Experimental Methods to Simulate and Evaluate Postsurgical Peripheral Nerve Scarring

As a consequence of trauma or surgical interventions on peripheral nerves, scar tissue can form, interfering with the capacity of the nerve to regenerate properly. Scar tissue may also lead to traction neuropathies, with functional dysfunction and pain for the patient. The search for effective antiadhesion products to prevent scar tissue formation has, therefore, become an important clinical challenge. In this review, we perform extensive research on the PubMed database, retrieving experimental papers on the prevention of peripheral nerve scarring. Different parameters have been considered and discussed, including the animal and nerve models used and the experimental methods employed to simulate and evaluate scar formation. An overview of the different types of antiadhesion devices and strategies investigated in experimental models is also provided. To successfully evaluate the efficacy of new antiscarring agents, it is necessary to have reliable animal models mimicking the complications of peripheral nerve scarring and also standard and quantitative parameters to evaluate perineural scars. So far, there are no standardized methods used in experimental research, and it is, therefore, difficult to compare the results of the different antiadhesion devices.


Introduction
Scar tissue around the nerve can arise as a consequence of traumatic injuries and surgical procedures on peripheral nerves. This condition easily worsens the capacity of the peripheral nerve to regenerate and can give rise to traction neuropathies. Nerve tethering in the surgical scar is still the main cause of symptoms related to perineural scarring [1].
Traction neuropathies can be the consequence of elective procedures, including nerve decompression, primary nerve repair, and so on. For instance, 7-20% of patients subjected to primary median nerve release report pain and symptom recurrence [2,3]. Thus, peripheral nerve injuries compromise the quality of life of affected people, with a consequent important socioeconomic impact [4][5][6].
This condition is difficult to manage; according to different reports, compression symptoms persist after 40-90% of revision procedures, and 20% of patients actually require a third operation [7]. Moreover, 5% of nerve sutures have been estimated to induce a pain syndrome [8].
A primary role in this pathological condition has been attributed to the formation of scar tissue around the injured nerve. In particular, extrinsic nerve scarring occurs at the periphery of the epineurium, while intrinsic nerve scarring occurs within the nerve and can surround neural structures at all levels (both perineurium and endoneurium) [9]. To Adhesion of peripheral nerve to surrounding tissues results in fibrosis in the nerve. Compound muscle action potentials were reduced in amplitude, and blood flow was significantly decreased at adhesion sites in Group IIb.
The methods used to induce scar formation are several ( Figure 1), but two of these are more frequently used. The first one consists of a direct injury (mechanical, epiperineurectomy, suture and repair, thermal, chemical, or physical) applied to the nerve surface. The second one consists of inducing an injury in the surrounding muscular bed by means of electrocoagulation, triggering the process of fibrosis from the surrounding tissue. Some researchers have induced a global injury to the nerve and surrounding tissues by scratching both nerve and muscles with irradiation or chemical injuries. Furthermore, the envelopment of the nerve in a silastic tube in order to let the scar tissue rise has been proposed. The methods used to induce scar formation are several ( Figure 1), but two of these are more frequently used. The first one consists of a direct injury (mechanical, epiperineurectomy, suture and repair, thermal, chemical, or physical) applied to the nerve surface. The second one consists of inducing an injury in the surrounding muscular bed by means of electrocoagulation, triggering the process of fibrosis from the surrounding tissue. Some researchers have induced a global injury to the nerve and surrounding tissues by scratching both nerve and muscles with irradiation or chemical injuries. Furthermore, the envelopment of the nerve in a silastic tube in order to let the scar tissue rise has been proposed. Some of the earlier papers [23][24][25] performed a two-stage procedure (first stage injury, second stage neurolysis and antiadhesion application), which is a more traumatic experience for animals, without evidence of increased efficacy compared to a one-stage procedure. With respect to the 3Rs statement [26], a single-stage experiment can have the same efficacy as a two-stage one.

