- Email: [email protected]
Prosthetic nerve grafts: A resorbable tube as an alternative to autogenous nerve grafting
Prosthetic nerve grafts: A resorbable tube as an alternative to autogenous nerve grafting The use of a prosthetic nerve graft, composed of a resorbable polyorthoester tube, as an alternative to free autogenous nerve grafting for the treatment of a gap in a peripheral nerve was studied, with a cat sciatic nerve as the model. The results demonstrate that regeneration will occur through a resorbable tube spanning a 1.5 cm gap and reinnervate end organ muscle. In those muscles showing evidence of reinnervation, nerve regeneration through the tubes as assayed by electrophysiologic examination demonstrated no difference compared with autogenous nerve grafts, with the exception that the initial rate of regeneration was delayed by 4 to 6 weeks. (J HAND SURG 1987;12A[2 Pt 1]:685-92.)
F. William Bora, Jr., M.D., John M. Bednar, M.D., A. Lee Osterman, M.D., Mark J. Brown, M.D., and Austin J. Sumner, M.D., Philadelphia, Pa.
PeriPheral nerve injuries cause disability because of incomplete recovery of function despite optimal surgical treatment. These injuries are an especially difficult problem when a segmental loss of peripheral nerve is present. The autogenous nerve graft is the most common method to treat patients with a significant gap in peripheral nerve. 1-5 The technique is technically demanding, destroys the function of another nerve, and has unpredictable results. These deficiencies have encouraged the development of other methods. Several investigators6-9 have used nonneural tissue to bridge nerve defects with variable results. Lundborg and Hansson 7 • 8 used a pseudosynovial sheath, produced by an implanted silicone rubber rod to bridge 10 mm gaps in the sciatic nerves of rats which reinnervated the short flexors of the foot by 3 months. Rat sciatic nerves, however, will not regenerate through a tube spanning a gap greater than 10 to 12 mm. Molander et aU reported that a resorbable From the Department of Orthopaedic Surgery, Hospital of the University of Pennsylvania, Philadelphia, Pa. Received for publication Aug. 11, 1986; accepted in revised form Jan. 2, 1987. The author or one or more of the authors have received or will receive benefits for personal or professional use from a commercial party related directly or indirectly to the subject of this article. In addition, benefits have been or will be directed to a research fund, foundation, education institution, or other nonprofit organization with which one or more of the authors are associated. Reprint requests: F. William Bora, Jr., M.D., Department of Orthopaedic Surgery, Hospital of the University of Pennsylvania, Silverstein Pavilion - Second Floor, 3400 Spruce St., Philadelphia, PA 19104.
Fig. 1. Sciatic nerve with resorbable tube spanning 1.5 em gap.
polyglactin mesh tube conducted axons across a 9 mm defect in the tibial nerve of a rabbit. Histologic examination in this experiment showed newly formed fascicles, but no functional assessment of useful reinnerTHE JOURNAL OF HAND SURGERY
685
686
The Journal of HAND SURGERY
Bora et al.
through the epineurium at each end. Ten of the cats received a tube containing 10% cis-hydroxyproline by weight and ten received a plain tube. This produced three groups of nerves for comparison: 20 containing a nerve graft, 10 containing a tube with 10% cis-hydroxyproline, and 10 plain tubes. The cats were left unrestrained in their cages and assayed by macroscopic and microscopic histologic examination, electrophysiologic testing, and fast axonal transport kinetics.
HI LEUCINE
SUTURE LINE
I
SCIATtC NERVE
-r-a "".
~==:g;jl8J:===_lj
I.. ---------....
'D·
,08
:I: IL
()
iii iii iii
El El El
'07
PLATEAU
",., ,03
j
I
SECTIOH VIALS
eREST
,03 '0'
Macroscopic histology
BACKGROUND
20
0
20
<10
eo
10
100
120
140
mm
Fig. 2. Schematic drawing depicts the technique of fast axonal transport kinetics and graphic presentation of data.
vation was perfonned. These studies demonstrate that nerve regeneration will occur over short distances through almost any type of conduit, but they do not compare their technique with the current standard of treatment, the autogenous nerve graft. The concept of a resorbable prosthetic nerve graft has many advantages including the following: (1) it is technically simple, (2) no donor nerve is required , (3) nerve ends are free of suture, (4) it is a potential vehicle for trophic factors, and (5) a method to carry a drug to control scar fonnation. The goal of this work was to study regeneration through a resorbable tube with and without an antiscarring agent (cis-hydroxyproline) to detennine if it could be used as an alternative to autogenous nerve graft for the treatment of a gap in a peripheral nerve. Materials and methods There were 20 adult female cats that weighed 3 to 4 kg in the study. Under sodium pentobarbital (Nembutal) anesthesia, each had a 1.5 cm segment of sciatic nerve removed bilaterally. The left nerve gap was spanned by a 1.5 cm nerve graft from the contralateral side and secured with five 10-0 nylon epineural sutures at each suture line. The right nerve gap was then spanned with a 2 cm long tube, with an inner diameter of 2.8 mm, with the cut ends of the nerve tucked into the tube to leave a 1.5 cm gap (Fig. I) . The ends were secured to the tube by two 9-0 nylon sutures placed
Nerves were operatively exposed and examined to detennine the percentage of tube remaining and the tissue present in the previous nerve gap. Two nerve grafts, one tube wi 'h and one tube without cishydroxyproline were examined at 6 weeks postoperatively. Eight nerve grafts, four tubes with and four tubes without cis-hydroxyproline were examined at 12 weeks postoperatively. (Note the 12-week cats were those assayed by fast axonal transport kinetics and were examined at autopsy.) Microscopic histology Twenty-eight weeks postoperatively ten nerve grafts, five nerves containing a tube with cis-hydroxyproline, and five containing a plain tube were assayed by microscopic histologic examination. The sciatic nerves were exposed from hip to knee and fixed in situ with 3.6% glutaraldehyde solution for 10 minutes . The nerves were then removed and further fixed in a 3.6% glutaraldehyde for 2 hours. A 5 mm section located 1 em distal to the distal anastomosis was then isolated and washed overnight in O.lm phosphate buffer solution, pH 7.4. The tissue was then postfixed with 2% osmium tetroxide for 2 hours, dehydrated in graded ethanol and propylene oxide, and embedded in epoxy. Toluidine blue-stained cross sections, one micron thick, were then prepared and studied by light microscopy. Electrophysiologic testing Nerve conduction velocity was measured by supramaximal stimulation of the sciatic nerve proximal to the nerve graft or tube. Insertional activity and fibrillation potentials were noted . The presence of an evoked multiphasic action potential recorded from the medial head of the gastrocnemius muscle and intrinsic muscles of the foot was interpreted as evidence of reinnervation of these muscles. The motor latency from stimulating signal to the onset of recorded action potential was measured in milliseconds. Ten cats were studied (10
Vol. l2A, No.5, Part I September 1987
Fig. 3. Sciatic nerve graft 12 weeks postoperatively.
