Keywords
peripheral nerve regeneration - nerve gap - heparin-binding growth factors
Several biomaterial conduits have been used to bridge nerve gaps,[1] either alone or filled with various substances.[2]
[3]
[4] Artery grafts present large quantities of laminin and some collagen. These substances
are also found in Schwann-cell basal membrane and are reported as axonal outgrowth
factors.[5] Calcipliq20 (OTR3, 4, rue Française 75001 Paris, France) is a product that contains
a molecule belonging to the family of the ReGeneraTing Agents (RGTAs). RGTA improves
tissue healing in several animal models by stabilizing and protecting heparin-binding
growth factors (HBGFs) and matrix proteins.[6]
[7]
[8] The purpose of this study was to use artery grafts[5] filled with CACIPLIQ20 and to see if they promote nerve regeneration.
Methods
Sixty male Wistar rats weighing approximately 280 g were used. The work has been reported
in accordance with the ARRIVE (Animals in Research: Reporting In Vivo Experiments)
guidelines.
Sample Size Estimation
It was calculated that a sample size of 15 male Wistar rats per group was required
in order to have 90% probability of demonstrating a between nerve regeneration's biomaterial
conduits (AG/NS [5,000] vs. AG/CACIPLIQ20 [6,000]) difference of >1,000 (standard
deviation [SD]: 650) in number of nerve fibers with significance of < 1% (two-tailed
test with Bonferroni correction ).
G-Power Analysis Output
It was calculated that a sample size of 15 male wistar rats per group was required in order to have 90% probability of demonstrating a between
nerve regeneration's biomaterial conduits [AG/NS( -60) vs. AG/CACIPLIQ20(-45) ] difference
of > 15 (SD: 10) in sciatic nerve function index with significance of < 1% (two-tailed
test with Bonferroni correction).
The rats were randomly divided into four experimental groups (n = 15): transected control group (negative control group [NCG]), sham-operated group
(positive control group [SO]) artery graft filled with saline group (AG/NS), and CACIPLIQ20-treated
group (AG/CACIPLIQ20). Fifteen rats were used as artery graft donors. The animals
were anesthetized with an intramuscular injection of a 3:2 mixture of 100 mg/dL of
ketamine hydrochloride and 20 mg/dL of xylazine hydrochloride. The animals were individually
housed, given food and water ad libitum, and exposed to a 12-hour light–dark cycle.
The left sciatic nerve was exposed through a gluteal muscle incision ([Fig. 1]). In the transected control group (NCG), the left sciatic nerve was transected proximal
to the tibioperoneal bifurcation, where a 7-mm segment was excised, leaving a 10-mm
gap due to retraction of nerve ends. Proximal and distal stumps were fixed in the
adjacent muscle with a 10–0 nylon epineurial suture ([Fig. 2]). No graft was interposed between the stumps. In the SO group, the nerve was dissected
from surrounding tissues and left intact. In the AG/NS group, a 7-mm nerve segment
was resected to produce a 10-mm nerve gap after retraction of the nerve transected
ends. The gap was bridged using an artery graft filled with saline, entubulating 2 mm
of the nerve stump at each end ([Fig. 3]). The artery graft was harvested from the abdominal aorta of donor animals. The
abdominal aorta was exposed through a midline abdominal incision ([Fig. 4]) and cannulated. Then, a 15-mm segment was harvested on the cannula. Donor animals
were sacrificed after graft harvesting. Harvested grafts were washed in physiological
solution and left at room temperature for 40 minutes with no preliminary treatment
to reduce their antigenicity. Two 10–0 nylon sutures were used to anchor the graft
to the epineurium at each end. In the AG/CACIPLIQ20 group, the graft was filled with
10-μL CACIPLIQ20.
Fig. 1 Exposure of the left sciatic nerve through a gluteal muscle incision.
Fig. 2 Excision of a 7-mm segment leaving a 10-mm gap due to retraction of nerve ends in
the transected control group (negative control group). Proximal and distal stumps
were fixed in the adjacent muscle with a 10–0 nylon epineurial suture.
Fig. 3 The gap was bridged using an artery graft filled with saline in the AG/NS group and
with CALCIPLIQ in the AG/CALCIPLIQ group, entubulating 2 mm of the nerve stump at
each end.
Fig. 4 The artery graft was harvested from the abdominal aorta of donor animals and exposed
through a midline abdominal incision.
Walking track analysis was performed at 4, 8, 12, and 16 weeks after surgery based
on the method of others.[8] Recovery assessment was performed using the weight ratio of the gastrocnemius muscles
16 weeks after surgery. Immediately after animal sacrifice, gastrocnemius muscles
were dissected and harvested carefully from intact and injured sides and weighed while
still wet using an electronic balance.
