Dean G. Sotereanos, Anthony V. Seaber, James R. Urbaniak, David A. Spiegel, Donna Sotereanos, and Douglas C. Anthony
REVERSING NERVE-GRAFT POLARITY IN A RAT MODEL: THE EFFECT ON FUNCTION Downloaded by: Universite Laval. Copyrighted material.
ABSTRACT To evaluate the effect of nerve-graft polarity on function, a 1-cm segment of sciatic nerve was excised and reoriented in three groups of 20 adult Sprague-Dawley rats. In one group, the nerve was cut and anastomosed in the original orientation to act as a control. In the second group, the nerve-graft polarity was reversed 180°. The final group underwent reversal of polarity 180° and was rotated 180°, {i.e., the posterior tibial nerve fascicles proximally were opposite to the peroneal nerve fascicles distally). Functional recovery was evaluated using Bain's modification of de Medinaceli's sciatic functional index (SFI). Rat-track analysis was performed over a 120-day period. Histologic correlation was also performed at the time of sacrifice. From our study, we concluded that reversing nerve-graft polarity, with or without rotation, does not influence subsequent function in this model.
The present study was designed to determine whether reversal of a nerve graft would result in decreased function, compared to that of the usual, normally polarized, nerve graft. We attempted to answer the question of what effect reversal of nerve-graft polarity would have on function, as determined by Bain's modification of de Medinaceli's sciatic functional index.1-3 Stromberg has shown that the nerve conduction velocity and nerve amplitude of reversed and non-reversed nervegraft are identical after six months.4 He also showed the grafted nerve to be histologically indistinguishable from the control, whether or not the segment was reversed. He did not, however, evaluate function in his experiments.
MATERIALS AND METHODS Three groups of 20 rats each were selected. After preoperative functional testing, a 1-cm segment of sciatic nerve prior to branching5 was excised and reoriented in two of the three groups. In one of the two
groups, the nerve-graft polarity was simply reversed 180°. The other group was reversed 180° and rotated 180°. In the third group, a 1-cm segment was cut and anastomosed in the original orientation, to act as a control (Figs. 1, 2). All surgeries were performed randomly. The 60 rats underwent exposure through a gluteal splitting incision by one surgeon on the right lower limb. The neurorrhaphies were also performed by one surgeon under the operating microscope using 10-0 nylon suture. The National Research Council's guide for the care and use of laboratory animals was utilized for this experiment. Walking tracks were recorded preoperatively, postoperative day 1, then sequentially every 10 days for a total of 90 days. Gait analysis was also performed randomly. One final run of all rats on postoperative day 120 prior to sacrifice was also performed. The track analysis was performed manually by two observers blinded to which group of rats was being tested. After conditioning runs, animals were tested in a confined walkway 8.7 cm wide by 43 cm long, with a dark shelter at the end. The tracks were recorded on Bromphenol Blue paper.6 Bain's modification of the
Department of Orthopaedic Surgery, University of Pittsburgh, PA and the Orthopaedic Research Laboratories, Department of Surgery and the Department of Pathology, Duke University Medical Center, Durham, NC Reprint requests-. Dr. Sotereanos, University Orthopaedics, Inc., Falk Clinic, 3601 Fifth Ave., Pittsburgh, PA 15213 Accepted for publication January 2,1992 Copyright © 1992 by Thieme Medical Publishers, Inc., 381 Park Avenue South, New York, NY 100U All rights reserved.