nerve grafts, five tubes with cis-hydroxyproline, and 5 plain tubes) every 2 weeks until the cats were killed at 28 weeks.
Fast axonal transport kinetics The rate and extent of regeneration through the tubes as compared with nerve grafts was assayed by fast axonal transport kinetics in eight nerve grafts, four tubes with cis-hydroxyproline, and four plain tubes at 12 weeks postoperatively. Forman and Berenberg lO have shown that regenerating axons can be labeled with axonally transported radioactive proteins to provide information about the location of the entire range ofaxons from the fastest growing ones to those trapped in scar at the suture site. The method of Forman and Berenberg, with modification by Politis and Spencer, II to adapt the technique to cats was used, 3H-leucine (NET 135H, New England Nuclear Corp, Boston, Mass.) 40 to 60 Ci/flmol was concentrated to 20 flCi/flL in Hank's balanced salt solution. The cat was immobilized in a stereotaxic frame, and under Nembutal anesthesia, a wide L4 and 5 laminectomy was performed followed by durectomy. Ten pairs of 0.5 flL injections of 3H_
Prosthetic nerve grafts
687
Fig. 4. Sciatic nerve regenerating through tube containing cis-hydroxyproline at 6 weeks postoperatively. Note 60% to 70% of tube has resorbed and is replaced by a nerve like cord.
leucine were made with 10 flL Hamilton syringe held in a micromanipulator attached to the stereotaxic frame. The injections began 2.5 cm rostral to the conus and were made 1 mm from the midline, 3 mm deep, with 2 mm intervals between injections in the rostral-caudal direction. The animal's body temperature was carefully monitored during and after the procedure to maintain the temperature in the range of 100° F to 102° F with external warming. Fourteen hours after isotope injection, the animals were deeply anesthetized with Nembutal, heparin was administered and perfused through the heart with a 2% formaldehyde saline solution. Both intact sciatic nerves, one median nerve, and one ulnar nerve were removed and cut into 5 mm segments. The corresponding segments of the peroneal and tibial nerves beyond the bifurcation were added together. The segments were solubilized for 16 hours in tissue solubilizer (Protosol, New England Nuclear Corp, Boston, Mass.) at 45° C, 5 cc of scintillation cocktail added, and radioactivity counted in an Intertechnique SL30 Liquid Scintillation
688
The Journal of HAND SURGERY
Bora et al.
Fig. 5. Sciatic nerve regenerating through tube containing cis-hydroxyproline at 12 weeks postoperatively. Note tube is completely resorbed and replaced by regenerated nerve.
Fig. 6. Sciatic nerve regenerating through plain tube at 12 weeks postoperatively. Note tube is completely resorbed and replaced by thin nerve like structure.
Counter. Results were expressed as counts per minute per 5 mm nerve segment and graphed versus position along the nerve to allow calculation of the rate of regeneration, degree of axonal trapping, and position of the axonal growth cone (Fig. 2). The median and ulnar nerves serve as a control to establish the level of background radiation.
6 weeks (Fig. 4) and totally resorbed at 12 weeks. The nerves spanned by tubes containing cis-hydroxyproline had a distinct "nerve like" cord present filling the gap, which was similar in diameter to the nerve grafts (Fig. 5). The nerves spanned by plain tubes also had nerve growth across the tube but with a consistently smaller diameter cord (Fig. 6).
Statistical analysis
Microscopic histologic examination
Chi-square analysis and Student's t-test statistical analysis were used to evaluate the data.
Microscopic analysis of sections taken distal to the grafts or tubes reveal regenerating axons in a fascicular pattern (Fig. 7). The number ofaxons, axonal area, and degree of myelination grossly appeared very similar in those specimens in which regeneration had occurred through a nerve graft as compared with those regenerating through a tube with or without cishydroxyproline. Quantitative axon analysis was not performed.
Results All cats recovered from the operation and none had infections in the wound. Neurotrophic plantar ulcers occurred on the heels of 10% of the cats randomly on the sides with both nerve grafts and tubes.
