Schnabel Cavernous Degeneration: A Vascular Change of the Aging Eye Open Access

Schnabel cavernous degeneration is an uncommon histologic entity characterized by the focal loss of myelin and axons with the preservation of septa, resulting in a spongiform appearance of the proximal optic nerve. Within the spaces created by the process, there is an accumulation of acid mucopolysaccharides (hyaluronic acid) in the optic nerve posterior to the lamina cribrosa. When first described by Schnabel1 in 1892, the condition was associated with glaucoma. Others further defined this notion as being a consequence of vitreous displacement through the lamina cribrosa during an episode of extremely increased intraocular pressure.2–6 Giarelli et al7 associated these changes with advanced age and generalized vascular abnormalities affecting the end arteries of a vulnerable area of the optic nerve supplied by the posterior ciliary circulation.

Little information is available about the clinical signs associated with the often-striking histologic lesions of the optic nerve. This study is a retrospective evaluation of 4500 eyes examined as part of a series of 50 000 autopsies performed between 1967 and 1991. Ninety-three persons had cavernous degeneration. Fifteen of the patients had had some type of clinical ophthalmic evaluation before death.

Trieste is the capital of the Friuli-Venezia-Giulia region of Italy, which is located at the northernmost tip of the Adriatic Sea approximately 700 km (430 miles) north of Rome. The area of the province is 211 km². More than 90% of the residents live in Trieste (212 327 people, 1993 census). The population has been influenced only marginally by immigration since the end of World War II. A high percentage of the population is elderly (almost 30% of the residents in 1991 were older than 60 years).

All autopsies within the province are conducted at the Department of Anatomic Pathology of the University of Trieste. Since 1968, the autopsy rate has increased rapidly from 17.5% to 96.7%, with an average rate of 95% since 1975.8 Italian law states that autopsies on patients dying in hospitals, medical university departments, and nursing homes are to be performed whenever the directors of the medical facilities decide that an autopsy is necessary for diagnosis verification or for solutions to medical or scientific problems. Embalming is not commonly practiced in Italy. As a result, nearly all of the autopsies were performed within 36 to 48 hours after death. Approximately 50 000 autopsies were performed between 1967 and 1991.

Ocular tissue was histopathologically processed when one of the following criteria was met: the person had been examined at the Eye Clinic of the University of Trieste, had a known history of ocular disease, or had ocular symptoms suggestive of ocular disease. The eyes of patients with severe systemic diseases, including diabetes mellitus, malignant hypertension, autoimmune disease (eg, rheumatoid arthritis, systemic lupus erythematosus), and malignancy, were processed. The eyes of all persons 90 years and older were examined. In addition, the eyes of approximately 10% of all cases of all ages with no apparent systemic disease were processed.

The eye, the optic nerve, and a sample of the orbital fat were removed through access created in the orbital roof after removal of the brain. The tissue was fixed in 4% formaldehyde for 2 weeks and processed for paraffin embedding as previously described.9 

The tissue sections of all eyes studied were stained with hematoxylin-eosin, periodic acid–Schiff, and van Gieson stains. All eyes studied that had characteristics of amyloid accumulation by hematoxylin-eosin staining were also stained with Congo red.

The study subpopulation consisted of all eyes removed at autopsy with optic nerve cavernous degeneration during a 25-year period (1967–1991). An electronic search of the outpatient records of the University Eye Clinic identified 15 patients with at least partial information about the clinical condition of the eyes at some time before death. These clinical findings were correlated with the histologic findings.

In addition to the routine histologic stains, eyes from the most severe cases of cavernous degeneration (approximately grade III/IV, see the following paragraph for details of how these lesions were graded) were stained with Alcian blue. Histochemical evaluation of smaller cavernous lesions was less effective for technical reasons. The eyes were examined specifically for histologic signs of glaucoma, other significant ocular diseases, and histologic signs of systemic disease, particularly vascular abnormalities. Because of the obvious nature of the cavernous degeneration in the sections, the observers could not be masked effectively to the presence or absence of cavernous degeneration. An effort was made to be as objective as possible in reporting the histologic findings.

