| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Editorials |
1 From the Department of Radiology, Massachusetts Eye and Ear Infirmary, 243 Charles St, Boston, MA 02114. Received October 15, 2002; accepted October 18. Address correspondence to the author (e-mail: hugh_curtin@meei.harvard.edu).
Index terms: Computed tomography (CT), multi–detector row, 2131.12118 • Computed tomography (CT), thin-section, 2131.12118 • Ear, abnormalities, 2131.218 • Editorials • Temporal bone, abnormalities, 2131.218 • Temporal bone, CT, 2131.12115, 2131.12118
Superior semicircular canal dehiscence syndrome is a new and exciting concept in the field of otology. The extraordinary observations in the original article by Minor et al (1) give an explanation for and allow successful treatment of a particular type of dizziness previously considered to be of unknown cause. Identification and description of the syndrome would not have been possible without the latest advances in computed tomography (CT) (1,2). Imaging frequently depicts abnormalities, but it is an uncommon event when imaging is actually the key to the original description of the disease entity itself.
The diagnosis of superior semicircular canal dehiscence syndrome depends on the demonstration of a very small defect in the bony wall of the superior semicircular canal. Even a sliver of intact bone excludes the diagnosis. In their article in this issue of Radiology, Dr Belden and colleagues (3) push their machines to the very limits of resolution in an effort to improve the accuracy of the diagnosis. Precision imaging is directly pitted against the partial volume effect.
Since superior semicircular canal dehiscence syndrome is just now finding its way into journals and textbooks, a description of the syndrome is appropriate (1,4,5). The syndrome appears to be the result of a minute hydraulic phenomenon within the inner ear. When sound vibrations are transmitted from the tympanic membrane by way of the ossicles to the oval window, the stapes acts as a piston, pushing into the perilymph of the inner ear. Fluids are not compressible, and to cause even minute movement of the fluid, there must be a compensatory displacement somewhere in the system that coincides with that of the stapes. In the cochlea, the outward movement of the round window membrane balances the inward movement of the stapes in the oval window. A pressure wave moves through the perilymph of the scala vestibuli and scala tympani to the round window membrane.
Normally, the oval and round windows are the only two openings in the hydraulic system of the inner ear. The semicircular canals represent a hydraulically closed system, and there is no substantial movement of fluid in the semicircular canals when the stapes vibrates in response to sound. However, if there is creation of a "third window" in one of the semicircular canals, the hydraulic purity is corrupted, and movement can occur. As the stapes pushes inward, the covering of this third window pushes outward. This causes slight movement of the perilymph. This movement or pressure wave in the perilymph apparently compresses the endolymph within the membranous semicircular canal. The brain interprets motion of the endolymph as movement of the body, and the patient feels dizzy.
Tullio (6) described the phenomenon of dizziness in response to a loud sound in 1929. He created a third mobile window in a pigeon by drilling a hole into a semicircular canal. When presented with a loud sound from the side of the surgery, the pigeon developed nystagmus. In humans, any process that can create a third window can create the Tullio phenomenon. When otosclerosis was treated with "fenestration," or creation of an opening in the lateral semicircular canal, some patients experienced the Tullio phenomenon. Cholesteatoma, which erodes the canal, can create the effect. Syphilis with an osteitis of the otic capsule has also been implicated. Not all cases of the Tullio phenomenon are due to the third window phenomenon. Inflammatory disease or fibrosis may form an abnormal connection between the stapes and the membranous labyrinth. Pressure on the stapes displaces the membranous labyrinth (7). The resulting distortion causes movement of the endolymph, which results in the dizziness. Belden et al (3) mention other associations with the Tullio phenomenon, including trauma, Ménière disease, perilymphatic fistula, and Lyme disease.
