| OCR Text |
Show VIEW POINT A Mechanical Theory to Account for Bitemporal Hemianopia From Chiasmal Compression Gawn G. Mcllwaine, FRCS, FRCOphth, Zia I. Carrim, MB, ChB, Christian J. Lueck, PhD, FRCPfVK), FRCP( Ed), FRACP, and T. Malcolm Chrisp, PhD Abstract: The association between bitemporal hemianopia and chiasmal compression is well recognized. The majority of chiasmal syndromes are caused by extrinsic compression from pituitary tumors, suprasellar meningiomas, craniopharyngiomas, and aneurysms. However, it is not clear why compressive lesions of the chiasm show a predilection for damage to nasal fibers with bitemporal hemianopia. Few experimental attempts at elucidating these mysteries have been reported and none has provided an adequate explanation. The authors postulate that the susceptibility of nasal fibers to preferential damage is explained by structural collapse theories as applied to crossing and noncrossing cylinders. By constructing a simplified mathematical model, the authors demonstrate that nasal fibers are subject to relatively greater pressures for any given external compressive force acting on the chiasm. ( J Neuro- Ophthalmol 2005; 25: 40- 43) ANATOMIC FEATURES The optic chiasm is a flattened quadrilateral bundle of nerve fibers located at the junction of the anterior wall of the third ventricle and its floor ( 1). It contains approximately 2.4 million afferent nerve fibers that arrive anteriorly from both eyes and leave posteriorly via the two optic tracts. It measures approximately 8 mm from anterior to posterior notch, 15 mm across, and 4 mm in height ( 1,2). It is now accepted that the nerve fibers arising from the temporal halves of the retinas of the two eyes pass directly backwards to the ipsilateral optic tracts, whereas the fibers arising from Department of Clinical Neurosciences, Western General Hospital, Edinburgh ( G. G. M.); Institute of Neurological Sciences ( Z. I. C.), Southern General Hospital, Glasgow; Department of Neurology ( C. J. L.), The Canberra Hospital, and the Australian National University, Canberra, Australia; School of the Built Environment ( T. M. C.), Heriot- Watt University, Edinburgh, Scotland. Address correspondence to Gawn Mcllwaine, MD, Department of Clinical Neurosciences, Western General Hospital, Crewe Road South, Edinburgh EH4 2XU, Scotland; E- mail: gawn. mcilwaine@ lunt. scot. nhs. uk the nasal halves of the retinas cross over to the contralateral tracts. However, this anatomic arrangement has not always been accepted or understood. The phenomenon of partial decussation was first proposed in 1704. In an attempt to explain the phenomenon of perceived singularity of vision derived from two separate eyes, Sir Isaac Newton ( 3) hypothesized that the optic chiasm was derived from the merger of both optic nerves with partial cross- over of nerve fibers. A few decades later, in 1723, Vater and Heinicke ( 3) suggested a similar theory to explain the phenomenon of " half vision." Today, exactly three centuries after Newton's original publication, the crossing of fibers at the optic chiasm is regarded as fundamental to an understanding of anatomy and disease of the visual pathways. Damage to the chiasm typically produces a bitemporal visual field loss caused by selective damage to the fibers from the nasal sides of the retina. Most chiasmal syndromes are caused by extrinsic compression from pituitary tumors, suprasellar meningiomas, craniopharyngiomas, and aneurysms ( 4). Although it is well- recognized that the exact pattern of field loss is related to the site of chiasmal damage, two questions remain unanswered. First, why do compressive lesions of the chiasm selectively damage nasal fibers, and second, how does this damage occur? Although it is common to encounter major thinning, distortion, and stretching of the chiasm and nerves at craniotomy, it is difficult to attribute the underlying mechanism of neural dysfunction either to damage by direct compression or to disruption of blood supply resulting in ischemia. PREVIOUS THEORIES Few experimental attempts at answering these questions have been reported. None has provided an adequate explanation. In 1969, Hedges ( 5) simulated the effects of a growing pituitary tumor on the chiasm by inserting and expanding the balloon tip of a Foley catheter into an empty sella turcica in fresh, normal, adult human necropsy material. He observed, macroscopically, that as the balloon 40 J Neuro- Ophthalmol, Vol. 25, No. 1, 2005 Bitemporal Hemianopia From Chiasmal Compression J Neuro- Ophthalmol, Vol. 25, No. 1, 2005 expanded and the chiasm was elevated, crossing lower nasal ( and subsequently upper nasal) fibers were most stretched, whereas temporal fibers remained relatively unaffected. He concluded that the relative sparing of nasal fields could be explained by the absence of stretch in noncrossing fibers. During the same year, Bergland and Ray ( 6), intrigued by the absence of altitudinal field defects in association with pituitary tumors, studied the arterial supply of the chiasm in 480 human brain specimens at autopsy. They observed that the blood supply of the optic chiasm arises from a superior and an inferior group of vessels. They concluded that crossing fibers in the chiasm receive their arterial supply solely from the inferior group of vessels. On these grounds, they concluded that infra- chiasmal vascular compression by the growing pituitary tumor accounts for bitemporal hemianopia. The anatomic findings of Bergland and Ray ( 6) are