303
JOURNAL OF RECONSTRUCTIVE MICROSURGERY/VOLUME 8, NUMBER 4
JULY 1992
In the experimental time span, 46 percent of the initial rat number (28/60) were removed from the ex1 cm periment secondary to toe chewing. Rats were removed sequentially from each experimental group Cutaneous Branch during the course of the experiment, as determined by Sural Nerve the extent of toe chewing. During the final analysis of Peroneal Nerve walking tracks on postoperative day 120, nine rats remained in the control group, 11 in the reversed and Tibial Nerve rotated group, and 12 in the reversed group. These rats incurred no toe loss secondary to chewing, and had good tracks for accurate analysis. Figure 1. A 1 -cm portion of sciatic nerve excised proxiDuring the 120 day trial, SFI improved in all mal to branching, as shown. groups, without significant difference in regard to nerve-graft orientation (p > .1). The SFI for all 60 rats on postoperative day 1 was -96.1 ± 22.4. This improved to -74.3 ± 25.1 on postoperative day 120 de Medinaceli sciatic functional index (SFI) was then (the final analysis) for the remaining 32 rats. utilized to analyze the rat tracks. Print length improved significantly over the 120 After sacrifice, Epon sections proximal, through, days, from an average of 42.3 ± 3 mm for all 60 rats and distal to the nerve grafts were taken on random on postoperative day 1 to 28.2 ± 2 mm for the 32 rats from each group. The animals were anesthetized remaining rats on day 120. Of note was the significant with chloral hydrate (400 mg/kg), and perfused through improvement in print length between postoperative the ascending aorta with an initial saline perfusate (0.9 day 90 and postoperative day 120. Preoperative print percent NaCl 0.2 percent sodium nitroprusside, 2.5 length for all 60 rats averaged 28.1 ± 4 mm. percent polyvinyl-pyrrolidone), followed by 4 percent Experimental total and intermediary toe spread glutaraldehyde in buffered 0.1 M sodium cacodylate. did not improve in any of the three groups from postAfter fixation for 24 hr at 4° C, the sciatic nerve was operative day 1 to postoperative day 120 (Figs. exposed, and microscopic sections were prepared 3, 4). from three locations; at the graft site, at a point 1 cm Histologic sections proximal to the graft site proximal to the graft, and at a point 1 cm distal to the showed a normal axonal population, with a typical graft. The tissue sections were post-fixed in osmium bimodal appearance of axonal diameters. At the graft tetroxide, dehydrated through graded ethanol, cleared site, there was a variable amount of perineural fibroin propylene oxide, and embedded in Epon 812. Sec- blastic scarring and, in some animals, there was a tions of one l-|xm thickness were prepared and stained population of axons within the perineural connective with toluidine blue. tissue. Perineural cells extended into the endoneural
180c
b)
±) ±)
180°+
c)
304
±)
Figure 2. Diagrammatic representation of nerve reorientation in each group, a—polarity reversal of 180°; b—polarity reversal of 180° and rotation of 180° (i.e., posterior tibial nerve fascicles proximally were opposite to the peroneal nerve fascicles distally); c—nerve simply cut and repaired in original orientation (control).
Downloaded by: Universite Laval. Copyrighted material.
RESULTS
REVERSING NERVE-GRAFT POLARIT//SOTEREANOS, SEABER, URBANIAK, ET AL.
PRE-OP
space and surrounded small groups of fibers. The overall axonal population within the grafts was diminished, compared to the proximal site; however, there were large numbers of fibers in each of the animal specimens. These fibers had extended through the graft in each of the animals examined, and had partially repopulated the nerve at I cm distal to the graft. At this site, however, regeneration was not yet complete, with the population composed of a smaller average axonal size (Fig. 5).
Figure 4. Graphic analysis of results (SFI = sciatic functional index, PL = print length, TS = toe spread, ITS = intermediary toe spread, C = control, 180+ = reversed and rotated, 180 = reversed, and T = total. PL, TS, ITS in mm). Note improvement in SFI and PL; also note lack of improvement in TS and ITS.
POD 120
POD 1
Downloaded by: Universite Laval. Copyrighted material.
Figure 3. Track samples preoperatively, POD l ? and POD 120. Note the improvement in print length and lack of toe-spreading over the 120 day experiment.
DISCUSSION To our knowledge, no author has attempted to study functional recovery post-reversal of autogenous nerve graft. Dellon and Mackinnon7 have shown that walking track analysis after neurotmesis is extremely reliable over a 3 to 4 month period. Longer follow-up may not be as reliable secondary to the development of chronic contracture.
preop
1
30
60
90
120
preop
1
30
60
90
120
CO
preop 1
30
60
Days
90
120
Days
305
JULY 1992
B
306
C !