Macroscopic histologic examination All nerve grafts studied at all time intervals were intact, with small neuromas at both suture lines (Fig. 3) . The tubes were 60% to 70% resorbed by
Electrophysiologic testing Electrical evidence of reinnervation of the gastrocnemius muscle at 8 to 10 weeks and the intrinsic mus-
Vol. 12A, No.5, Part I September 1987
Prosthetic nerve grafts
689
Fig. 7. Histologic sections of regenerating sciatic nerve taken distal to the repair site at 28 weeks postoperatively. A, Nerve graft. (Magnification x 60).) B, Tube with cis-hydroxyproline. (Magnification x 60.) C, Plain tube. (Magnification x 60.) D, Nerve graft. (Magnification x 200.) E, Tube with cis-hydroxyproline. (Magnification x 200.) F, Plain tube . (Magnification x 200.) Note fascicular pattern and uniformity of myelination .
cles of the foot by 18 to 20 weeks occurred in 100% of the nerve grafts (Fig. 8) . Eighty percent of those nerves containing a tube with cis-hydroxyproline demonstrated reinnervation of the gastrocnemius occurring
first at 12 to 14 weeks, with reinnervation of the foot by 24 to 26 weeks. Sixty percent of those nerves containing a plain tube demonstrated reinnervation of the gastrocnemius by 14 to 16 weeks, with reinnervation
690
The Journal of HAND SURGERY
Bora et al.
ELECTROPHYSIOLOGIC EVIDENCE OF REINNERVATION
Peripheral Nerve Regeneration Measured By Fast Axonal Transport
25
Nerve Graft
.
12 WEEKS POST OP
Gastrocneni.us
Foot Intrinsic Muscles
8 - 10 wks.
18 - 20 wks.
Tube with 10\ Cis-hydroxypro1 ine
12 - 14 wks.
24 - 26 wks.
Plain Tube
14 - 16 wks .
26 - 28 wks.
20
..
-
0
weeks post - operative
Fig. 8. Regeneration as detected by evoked muscle action potentials.
.
10 1. 5 em Nerve Or.tt
)(
c
l t'l
•
~
0
CI.
Peripheral Nerve Regeneration After Neurorrhaphy Measured By Fast Axonal Transport
:; c: :!
.
10
CI.
"' C :0
20
0
U 0
-
N
2.0 em Alzeme' Tube ContakMng 10" eta-Hydroxyproline Brlelgk1; 1. 5 em Gap
Q)
•
"
0
)(
10 10
E
0>
(/)
•
_
_
_
_
_
_
BACKGROUND
I»
CI.
:; c:
25
30 Ankle
Fig. 10. Fast axonal transport kinetics-axonal labeling profile for a nerve graft, tube containing cis-hydroxyproline and plain tube at 12 weeks after surgery. Note position of axonal growth cone.
4 Weeks Post Op
:i I»
Number of specimens
CI.
"E"':0
20
Segment Number
5 Weeks Post Op
"E
"
0
3 Weeks Post Op _ BACKGROUND
o
Sc iatic Notch
•
20
Suture Line
25
30 Ankle
Segment Number
Fig. 9. Fast axonal transport kinetics-axonal labeling profile for a simple neurorrhaphy at 3, 4, and 5 weeks after surgery.
of the foot by 26 to 28 weeks. Chi-square analysis demonstrated no significiant difference in percentage of nerve reinnervated for all three groups. Motor latency measured at the medial head of the gastrocnemius muscle at 28 weeks:
Average (msec) Standard deviation
Tubes with Nerve graft cis-hydroxyproline 3.6 4.1 1.2
1.4
4
3
Statistical analysis by Student's t-test showed no significant difference between either tube and nerve· graft or between tubes.
0
U
10
Plain tube 7.1
4.8
Motor latency measured at the intrinsic muscles of the foot at 28 weeks:
Average (msec) Standard deviation Number of specimens
Nerve Tubes with graft cis-hydroxyproline 8.0 12.9 1.9 10
Plain tube 11.8
6.1
7.2
4
3
Statistical analysis by Student's t-test showed no significant difference between either tube and nerve graft or between tubes.
Fast axonal transport kinetics This method uses the principle that nerves are composed of a cell body that supplies the materials for repair of injured axons by a sophisticated transport system.
Vol. 12A, No.5, Part 1 September 1987
Prosthetic nerve grafts
The proteins carried by this transport system can be labeled by stereotaxic injection of tritiated amino acids into the ventral motor horns of the lumbar spinal cord supplying the sciatic nerve. The labeled proteins are transported along the regenerating axons and allow the accurate determination of the position of the distal most axons by scintillation counting. Fig. 9 is a profile of the distribution of label along a sciatic nerve that has had a simple neurorrhaphy and was allowed to regenerate for 3 to 5 weeks before labeling. The peak present at segment five, the suture line, represents those axons in the neuroma at the suture line. This peak remains stationary at all time periods measured. The point at which the graph crosses baseline counts represents the distal most progress of the regenerating axons. As the time between repair and label increases from 3 to 5 weeks, this intersection with baseline or distal most point of regeneration, moved at a rate of 1 cm per week. This progression represents the rate of axonal regeneration. No attempt was made to correlate the area under the curve with the number of regenerating axons. Fig. 10 represents a composite of the labeling profile for a nerve graft, a tube with and without cis-hydroxyproline at 12 weeks. An average of profiles for all specimens was not plotted because of the variation of specific activity of the 3H-Ieucine between assays. Note that the regeneration has progressed well past the graft or tubes but at a differential rate consistent with the rates observed by electrophysiologic testing. (See previous section.) The point of the distal most regeneration was:
Nerve graft Tube with cis-hydroxyproline
Segment number 29.00
p < 0.05
21.00 P
Plain tube
18
< 0.05
SD 0.76
0.82 0.82
Discussion This study demonstrates that axons will pass through a tube spanning a gap in peripheral nerve and reinnervate end organ muscle. The gap spanned was 5 mm longer than that spanned in Lundborg's study using pseudosynovial tubes. The tubes (Alza Corp., Palo Alto, Calif.) used in this study were made of a polymer of polyorthoester that has the ability to incorporate excipients, such as cis-hydroxyproline, and release them with zero order kinetics via hydrolysis. The tubes are biodegradable and nonantigenic. The hydrolytic