Finally, immediately after sacrificing (16 weeks after surgery), nerve sections were
harvested for histopathology analysis. Nerve sections were fixed in 4% buffered formalin,
cut perpendicular to the longitudinal axis in 2-mm pieces, and embedded in paraffin.
Subsequently, 2-μm-thick serial sections were cut and placed on glass slides. Special
stains hematoxylin–eosin and Masson's trichrome, as well as immunohistochemical stains
for S100 protein and neurofilaments (DAKO autostainer, DAB as chromogen) in serial
sections were performed. The stained sections were examined under a light microscope,
and photographs were taken. For morphometric analysis, the Image-J software (Wayne
Rasband, NIH, LOCI, University of Wisconsin, WI) was used. Qualitative features noted
included axonal degeneration, empty myelin sheaths, degradation of the myelin sheath,
axonal regeneration with sprouting (inside and occasionally outside the graft), and
location and intensity of S100 and neurofilament staining. Quantitative information
included the number and density of nerve fibers and the diameter of myelin sheath
thickness in cross-sections.
Statistical Analysis
Statistical analysis was conducted through SPSS (edition 23, SPSS Inc.) software package.
The Shapiro–Wilk test was used to test normality for the quantitative variables (i.e.,
SO, AG/NS, NCG, CALC) in consequent times. Data were thoroughly examined and paired
samples t-test and analysis of variance test for multiple comparisons (Bonferroni) were used
to figure out the statistically significant difference between the multiple consecutive
measurements of the walking track analysis. The level of statistical significance
was determined with p-values < 0.05.
Results
One animal in the AG/NS group died at the end of the eighth week from infection. This
animal was excluded from the study. The animal was replaced. Fifteen animals were
included in the study from all the groups (AG/NS, NCG, SO, AG/CALCIPLIQ). The mean
number of nerve fibers in AG/CACIPLIQ20 group was approximately 74% in the AG/NS group,
57% in the SO group, and in NCG group 10% of the SO group ([Fig 5]). There was no significant difference in myelin sheath thickness between the AG/NS
and AG/CACIPLIQ20 groups. S100-positive staining was weak and diffuse in the NCG group
(evidence for nerve degeneration) and strong and clearly located in the myelin sheath
in the AG/NS and AG/CACIPLIQ20 groups, indicating a normal pattern of nerve regeneration.
This pattern was verified with neurofilament immunostaining, where regenerated axons
were visible in the center of myelin sheaths in the AG/NS and AG/CACIPLIQ20 groups
and absent in the NCG group. Interestingly, neurofilament staining was weaker in the
AG/NS and AG/CACIPLIQ20 groups compared with the SO group. Regarding gastrocnemius
muscle weight, the best results were as expected in the SO group. There was a statistically
significant difference between the SO and AG/NS groups (p = 0.000) as well as between the SO and AG/CACIPLIQ20 groups (p = 0.000). Muscle weight in the NCG group was statistically significant less (p = 0.000) compared with that in the SO, AG/NS, and AG/CACIPLIQ20 groups. Muscle weight
in the AG/CACIPLIQ20 group was higher but not statistically significant (p = 0.168) compared with that in the AG/NS group (results are summarized in [Table 1]). Regarding walking track analysis, at the end of 4 months, the control (SO) and
CALCIPLIQ groups had comparable results (p = 1.000) ([Fig. 6]). The measurements of the biggest improvement from the first to fourth month was
documented in the CALCIPLIQ group (mean = CALCIPLIQ [first month] = –60.63, mean = CALCIPLIQ
[fourth month] = –3.83; p = 0.001). The AG/NS group also presented statistically significant improvement from
the first to fourth month (p = 0.025), less however, than CALCIPLIQ group ([Fig. 7]). Regarding the fourth month comparison between AG/NS and CALCIPLIQ, there was no
statistically significant difference (p = 0.605); nevertheless, CALCIPLIQ (fourth month) mean was better than AG/NS (fourth
month) mean (mean CALCIPLIQ [fourth month] = –3.8385, mean AG/NS [fourth month] = –20.6321)
([Fig. 8]).
Fig. 5 Mean number of nerve fibers at the end of 16 weeks.
Fig. 6 Walking track analysis results at the end of 4 months.
Fig. 7 Walking track analysis results for each group.
Fig. 8 Comparison of walking track analysis at the end of 4 months among four groups.