Scar Evaluation: Methods to Evaluate Scar Formation in Experimental Models
Different procedures can be used to evaluate scar formation, including gross examination of the scar tissue, microscopical analysis of the nerve and surrounding tissue, functional tests, and electrophysiological and biomechanical evaluations ( Figure 2). All of Some of the earlier papers [23][24][25] performed a two-stage procedure (first stage injury, second stage neurolysis and antiadhesion application), which is a more traumatic experience for animals, without evidence of increased efficacy compared to a one-stage procedure. With respect to the 3Rs statement [26], a single-stage experiment can have the same efficacy as a two-stage one.

Scar Evaluation: Methods to Evaluate Scar Formation in Experimental Models
Different procedures can be used to evaluate scar formation, including gross examination of the scar tissue, microscopical analysis of the nerve and surrounding tissue, functional tests, and electrophysiological and biomechanical evaluations ( Figure 2). All of these evaluation methods are combined differently by authors. It must also be noted that the time points analyzed are very different among the studies, ranging from few days to several months from the induction of scar formation; the research also differs according to the employed animal model.

Macroscopical Analysis
Gross evaluation is the first fundamental step to macroscopically grade scar tissue; it aims to assess the enrolment of the surrounding tissues (including skin, muscles, and deep tissues) in the compression of the nerve and the collaboration with the newborn perineural scar tissue.
Different classifications have been proposed by different authors to evaluate the degree of scar formation. The most used and complete classification is the numeric grade scheme defined by Petersen [27]. This classification allows us to evaluate closure of skin and muscle fascia (Grade 1: skin or muscle fascia entirely closed; Grade 2: skin or muscle fascia partially open; Grade 3: skin or muscle fascia completely open) and to evaluate nerve adherence (Grade 1: no dissection or mild blunt dissection; Grade 2: some vigorous blunt dissection required; Grade 3: sharp dissection required). Another adopted grading scheme is the 4-point qualitative scale that evaluates the perineural adhesions and type of dissection required to achieve complete neurolysis, as follows: absent or thin adhesionsdelicate blunt (score 0); mild adhesion-vigorous blunt (score 1); moderate adhesiondelicate sharp (score 2); severe adhesion-difficult sharp (score 3) [13,14,23,28].
Abe [17] proposed a classification of nerve adhesion based on Fontana's band (an optical manifestation of axonal undulations characteristic of peripheral nerves). They classify nerve adhesion as Group I (nonadhesion group) when the bands appear and Group II if they are not visible. Additionally, Group II is divided into Group IIa when a thickening of the epineurium and perineurium is observed (but not endoneurial fibrosis) and Group IIb when endoneurial fibrosis is observed.
Finally, some authors have reported the presence or absence and qualitative observations of scar tissue around the nerve without grading it.

Macroscopical Analysis
Gross evaluation is the first fundamental step to macroscopically grade scar tissue; it aims to assess the enrolment of the surrounding tissues (including skin, muscles, and deep tissues) in the compression of the nerve and the collaboration with the newborn perineural scar tissue.
Different classifications have been proposed by different authors to evaluate the degree of scar formation. The most used and complete classification is the numeric grade scheme defined by Petersen [27]. This classification allows us to evaluate closure of skin and muscle fascia (Grade 1: skin or muscle fascia entirely closed; Grade 2: skin or muscle fascia partially open; Grade 3: skin or muscle fascia completely open) and to evaluate nerve adherence (Grade 1: no dissection or mild blunt dissection; Grade 2: some vigorous blunt dissection required; Grade 3: sharp dissection required). Another adopted grading scheme is the 4-point qualitative scale that evaluates the perineural adhesions and type of dissection required to achieve complete neurolysis, as follows: absent or thin adhesions-delicate blunt (score 0); mild adhesion-vigorous blunt (score 1); moderate adhesion-delicate sharp (score 2); severe adhesion-difficult sharp (score 3) [13,14,23,28].
Abe [17] proposed a classification of nerve adhesion based on Fontana's band (an optical manifestation of axonal undulations characteristic of peripheral nerves). They classify nerve adhesion as Group I (nonadhesion group) when the bands appear and Group II if they are not visible. Additionally, Group II is divided into Group IIa when a thickening of the epineurium and perineurium is observed (but not endoneurial fibrosis) and Group IIb when endoneurial fibrosis is observed.
Finally, some authors have reported the presence or absence and qualitative observations of scar tissue around the nerve without grading it.