691
products of the tubes are cyclohexanedimethanol and 4-hydroxybuterate. The former is excreted in the urine, and the latter is a natural metabolite. The rate of complete hydrolysis of the polymer can be delayed from its baseline of 10 to 14 days to 5 to 6 weeks by the addition of 2% NaC03 • Ten percent cis-hydroxyproline was incorporated into the tubes as an antiscar agent. Cis-hydroxyproline is incorporated into collagen in the place of proline at the ribosomal level without affecting other cellular processes. 12- 14 The formation of the collagen triple helix normally formed within the cell is prevented, and instead single chains are secreted from the cell that are degraded by collagenase and lysosomes in the extracellular space. 15- 17 Previous experiments l8 • 19 have shown that locally delivered cis-hydroxyproline will significantly decrease the amount of scar formed around a nerve after neurorrhaphy or neurolysis without interfering with nerve regeneration. Regeneration through the tubes as assayed by histologic examination and by nerve conduction velocity demonstrated no difference as compared with the autogenous nerve grafts, with the exception that the rate of regeneration was delayed by 4 to 6 weeks. The exact reason for this is to be investigated; however, this may represent the time needed for Schwann cells to migrate into the tube. By histologic examination a fascicular pattern distal to the tubes with a uniform degree of myelination demonstrated that the regenerating axons were able to reenter the distal nerve segment after trans versing a gap that initially had no neural tissue. Microscopic sections were not taken through the tubes or nerve grafts but, in retrospect, would have been helpful to determine if a fascicular pattern was reestablished within the tube. The addition of cis-hydroxyproline to the tubes had the effect of reducing the initial delay in regeneration by 2 weeks as compared with plain tubes. Those nerves that regenerated through plain tubes were smaller in diameter than the nerves that regenerated through tubes with cis-hydroxyproline, but histologic sections from the distal stump of both groups were similar. Conclusion The results of this study indicate 1. Axons in a cat sciatic nerve will pass through a resorbable tube and reinnervate end organ muscle. 2. In those muscles showing evidence of reinnervation, nerve regeneration through the tubes as assayed by electrodiagnostic examination demonstrated no difference as compared with autogenous nerve graft, with the exception that the initial rate of regeneration was delayed by 4 to 6 weeks.
692
The Journal of HAND SURGERY
Bora et al.
3. The addition of 10% cis-hydroxyproline may improve the rate of efficiency of regeneration through a tube.
11.
REFERENCES 1. Seddon HJ. The use of autogenous grafts for the repair of large gaps in peripheral nerve. Br J Surg 1947;35: 151-67. 2. Millesi H, Meissl G, Berger A. Experience with interfascicular grafting of the median, ulnar and radial nerves. J Bone Joint Surg [Am] 1976;58:209-18. 3. Moneim MD. Interfascicular nerve grafting. Clin Orthop 1982;163:65-74. 4. Seddon HJ. Nerve grafting. J Bone Joint Surg [Brl 1963;45:447-61. 5. Terzis JK, Farbisoff B, Williams HB. The nerve gap: Suture under tension vs graft. Plast Reconstr Surg 1975;56:166-70. 6. Chiu DTW, Janecka I, Krizek TJ, Wolff M, Lovelace RE. Autogenous vein graft as a conduit for nerve regeneration. Surgery 1982;91:226-33. 7. Lundborg G, Hansson H. Regeneration of peripheral nerve through a preformed tissue space. Preliminary observations of the reorganization of regenerating nerve fibers and perineurium. Brain Res 1979;178:573-6. 8. Lundborg G, Hansson H. Nerve regeneration through preformed pseudosynvoial tubes. A preliminary report of a new experimental model for studying the regeneration and reorganization capacity of peripheral nerve tissue. J HAND SURG 1980;5:35-8. 9. Molander H, Olsson Y, Engkvist 0, Bowald S, Eriksson I. Regeneration of peripheral nerve through a polyglactin tube. Muscle Nerve 1982;5:54-7. 10. Forman DS, Berenberg RA. Regeneration of motor axons
12.
13.
14.
15.
16.
17.
18.
19.
in the rat sciatic nerve studied by labelling with axonally transported radioactive proteins. Brain Res 1978;156: 213-25. Politis MJ, Spencer PS. A method to separate spatially the temporal sequence ofaxon-Schwann cell interaction during nerve regeneration. J Neurocytol 1981; 10:221-32. Bora FW, Lane JM, Prockop DS. Inhibitions of collagen biosynthesis as a means of controlling scar formation in tendon injury. J Bone Joint Surg [Am] 1972;54:1501-8. Uitto J, Prockop DJ. Incorporation of proline analogues into collagen polypeptides. Biochim Biophys Acta 1974; 336:234-51. Rosenblum J, Prockop DJ. Incorporation of cishydroxyproline into protocollagen and collagenCollagen containing cis-hydroxyproline in place of proline and transhydroxy-proline is not extruded at a normal rate. J Bioi Chern 1971;25:1549-55. Jimenez SA, Prockop DJ. Specific inhibition of collagen synthesis in rat granulomas by incorporation of the proline analogue 4-hydroxyproline into intercellular collagen. J Clin Invest 1971;50:49a. Takeuch T, Prockop DJ, Biosynthesis of abnormal collagens with amino acid to extrude collagen polypeptides containing L-ayetidine-2-carboxylic acid or cis-4-flouroL-proline. Biochem Biophys Acta 1969;175:142-55. Uitto J, Dehm P, Prockop DJ. Incorporation of cishydroxyproline into collagen by tendon cells. Failure of the intracellular collagen to assume a triple helical conformation. Biochim Biophys Acta 1972;278:601-5. Pleasure D, Bora FW, Lane JM, Prockop D. Regeneration after nerve transection: Effect of inhibition of collagen synthesis. Exp Neurol 1974;45:72-8. Pleasure D, BoraFW, LaneJM, ProckopD. Acceleration of nerve regeneration by inhibition of collagen synthesis. Neurology 1974;24:363.
