Despite over 150 years of experience in modern surgical management of the peripheral nerve, repair of a nerve defect remains a challenging problem in reconstruction surgery. Today, the widely accepted method used by most surgeons is bridging the defect with an autologous donor nerve. However, the requirement for a sacrifice of healthy nerves results in permanent denervation, associated sensory deficit and subsequently formation of painful neuroma at the donor site.1, 2 Furthermore, sufficient functionality may not be recovered when the distance to the target organ is long, such as in the case of injury proximal to limbs, or long defect cases. Problems with nerve regeneration following autografting include the fact that the regenerating axon must pass through two anastomosis sites, differences in the shape and diameter of nerve stumps, and the avascular nature of the graft tissue. There is much research being conducted to overcome these issues, and we also have developed a nerve-lengthening method as a new strategy for repair of a segmental peripheral nerve defect.3–5 This method involves grasping both nerve stumps and attaching traction sutures to an external fixator, then using the external fixator to gradually stretch both nerve stumps simultaneously to the point where they can be sutured together. With this nerve-lengthening method, there is only one anastomosis site, a low incidence of nerve stump shape/diameter mismatch, and no interposition of avascular tissue. So far, this method has been used to successfully treat 15-mm defects in rat sciatic nerves,4 as well as 20-mm defects in rabbit sciatic nerves6; these reports indicate superior nerve regeneration compared to autografting. Nerve tissue in both lengthened nerve stumps was not simply being elongated via visco-elasticity; it has been reported that this growth occurs through nerve regeneration: axonal sprouting and Schwann cell proliferations.7, 8 This method may be a novel regeneration technique that produces the necessary tissue to fill the defect through the mechanical stress of traction; distraction neurogenesis in vivo. This method is a promising therapeutic approach that may be implemented instead of autografting. However, before clinical application, we believe that it is necessary to confirm the efficacy and safety of this treatment in a species more closely related to humans. Thus, this study using cynomolgus monkeys was conducted to functionally, physiologically, and histologically evaluate nerve regeneration ability, as well as pain and other adverse events.
We have developed a new treatment for peripheral nerve defects: nerve-lengthening method, and confirmed the efficacy and safety of our method using cynomolgus monkeys. A 20-mm defect in the median nerve of monkey's forearms was repaired through the simultaneous lengthening of both nerve stumps with original nerve-lengthening device. To evaluate nerve regeneration after neurorrhaphy, electrophysiological, histological, and functional recovery were examined and compared to the standard autografting. Nerve conduction velocity, axon maturation, and the result of functional test were superior in the nerve-lengthening method than in the autografting. And there were no adverse events associated with our method. We concluded that this method is practical for clinical application. © 2011 Orthopaedic Research Society Published by Wiley Periodicals, Inc. J Orthop Res 30:153–161, 2012
Six adult cynomolgus monkeys (Macaca fascicularis, 8- to 10-year old, 3 kg) bred at the Tsukuba Primate Research Center were used. The animals were kept in a stainless steel cage individually under a 12-h light/dark cycle at 23–27°C and 50–70% humidity. Food was offered once a day, and water was ad libitum. All monkeys were healthy as assessed by annual examinations. The monkeys were maintained in accordance with the Rules for Animal Care and Management of the Tsukuba Primate Research Center and the Guiding Principles for Animal Experiments Using Nonhuman Primates formulated by the Primate Society of Japan.
All surgeries were performed under general anesthesia with endotracheal intubation. Vital signs were monitored with electrocardiography, blood pressure, oxygen saturation, and respiration. A skin incision was made aseptically in the forearm, the median nerve was exposed, and a 20-mm-long-segment was resected from 1.5 cm above the wrist creases.
Three monkeys were used. The radius was exposed, two 2.0-mm stainless steel half-pins were inserted into the radius, and a custom-made, external, nerve-lengthening device (Fig. 1A) was attached to the pins. The proximal and distal nerve stumps were fixed to a ring with a 4-0 nylon suture. Then, the traction sutures (3-0 polyester suture), which attached to the ring of each nerve stump, were attached to the external nerve-lengthening device via a small, stainless steel pipe (Fig. 1B). Then, the wound was closed with a 4-0 nylon suture. Both nerve stumps were distracted simultaneously via the nerve-lengthening device from the day after operation at a rate of 1 mm/day. Nerve lengthening was performed without anesthesia every day with the animals seated in a monkey chair. To avoid nerve-lengthening device removal, the device was covered with a plastic cast. After carrying out lengthening for 20 days, the nerve was exposed under general anesthesia, and the nerve-lengthening device and ring were removed. Both nerve stumps were refreshed, and direct end-to-end neurorrhaphy was performed with a 9-0 nylon suture (Fig. 1C).