Figure 5. A, Section from 1 cm proximal to graft showing normal axonal population. B, Section from graft showing variable perineural fibroblastic scarring. The axon population within the graft is diminished, compared to the proximal sites; however, large numbers of fibers are still seen. C, Section 1 cm distal to graft. Regeneration not yet complete, with population composed of smaller average axon size.
The present study attempted to quantify motor recovery post-reversal of sciatic nerve-graft polarity via gait analysis. Thirty-two rats (54 percent of the total) had adequate enough tracks to analyze 4 months postoperatively. Our analysis indicated a direct correlation between improved print length and improved
SFI. Of note was the marked improvement between postoperative day 90 and postoperative day 120 in print length and SFI. Toe spread and intermediary toe spread did not change between postoperative day 1 and day 120 (see Fig. 4). Complete sciatic nerve transection in the rat re-
Downloaded by: Universite Laval. Copyrighted material.
JOURNAL OF RECONSTRUCTIVE MICROSURGERY/VOLUME 8, NUMBER 4
REVERSING NERVE-GRAF POLARITY/SOTEREANOS, SEABER, URBANIAK, ET AL.
in rats (regardless of orientation) act as conduits for regenerating axons. In the reversed and rotated nerve group, inappropriate fascicular alignment resulted in sciatic functional indices similar to unrotated nerves. Some allowance for correction of inappropriate fascicular alignment must have occurred, to enable the reversed and rotated nerve group to have sciatic functional indices similar to unrotated nerves.
CONCLUSIONS This study demonstrated that reversing sciatic nerve-graft polarity in the rat, with or without rotation, does not influence subsequent motor function. The short-term (4-month) results of this study revealed significant return of proximal muscle-group function based on print length, and no return of intrinsic musculature based on toe-spread analysis. Downloaded by: Universite Laval. Copyrighted material.
suits in an increased print length, decreased toe spread, and increased ankle dorsiflexion. Isolated sectioning of the posterior tibial nerve results in a similar dysfunction. Isolated peroneal nerve sectioning results in a decrease in print length, only a slight decrease in toe spread, and decreased ankle dorsiflexion. This indicates a more significant contribution from the posterior tibial nerve than from the peroneal, as confirmed by Bain and colleagues.1 The improvement in SFI and print length was similar in all of our groups (control, reversed, reversed and rotated). This indicates a reinnervation of the posterior tibial innervated musculature, as determined by the decreased print length and decreased SFI in all of our groups. The occurrence of these recovery signs between 3 and 4 months postoperatively is noteworthy, since the nerve was sectioned approximately 1 to 2 cm above the sciatic nerve division. These findings indicate that it takes 3 to 4 months for the regenerating fibers from the sciatic and posterior tibial nerves to reach the posterior compartment musculature. Toe spread and intermediary toe spread did not improve in any of our groups, indicating a lack of intrinsic muscular return. Insufficient time for regenerating axons to reach these muscles could explain these findings. A race between reinnervation and intrinsic contracture could also result in the lack of toe spread. Another explanation could be the failure of specific axons to reach their destination to the motor end plates of the intrinsic muscles. These axons may have been channelled into other sciatic nerve divisions (peroneal, sural, cutaneous). Histologic sections obtained through the graft and distal to the graft in each group (control, reversed, reversed rotated) revealed large numbers of axon fibers in the grafts. Sections taken 1 cm distal to the graft site demonstrated repopulation of the sciatic nerve with axons in each of the three groups. Axon counts were not recorded and correlated with sciatic functional indices in this study. Shenaq has reported that qualitative histologic and quantitative axon counts did not correlate with de Medinaceli's SFI.8 This study indicates that autogenous nerve grafts
REFERENCES Bain JR, Mackinnon SE, Hunter DA: Functional evaluation of complete sciatic, peroneal, and posterior tibial nerve lesions in the rat. Plast Reconstr Surg 83:129, 1989 de Medinaceli L, Freed WJ, Wyatt RJ: An index of the functional condition of rat sciatic nerve based on measurements made from walking tracks. Exp Neurol 77:634, 1982 Brown CJ, Mackinnon SE, Evans PL et
REVERSING NERVE-GRAFT POLARITY IN A RAT MODEL: THE EFFECT ON FUNCTION Downloaded by: Universite Laval. Copyrighted material.
