Endoscopic orbital surgery represents a challenging arena for the endoscopic surgeon. The orbit is highly complex, with critical structures confined in a fat-filled, soft-tissue space, thereby limiting visibility, and restricting the necessary manipulation of muscles, nerves, and vessels required to resect primary orbital tumors. With the advent of increasing expertise and technology, the endoscope is now being used to transgress pneumatized sinuses and operate within the boundaries of nonpneumatized cavities. An unparalleled view of the medial orbital apex, with improved illumination, a spatial working corridor, a resection tailored to the size and location of the lesion, no external scar, and shorter hospitalization all represent advantages of the transnasal endoscopic approach to the medial orbit.
The endoscopic transnasal approach for orbital and optic nerve decompression was published almost three decades ago. The subsequent decade provided the first report on the transnasal endoscopic removal of an intraconal orbital cavernous hemangioma (OCH). This subject has lain relatively dormant for the next 20 years; however, a recent increase in publications would suggest a resurgence in this field. Despite this, the literature portrays only limited case series and case reports with only a recently evolving consensus on management strategies for intraconal lesions such as OCH.
This chapter reviews the endoscopic management of intraconal lesions and discusses the current techniques and outcomes.
Epidemiology and Etiology
The OCH is the most common primary orbital tumor of adults, with a reported incidence of 5% to 15% of all orbital tumors. It is more common in women and occurs in the fourth and fifth decades of life. Recent evidence regarding the immunohistochemical features of proliferative capacity, vascular differentiation, and hormone receptor status suggests that progesterone may play a role in the clinical course. This could also explain the sudden growth of OCHs during pregnancy and the reduction in size or stabilization in postmenopausal women. However, the exact role of progesterone is yet to be fully elucidated. The natural history of OCH remains elusive. A significant number of OCHs present as asymptomatic lesions incidentally found on computed tomography (CT) or magnetic resonance imaging (MRI) performed for unrelated reasons. Previous studies have shown that asymptomatic lesions often show no progression. In a retrospective comparative case series of OCH in 104 patients, 31 had an asymptomatic, incidental OCH on imaging. Seventy-nine patients underwent treatment and 11 of these had presented with an incidental, asymptomatic OCH that enlarged and produced symptoms or new clinical findings. In the 20 other patients, there was no or minimal change in the follow-up period of 1.2 to 20 years (mean 5.8 years, standard deviation 4.6 years). The investigators concluded that if an incidental OCH does not change over several years, it is unlikely to do so in more prolonged periods of follow-up.
Anatomic Location and Characteristics
The majority of OCHs are located between the optic nerve and the extraocular muscles and are therefore considered intraconal. It is well documented that OCH have a predilection for the intraconal space. The single most common anatomic site is lateral to the optic nerve, which may reflect the relationship between the optic nerve and the distribution of the ophthalmic vasculature. This area lateral to the optic nerve, within the intraconal space, is rich in small arteries and arterioles, However, OCHs can be found throughout the orbit, including the medial intraconal space, extraconal space, and within the optic canal. Rarely they can extend beyond the confines of the orbit into the pterygopalatine fossa, cavernous sinus, and intracranial space.
The International Society for the Study of Vascular Anomalies has classified the OCH as a slow-flow cavernous venous malformation. Histologically it is not clear if cavernous venous malformations contain exclusively venous vasculature. There is some suggestion of arterial flow on imaging studies, yet histologically there is no evidence of the elastic lamina associated with arterioles. The thicker-walled vessels are thought to be the result of thrombosis and recanalization. If any arterial component is present, it is thought to be inconsequential both histologically and clinically. OCHs are characterized as lesions with mature cellular components and do not tend toward dysplasia or hypercellularity. They have a fibrous capsule that can incorporate surrounding vessels and nerves, but they do not typically infiltrate into surrounding tissue. OCHs tend to be slow-growing vascular lesions with a radiologic growth rate of 10% to 15% per year. The expansion of OCH is thought to be a cycle of stasis and thrombosis with endothelial cellular proliferation and recanalization into multiple clefts and vascular channels.