Three monkeys were used. The autogenous sural nerve cable grafting (double bundled) procedure was performed for the 20-mm gap of the median nerve with 9-0 epineurial sutures. To ensure that the test conditions of the two groups were similar, casting was done postoperatively for 20 days.
Evaluation for Treatment-Related Side Effects
During nerve lengthening, especially after the nerve-lengthening operation, we examined whether pain-related behaviors, such as arm withdrawal, yells of pain, and autotomy were observed every day. After the lengthening period, observation of autotomy was continued. We checked for damage or removal of the lengthening device, as well as the presence of fractures by X-ray taken once a week during the lengthening period.
Nerve conduction studies were performed at 16 weeks after surgery under general anesthesia with the animal placed on polystyrene, and the temperature of the arms was maintained by a thermal gel pack. Abductor pollicis brevis (APB) needle electromyography was performed at 4, 8, 12, and 16 weeks after first surgery without anesthesia with the animals seated in a monkey chair.
Motor Conduction Studies
The compound muscle action potential (CMAP) was recorded from the APB muscle using a concentric needle electrode placed within the muscle. The median nerve was stimulated at the wrist distal to the nerve lesion and elbow joint level by rectangular pulses (duration 0.1 ms, frequency 1 Hz) with needle electrode. The voltage was set at the threshold required to obtain a supramaximal motor response. The CMAP was amplified and recorded with a Neuropack MEB2208 (Nihon-Kohden, Tokyo, Japan). The peak-to-peak amplitude of the CMAP was measured, and the latency of the response was measured to the first deflection from the baseline. The same procedure was performed on the contralateral side, and the amplitude of the CMAP and the motor nerve conduction velocity (MCV) were expressed as percentages of those of the contralateral side.
Nerve Conduction Studies
The median nerve was stimulated at the elbow joint level by rectangular pulses (duration 0.1 ms, frequency 1 Hz) with a stimulator (SEN-3401, Nihon-Kohden) using two stainless steel bipolar electrodes. The voltage was set at the threshold required to obtain a supramaximal nerve action response. The recording electrode (bipolar) was placed on the nerve at the wrist distal to the nerve lesion, and the compound nerve action potential (CNAP) was amplified with an amplifier (MEG-5200, Nihon-Kohden). The CNAP was recorded and analyzed with a PowerLab ML845 system (ADInstruments, Bella Vista, NSW, Australia) using Scope™3 v3.8.2 software (Physio-Tech Co., Tokyo, Japan). The same procedure was performed on the contralateral side, and the amplitude of the CNAP and the nerve conduction velocity (NCV) were expressed as percentages of those of the contralateral side.
Muscle Contraction Force and Wet Weight of the APB Muscle
The tetanic contraction force and wet weight of the APB muscle were measured bilaterally in both groups 16 weeks after the nerve resection. The APB muscle was exposed, and the insertion to the thumb proximal phalanx with the periosteum of the scaphoid was ligated with a 3-0 polyester suture and attached to a force transducer (TB-654T, Nihon-Kohden). The thumb and wrist were immobilized. The nerve was stimulated just proximal to the anastomosed site with supramaximal stimuli (rectangular shocks with a frequency of 100 Hz). The supramaximal isometric contraction force was recorded and analyzed with a PowerLab ML845 system using Chart™5 v5.5.1 software (Physio-Tech Co.). After measurement of muscle contraction forces, the APB muscle was harvested, and wet muscle weight was recorded. The tetanic contraction force and the wet weight of the APB muscle were expressed as percentages of those of the contralateral side.
Apple Pinch Test
A reach-pinch-retrieval task of a piece of apple on a reward stage was done to evaluate thumb abduction and upper limb function with voluntary use. A reward stage with a 1-cm-wide groove was constructed. Reward stages were attached to the outside of the cage on both sides. A piece of apple cut into a 7-mm cube was placed in the middle of the reward stage, 10 cm from the cage window. This test was repeated 30 times on each side of the cage for each monkey, with 10 pieces of apple being placed on the right side of the cage, then 10 pieces on the left side of the cage, and repeated. A digital video camera was used to record the reach-retrieval sequence from lateral view points. Still images were created from the digital videos, and thumb abduction during grasping of the apple was evaluated. The time from when the monkey's fingers left the cage until the fingers were completely inside the cage again were measured in 1/30 frame intervals. The mean time of the 20 trials between the 6th and 25th trials was calculated.