ABSTRACT To evaluate the effect of nerve-graft polarity on function, a 1-cm segment of sciatic nerve was excised and reoriented in three groups of 20 adult Sprague-Dawley rats. In one group, the nerve was cut and anastomosed in the original orientation to act as a control. In the second group, the nerve-graft polarity was reversed 180°. The final group underwent reversal of polarity 180° and was rotated 180°, {i.e., the posterior tibial nerve fascicles proximally were opposite to the peroneal nerve fascicles distally). Functional recovery was evaluated using Bain's modification of de Medinaceli's sciatic functional index (SFI). Rat-track analysis was performed over a 120-day period. Histologic correlation was also performed at the time of sacrifice. From our study, we concluded that reversing nerve-graft polarity, with or without rotation, does not influence subsequent function in this model.
The present study was designed to determine whether reversal of a nerve graft would result in decreased function, compared to that of the usual, normally polarized, nerve graft. We attempted to answer the question of what effect reversal of nerve-graft polarity would have on function, as determined by Bain's modification of de Medinaceli's sciatic functional index.1-3 Stromberg has shown that the nerve conduction velocity and nerve amplitude of reversed and non-reversed nervegraft are identical after six months.4 He also showed the grafted nerve to be histologically indistinguishable from the control, whether or not the segment was reversed. He did not, however, evaluate function in his experiments.
MATERIALS AND METHODS Three groups of 20 rats each were selected. After preoperative functional testing, a 1-cm segment of sciatic nerve prior to branching5 was excised and reoriented in two of the three groups. In one of the two
groups, the nerve-graft polarity was simply reversed 180°. The other group was reversed 180° and rotated 180°. In the third group, a 1-cm segment was cut and anastomosed in the original orientation, to act as a control (Figs. 1, 2). All surgeries were performed randomly. The 60 rats underwent exposure through a gluteal splitting incision by one surgeon on the right lower limb. The neurorrhaphies were also performed by one surgeon under the operating microscope using 10-0 nylon suture. The National Research Council's guide for the care and use of laboratory animals was utilized for this experiment. Walking tracks were recorded preoperatively, postoperative day 1, then sequentially every 10 days for a total of 90 days. Gait analysis was also performed randomly. One final run of all rats on postoperative day 120 prior to sacrifice was also performed. The track analysis was performed manually by two observers blinded to which group of rats was being tested. After conditioning runs, animals were tested in a confined walkway 8.7 cm wide by 43 cm long, with a dark shelter at the end. The tracks were recorded on Bromphenol Blue paper.6 Bain's modification of the
Department of Orthopaedic Surgery, University of Pittsburgh, PA and the Orthopaedic Research Laboratories, Department of Surgery and the Department of Pathology, Duke University Medical Center, Durham, NC Reprint requests-. Dr. Sotereanos, University Orthopaedics, Inc., Falk Clinic, 3601 Fifth Ave., Pittsburgh, PA 15213 Accepted for publication January 2,1992 Copyright © 1992 by Thieme Medical Publishers, Inc., 381 Park Avenue South, New York, NY 100U All rights reserved.