Clinical Presentation and Investigations
The presentation of intraconal OCHs is variable. Most studies reveal the most common presenting symptoms are visual impairment and proptosis, followed by pain and diplopia. In contrast, in a series of 214 patients painless, progressive proptosis was the most frequent presenting symptom, occurring in 76.6% of patients, lasting on average 4.0 years, and ranging from 2 months to 30 years. In another case series of 39 patients, 75% of whom had intraconally located OCHs, pain was the most common presenting feature in 15 patients (38.5%), followed by visual impairment in 13 (33.3%). Diplopia occurred in 4 patients (10%). The visual impairment was related to a compressive optic neuropathy in 10 patients (25.6%). Abnormal proptosis was identified in 79.5% on clinical examination, and duction deficits were seen in 20%. Papilledema, choroidal folds, and tropias were seen less commonly. A more posteriorly located lesion was associated with a relative afferent pupillary defect in 33.3%, and these individuals also had evidence of optic nerve compression on imaging.
Radiologic imaging is a fundamental component of the preoperative workup, as a routine biopsy is not typically performed, and therefore there is a reliance on the characteristic features of OCH on imaging. Preoperative MRI and or CT are nearly universally performed. The MRI provides superior detail on the intraorbital anatomy and the relationship of the OCH to adjacent structures. OCHs are isointense or slightly hypointense on T1-weighted images and hyperintense to muscle on T2-weighted sequences ( Fig. 27.1 ). The CT scan with contrast demonstrates a well-circumscribed round or elliptical, smooth mass, rarely lobulated in shape. OCHs appear as a soft-tissue density with contrast enhancement that varies depending on the phase of the study. Focal enhancement is seen in the early phase with a diffuse enhancement seen in the later phase. Occasionally a heterogeneous appearance can be seen, signifying irregular blood flow within the OCH. Bony erosion or demineralization is not common but has been documented. If the OCH abuts the globe, it will tend to indent the globe rather than mold or infiltrate it. OCHs do not expand with the Valsalva maneuver, highlighting the lack of both distensible structures and arteriovenous shunting, which is characteristic of slow-flow venous malformations. The MRI and CT scans have characteristic properties that secure the diagnosis in most cases; therefore although this feature can be useful to discriminate between true neoplasms and other vascular tumors, it is seldom required. Angiography is also not necessary for the reasons noted earlier. Furthermore, CT scanning enables intraoperative image guidance options and visualization of the surgical trajectory, thereby assisting with preoperative planning. The use of ultrasonography to assist with the diagnosis has also been documented in the literature ; this would seem more of a historical investigation, made redundant by MRI and CT.
Indications and Surgical Approach
Surgical resection is indicated for symptomatic lesions, whereas smaller asymptomatic lesions can be observed. The goal of surgery is definitive resection; however, given the benign nature, complete resection must be balanced against iatrogenic morbidity. Partial resections of intraconal OCHs have been reported in the literature, but long-term outcomes remain unknown.
The location of the intraconal OCH within the orbit relative to the optic nerve dictates the choice of the approach. OCHs with an epicenter medial to the optic nerve or below a plane of resectability, which represents a plane subtended by the contralateral nostril and the long axis of the optic nerve, are amenable to the endoscopic approach. The feasibility and safety of this approach has been consistently demonstrated in the literature. OCHs located lateral and superior to the plane of resectability are not candidates for a exclusively endoscopic resection. The surgical team involved in these procedures varies depending on institutions; however, a multidisciplinary team approach is invaluable and should include an otolaryngologist and oculoplastic surgeon and, in some cases, a neurosurgeon.
Surgical Equipment and Techniques—Hands/Surgeons
The rigid 0-degree endoscope is most commonly used initially. Other angled scopes (30-, 45-, and 70-degree scopes) can also be of assistance. Although image guidance is used in the majority of cases, it is not an absolute requirement as the tumor position may shift within the orbit during the approach and dissection. In one report, intraoperative MRI was used when image guidance was unable to locate a small apical intraconal hemangioma.
The standard single-nostril approach, using a complete uncinectomy, wide maxillary antrostomy, and sphenoethmoidectomy to create a working space and define the orbital wall and orbital axis, is a fundamental part of the surgery. The middle turbinate may be resected to increase access and visibility. A recent international multi-institutional study demonstrated the single-nostril, three-handed, two-surgeon approach or a binarial, four-handed transseptal approach is more commonly used in intraconal OCH resections, with only 31% resected using the single-nostril, two-handed approach. In the binarial approach, a posterior septectomy is needed. This is associated with minimal morbidity, allows for maneuvering endoscopes and instruments, and may be incorporated into the elevation of a nasoseptal flap for medial wall reconstruction. This suggests that intraconal OCH resections require consideration in the preoperative planning and operating room setup to optimize positioning and ergonomics of the second surgeon ( Fig. 27.2 ).