Gross Observation of the Lengthened Nerve
Gross observations of the lengthened nerve of monkeys in the nerve-lengthening group at the end of nerve lengthening and during final inspections were recorded. In addition, the distance that the epineurium lengthened was also measured. A marking suture was placed 1 mm apart from the ring, and the distance between the ring and the marking suture was measured during neurorrhaphy following completion of nerve lengthening (Fig. 1C).
Nerve regeneration was evaluated 16 weeks after nerve resection. After perfusion with 4% paraformaldehyde in 0.2 M sodium phosphate buffer (pH 7.2) via the left ventricle, the whole nerve segment was dissected out in both groups. At a point 10-mm proximal to the median nerve entrance into the APB muscle, 1-µm-thick transverse sections were cut and stained with toluidine blue. The total number, fiber diameter (D), and axonal diameter (d) of myelinated fibers were calculated in all fields at ×400 magnification. The mean axonal diameter was calculated. Myelin index; the ratio of axonal diameter to fiber diameter (d/D) was calculated for assessment of axonal maturation.
All data are presented as means ± SD. For the electrophysiological evaluation, Student's t-test was used to evaluate the differences between the two groups. For the apple pinch test and the histological evaluation, one-factor ANOVA and the Tukey–Kramer post hoc test were used. p < 0.05 was considered significant differences.
Evaluation for Treatment-Related Side Effects
There was no evidence of pain-related behaviors such as hand withdrawal or vocalization during nerve lengthening. No animals in either group developed wounds due to finger ulceration or autotomy. In the weekly X-ray examinations, none of the three animals in the nerve-lengthening group had fractures or damaged/removed lengthening devices.
APB needle electromyography was performed at 4, 8, 12, and 16 weeks after surgery. In all animals, only a denervation pattern was present at 8 weeks (Fig. 2a); however, at 12 weeks, a polyphasic wave that indicated re-innervation was detected (Fig. 2b). Because of this, 16 weeks (Fig. 2c) was established as the evaluation date; results of electrophysiological assessments at the 16th week are shown in Figures 2 and 3. Compared to the contralateral (healthy) sides, CMAP and CNAP wave patterns in both groups were low-amplitude with a long duration (Fig. 2d–g). At 16 weeks, MCV and NCV in the lengthening group were significantly higher than in the autografting group (Fig. 3).
Apple Pinch Test
Pictures of thumb abduction during grasping for all animals are presented in Figure 4. Before the operation, all animals were able to quickly grab the apple from the groove with the index finger and thumb. Thumb abduction was fully possible, and pictures of the moment when the apple was grasped reveal that the index finger and thumb form an “O-shape.” At 8 weeks following the operation, all subjects were unable to use the thumb to grasp the apple in the narrow groove because they were unable to perform sufficient thumb abduction. Instead, subjects used their index and middle fingers to scoop the apple out of the groove in order to pick up the apple. However, after 16 weeks, subjects were once again able to use their thumbs to pick up the apple, although movement was awkward. There were 3 animals in the nerve-lengthening group (Fig. 4a) and 1 in the autografting group (Fig. 4b) that recovered enough to make an O-shape with the thumb and index finger in a similar manner to that before the operation. However, two monkeys (No. 5 and 6 animals) in the autografting group are unable to grasp the apple with thumb and index finger. The required time for pinch task was shown in Figure 4c. In the lengthening group, the time required was significantly shorter at 16 weeks than at 8 weeks; there were no other significant differences between 8 and 16 weeks in the autografting group. These results suggest that the ability to perform the pinch task is recovered in the lengthening group.
Gross observation of the nerve in the nerve-lengthening group indicated that, following nerve lengthening, the rings of the proximal and distal stumps were crossing over at a sufficient distance for direct neurorrhaphy (Fig. 5a). The distance between the ring and marking suture grew from 1 mm at the beginning to an average of 11.5 mm in the proximal stump and 5.84 mm in the distal stump. When both stumps were refreshed, there was satisfactory bleeding from the stumps, indicating that intraneural blood flow was maintained.
Both stumps could be sutured without tension (Fig. 5b). During final observations, the nerve trunk appeared to be more-or-less normal and was recovered to the point where the anastomosis site was indiscernible (Fig. 5c).
Examples of tissue specimens at 16 weeks are shown in Figure 6. A large number of regenerated fibers was observed in both groups. Axon diameter was slightly larger in the nerve-lengthening group; the axon density also appeared to be higher. On quantitative histological evaluation, in the total number of axons and axon diameter, there were no statistical differences between two groups (Fig. 7a and b). The myelin index in the lengthening group was 0.06 ± 0.006 compared to 0.61 ± 0.008 in the autografting group. The Tukey–Kramer test p-value was 0.01, indicating significantly better axonal maturation in the nerve-lengthening group (Fig. 7c).
