Compression of nerve fibers can cause various clinical symptoms depending on the cause, magnitude and duration of the compression trauma (Rydevik and Lundborg, 1977). The pathophysiology of these various lesions is not fully understood and it has been debated whether the ischemia, secondary to compression, or the mechanical deformation of nerve fibers per se is the more significant etiological factor. Compression injury was, at first, attributed to ischemia due to following reasons: (1) nerve function was known to be dependent on blood supply and perfusion was impaired during compression; (2) nerve conduction failure did not occur even when in vitro nerve was compressed under high pressures (Powell and Myers, 1986); (3) nerve injury was greater with increased durations of compression; and (4) large myelinated fibers were especially vulnerable to ischemic injury. While the role of the ischemia has been stressed by some authors (Merrington and Nathan, 1949; Powell and Myers, 1986), the significance of the mechanical nerve fiber deformation
was emphasized by others (Gelfan and Tarlov, 1956; Ochoa et al., 1972). The optic nerve axons may be severed by either crushing or cutting the optic nerve. In rabbits and rats this operation does not usually result in ischemic death of the retina, provided that the lesion is more than about 1 mm behind the eyeball. The central retinal artery in these animals enters the optic nerve just behind the optic disc. Posteriorly, the optic nerve contains no large internal vessels and appears to receive its blood supply through its meningeal covering (Ruskell, 1964). Postoperative changes in the retina and optic pathways
can therefore be attributed with some confidence to the degeneration of axon (Kiernan, 1985). In the present study, the clip was placed 2-3 mm away from the optic nerve head and care was taken to ensure that the crush did not compromise the ophthalmic artery. Therefore, we think that the central retinal artery did not get affected from the compression.
由于压伤原因、强度、持续时间不同，压迫神经纤维可以引起不同临床症状(Rydevik and Lundborg, 1977). 目前尚无法完全了解各种损伤的病理生理学，以往曾就压迫神经后引起的局部缺血和神经纤维本身机械变形两者那个是更重要的致病因素做过讨论。起初，把挤压伤归因于局部缺血，原因在于：（1）神经功能依靠血液供给，挤压使灌注受损；（2）即使活体外高压力挤压神经，神经传导也未发生故障(Powell and Myers, 1986)；（3）挤压持续时间延长，神经损伤加重；（4）巨大的有髓神经纤维在局部缺血情况下特别容易损伤。一些著作者已经强调过局部缺血的作用(Merrington and Nathan, 1949; Powell and Myers, 1986)，也有人强调过神经纤维机械变形的重要性(Gelfan and Tarlov, 1956; Ochoaet al., 1972)。挤压或切割均可以离断视神经纤维轴索。在对兔子和大鼠所做实验中，只要损伤在眼球下超过1mm，通常不会导致视网膜局部缺血坏死。首先，这些动物的视网膜中央动脉于视神经盘后进入视神经；其次，视神经内部没有大血管，可能是通过脑脊膜接受血液供应(Ruskell, 1964)。视网膜和视路的术后改变一定程度上可归因于轴索退化(Kiernan, 1985)。此次研究动脉瘤夹置于距离视神经乳头2－3mm处，同时注意确保挤压不会损伤眼动脉，因此我们认为视网膜中央动脉未受到挤压的影响。
Although the crush itself is an acute injury, the propagation of damage is a chronic event. This model can therefore simulate the spreading of neurodegenerative diseases, but not their initiation. Existing models for glaucoma are associated with increased intraocular pressure (Schwartz, 2000; Yoles et al.,
1997). Each model has its specific advantages and disadvantages, and it is only by accumulating the relevant data from several models that it will be possible to make valid inferences for clinical situations. It is worth noting that whatever information was obtained from the optic nerve crush model turned
out to be valid for other models as well (Schwartz, 2000; Schwartz and Cohen, 2000). It has been reported that chronic degeneration caused by an increase in intraocular pressure would better simulate the characteristics of glaucoma than crush injury (Schwartz, 2000; Schwartz and Cohen, 2000).
虽然挤压伤本身是急性损伤，损伤的传播却是慢性过程。因此本模型可以模拟神经变性疾病的扩展，但是不能模拟神经变性疾病的起始反应。而现有的青光眼模型伴随眼内压增加(Schwartz, 2000; Yoles et al., 1997).。各个模型各有利弊，只有通过积累几个模型的相关数据才能对临床表现做有效推断。值得注意的是，从视神经挤压伤模型中所获得的信息经证明对其他模型也有确实根据(Schwartz, 2000; Schwartz and Cohen, 2000)。据报导，眼内压增加引起慢性退化比挤压伤能更好地模拟青光眼的特征(Schwartz, 2000; Schwartz and Cohen, 2000)。
It is often difficult to compare the results obtained by different investigators on nerve compression injuries, owing to differences in method of pressure application and noncomparable pressure levels. In the present study, we tried to overcome some of these problems by using a specially designed and commercially available device. In our previous study (Sarikcioglu and Ozkan, 2000), this device was used to produce quantitative crush injury to the rat sciatic nerve. One of the most important advantages of this device is that, due to its relatively small size, it can reach the narrow approached areas, such as spinal nerve roots, facial nerve, and optic nerve as studied in the present study. The most important disadvantage of this device is that it cannot be used for graded compression injury. The investigator could only change the duration of the compression. We think that standardization of the compression is necessary to compare interlaboratory results.