irrefutable. However, despite their excellently researched and presented article, there will always be a dispute as to which anastomoses occur within the chiasm. It seems unlikely that the developing blood vessels would selectively and precisely avoid anastomosing across the crossed/ uncrossed boundary. It seems unlikely that anastomoses between the superior and inferior circulations exist solely for the noncrossing fibers. It is generally accepted that early chiasmal compression causes a bitemporal hemianopia that " respects" the midline. This implies that, assuming the vascular theory is correct, anastomoses do not occur between the circulation of the crossed and the uncrossed fibers. Any deviation from this arrangement would mean that early bitemporal hemi-anopias would not " respect" the midline. Selective obliteration of the inferior blood supply and observation of the consequent visual field defects would confirm that this group of vessels was the sole blood supply of the crossing fibers. Such an experiment is clearly impractical. Even if the inferior blood vessels comprise the sole supply to the crossing fibers, this phenomenon does not preclude direct compression as an explanation for bitemporal hemianopias in chiasmal compression. Bergland and Ray ( 6) provide a seemingly credible anatomic explanation, but their theory does not explain why compression from above also causes bitemporal hemianopia. Furthermore, if the temporal uncrossed fibers have a dual blood supply from superior and inferior systems, one might expect damage to the inferior supply to result in an altitudinal defect of the nasal visual fields at the same time as causing a bitemporal hemianopia. This manifestly does not occur. Likewise, the susceptibility of crossing fibers to injury in well- documented cases of traumatic chiasmal syndrome is not explained by the vascular hypothesis ( 7). Although the experiment of Hedges ( 5) successfully models the progression of events in vivo, it does not shed light on why nasal fields are selectively spared, or how damage to nasal fibers actually occurs. It has long been recognized that neurologic damage can be caused by interference with the blood supply secondary to intracranial masses. However, the chiasm is unique because its location predisposes it to tumor compression and, for the reasons highlighted, is vulnerable to purely mechanical damage. OUR THEORY We postulate that the preferential susceptibility of nasal crossing fibers can be explained by the fact that the pressure generated by the force of external compression is inversely proportional to the area over which that force is applied, and that the area over which adjacent crossing fibers are in contact with each other is less than the area over which uncrossed fibers are in contact. Therefore, for a given compressive force, the resulting pressure applied to crossing fibers is greater than that applied to uncrossed fibers, resulting in greater damage and impairment of nerve function in the crossed fibers. Consider a simple thought experiment involving a segment of nerve fiber whose length equals its diameter d ( Fig. 1). For the purposes of simplicity, we will assume that nasal crossing fibers cross each other perpendicularly, FIG. 1. A comparison of the different deforming actions of compression applied to crossing ( perpendicular) and non-crossing ( parallel) fibers. D, diameter of axon and length of nerve segment, p, proportion of circumference flattened by pressure. 41 J Neuro- Ophthalmol, Vol. 25, No. 1, 2005 Mcllwaine et al whereas temporal fibers are aligned in parallel. As compression is applied to adjacent nerve fibers in these two situations, the resulting deformity will depend on the area of contact between the two nerves. If the compressing load can be spread over a larger area, there will be less deformity. In reality, the precise details of the contact area are somewhat more complex than presented here and require the use of Hertzian contact theory for precise definition. However, for the purposes of this thought experiment, the areas of contact have been simplified to the shaded rectangular areas. As a result of compression, adjoining nerve fibers will press on each other, and a proportion P of their circumferences ( IT X d) will be flattened. The arrangement of the nerve fibers makes a crucial difference to how much flattening occurs. In the case of crossing fibers, the area of contact will be approximated by a small square ( of sidep X ir X d), which gradually increases in area as further compression occurs ( Fig. 1). However, for the parallel, noncrossing fibers, the area of contact will run along the length of the abutting nerves and will be approximated by the shape of an elongated rectangle ( Fig. 1). Because the segments of nerve have a length that is the same as their diameter, the length of the rectangle will be d ( the entire length of the segment considered), whereas its width will be p X 7T X d. Thus, for the crossing fibers, the area of contact { ANUSUI) is ( p X 7T X d) 2, whereas for the parallel fibers (^ Temporal), the area is d X ( p X TT X d), so that and ANasai = ( pXirXdy ^ Temporal = dX{ pXirXd) Because pressure is inversely proportional to area of contact, for a given compressive force, the ratio of pressure on nasal fibers ( PNasai) to that on temporal fibers { PTemporal) would be given by: Nasal Temporal * Temporal ^ L Nasal irXp It is reasonable to assume that for the normal and minimally compressed chiasm, the proportion of the