Table 1
Muscle weight group statistics
|
Animal models muscle weight (grams)
|
N
|
Group A (NCG), N = 15
|
Group B (SO), N = 15
|
Group C (AG/NS), N = 15
|
Group D (CALCIPLIQ20), N = 15
|
|
1
|
0.68
|
3.97
|
1.95
|
1.96
|
|
2
|
1.17
|
4.17
|
2.28
|
2.25
|
|
3
|
0.74
|
3.90
|
2.22
|
2.45
|
|
4
|
0.66
|
2.90
|
1.80
|
1.73
|
|
5
|
1.17
|
3.79
|
1.90
|
2.15
|
|
6
|
0.67
|
3.89
|
2.10
|
2.20
|
|
7
|
0.77
|
3.91
|
0.95
|
2.30
|
|
8
|
0.69
|
2.89
|
2.02
|
1.77
|
|
9
|
0.55
|
3.89
|
1.80
|
1.96
|
|
10
|
0.77
|
3.73
|
1.95
|
2.43
|
|
11
|
0.61
|
3.54
|
2.17
|
2.28
|
|
12
|
0.48
|
4.20
|
1.90
|
2.30
|
|
13
|
0.59
|
4.00
|
0.95
|
1.95
|
|
14
|
0.65
|
2.59
|
2.05
|
2.07
|
|
15
|
0.58
|
4.35
|
0.92
|
1.86
|
|
Mean muscle weight
|
|
0.71 ± 0.19
|
3.71 ± 0.51
|
1.79 ± 0.46
|
2.11 ± 0.23
|
|
Group
|
N
|
Mean
|
Standard deviation
|
Standard error mean
|
|
NCG
|
15
|
0.7187
|
0.19982
|
0.05159
|
|
SO
|
15
|
3.7147
|
0.51908
|
0.13403
|
|
AG/NS
|
15
|
1.7973
|
0.46513
|
0.12010
|
|
AG/CALCIPLIQ
|
15
|
2.1107
|
0.23039
|
0.05949
|
Abbreviations: AG/CACIPLIQ20, artery graft filled with CACIPLIQ20 group; AG/NS, artery
graft filled with saline group; NCG, negative control group; SO, sham-operated group.
Discussion
Nerve gaps can result either traumatically or iatrogenically. To repair them, numerous
conduits have been used. Silicon tubes were used quite early to bridge nerve gaps[9] as they have moderate cost and the absence of donor-site morbidity. However, they
generate an inflammatory reaction and may cause chronic nerve compression.[10] To overcome the disadvantages of silicon tubules, biodegradable nerve conduits have
been developed. Several materials have been used, such as poly(caprolactone)[11] and copolymers of DL-lactide and epsilon-caprolactone [p(DLLA-epsilon-CL)],[12] either solely as conduits or as tubules, that can synchronously administrate factors
to improve the regeneration of peripheral nerves. The disadvantage of biodegradable
guides is that they collapse easily. Vein grafts filled with muscle are resistant
to collapse and provide support to both regenerating axons and migratory Schwann cells.[13] Artery walls are rigid compared to vein walls and do not need to be filled with
muscle to prevent collapse. Moreover, they present quantities of laminin and collagen
in analogy to Schwann-cell basal membrane.[5] Artery grafts have been used previously to bridge nerve gaps in experimental protocols.[5] RGTAs are biodegradable polymers engineered to mimic heparan sulfate in the extracellular
matrix of damaged tissue. CACIPLIQ20 contains a molecule belonging to the family of
the RGTAs and has been used to improve tissue healing in both animal models and patients
by stabilizing and protecting HBGFs and matrix proteins.[6]
[7]
[8] Nerve and skin tissue wound-healing process present similarities.[14] Therefore, we thought of using an agent that promotes skin wound healing and investigating
whether it enhances nerve regeneration.
The number of nerve fibers was higher in the AG/CACIPLIQ20 group compared with the
AG/NS group. However, arrival of sprouts from the proximal to the distal nerve stump
does not necessarily imply recovery of nerve function.[15] Instead, walking track analysis has frequently been used to reliably determine functional
recovery following nerve repair in rat models. Our walking track analysis demonstrated
better results in the AG/CACIPLIQ20 group compared with the AG/NS group, although
not statistically significant, and this was in accordance with muscle weight results
analysis. This could be an indication that CACIPLIQ20 improves functional recovery
of a transected peripheral nerve through a nerve gap bridged with an artery graft.
As with any experimental model, however, there are weaknesses. Rigidity of artery
grafts has been inculpated to cause excessive pressure on the nerve diameter and hypoxia,
thus impairing nerve regeneration. Different grafts and bigger animal models could
be used to investigate the efficacy of RGTAs in nerve functional recovery.