A subjective scale was used to grade the severity of the cavernous degeneration as viewed on a single, random longitudinal section of the optic nerve. In grade I lesions, cavernous degeneration was clearly present, but it involved less than 10% of the diameter of the optic nerve at its point of most severe involvement. Grade II cavernous degeneration involved 10% to 25% of the diameter; grade III cavernous degeneration, 50% to 75% of the diameter; and grade IV cavernous degeneration, more than 75% of the diameter.

Approximately 4500 pairs of eyes were studied between 1967 and 1991. In 93 individuals, optic nerve specimens showed cavernous spaces in the retrolaminar section. In 2 cases, the spaces were large enough to be seen by gross examination.

The spaces in the most severe cases contained acid mucopolysaccharide that was sensitive to hyaluronidase and stained positively with Alcian blue.

Of the 93 cases, 75 (81%) were women, and 18 (19%) were men (male-female ratio, 1:4.2) in contrast with the entire autopsy series of 50 000 cases in which the sex ratio was nearly equal (male-female ratio, 1:1.2). The age range for patients with cavernous degeneration was 54 to 103 years, with a mean age of 88 years for women and 81 years for men. Twenty-eight of the patients were older than 90 years, and 2 patients were older than 100 years.

Cavernous degeneration was unilateral in 77 persons (82.8%). A moderate-to-severe grade of cavernous degeneration was found in 74%. The findings were limited to the retrolaminar portion of the optic nerve in most cases (Figures 1 through 4). Pools of optically clear substance made up of mucopolysaccharides could be recognized at high power (Figure 2). Grade IV lesions predominated. In 4 eyes, prelaminar optic nerve degeneration could be identified. There appeared to be an increase in prevalence of cavernous degeneration with age, particularly in women. However, there appeared to be no trends established between grade of cavernous degeneration and sex or age. The severity of systemic vascular disease could not be graded with the available autopsy information and compared with the severity of involvement of cavernous degeneration.

In 51 people of the cavernous degeneration group, arteriolosclerosis involving the arteries that supply the optic nerve was usually bilateral. In these cases, the central retinal artery was patent but with a variable degree of narrowing and vessel wall hyalinization due to arteriosclerotic change. The parapapillary choroidal vascular lumen appeared to be narrowed in most of these eyes. Similarly, the branches of the short posterior ciliary arteries often showed moderate-to-severe luminal narrowing.

Findings due to vascular occlusive disease, including loss of nerve fiber layer and gliosis of the optic disc, were much more prevalent in the cavernous degeneration group.

Other ocular vascular diseases included diabetic retinopathy (background and proliferative) in 15 persons (16%) of the cavernous degeneration group. Histologic changes suggestive, but not diagnostic, of glaucoma, including loss of ganglion and nerve fiber layer associated with some degree of retrodisplacement of the optic nerve, were found in 22 persons (23.7%). In most cases, the degree of cavernous degeneration and glaucomatous damage was not correlated. A positive correlation of cavernous degeneration and glaucomatous damage was found in only 4 persons (4%). In no case could disruption of the glial surface of the optic disc or other evidence of posterior vitreous displacement be identified.

Examination of autopsied tissue revealed a high prevalence of atherosclerosis (Table 1). Other diseases commonly observed included diabetes mellitus, nondiabetic renal disease, and malignant neoplasms comparable to those in the general autopsy population.10 

Concomitant vascular abnormalities of the brain and eye were recorded in 27 (29%) persons. Recent and old cerebral infarction was observed in 29 (31%) persons. Myocardial infarction with severe coronary artery stenosis was found in 5 (5%) persons.

Some clinical information concerning the eye antemortem was available for 15 (16%) persons (Table 2). Of these 15 persons, 13 (87%) were women, and 2 (13%) were men.

The primary ophthalmic diagnosis was chronic open-angle glaucoma in 7 (47%) persons. The criteria for establishing the diagnosis of chronic open-angle glaucoma were not specified in most cases, although intraocular pressure elevation characterized most of the patients. When the clinical appearance of the optic nerve was described, it was stated to be “pale” or “cupped” without any attempt at quantification. There was no description of sector atrophy of the optic disc in any of the case histories. A Humphrey visual field test was performed in one case, described as “glaucoma with mild cupping,” in this person who was seen on only one occasion in the outpatient clinic. The visual field examination showed only nonspecific generalized depression of visual function. In the clinical chronic open-angle glaucoma group, the visual acuity in 5 (71%) persons was less than 20/200 (Table 3). In 3 (43%) of these cases, the visual acuity was described as “light perception or blind.” Grade IV cavernous degeneration was found in 5 (71%) of the persons with the clinical diagnosis of glaucoma. Grade IV cavernous degeneration was found in 2 persons without the clinical diagnosis of chronic open-angle glaucoma. Associated severe systemic vascular abnormalities were identified in 5 of the cases with the clinical diagnosis of chronic open-angle glaucoma. One case of “acute glaucoma” was identified; however, details of the episode or episodes were vague.