由于挤压方法和压力水平不同，通常很难比较不同研究者对神经挤压性损伤得出的结果。本研究中，我们使用专门设计且市面有售的设备克服这些问题。在我们以往的研究中(Sarikcioglu and Ozkan, 2000)，曾使用这一设备制造小鼠坐骨神经定量损伤。此设备最大的优点之一在于由于其尺寸相对较小，可以到达入路狭窄的区域，如脊神经根，面神经和本次研究中的视神经。其最大的缺点在于不能用来制作不同等级的挤压伤。研究者仅仅能够改变挤压的持续时间。我们认为为了比较不同试验得出的结果，有必要将压力标准化。
Although crush of the intraorbital part of the optic nerve has been extensively performed (Becker et al., 2000; Buys et al., 1995; Campbell et al., 1999; Chen and Weber, 2000; Freeman and Grosskreutz, 2000; Meyer and Miotke, 1990; Okada et al., 2000), crush of the intracranial part of the optic nerve has also been reported. Chierzi et al. (1999) performed intracranial optic nerve crush injuries in wild-type mice and in bcl-2 transgenic mice, which overexpress bcl-2 in neurons, including retinal ganglion cells. These investigators found that 100% of retinal ganglion cells were still viable 1 month after optic nerve crush in bcl-2 transgenic mice, compared with 45% retinal ganglion cell survival after crush in wild-type mice.
虽然视神经眶内挤压伤模型已广泛完成(Becker et al., 2000; Buys et al., 1995; Campbell et al., 1999; Chen and Weber, 2000; Freeman and Grosskreutz, 2000; Meyer and Miotke, 1990; Okada et al., 2000),然而颅内部视神经挤压伤模型也有报导。Chierzi等（1999）曾用野生鼠和神经元中过度表达bcl-2的转基因小鼠制作视神经眶内挤压伤模型，其中包括视网膜神经节细胞。研究者发现bcl-2转基因小鼠在视神经挤压伤后一个月视网膜神经节细胞仍然100％存活，而野生鼠经挤压伤后视网膜神经节细胞仅有45％存活。
In the literature, there are several types of indirect or direct injuring models, neither of which allows quantitative or standard application of compression. The balloon method was used on the optic nerve of the cat (Burke et al., 1985, 1986; Cottee et al., 1991) and micro-sling was used on the optic nerve of Japanese monkeys (Matsuzaki et al., 1982). Forceps was also used to cause a defined trauma to the optic nerve of rats (Buys et al., 1995; Duvdevani et al., 1990; Kiernan, 1985; Sautter and Sabel, 1993), mice (Li et al., 1999; Tezel et al., 2000), and opossum (Araujo Couto et al., 2000).
文献中，直接或间接损伤模型有好几种类型，但是都没有对压力定量或标准化。过去曾使用气球法制作猫视神经损伤模型 (Burke et al., 1985, 1986; Cottee et al., 1991)、采用微悬带制作日本猴视神经损伤模型 (Matsuzaki et al., 1982)、用镊子制作大鼠 (Buys et al., 1995; Duvdevani et al., 1990; Kiernan, 1985; Sautter and Sabel, 1993) 、小鼠(Li et al., 1999; Tezelet al., 2000)及负鼠(Araujo Couto et al., 2000)视神经损伤模型。
Chen and Weber (2000) used a smooth-faced bulldog clamp that exerts approximately 1000 gf (grams force) to crush the optic nerve for 15 s. They used this clamp in the cat. Additionally, ligature is also a popular method to crush the optic nerve as Okada et al. (2000) performed a pulleyeweight system to crush the optic nerve of the cat. They used a 6-0 polyfilament silk suture placed around the optic nerve and tightened the suture with a 20 g weight (0.2 N) for 60 s. They were able to accomplish
the optic nerve approach by removal of the orbital roof. We think that these clamps and the pulleyeweight system cannot be used for optic nerve crush of the small laboratory animals
Several reports on optic nerve injury based on crushing of the optic nerve have been published in which the force applied was not stated (Becker et al., 2000; Buys et al., 1995; Campbell et al., 1999; Freeman and Grosskreutz, 2000; Meyer and Miotke, 1990). Taking into account that the optic nerve of
the rat is a very vulnerable structure, a standardized crush model is mandatory to guarantee defined and reliable conditions for later microscopic or electrophysiological evaluation
of the post-traumatic changes in the retinal ganglion cell layer (Gellrich et al., 2000). To solve the standardization problem few attempts have been reported. Duvdevani et al. (1990) developed a Castroviejo’s cross-action (self-closing) capsule forceps modified by attaching a screw to calibrate the force applied to the optic nerve. They changed the screw nut position to achieve various crush forces. They studied the consequences of the lesion with different severities both electrophysiologically and behaviorally. On the electrophysiological examination, they observed a lesion-dependent loss of
conduction of the compound action potential across the crush site; and more severe the crush, the more severe the initial deficit and the smaller the recovery of the compound action potential. On behavioral examination they observed that the loss of visual function, as defined by rat’s ability to orient
toward a visual stimulus, and the subsequent recovery also depended on the severity of the lesion. To improve their method, Sautter et al. (1991) attached a micrometer screw to the handle cross-action forceps. These cross-action forceps have also been used by numerous studies (Hanke, 2000; Kipnis et al., 2000; Mawrin et al., 2000; Sautter and Sabel, 1993; Schmitt and Sabel, 1996).