F. William Bora, Jr., M.D., John M. Bednar, M.D., A. Lee Osterman, M.D., Mark J. Brown, M.D., and Austin J. Sumner, M.D., Philadelphia, Pa.
PeriPheral nerve injuries cause disability because of incomplete recovery of function despite optimal surgical treatment. These injuries are an especially difficult problem when a segmental loss of peripheral nerve is present. The autogenous nerve graft is the most common method to treat patients with a significant gap in peripheral nerve. 1-5 The technique is technically demanding, destroys the function of another nerve, and has unpredictable results. These deficiencies have encouraged the development of other methods. Several investigators6-9 have used nonneural tissue to bridge nerve defects with variable results. Lundborg and Hansson 7 • 8 used a pseudosynovial sheath, produced by an implanted silicone rubber rod to bridge 10 mm gaps in the sciatic nerves of rats which reinnervated the short flexors of the foot by 3 months. Rat sciatic nerves, however, will not regenerate through a tube spanning a gap greater than 10 to 12 mm. Molander et aU reported that a resorbable From the Department of Orthopaedic Surgery, Hospital of the University of Pennsylvania, Philadelphia, Pa. Received for publication Aug. 11, 1986; accepted in revised form Jan. 2, 1987. The author or one or more of the authors have received or will receive benefits for personal or professional use from a commercial party related directly or indirectly to the subject of this article. In addition, benefits have been or will be directed to a research fund, foundation, education institution, or other nonprofit organization with which one or more of the authors are associated. Reprint requests: F. William Bora, Jr., M.D., Department of Orthopaedic Surgery, Hospital of the University of Pennsylvania, Silverstein Pavilion - Second Floor, 3400 Spruce St., Philadelphia, PA 19104.
Fig. 1. Sciatic nerve with resorbable tube spanning 1.5 em gap.
polyglactin mesh tube conducted axons across a 9 mm defect in the tibial nerve of a rabbit. Histologic examination in this experiment showed newly formed fascicles, but no functional assessment of useful reinnerTHE JOURNAL OF HAND SURGERY
685
686
The Journal of HAND SURGERY
Bora et al.
through the epineurium at each end. Ten of the cats received a tube containing 10% cis-hydroxyproline by weight and ten received a plain tube. This produced three groups of nerves for comparison: 20 containing a nerve graft, 10 containing a tube with 10% cis-hydroxyproline, and 10 plain tubes. The cats were left unrestrained in their cages and assayed by macroscopic and microscopic histologic examination, electrophysiologic testing, and fast axonal transport kinetics.
HI LEUCINE
SUTURE LINE
I
SCIATtC NERVE
-r-a "".
~==:g;jl8J:===_lj
I.. ---------....
'D·
,08
:I: IL
()
iii iii iii
El El El
'07
PLATEAU
",., ,03
j
I
SECTIOH VIALS
eREST
,03 '0'
Macroscopic histology
BACKGROUND
20
0
20
<10
eo
10
100
120
140
mm
Fig. 2. Schematic drawing depicts the technique of fast axonal transport kinetics and graphic presentation of data.
vation was perfonned. These studies demonstrate that nerve regeneration will occur over short distances through almost any type of conduit, but they do not compare their technique with the current standard of treatment, the autogenous nerve graft. The concept of a resorbable prosthetic nerve graft has many advantages including the following: (1) it is technically simple, (2) no donor nerve is required , (3) nerve ends are free of suture, (4) it is a potential vehicle for trophic factors, and (5) a method to carry a drug to control scar fonnation. The goal of this work was to study regeneration through a resorbable tube with and without an antiscarring agent (cis-hydroxyproline) to detennine if it could be used as an alternative to autogenous nerve graft for the treatment of a gap in a peripheral nerve. Materials and methods There were 20 adult female cats that weighed 3 to 4 kg in the study. Under sodium pentobarbital (Nembutal) anesthesia, each had a 1.5 cm segment of sciatic nerve removed bilaterally. The left nerve gap was spanned by a 1.5 cm nerve graft from the contralateral side and secured with five 10-0 nylon epineural sutures at each suture line. The right nerve gap was then spanned with a 2 cm long tube, with an inner diameter of 2.8 mm, with the cut ends of the nerve tucked into the tube to leave a 1.5 cm gap (Fig. I) . The ends were secured to the tube by two 9-0 nylon sutures placed
Nerves were operatively exposed and examined to detennine the percentage of tube remaining and the tissue present in the previous nerve gap. Two nerve grafts, one tube wi 'h and one tube without cishydroxyproline were examined at 6 weeks postoperatively. Eight nerve grafts, four tubes with and four tubes without cis-hydroxyproline were examined at 12 weeks postoperatively. (Note the 12-week cats were those assayed by fast axonal transport kinetics and were examined at autopsy.) Microscopic histology Twenty-eight weeks postoperatively ten nerve grafts, five nerves containing a tube with cis-hydroxyproline, and five containing a plain tube were assayed by microscopic histologic examination. The sciatic nerves were exposed from hip to knee and fixed in situ with 3.6% glutaraldehyde solution for 10 minutes . The nerves were then removed and further fixed in a 3.6% glutaraldehyde for 2 hours. A 5 mm section located 1 em distal to the distal anastomosis was then isolated and washed overnight in O.lm phosphate buffer solution, pH 7.4. The tissue was then postfixed with 2% osmium tetroxide for 2 hours, dehydrated in graded ethanol and propylene oxide, and embedded in epoxy. Toluidine blue-stained cross sections, one micron thick, were then prepared and studied by light microscopy. Electrophysiologic testing Nerve conduction velocity was measured by supramaximal stimulation of the sciatic nerve proximal to the nerve graft or tube. Insertional activity and fibrillation potentials were noted . The presence of an evoked multiphasic action potential recorded from the medial head of the gastrocnemius muscle and intrinsic muscles of the foot was interpreted as evidence of reinnervation of these muscles. The motor latency from stimulating signal to the onset of recorded action potential was measured in milliseconds. Ten cats were studied (10
Vol. l2A, No.5, Part I September 1987
Fig. 3. Sciatic nerve graft 12 weeks postoperatively.