Microscopical Analysis: Histological Staining and Immunohistochemistry
Microscopical analysis is employed by most authors and consists mainly of the use of different histological stainings to visualize and describe the different structures involved (not only the scar tissue but also the nerve and surrounding tissues, such as muscle), both qualitatively and quantitatively (see Tables 1-6).

Analysis of the Scar Tissue
The most employed method to highlight scar tissue is Masson's trichrome staining because it specifically marks collagen in green/blue and can be easily distinguished from other structures such as muscle fibers (stained in red), cytoplasm (light red or pink), and cell nuclei (dark brown to black).
Haematoxylin and eosin (H&E) is the most widely used staining for histological purposes because it provides a general overview of the tissue, and it is mainly used to distinguish nerves from surrounding tissues. It is the combination of two histological stains: hematoxylin, which stains cell nuclei in blue/dark-purple, and eosin, which stains cytoplasm in pink, and other structures, including extracellular matrix and collagen, in shades of pink.
Picrosirius red is often used since it selectively highlights collagen fibers [29]; indeed, this dye allows us to visualize collagen fibers in red (specific for collagen types I and III), while the other structures are stained in yellow (nuclei, cytoplasm, muscular fibers, red blood cells). The Gordon and Sweet technique, used to reveal reticulinic acid, a collagenous tissue marker, has also been adopted, as well as chromotrop-aniline-blue, which stains collagen in blue and muscle fibers in red.
The classification of Ornelas [45] has also been adopted [46,47], and it allows us to classify extraneural and intraneural fibrosis; extraneural fibrosis is classified into Grade 1-absent or minimal fibrosis; Grade 2-moderate fibrosis; Grade 3-major fibrosis. Intraneural fibrosis is classified into Grade 1-the presence of fibrous tissue between the nerve fibers; Grade 2-fibrous tissue partially blocking the passage of nerve fibers; Grade 3-fibrous tissue completely interrupting the passage of nerve fibers.
Dam-Hieu [28] calculated the thickness of the dense scar surrounding the nerve. The largest thickness of the scar ring (ST) was measured. This value was then normalized by dividing it by the nerve diameter (ND). The authors call this value the fibrotic index (fibrotic index = 2 ST/ND).
Other quantitative or semiquantitative analyses have also been proposed, such as the average thickness of collagen in the epineurium [14,48], the count of fibroblasts and inflammatory cells [36,38,46], the thickness of the epifascicular epineurium, the amount of connective tissue in the interfascicular and epifascicular epineurium [49], and the percentage of area of staining (PAS) calculation by outlining the intraneural tissue [50][51][52][53].

Analysis of the Nerve Tissue
To investigate the nerve tissue specifically, Luxol fast blue, P-phenylenediamine, silver staining, the Weil method, and toluidine blue were used to describe axon distribution and highlight the myelin sheath. In particular, toluidine blue staining offers the possibility of performing morphoquantitative analyses to estimate the number of myelinated fibers, axon density, fiber and axon diameter, myelin thickness, and g-ratios (axon-diameter/fiberdiameter), which can be correlated with functional recovery [5,54].
Moreover, the longitudinal histomorphological organization of the axon at the nerve repair site can be evaluated according to the scale developed by Brown et al. [55] and adopted by several authors [31,35,36,38,56]: Grade 1-failure, no continuity of the axons from the proximal to the distal ends; Grade 2-poor organization (interlacing or whirling appearance of the nerve fibers); Grade 3-fair organization (focal whirling appearance, focal parallel alignment); Grade 4-good organization, approaching normal (mostly parallel, without a whirling or wavy appearance); Grade 5-excellent organization of the repair site, indistinguishable from the norm.

Functional Analysis
Some studies performed functional analyses, even though these analyses are not precise for scar quantification and we are not sure that they can be directly correlated to the amount of scarring observed around the nerve.
Most of the studies dealing with peripheral scarring use the sciatic nerve model, foot print analysis, and the Sciatic Function Index (SFI) as the most adopted tests [63]. Other parameters evaluated are allodynia by means of von Frey filaments [62] and walking patterns induced by pain with the CatWalk system [21,51,62].
The only study that used the median nerve model tested the function of the nerve by means of the grasping test [16].