In 1892, there was only a rudimentary understanding of the relationship between increased intraocular pressure and optic neuropathy. In that year and in 1900, Schnabel1,11 proposed a theory of pathophysiology of optic nerve damage in glaucoma derived from the antemortem feature of increased intraocular pressure and microscopic evidence of “kettle-like” excavation of the prelaminar optic nerve, attributed to a loss of axons in the posterior nerve, and cavernous spaces imparting a spongelike appearance to the optic nerve.

Later, Schnabel12 described stages of development of glaucomatous optic nerve damage and concluded that the primary pathologic process was in the tissues of the optic nerve, as evidenced by the creation of cavernous spaces due to tissue loss. During the next 50 years, the matter was reassessed by several investigators, some of whom expressed doubts about Schnabel's theory, suggesting an alternative pathophysiologic mechanism linked to poor vascular supply of the optic nerve.13–22 

In 1967 and 1968,2,23 Zimmerman and colleagues again brought attention to the relationship between glaucoma and cavernous atrophy of the optic nerve by studying an animal model of acute glaucoma. They concluded that the caverns formed because of a “peculiar form of ischemic necrosis developing as a consequence of severe acute glaucoma.” The hyaluronidase-sensitive acid mucopolysaccharide was thought to be vitreous forced into the cavernous spaces by increased intraocular pressure.

In a later case report,3 in which Zimmerman was again one of the authors, a 26-year-old man with uveal metastatic malignant melanoma (but no history of acute or chronic elevation of intraocular pressure) was found to have cavernous degeneration at autopsy. Sepsis, whole head radiation, and periods of hypotension had complicated his terminal clinical course. Hyaluronidase-sensitive acid mucopolysaccharide was identified in the cavernous spaces. The origin of the optic nerve change in this case was not determined with certainty, although unrecognized acute glaucoma, ischemia due to hypotension and terminal pulmonary failure, and a possible congenital defect in optic nerve function were considered.

In 1997, Gong et al24 concluded that in Schnabel cavernous optic atrophy, the hyaluronic acid in the atrophic area of the optic nerve was probably produced in situ and not necessarily from some external source.

In a previous study5 published in 1977, one of us (L.G.) and collaborators again emphasized the role of ischemia in the formation of cavernous degeneration. In that series, there was histologic evidence of arterio-arteriolosclerotic compromise of the ciliary vessels. Most of the patients did not have a history of glaucoma. In 3 of the 4 patients with glaucoma, cavernous degeneration was unilateral. We concluded that cavernous degeneration is not rare and that “the lesion that always and only affects the retro-laminar segment of the optic nerve appears to be attributable to vascular phenomena that cause chronic ischemia.” Other single case reports have also confirmed the relationship between ischemia and cavernous degeneration.6,25 

In a retrospective review by Knox et al14 of 193 eyes with the histopathologic diagnosis of ischemic optic neuropathy, 36% (69 eyes from 55 persons) had cavernous lesions. Swelling of nerve bundles with separation and occasional rupture of septa, loss of axons and myelin, and formation of caverns characterized the cavernous lesions. The caverns distended the area of infarction, stretched pial membranes on the surface, and compressed adjacent, more normal, often central nerve fibers. Mucopolysaccharide was found in most the cavernous lesions (37/69 nerves). The earliest evidence of mucopolysaccharide accumulation was in a lesion of at least 1 month in duration. Knox et al concluded that a dense accumulation of mucopolysaccharide indicates a process extending over a period of weeks or longer and that the mucopolysaccharide was probably produced locally. Only 5 eyes had a history of glaucoma or showed histologic evidence consistent with glaucoma. They noted that progressive visual loss in ischemic optic neuropathy might be due to the compression of intact nerves from ischemic edema and cavernous swelling or a second ischemic lesion. They therefore deduced that if the cavernous lesions could be identified clinically (with modern imaging techniques), then possible therapeutic approaches to prevent further optic nerve damage would include nerve sheath decompression or local pharmacologic inhibition of mucopolysaccharide formation.