nerve grafts, five tubes with cis-hydroxyproline, and 5 plain tubes) every 2 weeks until the cats were killed at 28 weeks.
Fast axonal transport kinetics The rate and extent of regeneration through the tubes as compared with nerve grafts was assayed by fast axonal transport kinetics in eight nerve grafts, four tubes with cis-hydroxyproline, and four plain tubes at 12 weeks postoperatively. Forman and Berenberg lO have shown that regenerating axons can be labeled with axonally transported radioactive proteins to provide information about the location of the entire range ofaxons from the fastest growing ones to those trapped in scar at the suture site. The method of Forman and Berenberg, with modification by Politis and Spencer, II to adapt the technique to cats was used, 3H-leucine (NET 135H, New England Nuclear Corp, Boston, Mass.) 40 to 60 Ci/flmol was concentrated to 20 flCi/flL in Hank's balanced salt solution. The cat was immobilized in a stereotaxic frame, and under Nembutal anesthesia, a wide L4 and 5 laminectomy was performed followed by durectomy. Ten pairs of 0.5 flL injections of 3H_
Prosthetic nerve grafts
687
Fig. 4. Sciatic nerve regenerating through tube containing cis-hydroxyproline at 6 weeks postoperatively. Note 60% to 70% of tube has resorbed and is replaced by a nerve like cord.
leucine were made with 10 flL Hamilton syringe held in a micromanipulator attached to the stereotaxic frame. The injections began 2.5 cm rostral to the conus and were made 1 mm from the midline, 3 mm deep, with 2 mm intervals between injections in the rostral-caudal direction. The animal's body temperature was carefully monitored during and after the procedure to maintain the temperature in the range of 100° F to 102° F with external warming. Fourteen hours after isotope injection, the animals were deeply anesthetized with Nembutal, heparin was administered and perfused through the heart with a 2% formaldehyde saline solution. Both intact sciatic nerves, one median nerve, and one ulnar nerve were removed and cut into 5 mm segments. The corresponding segments of the peroneal and tibial nerves beyond the bifurcation were added together. The segments were solubilized for 16 hours in tissue solubilizer (Protosol, New England Nuclear Corp, Boston, Mass.) at 45° C, 5 cc of scintillation cocktail added, and radioactivity counted in an Intertechnique SL30 Liquid Scintillation
688
The Journal of HAND SURGERY
Bora et al.
Fig. 5. Sciatic nerve regenerating through tube containing cis-hydroxyproline at 12 weeks postoperatively. Note tube is completely resorbed and replaced by regenerated nerve.
Fig. 6. Sciatic nerve regenerating through plain tube at 12 weeks postoperatively. Note tube is completely resorbed and replaced by thin nerve like structure.
Counter. Results were expressed as counts per minute per 5 mm nerve segment and graphed versus position along the nerve to allow calculation of the rate of regeneration, degree of axonal trapping, and position of the axonal growth cone (Fig. 2). The median and ulnar nerves serve as a control to establish the level of background radiation.
6 weeks (Fig. 4) and totally resorbed at 12 weeks. The nerves spanned by tubes containing cis-hydroxyproline had a distinct "nerve like" cord present filling the gap, which was similar in diameter to the nerve grafts (Fig. 5). The nerves spanned by plain tubes also had nerve growth across the tube but with a consistently smaller diameter cord (Fig. 6).
Statistical analysis
Microscopic histologic examination
Chi-square analysis and Student's t-test statistical analysis were used to evaluate the data.
Microscopic analysis of sections taken distal to the grafts or tubes reveal regenerating axons in a fascicular pattern (Fig. 7). The number ofaxons, axonal area, and degree of myelination grossly appeared very similar in those specimens in which regeneration had occurred through a nerve graft as compared with those regenerating through a tube with or without cishydroxyproline. Quantitative axon analysis was not performed.
Results All cats recovered from the operation and none had infections in the wound. Neurotrophic plantar ulcers occurred on the heels of 10% of the cats randomly on the sides with both nerve grafts and tubes.
Macroscopic histologic examination All nerve grafts studied at all time intervals were intact, with small neuromas at both suture lines (Fig. 3) . The tubes were 60% to 70% resorbed by
Electrophysiologic testing Electrical evidence of reinnervation of the gastrocnemius muscle at 8 to 10 weeks and the intrinsic mus-
Vol. 12A, No.5, Part I September 1987
Prosthetic nerve grafts
689
Fig. 7. Histologic sections of regenerating sciatic nerve taken distal to the repair site at 28 weeks postoperatively. A, Nerve graft. (Magnification x 60).) B, Tube with cis-hydroxyproline. (Magnification x 60.) C, Plain tube. (Magnification x 60.) D, Nerve graft. (Magnification x 200.) E, Tube with cis-hydroxyproline. (Magnification x 200.) F, Plain tube . (Magnification x 200.) Note fascicular pattern and uniformity of myelination .
cles of the foot by 18 to 20 weeks occurred in 100% of the nerve grafts (Fig. 8) . Eighty percent of those nerves containing a tube with cis-hydroxyproline demonstrated reinnervation of the gastrocnemius occurring
first at 12 to 14 weeks, with reinnervation of the foot by 24 to 26 weeks. Sixty percent of those nerves containing a plain tube demonstrated reinnervation of the gastrocnemius by 14 to 16 weeks, with reinnervation
690
The Journal of HAND SURGERY
Bora et al.