In this study, a 20-mm defect in the median nerve of monkeys' forearms was successfully repaired through the simultaneous gradual lengthening of both nerve stumps. In order to determine the efficacy of this method, the nerve regeneration ability after repair was compared between the nerve-lengthening method and the standard autografting method. In addition, we also investigated possible adverse events associated with this treatment, including observation for pain-related behaviors.
Nerve regeneration was evaluated at 16 weeks after nerve resection in both groups. In other words, the final day of observation was 16 weeks after the nerve-lengthening operation and 13 weeks after neurorrhaphy in the nerve-lengthening group, and 16 weeks following nerve transplantation in the autografting group. This was chosen to simulate actual clinical settings where it would be necessary to determine which method produces faster, higher quality nerve regeneration when patients with nerve defects arrive in a hospital.
Monkeys are an extremely important experimental animal in Japan, and the number of monkeys usable in this experiment was set by the monkey research center at 3 in each group; this greatly limited the statistical power of our study. However, despite these conditions, conduction velocity and the myelin index, which reflect axon maturation, were still significantly higher in the nerve-lengthening group. The nerve-lengthening group showed tendencies to be good regeneration on all parameter. In comparing individuals, there were no animals in the nerve-lengthening group with lower values than any animal in the autografting group, suggesting that recovery is superior with nerve lengthening than with autografting. Although the nerve-lengthening group received neurorrhaphy later than the autografting group, nerve regeneration might be already occurring during lengthening. These results show that there was no loss of time compared to autografting with regard to axon regeneration speed.
Pain associated with this treatment has been previously investigated in rats.9 Previous reports in a rat model of proximal stump lengthening of the sciatic nerve indicated no incidence of autotomy, allodynia, increased pain-related cytokine expressions, or neuropathic pain from abnormal termination of A-β and C fibers during nerve lengthening until nerve regeneration. In this monkey study, there was no evidence of pain-related behavior or allodynia during nerve lengthening or during the observation period until the final check following neurorrhaphy. In addition, there were no adverse events related to the lengthening device, such as fractures or damage to the device. From these results, we are able to confirm the efficacy and safety of this treatment.
Poor outcome in the treatment of long nerve defects is a major problem in neurosurgery. Furthermore, it is preferable to avoid sacrificing other nerves if possible. Research has been conducted to develop new treatment methods using neurotrophic factors, cultured cells, artificial nerves, various nerve conduits, allografts, or a combination of these in order to increase axon regenerating speed.10 We believe that reducing the number of anastomosis sites to one is the easiest method to improve nerve regeneration in the treatment of peripheral nerve defects. This concept is fundamentally different from using artificial nerves. There is a recent report of satisfactory nerve regeneration in repairing a long defect through the combined use of neurotrophic factors, introduced cells, and allografting11, 12; however, our method uses no medications, and there is no need to suppress the immune response. Our method uses mechanical stress, effectively using own tissue to produce nerve regeneration in vivo.
There are reports from other research groups who have investigated treatments using nerve lengthening in order to repair defects13–21; however, our method is unique in that it uses external fixation to linearly lengthen both nerve stumps. This method has the advantage of control of speed, distance, and direction. The elongation speed for both nerve stumps was set at 1 mm/day, the same rate as for regenerating axon growth. Based on this research, we believe that it is possible to repair a 20-mm nerve defect in humans using this same speed; however, it may be necessary to reduce the elongation rate for repairing longer gaps. In humans, it may be possible to divide up 1 mm/day elongations over several time periods, which may be a more effective method, as is the case in distraction osteogenesis.22 Although the maximum length of defects repairable through this method is still uncertain, we have succeeded in repairing a 30-mm defect in monkeys with similar results as for 20-mm defects. It is still necessary to address these remaining issues before turn our method to general use, clinical trials have already begun. We are conducting further research in order to develop this method as an effective new treatment for nerve defects.
In conclusion, we treated a 20-mm defect in monkeys and observed functional recovery with no adverse events through multiple types of evaluation, suggesting that this method is practical for clinical application.
This work was supported by Grant-in-Aid for Scientific Research (Kakenhi No. 19390387) from Japan Society for the Promotion of Science. We thank Fumiko Ono, Naoyuki Saito, Hayato Narita, and their colleague, in the Tsukuba Primate Research Center for assistant of animal experiment. We have no conflicts of interest about this work.