Electrophysiological Study
Electrophysiology is another test that can be used in order to analyze the formation of scars around the nerve. Indeed, compression around a nerve causes pathophysiological changes that can be registered with this assessment.
The electrophysiological analysis is based on compound motor action potential (CMAP). Different aspects of electrical activity can be registered, such as latency, signal amplitude, and speed conduction. These parameters correlate with nerve conduction; the more the scar is present, the more these parameters are altered.
Another assessment that is useful to evaluate is the frequency of spontaneous firing because it has been demonstrated to be related to nerve suffering: the more frequent the firing is, the more the nerve is suffering [64].
Zuijdendorp [22] and its colleagues performed the evaluation by means of magnetoneurography of first peak amplitude, peak-peak amplitude, area, and conduction velocity over the nerve segment between the stimulation and the recording site.
Recently, the combination of electrodiagnostic evaluation, with the commonly used grasping test (reflex-based gross motor function) and the staircase test (skilled forelimb reaching), has been found to produce results with high translatability [65].

Biomechanical Analysis
Biomechanical analysis gives an objective evaluation of scar tissue formation and consists of measuring the force required to overcome the adhesion conjunctions between the nerve, scar tissue, and surrounding tissues. Different methods have been proposed, but the results obtained represent, with variability according to the precision of the utilized instruments, a quantitative expression of newborn scar tissue.
Dumanian [66] and his colleagues were the first to describe a method and device to measure the strength of nerve adhesion to surrounding muscles. They used a standard alligator clamp placed on the nerve and a force transducer connected, in turn, to a micrometer. The micrometer was distracted in 1 mm increments. The measurement is continued until final failure of the nerve or nerve pullout from the clamp.
Another method is to mount the nerve proximal stump on a digital force gauge using a suture connected to the load cell; then, the nerve is subjected to traction at a rate of 2 cm/min (or 1 cm/s) [24] until its complete detachment from the neural bed; the ultimate strength is recorded [57,[67][68][69].
A different way consists of transecting both the proximal and distal ends of the nerve; the proximal end is then interconnected to a force transducer, which is connected, in turn, to a motorized drive with a constant extension rate of 29 mm/min. The force required to pull the nerve segment out of its tissue bed is recorded [10,22].
In another paper, after nerve and surrounding tissue removal from the animal, the distal end of the nerve was held by a clamp to the cross-head of an Instron machine. The subsequent cross-head movement (at a rate of 10 mm/min) then gradually peeled the nerve away from the adhesion site, and the maximum peeling force was recorded [15].
Finally, some recent papers have described a simple and cheap method that consists of connecting the nerve to a plastic can that is gradually filled with water at a constant flow of 100 mL/min. The adhesion force is obtained from the grams of water at the break moment [20,70,71].

Other Analysis
Other types of analyses have also been described, such as enzyme-linked immunosorbent assays (ELISAs) to evaluate neurothophic factor concentration [62], RNA and protein analysis [61], assessment of autotomy [37], functional analysis of the blood-nerve barrier and the perineurial barrier [69], and hydroxyproline and collagen assays [60].
In vitro culture of rat skin fibroblasts to test the efficacy of drug administration has also been described [58].
Histological staining can also be used to describe muscle tissue organization in order to evaluate atrophy and fibrous degeneration of the innervated muscles [13,49,57]. Moreover, atrophy is often investigated by measuring muscle wet weight. Finally, transmission electron microscopy has been adopted to describe the ultrastructure of nerve tissue and surrounding tissues [30,39,48,60,72].

How to Prevent Scar Formation? An Overview on Different Antiadhesion Devices
Every surgical practice on peripheral nerves is followed by postsurgical scar tissue formation. In order to limit this event, surgeons apply different procedures such as local or free tissue transfer and antiadherent items of different origins. There are many different kinds of antiadhesion devices, composed of different materials with different ways of application, but there is no evidence of their efficacies. Below is an overview of the different antiadhesion devices tested so far in experimental models.