303
JOURNAL OF RECONSTRUCTIVE MICROSURGERY/VOLUME 8, NUMBER 4
JULY 1992
In the experimental time span, 46 percent of the initial rat number (28/60) were removed from the ex1 cm periment secondary to toe chewing. Rats were removed sequentially from each experimental group Cutaneous Branch during the course of the experiment, as determined by Sural Nerve the extent of toe chewing. During the final analysis of Peroneal Nerve walking tracks on postoperative day 120, nine rats remained in the control group, 11 in the reversed and Tibial Nerve rotated group, and 12 in the reversed group. These rats incurred no toe loss secondary to chewing, and had good tracks for accurate analysis. Figure 1. A 1 -cm portion of sciatic nerve excised proxiDuring the 120 day trial, SFI improved in all mal to branching, as shown. groups, without significant difference in regard to nerve-graft orientation (p > .1). The SFI for all 60 rats on postoperative day 1 was -96.1 ± 22.4. This improved to -74.3 ± 25.1 on postoperative day 120 de Medinaceli sciatic functional index (SFI) was then (the final analysis) for the remaining 32 rats. utilized to analyze the rat tracks. Print length improved significantly over the 120 After sacrifice, Epon sections proximal, through, days, from an average of 42.3 ± 3 mm for all 60 rats and distal to the nerve grafts were taken on random on postoperative day 1 to 28.2 ± 2 mm for the 32 rats from each group. The animals were anesthetized remaining rats on day 120. Of note was the significant with chloral hydrate (400 mg/kg), and perfused through improvement in print length between postoperative the ascending aorta with an initial saline perfusate (0.9 day 90 and postoperative day 120. Preoperative print percent NaCl 0.2 percent sodium nitroprusside, 2.5 length for all 60 rats averaged 28.1 ± 4 mm. percent polyvinyl-pyrrolidone), followed by 4 percent Experimental total and intermediary toe spread glutaraldehyde in buffered 0.1 M sodium cacodylate. did not improve in any of the three groups from postAfter fixation for 24 hr at 4° C, the sciatic nerve was operative day 1 to postoperative day 120 (Figs. exposed, and microscopic sections were prepared 3, 4). from three locations; at the graft site, at a point 1 cm Histologic sections proximal to the graft site proximal to the graft, and at a point 1 cm distal to the showed a normal axonal population, with a typical graft. The tissue sections were post-fixed in osmium bimodal appearance of axonal diameters. At the graft tetroxide, dehydrated through graded ethanol, cleared site, there was a variable amount of perineural fibroin propylene oxide, and embedded in Epon 812. Sec- blastic scarring and, in some animals, there was a tions of one l-|xm thickness were prepared and stained population of axons within the perineural connective with toluidine blue. tissue. Perineural cells extended into the endoneural
180c
b)
±) ±)
180°+
c)
304
±)
Figure 2. Diagrammatic representation of nerve reorientation in each group, a—polarity reversal of 180°; b—polarity reversal of 180° and rotation of 180° (i.e., posterior tibial nerve fascicles proximally were opposite to the peroneal nerve fascicles distally); c—nerve simply cut and repaired in original orientation (control).
Downloaded by: Universite Laval. Copyrighted material.
RESULTS
REVERSING NERVE-GRAFT POLARIT//SOTEREANOS, SEABER, URBANIAK, ET AL.
PRE-OP
space and surrounded small groups of fibers. The overall axonal population within the grafts was diminished, compared to the proximal site; however, there were large numbers of fibers in each of the animal specimens. These fibers had extended through the graft in each of the animals examined, and had partially repopulated the nerve at I cm distal to the graft. At this site, however, regeneration was not yet complete, with the population composed of a smaller average axonal size (Fig. 5).
Figure 4. Graphic analysis of results (SFI = sciatic functional index, PL = print length, TS = toe spread, ITS = intermediary toe spread, C = control, 180+ = reversed and rotated, 180 = reversed, and T = total. PL, TS, ITS in mm). Note improvement in SFI and PL; also note lack of improvement in TS and ITS.
POD 120
POD 1
Downloaded by: Universite Laval. Copyrighted material.
Figure 3. Track samples preoperatively, POD l ? and POD 120. Note the improvement in print length and lack of toe-spreading over the 120 day experiment.
DISCUSSION To our knowledge, no author has attempted to study functional recovery post-reversal of autogenous nerve graft. Dellon and Mackinnon7 have shown that walking track analysis after neurotmesis is extremely reliable over a 3 to 4 month period. Longer follow-up may not be as reliable secondary to the development of chronic contracture.
preop
1
30
60
90
120
preop
1
30
60
90
120
CO
preop 1
30
60
Days
90
120
Days
305
JULY 1992
B
306
C !
