5 week multiple-stage procedure. In an effort to present a more desired postsurgical aesthetic appeal and also preserve the cartilage of the septum, Albert Halstead pioneered the sublabial gingival approach to the sphenoid. Although this was adopted by Harvey Cushing’s, he reverted to the transcranial approach and many surgeons followed suit.
Hirsch, despite unpopular reception from most surgeons at that time, continued as “an obscure voice in the wilderness.” Fortunately, this voice was not as obscure as thought when pupils of Cushing’s, particularly Norman Dott, modified (with attachment of a light source) and continued this approach in Scotland. Gerard Guiot from France learned the technique from Dott and in the late 1950s pioneered trajectory confirmation using radiofluoroscopic guidance and finally introduced the use of the microscope for dissection. With the advancement of medical endocrinology, technology, and general medicine, transsphenoidal surgery gained momentum in the 1960s with Jules Hardy expanding surgical intervention to include curative microadenectomies for endocrinologically active tumors versus debulking surgery for primarily visual relief.
In the late 1980s, further refinement of the technique was performed by Martin H. Weiss with the “extended” approach describing the removal of the tuberculum sellae and the posterior planum sphenoidale to overcome the difficulty in tumors with parasellar, suprasellar, and anterior skull base extension. Since the 1990s the use of the endoscope, stereotaxy, and frameless fluoroscopic guidance has allowed surgical options to evolve significantly. This evolution has provided surgeons with better techniques to overcome a very precariously located tumor. The need to preserve, restore, or minimize loss of endocrine and visual function while providing optimal removal of pituitary tumors remains challenging.
Epidemiology and Histopathology
Pituitary tumors account for 10% to 15% of all brain lesions from varied studies across several populations. These tumors are broadly classified as either nonfunctioning or clinically secretory adenomas. Nonfunctioning adenomas produce no clinically relevant pituitary hormones. This has presented a challenge in early diagnosis, often occurring with symptoms of mass effect. Nonfunctioning adenomas comprise 30% of all pituitary adenomas. These broad classes of tumors include the silent somatotroph adenoma, the silent corticotroph adenomas subtypes 1 and 2, silent subtype 3, null cell adenomas, and oncocytomas. Gonadotrophic hormones are also included in this category of pituitary tumors due to their clinically benign secretions.
When diagnosed incidentally by computed tomography and/or magnetic resonance imaging, at least one third will continue to progress. The utility of serial imaging of patients who wish to delay surgery due to the small size of tumor, lack of endocrinopathy, and no visual compromise from chiasmal compression was evaluated by Coulter et al. With 41 patients analyzed for a median duration of 70 months, a “steady state of growth” was achieved within 88 months by 90% of the tumors. Furthermore, they advocate a visual field test with routine endocrine monitoring as more economical for follow-up. An argument against that will be a young patient who may encounter a biphasic growth pattern and will not be outlived by his/her tumor.
Although attempts have been made at pathological classification of pituitary tumors, it has not lent itself to routine markers currently used in neuro-oncology to identify indolent versus malignant types (pleomorphism, nuclear atypia, increased cellularity, and mitotic activity). Imaging classification has for more than 40 years relied on the initial Hardy classification. This was a five tiered classification system identifying microadenomas (<1cm) and macroadenomas (>1cm). Tumor was further classified in terms of invasiveness. Microadenomas with normal sella were designated as 0 or grade 1 tumors. Grade 2, 3 and 4 tumors categorized macroadenomas with diffuse enlargement, focal destruction and extensive destruction, respectively. In an attempt to link endocrine, surgical morphology, pathological, and also radiographic classification, Kovacs et al in 1996 presented a World Health Organization (WHO) system that grouped all pituitary tumors into five types based on: (1) presentation and secretory activity, (2) size and invasiveness, (3) histological features, (4) immunohistochemical profile, and (5) ultrastructural subtype.
In a series of papers, attempts have been made to follow nonfunctioning adenoma patients to determine optimal time for intervention. However, the number of patients who eventually receive operative intervention makes the estimation of the true natural history difficult to characterize. In these above studies, over a period of 22 to 85 months, there was an increase in tumor growth in 50% of cases. Again visual impairment (mass effect, apoplexy), and pituitary endocrine axis suppression led to surgical intervention. In one of the papers, after following a select group of 28 conservatively managed patients for approximately 85 months, 29% of tumors showed regression in size; however, no independent predictors of tumor increase or decrease were identified. Hence most treatment centers caution careful monitoring of patients, even those noted to have regression over a period of time. Finally, the possibility of “silent apoplexy” may cloud the true natural course of nonfunctioning adenomas.
