Lifeng Yan1, Xiuchun Zhao1, Yaqiong Wang1, Lin Wang1, Xiao Cheng1, Lihua Zhou1, Xia Feng2

1Department of Human Anatomy, Zhongshan School of Medicine, Sun Yat-sen University, Guangzhou  510080, Guangdong Province, China
2Department of Anesthesiology, First Affiliated Hospital of Sun Yat-sen University, Guangzhou  510080, Guangdong Province, China

Lifeng Yan☆, Studying for doctorate, Attending physi-cian, Department of Human Anatomy, Zhongshan School of Medicine, Sun Yat-sen University, Guangzhou  510080, Guangdong Province, China

Corresponding author: Lihua Zhou, M.D., Ph.D., Professor, Department of Human Anatomy, Zhongshan School of Medicine, Sun Yat-sen University, Guangzhou  510080, Guangdong Prov-ince, China; Xia Feng, Master, Associate professor, Department of Anesthesiology, First Affiliated Hospital of Sun Yat-sen University, Guangzhou  510080, Guangdong Province, China
zhoulih@mail.sysu.edu.cn; fengxiar@sina.com

Supported by: a Grant from Health Department of Guangdong Province in China, No. A2007169*

www.crter.cn
www.nrronline.org

doi:10.3969/j.issn.1673-5374.2010.10.008

Abstract
BACKGROUND: Animal models of brachial plexus root avulsion are required for the study of brachial plexus root injuries. The established ventral approach results in slight injuries, and is similar to mechanisms underlying human brachial plexus root avulsion.
OBJECTIVE: To analyze the effects of weight, age, and species on the success rate of brachial plexus root avulsion, and to determine the perfect method for establishing models of brachial plexus root avulsion.
DESIGN, TIME AND SETTING: A randomized, block design was performed at the Laboratory of Professor Lihua Zhou, Zhongshan School of Medicine, Sun Yat-sen University, China from June 2008 to June 2009.
MATERIALS: Sprague Dawley (SD) rats, golden hamsters, and BALb/C mice were used in the present study.
METHODS: All animals were randomly subjected to classical brachial plexus root avulsion and modified brachial plexus root avulsion.
MAIN OUTCOME MEASURES: Success rate of brachial plexus root avulsion.
RESULTS: The success rate of brachial plexus root avulsion was greater in the modified group than in the classical group (P < 0.01). Moreover, the difference was significant in 15-day-old SD rats, 5-week-old SD rats, and 3-month-old BALb/C mice (P < 0.01). The success rate of brachial plexus root avulsion was greater in the same weight, 15-day-old juvenile SD rats, than in the 3-month-old BALb/C mice (classical group, P < 0.01; modified group, P < 0.05). The success rate of brachial plexus root avulsion was significantly greater in 3-month-old golden hamsters than in 5-week-old SD rats in the classical group (P < 0.05). The success rate of brachial plexus root avulsion was significantly lower in the 15-day-old SD rats compared with the 5-week-old and 3-month-old SD rats in the classical group (P < 0.01). However, there was no significant difference in the success rate of brachial plexus root avulsion between various ages of SD rats in the modified group (P > 0.05).
CONCLUSION: Modified surgery to induce brachial plexus root avulsion significantly increases the success rate of model establishment. Species, age, and weight affect the success rate of brachial plexus root avulsion, and species plays an important role in the success rate.
Key Words: brachial plexus root injury; species; age; animal; disease model; modified surgery; peripheral nerve injury; neural regeneration