ELECTROPHYSIOLOGIC EVIDENCE OF REINNERVATION
Peripheral Nerve Regeneration Measured By Fast Axonal Transport
25
Nerve Graft
.
12 WEEKS POST OP
Gastrocneni.us
Foot Intrinsic Muscles
8 - 10 wks.
18 - 20 wks.
Tube with 10\ Cis-hydroxypro1 ine
12 - 14 wks.
24 - 26 wks.
Plain Tube
14 - 16 wks .
26 - 28 wks.
20
..
-
0
weeks post - operative
Fig. 8. Regeneration as detected by evoked muscle action potentials.
.
10 1. 5 em Nerve Or.tt
)(
c
l t'l
•
~
0
CI.
Peripheral Nerve Regeneration After Neurorrhaphy Measured By Fast Axonal Transport
:; c: :!
.
10
CI.
"' C :0
20
0
U 0
-
N
2.0 em Alzeme' Tube ContakMng 10" eta-Hydroxyproline Brlelgk1; 1. 5 em Gap
Q)
•
"
0
)(
10 10
E
0>
(/)
•
_
_
_
_
_
_
BACKGROUND
I»
CI.
:; c:
25
30 Ankle
Fig. 10. Fast axonal transport kinetics-axonal labeling profile for a nerve graft, tube containing cis-hydroxyproline and plain tube at 12 weeks after surgery. Note position of axonal growth cone.
4 Weeks Post Op
:i I»
Number of specimens
CI.
"E"':0
20
Segment Number
5 Weeks Post Op
"E
"
0
3 Weeks Post Op _ BACKGROUND
o
Sc iatic Notch
•
20
Suture Line
25
30 Ankle
Segment Number
Fig. 9. Fast axonal transport kinetics-axonal labeling profile for a simple neurorrhaphy at 3, 4, and 5 weeks after surgery.
of the foot by 26 to 28 weeks. Chi-square analysis demonstrated no significiant difference in percentage of nerve reinnervated for all three groups. Motor latency measured at the medial head of the gastrocnemius muscle at 28 weeks:
Average (msec) Standard deviation
Tubes with Nerve graft cis-hydroxyproline 3.6 4.1 1.2
1.4
4
3
Statistical analysis by Student's t-test showed no significant difference between either tube and nerve· graft or between tubes.
0
U
10
Plain tube 7.1
4.8
Motor latency measured at the intrinsic muscles of the foot at 28 weeks:
Average (msec) Standard deviation Number of specimens
Nerve Tubes with graft cis-hydroxyproline 8.0 12.9 1.9 10
Plain tube 11.8
6.1
7.2
4
3
Statistical analysis by Student's t-test showed no significant difference between either tube and nerve graft or between tubes.
Fast axonal transport kinetics This method uses the principle that nerves are composed of a cell body that supplies the materials for repair of injured axons by a sophisticated transport system.
Vol. 12A, No.5, Part 1 September 1987
Prosthetic nerve grafts
The proteins carried by this transport system can be labeled by stereotaxic injection of tritiated amino acids into the ventral motor horns of the lumbar spinal cord supplying the sciatic nerve. The labeled proteins are transported along the regenerating axons and allow the accurate determination of the position of the distal most axons by scintillation counting. Fig. 9 is a profile of the distribution of label along a sciatic nerve that has had a simple neurorrhaphy and was allowed to regenerate for 3 to 5 weeks before labeling. The peak present at segment five, the suture line, represents those axons in the neuroma at the suture line. This peak remains stationary at all time periods measured. The point at which the graph crosses baseline counts represents the distal most progress of the regenerating axons. As the time between repair and label increases from 3 to 5 weeks, this intersection with baseline or distal most point of regeneration, moved at a rate of 1 cm per week. This progression represents the rate of axonal regeneration. No attempt was made to correlate the area under the curve with the number of regenerating axons. Fig. 10 represents a composite of the labeling profile for a nerve graft, a tube with and without cis-hydroxyproline at 12 weeks. An average of profiles for all specimens was not plotted because of the variation of specific activity of the 3H-Ieucine between assays. Note that the regeneration has progressed well past the graft or tubes but at a differential rate consistent with the rates observed by electrophysiologic testing. (See previous section.) The point of the distal most regeneration was:
Nerve graft Tube with cis-hydroxyproline
Segment number 29.00
p < 0.05
21.00 P
Plain tube
18
< 0.05
SD 0.76
0.82 0.82
Discussion This study demonstrates that axons will pass through a tube spanning a gap in peripheral nerve and reinnervate end organ muscle. The gap spanned was 5 mm longer than that spanned in Lundborg's study using pseudosynovial tubes. The tubes (Alza Corp., Palo Alto, Calif.) used in this study were made of a polymer of polyorthoester that has the ability to incorporate excipients, such as cis-hydroxyproline, and release them with zero order kinetics via hydrolysis. The tubes are biodegradable and nonantigenic. The hydrolytic
691