Figure 5. A, Section from 1 cm proximal to graft showing normal axonal population. B, Section from graft showing variable perineural fibroblastic scarring. The axon population within the graft is diminished, compared to the proximal sites; however, large numbers of fibers are still seen. C, Section 1 cm distal to graft. Regeneration not yet complete, with population composed of smaller average axon size.
The present study attempted to quantify motor recovery post-reversal of sciatic nerve-graft polarity via gait analysis. Thirty-two rats (54 percent of the total) had adequate enough tracks to analyze 4 months postoperatively. Our analysis indicated a direct correlation between improved print length and improved
SFI. Of note was the marked improvement between postoperative day 90 and postoperative day 120 in print length and SFI. Toe spread and intermediary toe spread did not change between postoperative day 1 and day 120 (see Fig. 4). Complete sciatic nerve transection in the rat re-
Downloaded by: Universite Laval. Copyrighted material.
JOURNAL OF RECONSTRUCTIVE MICROSURGERY/VOLUME 8, NUMBER 4
REVERSING NERVE-GRAF POLARITY/SOTEREANOS, SEABER, URBANIAK, ET AL.
in rats (regardless of orientation) act as conduits for regenerating axons. In the reversed and rotated nerve group, inappropriate fascicular alignment resulted in sciatic functional indices similar to unrotated nerves. Some allowance for correction of inappropriate fascicular alignment must have occurred, to enable the reversed and rotated nerve group to have sciatic functional indices similar to unrotated nerves.
CONCLUSIONS This study demonstrated that reversing sciatic nerve-graft polarity in the rat, with or without rotation, does not influence subsequent motor function. The short-term (4-month) results of this study revealed significant return of proximal muscle-group function based on print length, and no return of intrinsic musculature based on toe-spread analysis. Downloaded by: Universite Laval. Copyrighted material.
suits in an increased print length, decreased toe spread, and increased ankle dorsiflexion. Isolated sectioning of the posterior tibial nerve results in a similar dysfunction. Isolated peroneal nerve sectioning results in a decrease in print length, only a slight decrease in toe spread, and decreased ankle dorsiflexion. This indicates a more significant contribution from the posterior tibial nerve than from the peroneal, as confirmed by Bain and colleagues.1 The improvement in SFI and print length was similar in all of our groups (control, reversed, reversed and rotated). This indicates a reinnervation of the posterior tibial innervated musculature, as determined by the decreased print length and decreased SFI in all of our groups. The occurrence of these recovery signs between 3 and 4 months postoperatively is noteworthy, since the nerve was sectioned approximately 1 to 2 cm above the sciatic nerve division. These findings indicate that it takes 3 to 4 months for the regenerating fibers from the sciatic and posterior tibial nerves to reach the posterior compartment musculature. Toe spread and intermediary toe spread did not improve in any of our groups, indicating a lack of intrinsic muscular return. Insufficient time for regenerating axons to reach these muscles could explain these findings. A race between reinnervation and intrinsic contracture could also result in the lack of toe spread. Another explanation could be the failure of specific axons to reach their destination to the motor end plates of the intrinsic muscles. These axons may have been channelled into other sciatic nerve divisions (peroneal, sural, cutaneous). Histologic sections obtained through the graft and distal to the graft in each group (control, reversed, reversed rotated) revealed large numbers of axon fibers in the grafts. Sections taken 1 cm distal to the graft site demonstrated repopulation of the sciatic nerve with axons in each of the three groups. Axon counts were not recorded and correlated with sciatic functional indices in this study. Shenaq has reported that qualitative histologic and quantitative axon counts did not correlate with de Medinaceli's SFI.8 This study indicates that autogenous nerve grafts
REFERENCES Bain JR, Mackinnon SE, Hunter DA: Functional evaluation of complete sciatic, peroneal, and posterior tibial nerve lesions in the rat. Plast Reconstr Surg 83:129, 1989 de Medinaceli L, Freed WJ, Wyatt RJ: An index of the functional condition of rat sciatic nerve based on measurements made from walking tracks. Exp Neurol 77:634, 1982 Brown CJ, Mackinnon SE, Evans PL et