INTRODUCTION
  
Brachial plexus root avulsion, which is a severe trauma induced by brachial plexus injury, results in partial or entire loss of upper limb sensory and motor functions[1].
Animal models of brachial plexus root avulsion are needed for basic research. Two animal models of brachial plexus root avulsion exist: extraspinal nerve root avulsion via a ventral approach[2-3] and intraspinal nerve root avulsion via a dorsal approach[4]. The ventral surgical approach is simple and easy to expose, and results in minimal damage to the animals, but the success rate and reliability are low[5]. The dorsal approach surgery can provide a direct and clear view of the surgical area to avulse accurately, and success rate is high. However, the surgery is complicated and results in great traumas, severe paraspinal muscle injury, bleeding in the paraspinal venous plexus, and poor stability in the cervical spinal cord[6]. The dorsal approach is used to avulse the distal stump immediately following surgery or when nerve roots are reimplanted or implanted immediately following surgery[6-8]. Injury mechanisms in animal avulsions via the ventral approach are identical to human nerve root avulsion, and injury to the animals is slight[2, 9-10]. Therefore, numerous studies have utilized nerve root avulsion at 
the intervertebral foramina via the ventral approach[2, 9-10]. Sprague Dawley (SD) or Wistar rats have been widely used in previous studies inducing brachial plexus avulsion injury[10-11]. However, neonatal SD rats, BALb/C mice, and other mouse species are rarely employed. If mice are utilized, the avulsion model is established only at the C7 nerve root[2, 12], and an entire brachial plexus avulsion does not result.
Previous results[13] were obtained from the classical whole brachial plexus root avulsion via the ventral approach, which demonstrated large differences in surgical difficulties of model establishment and success rate of avulsion in SD rats, golden hamsters, and BALb/C mice aged 3 months. Models using golden hamsters of this age are relatively easy to establish, whereas it is difficult to generate BALb/C mouse models aged 3 months, because of the high failure rate. Body weight varies greatly between the above-mentioned murine species. Therefore, it remains unknown whether difficulties of establishing avulsion models are associated with species or weight, and whether modified brachial plexus root avulsion increases the success rate of nerve root avulsion.
In the present study, 3-month-old SD rats, 3-month-old golden hamsters, 3-month-old BALb/C mice, and 5-week-old SD rats, as well as 15-day-old, juvenile SD rats were utilized to establish models of classical avulsion and modified avulsion. The present study analyzed the success rate and complications of post-surgical brachial plexus root avulsion to determine the best method for establishing a model of brachial plexus root injury.

MATERIALS AND METHODS

Design
A randomized, block design study.
Time and setting
This experiment was performed at the Laboratory of Professor Lihua Zhou, Zhongshan School of Medicine, Sun Yat-sen University, China from June 2008 to June 2009.
Materials
A total of 40 SD juvenile rats aged 15 days, 40 SD rats aged 5 weeks, 40 SD rats aged 3 months, 40 golden hamsters aged 3 months, and 40 BALb/C mice aged 3 months old, which were specific pathogen-free and comprised both genders, were randomly assigned to classical and modified groups (n = 20). The animals were supplied by the Experimental Animal Center, Sun Yat-sen University, China (license No. SCXK 22004-0011). Protocols were approved by the Animal Ethics Committee, Zhongshan School of Medicine, Sun Yat-sen University, China and were conducted in accordance with the Guidance Suggestions for the Care and Use of Laboratory Animals, formulated by the Ministry of Science and Technology of the People’s Republic of China [14].
Precise animal grouping is as follows:

Reagents and equipment are as follows:

Methods
The study was performed by the same person, who had microscopic, surgical and research experience with brachial plexus root injury.
Establishment of models of brachial plexus root avulsion
All animals were intraperitoneally anesthetized with 10% chloral hydrate (350 mg/kg), and then fixed in a supine position. The right side served as the surgery side, and an incision was made at the right sternal border along the clavicle to isolate subcutaneous tissue and to locate the cutaneous nerve, which passes through the chest muscle. Following separation of the chest muscle, the white nerve plexus was located and pulled apart using a self-made retractor to expose the surgical field. The white nerve plexus consisted of brachial plexus, which was accompanied by an axillary artery and veins. The fascia was separated utilizing blunt, rounded, micro-hemostat. Blood vessels were retracted to prevent bleeding, and the brachial plexus was fully exposed.
Classical group
Following exposure of the brachial plexus, the superior (C5-6), middle (C7), and inferior (C8–T1) nerve trunks were dissociated from the brachial plexus towards the proximal end. Micro-hemostatic forceps was employed to clamp the proximal nerve trunk end to avulse the nerve root.
Modified group
In the 5-week-old SD rats, 3-month-old SD rats, and 3-month-old golden hamsters, following exposure of the brachial plexus trunk, the proximal end was further dissociated until the C5–8 and T1 branches were visible. The C5–8 and T1 spinal nerve roots were separately dissociated until the vertebra was located. Subsequently, micro-hemostatic forceps was applied to clamp the C5–8 and T1 nerve roots and remove them.
In the 15-day-old, juvenile SD rats and 3-month-old BALb/C mice, following exposure of the brachial plexus, five branches of the spinal nerve root were dissociated until the vertebra was located. A total of five 1-cm long slivers, identical to the nerve root thickness, were made, which twisted around the proximal end of the nerve root. Ophthalmic forceps was used to nip this tissue and to remove the brachial plexus.
The nerve root was avulsed through the use of two surgical methods, together with the nerve trunk, and both were immersed in double-distilled water for 10 minutes. Under a surgical microscope, the anterior spinal root filaments and dorsal root ganglion were successfully avulsed, resulting in pre-ganglionic injury[2, 15-16]   (Figure 1).