Polysaccharide-Based Devices
Different polysaccharides were used as antiscarring agents, and the available preclinical studies on the polysaccharides-based devices are reported in Table 2. Reduction scar around the nerve, both macroscopically and microscopically. Increased nerve diameter. Higher gastrocnemius mass. Improved microstructural organization. Higher expression of S100.

Hyaluronic acid
Hyaluronic acid is a glycosaminoglycan that is widely found in the body of all living organisms as it is an important extracellular matrix component. Since it does not exhibit species or tissue specificity and is biodegradable in vivo, it is often used as an ideal biomaterial. It has been demonstrated that hyaluronic acid reduces epineural and extraneural scar formation [36,38,67]. Additionally, biomechanical reduction of scar tissue has been documented [10], together with an improvement of latency [24].

Carboxymethylcellulose
Carboxymethylcellulose is another biocompatible polysaccharide that acts as a physical barrier and can reduce scar formation in the central nervous system; it has been demonstrated that carboxymethylcellulose, in association with phosphatidylethanolamine, reduces peripheral nerve scarring and biomechanical resistance [67,71]. It has also been used in association with hyaluronic acid; additionally, in this case, it reduces scar formation, reduces inflammation cells and fibroblasts, and leads to better axonal organization [38,72], together with an increase in the quality of myelin sheets and the number of axons [49].

Chitosan
Chitosan is a polysaccharide obtained by partial deacetylation of chitin. It has wellknown advantageous properties, such as lack of toxicity and biocompatibility, biodegradability, and antimicrobial properties. Various forms of chitosan can be produced and microcrystallic chitosan gel applied to the proximal stump of a transected sciatic nerve has been shown to reduce the incidence and size of the neuroma and the formation of extraneural fibrosis [37]. It has also been used in the form of conduit in association with hyaluronic acid, and it has been demonstrated to reduce nerve scarring and promote nerve regeneration and recovery [48].
Oxidized regenerated cellulose is a chemically altered form of cellulose used mainly as a hemostatic agent. It has been shown to not give an advantage to the prevention of nerve fibrosis; on the contrary, it interferes with healing by increasing inflammatory phenomena and granulomatous reactions [74]. Other polysaccharides have demonstrated their efficacy in the reduction of scarring through biomechanical and macro-and microscopical testing [22,28,69,74].

Collagen-Based Devices
The available preclinical studies on collagen-based devices are reported in Table 3.  The use of ADCON-T/N, a bioabsorbable gel composed of a polyglycan ester in a phosphate-buffered saline solution, showed a significant reduction of scar formation with no residual implant material [13,23,27]. Additionally, the use of collagen-based film wrapped around the suture stitches showed a reduction in epineural and perineural scar tissue formation [47,75,76]. Finally, a recent study showed that a collagen sheath derived from an acellular hypoallergenic dermal matrix wrapped around the suture leads to better nerve regeneration in terms of axon diameter [16].

Amniotic Membrane
The amniotic membrane is the inner layer of fetal membranes; it is composed of an inner layer of epithelial cells on a thick basement membrane. It is nonimmunogenic, and it has been demonstrated to reduce inflammation, inhibit vascularization, combat infection, and reduce scarring. It is widely used in multiple fields of surgery and medicine, including skin substitute, wound care, urethral reconstruction, and repair of corneal and other tissues [77]. Its use in reducing peripheral nerve scarring has been demonstrated in different papers [14,39,40].

Fat Grafting
In the last decade, adipose tissue has been widely studied in the field of regenerative medicine due to the presence of adipose-tissue-derived mesenchymal stem cells (which can differentiate into different cellular lineages) and its endocrine activity (release of adipocytokines, cytokines, transcriptional and growth factors). It is easy to access and harvest with painless procedures. The use of fat grafting in the prevention of peripheral scar tissue formation has had different results: it produces nerve stiffness reduction in biomechanical testing [66], but no significant differences were reported when compared to other antiadhesion devices. Moreover, in microscopical analysis, it appears to be able to reduce scar thickness [70].