Ischemic involution occurs inevitably in all tissues with time, including the optic nerve head area.26,27 The diagnostic difficulty arises in being able to clinically recognize the effect.28 The persons affected are generally elderly and have multiple simultaneous reasons (cataract, glaucoma, and macular degeneration) for their slowly progressive loss of vision. The optic nerve tissue affected is functionally vital to the eye and is supplied by a complex and individually variable and particularly vulnerable vascular system.29 Finally, the diseases leading to ischemia (with the exception of the arteritides) are generally slowly, but irregularly, progressive and are not heralded by pain or acute ocular dysfunction (except acute, nonarteritic, anterior ischemic optic neuropathy). The ischemic processes leading to the creation of physical caverns in the retrolaminar optic nerve are found in the elderly, progressing at an irregular rate involving variable sites of vulnerable tissue, and leading to minimal or no recognized ocular dysfunction. The end result is recognized with certainty only at autopsy.

This study demonstrates that, in a large autopsy series, cavernous degeneration is a rare but not an exceptional disorder. Less than 0.1% of the subpopulation studied had these specific retrolaminar optic nerve abnormalities. The condition is found mainly in women older than 90 years and is often unilateral. This sex distribution was present in spite of the nearly equal sex distribution of the general autopsy population. The reason for the sex specificity is not known.

The most frequently found systemic disorder was vascular disease, with a prevalence of 75%. The portion of the retrolaminar optic nerve affected by cavernous degeneration is nourished by end arteries of the short posterior ciliary arteries forming the circle of Zinn-Haller. It is possible, given the capricious nature of systemic vascular occlusive diseases, that the proximal retrolaminar optic nerve is selectively, although infrequently, involved. The relatively greater representation of the macular fibers, preserving central vision, and involvement of the peripheral vision would also explain the lack of symptoms in a nonspecific manner. Additionally, the insidious nature of vaso-occlusive disease involving a small proportion of tissue might explain the lack of pain or other nonvisual symptoms. The principal difference between cavernous degeneration and giant cell arteritis may only be the rate and the degree of occlusion.

Histologic changes of vascular occlusive disease were identified frequently in eyes with cavernous degeneration. In most cases, there was no correlation between the degree of glaucomatous damage, in those cases in which glaucoma had been diagnosed, and the degree of cavernous degeneration. In no case was there a disruption of the glial surface of the optic disc to suggest retrodisplacement of vitreous or other material through the lamina cribrosa into the postlaminar optic nerve. The finding of large cavernous spaces filled with mucopolysaccharide is consistent with ischemic processes elsewhere in the central nervous system. The most common example is the “lacunae” associated with systemic hypertension. Ischemic damage in the central nervous system is not repaired with the production of extracellular materials, as is found in tissue outside the central nervous system.

The clinical findings correlating with cavernous degeneration have always been elusive. In most cases, attention was drawn to those with prelaminar optic disc changes, assuming that the retrolaminar changes were causally related. Even in the index case described by Schnabel, the visual acuity was preserved at a level of 20/40, despite abnormalities of the optic nerve being present “to the chiasm.” It is possible that this process involves the macular fibers of central vision only late in the course of the disease. In this series, only a small number of patients had records of receiving ophthalmic care. Approximately half of the patients were diagnosed as having glaucoma. Several of the people seeking ophthalmic care had extensive visual loss. There was no specific reference to any unusual appearance of the optic disc out of proportion to the clinical degree of glaucoma or other pathologic change. Extensive cavernous degeneration was found in patients with and without the clinical diagnosis of chronic glaucoma. In those cases in which formal visual fields were available, only generalized, nonspecific constriction was found that was not explained by elevated intraocular pressure or intraocular vaso-occlusive disease.

Cavernous degeneration appears to remain a diagnosis used by the histopathologist for mucopolysaccharide-filled spaces in regions of axonal loss. Perhaps, with more sophisticated imaging, this process could be established antemortem. The likely cause of the pathologic process appears to be slowly progressive vascular occlusion independent of changes in the intraocular pressure.

Helmut Buettner, MD, of the Mayo Clinic, Rochester, Minn, translated the original work by I. Schnabel into English.