Surgery for Nonfunctioning Adenomas
Indications for surgery for nonfunctioning adenomas include apoplexy, visual compromise from optic chiasm compression, and also disruption of the pituitary endocrine axis.
Due to the often large size of these tumors on presentation, careful preoperative assessment and establishment of goals of surgery must be discussed before surgical treatment. At times it may be more prudent to aim for debulking relief of chiasmal compression in an elderly patient than to resort to “gross total resection” that might involve more risk to vascular structures or pituitary axis.
The basic principles of transsphenoidal surgery for nonfunctioning adenomas are no different than those for other lesions. We will briefly describe the main steps of the most common approach variants and the reader is referred to other chapters in this book for a more detailed description of the different surgical steps. Modifications of the transsphenoidal approach to the sella include the endoscopic transsphenoidal approach, sublabial transseptal transsphenoidal, endonasal submucosal transseptal, and submucosal septal “pushover” technique. Figure 22-1 shows the optimal operating room setting for the transsphenoidal approach to the sella. The patient is in a beach chair (semirecumbent) position and head support in a Mayfield head holder after perioperative anesthetic needs, including the administration of exogenous “stress dose” steroids, are addressed. Note the orientation of the head with the left ear pointed toward the left shoulder. A horseshoe or three-point fixation Mayfield attachment is used. Recently, frameless stereotactic registration has been performed as an added safety measure for preventing carotid injury and also to better define tumor extension. However, this step is not universally performed by surgeons. Another step is the use of a lumbar drain to insufflate air, allowing descent of the superior portion of the tumor into the sella if necessary. Preparation for sella reconstruction includes prepping the abdomen for fat graft harvest. This is more often used than fascia lata in conjunction with synthetic and autologous grafts.
Sublabial approach: Initial exposure is made after prepping with prep of choice and draping by first injecting the upper mucosa of the buccogingival junction and also the inferior nasal passage. An incision is then made from one canine to the other in the buccogingival junction ( Figure 22-2 ) and then via a subperiosteal dissection, elevation of the mucosa from the anterior nasal spine and the maxillary ridge inferiolateral to the nasal spine ( Figure 22-3 ). The sharp dissection is carried posteriorly to finally disconnect the cartilaginous portion of the nasal septum. The retractor is then advanced, visualizing the keel of the vomer bone. Care must be taken to avoid over distracting the retractor ( Figure 22-4 ). At this point the microscope is introduced.
Endonasal approach: After prepping with the antiseptic of choice, most surgeons insert cotton pledgets soaked with a vasoconstrictive agent introduced with the help of loupe magnification. This is followed by draping of the patient, thus preparing a site for fat harvest for sella reconstruction. Universally a local anesthetic with epinephrine is injected submucosally in the nasal mucosa (inferior and lateral nasal septum) and upper gingiva. Mucosal sharp dissection is carried out to the junction of the bony and cartilaginous septum and then both nasal tunnels are connected with the aid of retractors. Mucosal dissection must be carried out first with a right-sided hemitransfixion incision in the nostril. A septal pushover technique is preferred in revision surgery and in pediatric patients. With this technique, the initial mucosal incision is made at the junction of the bony and cartilaginous septa and both tunnels connected.
Endoscopic approach: Currently increasingly favored among pituitary surgeons is the use of the endoscope alone as a visualization and magnification tool during pituitary surgery. Rigid scopes commonly used are 4 mm in diameter, 18 cm in length, and with a zero-degree lens. A 45-degree and 30-degree angled lens are used mostly in select portions of the case. Standard patient positioning as previously described is used. With the introduction of the endoscope, the sphenoid ostium, located 1.5 cm above the room of the choana, is then entered. More often than not the superior turbinates may have to be lateralized to gain access. Complete connection of the two nasal passages is obtained with the repeating of the ostium entrance in the contralateral nostril. A microdrill with a small cutting burr is then used after further soft tissue removal to thin the walls of the anterior sphenoid sinus following detachment of the nasal septum. It is important not to perform a too wide inferolateral opening to avoid the sphenopalatine artery or to prevent a major arterial injury.
During the sella portion of endoscopic work, we prefer an assistant with the capability of driving an endoscope to a fastened holder. Major benefits of the endoscope include the ability to have a close up panoramic view and, with the aide of the angled lens, the ability to see within the suprasellar and parasellar regions ( Figure 22-5 ). Disadvantages include the technical ability to operate in two dimension, and also the ability to visualize and not have instruments in conflict during work.