products of the tubes are cyclohexanedimethanol and 4-hydroxybuterate. The former is excreted in the urine, and the latter is a natural metabolite. The rate of complete hydrolysis of the polymer can be delayed from its baseline of 10 to 14 days to 5 to 6 weeks by the addition of 2% NaC03 • Ten percent cis-hydroxyproline was incorporated into the tubes as an antiscar agent. Cis-hydroxyproline is incorporated into collagen in the place of proline at the ribosomal level without affecting other cellular processes. 12- 14 The formation of the collagen triple helix normally formed within the cell is prevented, and instead single chains are secreted from the cell that are degraded by collagenase and lysosomes in the extracellular space. 15- 17 Previous experiments l8 • 19 have shown that locally delivered cis-hydroxyproline will significantly decrease the amount of scar formed around a nerve after neurorrhaphy or neurolysis without interfering with nerve regeneration. Regeneration through the tubes as assayed by histologic examination and by nerve conduction velocity demonstrated no difference as compared with the autogenous nerve grafts, with the exception that the rate of regeneration was delayed by 4 to 6 weeks. The exact reason for this is to be investigated; however, this may represent the time needed for Schwann cells to migrate into the tube. By histologic examination a fascicular pattern distal to the tubes with a uniform degree of myelination demonstrated that the regenerating axons were able to reenter the distal nerve segment after trans versing a gap that initially had no neural tissue. Microscopic sections were not taken through the tubes or nerve grafts but, in retrospect, would have been helpful to determine if a fascicular pattern was reestablished within the tube. The addition of cis-hydroxyproline to the tubes had the effect of reducing the initial delay in regeneration by 2 weeks as compared with plain tubes. Those nerves that regenerated through plain tubes were smaller in diameter than the nerves that regenerated through tubes with cis-hydroxyproline, but histologic sections from the distal stump of both groups were similar. Conclusion The results of this study indicate 1. Axons in a cat sciatic nerve will pass through a resorbable tube and reinnervate end organ muscle. 2. In those muscles showing evidence of reinnervation, nerve regeneration through the tubes as assayed by electrodiagnostic examination demonstrated no difference as compared with autogenous nerve graft, with the exception that the initial rate of regeneration was delayed by 4 to 6 weeks.
692
The Journal of HAND SURGERY
Bora et al.
3. The addition of 10% cis-hydroxyproline may improve the rate of efficiency of regeneration through a tube.
11.
REFERENCES 1. Seddon HJ. The use of autogenous grafts for the repair of large gaps in peripheral nerve. Br J Surg 1947;35: 151-67. 2. Millesi H, Meissl G, Berger A. Experience with interfascicular grafting of the median, ulnar and radial nerves. J Bone Joint Surg [Am] 1976;58:209-18. 3. Moneim MD. Interfascicular nerve grafting. Clin Orthop 1982;163:65-74. 4. Seddon HJ. Nerve grafting. J Bone Joint Surg [Brl 1963;45:447-61. 5. Terzis JK, Farbisoff B, Williams HB. The nerve gap: Suture under tension vs graft. Plast Reconstr Surg 1975;56:166-70. 6. Chiu DTW, Janecka I, Krizek TJ, Wolff M, Lovelace RE. Autogenous vein graft as a conduit for nerve regeneration. Surgery 1982;91:226-33. 7. Lundborg G, Hansson H. Regeneration of peripheral nerve through a preformed tissue space. Preliminary observations of the reorganization of regenerating nerve fibers and perineurium. Brain Res 1979;178:573-6. 8. Lundborg G, Hansson H. Nerve regeneration through preformed pseudosynvoial tubes. A preliminary report of a new experimental model for studying the regeneration and reorganization capacity of peripheral nerve tissue. J HAND SURG 1980;5:35-8. 9. Molander H, Olsson Y, Engkvist 0, Bowald S, Eriksson I. Regeneration of peripheral nerve through a polyglactin tube. Muscle Nerve 1982;5:54-7. 10. Forman DS, Berenberg RA. Regeneration of motor axons
12.
13.
14.
15.
16.
17.
18.
19.
in the rat sciatic nerve studied by labelling with axonally transported radioactive proteins. Brain Res 1978;156: 213-25. Politis MJ, Spencer PS. A method to separate spatially the temporal sequence ofaxon-Schwann cell interaction during nerve regeneration. J Neurocytol 1981; 10:221-32. Bora FW, Lane JM, Prockop DS. Inhibitions of collagen biosynthesis as a means of controlling scar formation in tendon injury. J Bone Joint Surg [Am] 1972;54:1501-8. Uitto J, Prockop DJ. Incorporation of proline analogues into collagen polypeptides. Biochim Biophys Acta 1974; 336:234-51. Rosenblum J, Prockop DJ. Incorporation of cishydroxyproline into protocollagen and collagenCollagen containing cis-hydroxyproline in place of proline and transhydroxy-proline is not extruded at a normal rate. J Bioi Chern 1971;25:1549-55. Jimenez SA, Prockop DJ. Specific inhibition of collagen synthesis in rat granulomas by incorporation of the proline analogue 4-hydroxyproline into intercellular collagen. J Clin Invest 1971;50:49a. Takeuch T, Prockop DJ, Biosynthesis of abnormal collagens with amino acid to extrude collagen polypeptides containing L-ayetidine-2-carboxylic acid or cis-4-flouroL-proline. Biochem Biophys Acta 1969;175:142-55. Uitto J, Dehm P, Prockop DJ. Incorporation of cishydroxyproline into collagen by tendon cells. Failure of the intracellular collagen to assume a triple helical conformation. Biochim Biophys Acta 1972;278:601-5. Pleasure D, Bora FW, Lane JM, Prockop D. Regeneration after nerve transection: Effect of inhibition of collagen synthesis. Exp Neurol 1974;45:72-8. Pleasure D, BoraFW, LaneJM, ProckopD. Acceleration of nerve regeneration by inhibition of collagen synthesis. Neurology 1974;24:363.