Post-surgical complications and management
Following surgery, ulcers in the injured extremities of the animals were treated with erythromycin ointment and were bandaged daily until completely healed.
Main outcome measures
Success rate of brachial plexus root avulsion.
Statistical analysis
The proportion of avulsed nerve roots to total nerve roots was statistically analyzed in each group. The success rate of nerve root avulsion was calculated according to the number of avulsed nerve roots/total number of nerve roots. All results were expressed as Mean ± SD. The mean value of the groups was compared using analysis of variance (randomized block design). Data were analyzed utilizing SPSS 16.0 software (SPSS, Chicago, IL, USA). A P value < 0.05 was considered statistically significant.

RESULTS

Quantitative analysis of experimental animals
All animals survived post-surgery, exhibiting right upper limb paralysis. The remaining limbs exhibited normal activity. Ventral wounds healed in all animals, without inflammation, suppuration, or discharge. Four of the 40 3-month-old golden hamsters experienced ulcers in the injured extremities, which healed. Six of the 40 3-month-old BALb/C mice experienced ulcers in the injured extremities. These also healed. The injured extremities of three 3-month-old BALb/C mice fell off.
Brachial plexus root avulsion
SD rats
Classical group: the C5 nerve root in 12 of the 3-month-old SD rats, as well as the T1 nerve root in four of the 3-month-old SD rats were not removed. The C5 nerve root in sixteen 5-week-old SD rats and the T1 nerve root in six 5-week-old SD rats were not removed. The C5 nerve root in 20 of the 15-day-old, juvenile SD rats, the C6 nerve root in two of the 15-day-old, juvenile SD rats, and the T1 nerve root in 12 of the 15-day-old, juvenile SD rats were not extracted. Ulcers did not form in the injured extremities following avulsion.
Modified group: the C5 nerve root was not removed in four of the 3-month-old SD rats, in six of the 5-week-old SD rats, and in six of the 15-day-old, juvenile SD rats (Figure 2). Ulcers did not form in the injured extremities during the 2 weeks of observation following surgery.

Adult golden hamster
Classical group: the C5 nerve root was not removed in two of the 3-month-old golden hamsters.
Modified group: all nerve roots were extracted (Figure 2).
After 2 weeks of observation, ulcers formed in the injured extremities in two animals from each group; these subsequently healed.
Adult BALb/C mice
Classical group: the C5 and T1 nerve roots were not removed from 20 of the 3-month-old BALb/C mice. Moreover, the C6-8 nerve roots were not completely extracted.
Modified group: the C5 nerve root was not removed from 12 of the 3-month-old BALb/C mice, and the T1 nerve root was not removed from six of the 3-month-old BALb/C mice. However, the C6-8 nerve roots were completely avulsed (Figure 2).
After 2 weeks of post-operative observation, ulcers were observed in the injured extremities of three mice; these ulcers subsequently healed, and two injured extremities fell off in the classical group. Ulcers formed in a total of three animals, and one injured extremity fell off in the modified group.
Success rate of nerve root avulsion between modified and classical groups
The success rate of brachial plexus root avulsion was greater in the modified group than in the classical group (F = 28.32, P < 0.01). Moreover, this difference was significant in 15-day-old, juvenile SD rats, 5-week-old SD rats, and 3-month-old BALb/C mice (P < 0.05 or P < 0.01) (Figure 2).
Success rate of brachial plexus root avulsion in animals of the same weight and different species
The success rate of brachial plexus root avulsion was significantly greater in 15-day-old, juvenile SD rats than in 3-month-old BALb/C mice of equal weight (classical group P < 0.01, modified group P < 0.05).
In the classical group, the success rate of brachial plexus root avulsion was significantly greater in 3-month-old golden hamsters than in 5-week-old SD rats of the same weight (P < 0.05). The success rate of brachial plexus root avulsion was similar between 3-month-old golden hamsters and 5-week-old SD rats in the modified group (P > 0.05) (Figure 2).
Success rate of brachial plexus root avulsion in SD rats of different ages and weights
In the classical group, the success rate of brachial plexus root avulsion was significantly greater in the 5-week-old and 3-month-old SD rats than in the 15-day-old, juvenile SD rats (P < 0.01). However, there was no significant difference in success rate of brachial plexus root avulsion between the 5-week-old and 3-month-old SD rats (P > 0.05). In addition, there were no significant differences in success rate of brachial plexus root avulsion between SD rats of various ages and weights in the modified group  (P > 0.05) (Figure 2).