以前出版过一些关于视神经挤压性损伤的报导，其中应用的压力没有定量(Becker et al., 2000; Buys et al., 1995; Campbell et al., 1999; Freeman and Grosskreutz, 2000; Meyer and Miotke, 1990)。要考虑到大鼠视神经结构非常脆弱，必须采用标准模型，以确保视网膜神经节细胞层损伤后改变的镜下或电生理学评定条件明确可靠(Gellrich et al., 2000)。报导解决标准化问题的方法很少。Duvdevani等人（1990）发明Castroviejo’s反作用（闭路）密封舱镊子，安装螺杆以校准加在视神经上的作用力。他们改变螺母位置以获得不同挤压力，研究不同电生理学和行为学严重程度带来的损伤后果。电生理学试验中，他们观察到复合动作电位传导的一种损伤依赖性丧失，压伤越严重，最初的丧失越严重，复合动作电位恢复的程度越少。行为学试验中，观察到视觉功能的丧失似乎取决于大鼠对视觉刺激物的适应能力，随后的恢复也依赖于损伤的严重程度。为改良方法，Sautter等人（1991）将测微螺旋安装到手柄反作用镊上。这种反作用镊也已经被应用到众多研究中(Hanke, 2000; Kipnis et al., 2000; Mawrin et al., 2000; Sautter and Sabel, 1993; Schmitt and Sabel, 1996)。
Another attempt to calibrate the optic nerve crush was performed by Klocker et al. (2000). They described a model of crush lesion of the rat optic nerve inducing retrograde retinal ganglion cell degeneration that can be carefully controlled in its extent by a newtonmeter device. They studied the histological consequences of the injury, and also monitored the functional integrity of the retinal ganglion cell projection. They observed that the extent of secondary retinal ganglion cell death increased linearly with the applied crush force. Moreover, visually evoked potentials were used to characterize the consequences of controlled optic nerve lesion on the functional integrity of the visual projection. Their method is based on a calibrated device in combination with the supraorbital approach. Their method has several advantages in comparison with previously reported crush methods and latero-
orbital approaches. Gellrich et al. (2000) used the same microinjuring device for making optic nerve injury, and quantified the histological changes after this calibrated crush.
To perform the same crush injury model described above, investigators need to obtain the same calibrated device. However, Yasargil aneurysm clip is a specially designed device and is also commercially available all over the world. The results of the present study revealed that the crush injury of the optic nerve resulted in degeneration, approved by appearance of the numerous damaged myelin residues in the optic nerve. The present device would be sterilized and successively applied.
No inflammatory response was observed during the postoperative days. Another advantage of this device is that, due to its relatively small size, it can be reached to the narrow approached
areas. The most important disadvantage of this device is that it cannot be used for graded compression injury. In our previous report (Sarikcioglu and Ozkan, 2000), we used Yasargil aneurysm clip to create 5, 10, 20 min compressions to the rat sciatic nerve and found a strong relationship between nerve damage and subsequent recovery. Increment of the time resulted in severe axonal degeneration in sciatic nerve. We think that such increment of time might also be studied for the optic nerve.
要制作以上所述同样的视神经挤压伤模型，就要获得同样的标刻度设备。Yasargil(脑)动脉瘤夹是一个专门设计的设备，全世界都可以买到。目前研究结果表明，视神经挤压伤导致退化，视神经中大量损坏的髓鞘残余的形态可以证实这一点。这些设备经灭菌后使用，手术后数天没有观察到炎症反应。这种设备另一优势在于，由于其尺寸较小，可以到达入路狭窄的区域。该设备最大劣势在于，不能用于分级的挤压性损伤。在以往报导中(Sarikcioglu and Ozkan, 2000)，我们使用Yasargil（脑）动脉瘤夹压迫大鼠坐骨神经5、10、20分钟，发现神经损伤和随后的恢复之间存在密切联系。挤压时间增加导致坐骨神经严重的轴索变性。我们认为，这种时间的增加也可以用于视神经的研究。
The present study was supported by Akdeniz University Research Fund. The authors thank Mr. Oktay Kuru for his constrictive comments and help with manuscript. We also thank the anonymous reviewers of the Experimental Eye Research for their constrictive comments and help with manuscript.