Depending on which canal is stimulated, there are specific eye movements associated with the presence of a third mobile window (4,8–10). For instance, many patients with the Tullio phenomenon that resulted from fenestration surgery had abnormal stimulation of the lateral semicircular canal, with horizontal eye movements. There are, however, patients with no history of surgery that clearly have sound-induced vertigo and nystagmus, implicating the superior semicircular canal. In the case of superior semicircular canal stimulation, the ipsilateral eye moves vertically but with an inward torsion, reflecting the exact orientation of the canal within the body. Moreover, when these patients strain or bear down (Valsalva maneuver), some have nystagmus in the opposite direction but still along the plane of the superior semicircular canal. The transient elevation of intracranial pressure pushes the covering of the third window inward rather than pushing it outward, as would be the deviation caused by exposure to a loud sound.
A defect in the osseous semicircular canal can also cause apparent conductive hearing loss. Displacement of the covering of the third window dissipates energy and thus diminishes the amplitude of the pressure wave in the cochlea. The patient interprets this decrease in acoustic energy as a diminished sound volume. This would represent one of the few causes of conductive hearing loss that results from an abnormality within the labyrinth in a location more central than the oval window. Therefore, superior semicircular canal dehiscence syndrome might be considered a diagnostic possibility for conductive hearing loss, along with other causes, such as otosclerosis.
In 1998, Minor et al (1) described the specific eye movements and symptoms of superior semicircular canal syndrome and correlated the findings with a defect in the canal seen with thin-section CT. In two patients in whom Dr Minor was convinced that a small bony defect was present on the basis of the clinical symptoms and imaging findings, he patched the canal defect by using a middle cranial fossa approach, and the symptoms of both patients improved. A new clinical entity was established. Since that initial study, many surgeons have performed similar surgical procedures with excellent results (11).
In potential candidates for this syndrome, reliable documentation of the presence or absence of a defect becomes crucial. The surgery, though highly effective, requires a craniotomy—far from a trivial procedure. The optical findings are typical, but other abnormalities can overlap. How then does one firmly establish the diagnosis? Even a very thin layer or sliver of bone is sufficient to close the canal and prevent the problem. How can we be sure it is dehiscent, or, alternatively, prove that the bone is intact? Belden et al (3) use several strategies. They combine a thin-section CT technique with off-axis oblique reformations and finally examine the actual attenuation numbers within individual pixels to make the diagnosis more precise.
The superior semicircular canal lies in a plane approximately 45° divergent from both the sagittal and coronal planes. It is oblique to the routine transverse and coronal planes used in CT. Now, by using multi–detector row machines and thin collimation, the section width finally approaches the in-plane dimensions of the pixels used to make the transverse image. Thus, the multiplanar reformation achieves the same resolution as does the direct scan. Thin-section images are possible in virtually any plane. But what plane is optimal?
Before CT, pluridirectional tomography was used to evaluate the temporal bone. This technology allowed visualization of very small structures by blurring everything in front of or behind the object in question. The blurring was inexact, and shadows of other structures were superimposed. To minimize this effect and to most accurately display a particular minute structure, great efforts were made to define the orientation of fine cortical bone landmarks in space. A basic principle evolved that the optimal plane for displaying a particular plate of bone or a particular tubular canal was perpendicular to the plane of the structure of interest. Particular image orientations or projections that optimally displayed certain structures adopted names of their developers (12). The Guillen projection, for example, gave a plane perpendicular to the oval window and the medial wall of the middle ear. Two planes are particularly pertinent to our present discussion: the planes of Pöschl and Stenvers.
The Pöschl transverse pyramidal plane was 45° oblique from both the sagittal and coronal planes and sectioned the petrous bone in a plane perpendicular to its axis. The Pöschl plane gave a perfect section along the axis of the cochlear modiolus and optimally showed the vestibular aqueduct in longitudinal section. The superior semicircular canal appeared as a ring, with the entire arc of its outer wall displayed on one image. The Stenvers plane was perpendicular to that of Pöschl. This plane, originally described for radiography and incorporated into pluridirectional tomography, was also 45° oblique to the coronal and sagittal planes but was rotated 90° to the plane of Pöschl. This plane shows the turns of the cochlea. The Stenvers plane also images the superior cortex of the superior semicircular canal in perfect cross section.