DISCUSSION

The present study utilized the ventral approach, as well as modified classical models, of brachial plexus root avulsion. (1) In the classical surgery model, the superior (C5-6), middle (C7), and inferior (C8–T1) nerve trunks were dissociated from the brachial plexus towards the proximal end, and the three nerve trunks were separately clamped. Surgery was simple, time efficient, and resulted in slight injury. However, the success rate was low, in particular in BALb/C mice. In the modified surgery model, a total of five spinal nerve branches were dissociated towards the proximal end. The micro-hemostatic forceps clamped the spinal nerve roots, rather than the nerve trunks. (2) During dissociation of the nerve roots, the forceps were not able to remove the brachial plexus due to axonal breakage or nerve apraxia. In addition, in BALb/C mice, it was fairly easy to induce nerve breakage. (3) During surgery, the surgical field should be fully exposed for clear identification of anatomical structures. One hand should maintain a tight grip on the tissues or muscles surrounding the dissociated nerve, and a forceps in the other hand should be used to softly remove fascia and other tissues surrounding the nerves. (4) In the 3-month-old BALb/C mice and 15-day-old, juvenile SD rats, the spinal nerves were thin, but the microforceps were relatively thick. Direct clamping of the nerve could result in nerve slippage and injury, and re-clamping could lead to nerve breakage. Therefore, during model establishment, sterile small cotton slivers were wrapped around the proximal end of the dissociated spinal nerve branches in the 3-month-old BALb/C mice and 15-day-old, juvenile SD rats. This led to increased nerve thickness, as well as increased friction between the nerve and surgical instruments, resulting in significantly improved success rate of nerve root avulsion.
The present study compared the success rate of models of the same weight, different species, and varied ages. The success rate was highest in 3-month-old golden hamsters, and no significant difference was detected between the use of classical or modified methods. The brachial plexus in 3-month-old golden hamsters was tough and could be easily avulsed from the root, which suggested that 3-month-old golden hamsters could be used to establish the classical method. The success rate of nerve root avulsion using the classical method was extremely low in 3-month-old BALb/C mice. Comparison of various species of equal weight verified that the success rate was significantly greater in 15-day-old, juvenile SD rats compared with 3-month-old BALb/C mice, which could be associated with fragile nervous tissue of 3-month-old BALb/C mice. Therefore, species played an important role in brachial plexus root avulsion.
It is difficult to avulse brachial plexus roots from younger animals of lighter weight compared with older animals. Nevertheless, after modifying the surgical method, 15-day-old, juvenile SD rats could be utilized to successfully establish models of brachial plexus root avulsion. The 15-day-old, juvenile SD rats, which were similar to 1-year human infants, could be employed to simulate neonatal brachial plexus injury. These models could provide a greater understanding of the mechanisms involved in the onset of palsy.
In the present study, ulcers or shedding were detected in injured extremities of 3-month-old golden hamsters and 3-month-old BALb/C mice, but not in SD rats of the same age. Post-surgical observations demonstrated that ulcers or shedding were self-induced by biting injured extremities with the teeth. Nerve root avulsion can cause pathological pain[17-19]. In fact, studies have commonly used brachial plexus root avulsion to establish animal models of pain[20]. In the clinic, patients with pain are sometimes treated for brachial plexus injury[21]. However, there are no reports that address the differences in pain threshold among the three murine species, and ulcer or shedding was not detected in injured extremities of the SD rats, which was likely due to high pain threshold.
In conclusion, the present study verified differences in models of brachial plexus root injury using the ventral approach in different rodent species, ages, and weights. The factor of species, weight, or age influenced the degree of brachial plexus root avulsion in murine species was a major influential factor. Adult golden hamsters could be subjected to classical brachial plexus root avulsion, whereas modified brachial plexus root avulsion should be utilized in juvenile SD rats and adult BALb/C mice.