circumference P would be less than 1/ TT ( when P = 1/ TT there is no pressure difference between crossing and noncrossing fibers) ( Fig. 2). It seems unlikely that nerve fibers could adopt the configuration shown in Figure 2 without becoming dysfunctional. Moreover, nerves are in contact with many other nerves, not just one, as shown in Figure 2, and each of the many nerve- to- nerve contacts will potentially cause deformation. Once deformation has occurred, the fibers become even more vulnerable to further compression. Thus, the pressure affecting the crossing nasal fibers will be greater than that affecting the parallel p = 0 p = 1/ jt FIG. 2. Deformation of two cylinders ( representing stylized axons) compressed against each other. Before compression starts ( A), neither cylinder is deformed. As the cylinders are compressed against each other ( B), there is flattening of the interface ( shown in blue), which occupies a fraction, p, of the circumference. When the length of the interface reaches the diameter of the circle, P = 1 / IT. temporal fibers. For example, if 5% of nerve circumference were in contact with adjacent nerves ( P = 0.05), this would result in an approximate six- fold pressure difference between the crossing nasal fibers and the non crossing temporal fibers ( Fig. 3). It is not possible to quantify the precise magnitude of this pressure difference without determination of P in normality and in pathologic situations. Furthermore, in formulating this hypothesis, we have assumed that nasal fibers are perpendicular to each other. In reality, the crossing angle varies and this will therefore significantly influence the area of contact. Nevertheless, assuming that nerve fibers conduction is compromised as a result of architectural distortion ( Fig. 4), we can qualitatively deduce that when an external compressive force is applied to the chiasm, nasal fibers will undergo preferential damage simply by virtue of the fact that they cross each other. It is this phenomenon that results in the clinical manifestation of a bitemporal field defect. The same phenomenon may also explain bitemporal hemianopia in patients sustaining head trauma. Large transmitted shearing forces will cause a massive increase in 42 © 2005 Lippincott Williams & Wilkins Bitemporal Hemianopia From Chiasmal Compression J Neuro- Ophthalmol, Vol. 25, No. 1, 2005 3G 25 20 Nasal p 15 Temporal 10 I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I 0.05 0.1 0.15 0.2 0.25 0.3 Circumferential contact p FIG. 3. Relationship between ratio of pressure on nasal fibers ( PNasai) to that on temporal fibers ( PTemPorai) with P, proportion of circumferential contact. External applied force Local buckling/ crushing of central nerve conduit FIG. 4. Crossing nasal fibers undergoing preferential damage with local buckling. pressure and, if applied ( approximately) vertically through the fiber arrangement, would result in selective damage to the fibers that cross ( 7). TESTING OUR THEORY Although this hypothesis cannot be easily tested under laboratory conditions, it is possible to study the effects of a growing tumor on the optic chiasm and its nerve fibers by computer- based virtual simulation. This would use well- proven mathematical formulations and techniques, collectively known as finite element analysis, to represent the behavior of fibers in pathologic states. Finite element analysis is an established method in engineering and has also been very successfully applied in the field of medical biomechanics. A recent application has been in modeling the biomechanics of porcine corneas ( 8). Nevertheless, building a reliable model to test this hypothesis requires, among other things, precise data on the geometry of the optic chiasm and the material properties of the nerve fibers and their surrounding medium. Existing imaging studies of the optic chiasm may be used to define geometry ( 1,2), but specimens of the optic chiasm will have to be tested to establish material elasticity- a property fundamental to describing material behavior undergoing loading. Once a basic model is operational, changes to variables such as chiasm size, nerve fiber arrangement, and tumor growth can be made to assess the validity of the hypothesis under various conditions. If such modeling proves to be consistent with this hypothesis, it would then be possible to undertake limited testing of real chiasms in a manner analogous to earlier studies ( 5,6). REFERENCES Daniels DL, Haughton VM, Williams AL, et al. Computed tomography of the optic chiasm. Radiology. 1980; 137: 123- 7. Tamraz JC, Outin- Tamraz C, Saban R. MR imaging anatomy of the optic pathways. Radiol Clin North Am. 1999; 37: 1- 36. Rucker CW. The concept of a semi- decussation of the optic nerves. AmaArch Opthalmol. 1958; 59: 159- 71. Miller NR, Newman NX Topical diagnosis of lesions in the visual sensory pathway. In: Miller NR, Newman NJ, editors. Clinical Neuro- Ophthalmology, 5th ed. Baltimore: Williams & Wilkins; 1998: 307- 322. Hedges TR. Preservation of the upper nasal field in the chiasmal syndrome: an anatomic explanation. Trans Am Ophthalmol Soc. 1969; 67: 131- 41. Bergland R, Ray BS. The arterial supply of the human optic chiasm. J Neurosurg. 1969; 31: 327- 34. Heinz GW, Nunery WR, Grossman CB. Traumatic chiasmal syndrome associated with midline basilar skull fractures. Am J Ophthalmol. 1994; 117: 90- 6. Anderson K, El- Sheikh A, Newson T. FEA of the biomechanics of porcine corneas. The Structural Engineer. 2004; 82: 20- 5. 43 |