The Pöschl and the Stenvers planes, which are perpendicular to each other, give optimal demonstration of the crucial wall of the upper arc of the superior semicircular canal. Belden et al (3) use approximations of these planes with excellent results. Demonstration of the cortical line is very accurate. There are no false-negative findings. If the bone is seen, then it is there, and the diagnosis of superior semicircular canal dehiscence syndrome is excluded. However, Belden et al (3) take the task to another level—that of the partial volume effect.
The partial volume effect results when a structure is smaller than the individual voxel used to generate a CT image. The attenuation value of the structure is averaged with that of the contiguous tissues within the voxel, and the resultant attentuation value is changed, altering the appearance on the final image. For example, the attenuation of a small bony plate averaged with that of a small amount of soft tissue may cause the plate to become invisible and appear to be dehiscent. As resolution improves with smaller and smaller voxels, partial volume effects are overcome. As the voxel becomes smaller than the target structure, the partial volume effect disappears. With each new iteration of a scanner, we resolve finer and finer detail. Tiny structures, such as the crura of the stapes, become visible, and diagnoses based on minute detail of these structures become possible.
The voxel dimension of the machine used in the present study was 0.5 mm. Conceivably less than 0.5 mm of otic capsular bone would be enough to form a protective barrier for the canal and prevent the diagnosis of superior semicircular canal dehiscence syndrome. The authors reason that if at least one-fifth of the pixel is filled with bone, the attenuation would still be high enough to be detectable by examining the actual attenuation measurements within a voxel, even if the image did not display the structure as the expected white line of bone. Even a small amount of bone would be detectable beside the substantially lower attenuation of the dura or against the fluid in the semicircular canal and the contiguous subarachnoid space. By measuring the attenuation values in each pixel, the authors changed the findings in their study for several temporal bones from false-positive to true-negative. Thus, the authors increase the reliability of the technique. They point out that it is still theoretically possible to have false-positive CT findings, but the frequency of such a finding diminishes with each step toward better spatial resolution. Finally, true-positive CT findings that demonstrate a real bony defect in patients with no symptoms remain theoretically possible. Perhaps the dura is thick enough to prevent fluid movement, or perhaps some other explanation will evolve. Future study and accumulated experience will allow further exploration of this possibility.
With each generation of CT, resolution will indeed improve, perhaps to a point where a tiny plate of bone even less than a 10th of a millimeter will become visible. In the meantime, the authors are to be congratulated on their innovative approach to optimizing their assessment.
Superior semicircular canal dehiscence syndrome is now an accepted cause of substantial patient symptoms. Evaluation of patients with superior semicircular canal dehiscence syndrome pushes the resolution limits of modern CT scanner technology. As shown by the Belden et al (3), reliable demonstration of the presence of a defect requires very precise imaging with the highest resolution available.
FOOTNOTES
See also the article by Belden et al in this issue.
REFERENCES
Related Article
This article has been cited by other articles:
|
|
 |
|
  C E OFFIAH, R T RAMSDEN, and J E GILLESPIE Imaging appearances of unusual conditions of the middle and inner ear Br. J. Radiol., June 1, 2008; 81(966): 504 - 514. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Ozgen, M.E. Cunnane, P.A. Caruso, and H.D. Curtin Comparison of 45{degrees} Oblique Reformats with Axial Reformats in CT Evaluation of the Vestibular Aqueduct AJNR Am. J. Neuroradiol., January 1, 2008; 29(1): 30 - 34. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. F. Branstetter IV, C. Harrigal, E. J. Escott, and B. E. Hirsch Superior Semicircular Canal Dehiscence: Oblique Reformatted CT Images for Diagnosis Radiology, March 1, 2006; 238(3): 938 - 942. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