REFERENCES

[1] Malin JP, Winkelmüller W. Phantom phenomena (phantom arm) following cervical root avulsion. Effect of dorsal root entry zone thermocoagulation. Eur Arch Psychiatry Neurol Sci. 1985;235(1): 53-56.
[2] Li L, Wu W, Lin LF, et al. Rescue of adult mouse motoneurons from injury-induced cell death by glial cell line-derived neurotrophic factor. Proc Natl Acad Sci U S A. 1995;92(21): 9771-9775.
[3] Gu YD, Ma MK. Nerve transfer for treatment of root avulsion of the brachial plexus: experimental studies in a rat model. J Reconstr Microsurg. 1991;7(1):15-22.
[4] Koliatsos VE, Price WL, Pardo CA, et al. Ventral root avulsion: an experimental model of death of adult motor neurons. J Comp Neurol. 1994;342(1):35-44.
[5] Zheng SD, zhang GM, Gu YD, et al. Anterior and posterior cervical spine surgery rat model of brachial plexus root avulsion. Zhonghua Xianwei Waike Zazhi. 2003;26(4):191-192.
[6] Fournier HD, Menei P, Khalifa R, et al. Ideal intraspinal implantation site for the repair of ventral root avulsion after brachial plexus injury in humans: A preliminary anatomical study. Surg Radiol Anat. 2001;23(3):191-195.
[7] Lang EM, Asan E, Plesnila N, et al. Motoneuron survival after C7 nerve root avulsion and replantation in the adult rabbit: effects of local ciliary neurotrophic factor and brain-derived neurotrophic factor application. Plast Reconstr Surg. 2005;115(7):2042-2050.
[8] Han L, Kan S, Yuan J. Experimental study on reimplantation of ventral root into spinal cord after brachial plexus avulsion. Zhongguo Xiufu Chongjian Waike Zazhi. 2007;21(9):948-952.
[9] Ko SJ, Chien CY, Lin CT, et al. Phrenic nerve transfer in the repair of brachial plexus injuries: an animal model. Chin J Physiol. 1997;40(1):51-55.
[10] Zhou LH, Han S, Xie YY, et al. Differences in c-jun and nNOS expression levels in motoneurons following different kinds of axonal injury in adult rats. Brain Cell Biol. 2008;36(5-6):213-227.
[11] Yuan Q, Hu B, Su H, et al. GAP-43 expression correlates with spinal motoneuron regeneration following root avulsion. J Brachial Plex Peripher Nerve Inj. 2009;4:18.
[12] Chan YM, Wu W, Yip HK, et al. Caspase inhibitors promote the survival of avulsed spinal motoneurons in neonatal rats. Neuroreport. 2001;12(3):541-545.
[13] Wang J, Yan L, Zhao X, et al. The diversity of nNOS gene expression in avulsion-injured spinal motoneurons among laboratory rodents. Nitric Oxide. 2010;22(1):37-42.
[14] The Ministry of Science and Technology of the People’s Republic of China. Guidance Suggestions for the Care and Use of Laboratory Animals. 2006-09-30.
[15] Wu W. Expression of nitric-oxide synthase (NOS) in injured CNS neurons as shown by NADPH diaphorase histochemistry. Exp Neurol. 1993;120(2):153-159.
[16] Chu TH, Wu W. Neurotrophic factor treatment after spinal root avulsion injury. Cent Nerv Syst Agents Med Chem. 2009;9(1): 40-55.
[17] Bigbee AJ, Hoang TX, Havton LA. At-level neuropathic pain is induced by lumbosacral ventral root avulsion injury and ameliorated by root reimplantation into the spinal cord. Exp Neurol. 2007;204(1):273-282.
[18] Berman JS, Birch R, Anand P. Pain following human brachial plexus injury with spinal cord root avulsion and the effect of surgery. Pain. 1998;75(2-3):199-207.
[19] Lai HY, Lee CY, Lee ST. High cervical spinal cord stimulation after failed dorsal root entry zone surgery for brachial plexus avulsion pain. Surg Neurol. 2009;72(3):286-289.
[20] Zheng Z, Hu Y, Tao W, et al. Dorsal root entry zone lesions for phantom limb pain with brachial plexus avulsion: a study of pain and phantom limb sensation. Stereotact Funct Neurosurg. 2009; 87(4):249-255.
[21] Sindou MP, Blondet E, Emery E, et al. Microsurgical lesioning in the dorsal root entry zone for pain due to brachial plexus avulsion: a prospective series of 55 patients. J Neurosurg. 2005;102(6): 1018-1028.
 (Edited by Jiang JM/Qiu Y/Song LP)