Vein Wrapping and Buccal Mucosa Graft
Another autologous tissue that has been tested for scar formation prevention is vein tissue, which is harvested from the same animal (femoral vein) and wrapped in a spiral pattern around the nerve [25] or harvested from the abdominal portion of the donor animal vena cava and wrapped around the ligated nerve [62]. In both cases, it has been demonstrated to reduce scar formation and improve nerve function recovery.
Finally, the use of a buccal mucosa graft has also been proposed as an antiadherent device since it is composed of nonkeratinized epithelium with underlying connective tissue and includes type I and III collagen. It has been shown that when wrapped around the nerve, it decreases adhesion and scar tissue formation but leads to higher inflammation in the early postoperative period [46].

Others
Many other devices/techniques have been investigated as antiscarring agents, and the preclinical results of these devices are reported in Table 6. A very recent study compared the efficacy of two novel biodegradable wraps made of synthetic 1% oxidized polyvinyl alcohol (OxPVA) and a leukocyte-fibrin-platelet membrane (LFPm) with the commercial product NeuraWrap, demonstrating their effectiveness in sustaining nerve regeneration, together with an absence of scar tissue/neuroma formation and significant inflammatory infiltrate [81].   [83] and a nerve conduit composed of PLA and poly(e-caprolactone) (PCL) and enriched with hyaluronic acid [57] have been shown to prevent nerve adhesion. Additionally, a novel multilayer membrane made of PLA-based biodegradable polymer (E8002) containing L-ascorbic acid has been demonstrating to reduce scar formation compared to the same membrane without ascorbic acid [82]. Atkins et al. [52] proposed the local administration of neutralizing antibodies to TGF-β1 and TGF-β2, showing a significant reduction in intraneural scar formation. The use of Ankaferd blood stopper resulted in better healing and better results in the histopathological evaluations [59], and the use of an absorbable oxidized regenerated cellulose sheet showed the prevention of adhesion in a histological study [84]. Finally, low-dose radiation therapy [34] and early mobilization [15] have also been proposed.

Discussion
Traction neuropathies are diffused and frequent consequences of injuries or surgical procedures on peripheral nerves [2,7]. Surgeons and researchers have been trying to prevent scar tissue formation, especially by applying antiadhesion devices on the surgical site. Before human implantation, preclinical studies are of crucial importance for assessing the effectiveness of antiadhesion strategies; however, the results reported in the literature are not easily comparable due to the many different methods to induced scar formation as well as quantitatively evaluate the amount of scar tissue and its impact on peripheral nerve regeneration and function. The reason is connected to the absence of a shared, effective, reliable, reproducible, and standardized protocol to induce and test the scar tissue around the peripheral nerves [18,[20][21][22].
The purpose of this review was, therefore, to resume and show the different strategies adopted in the last few years to simulate and evaluate scar tissue formation.
Different kinds of injuries have been proposed by researchers to simulate perineural scar tissue formation: nerve injury, injury to surrounding tissues, global injury by means of chemical or physical agents, and so on. Some authors have designed studies to simulate and evaluate perineural scars [17][18][19][20][21][22], but these works are incomplete because they have not considered all the aspects of induction and the evaluation methods available. Our review reveals that the widest protocol used to induce scar tissue is represented by section and suture of the nerve. This partially represents what really happens in clinical settings and results in a partial injury. In our opinion, an injury to the perineural tissue should always be associated in clinical settings by means of burning or chemical injury to the nerve. None of the papers have combined this kind of injury. It would, therefore, be very interesting and useful to test the combination of these methods in an experimental model that better mimics the clinic. Furthermore, researchers should consider that a different pattern of scar tissue arises if the nerve is transected or not. Without a nerve section and suture or without a crush injury, no internal scar will form, especially when there has been only a short period between the surgery and the analysis. In this way, no or only minimal impairment of nerve function will be present.
This aspect links to the other main aspect of this review: so many different types of analysis were performed to quantitatively or qualitatively assess the perineural scar and its reduction. We strongly encourage the use of quantitative analysis. Gross evaluation is certainly a fundamental step to macroscopically grading the scar, and the adoption of a numerical grading scheme is necessary to quantify (or semiquantify) the amount of scar tissue. Different grading schemes have been adopted [13,14,17,23,27,28] and most of the authors have used these schemes to score scar tissue during macroscopical inspection. Nevertheless, other authors have described only the presence or absence of scar tissue, and, sometimes, they qualitatively describe the scar tissue by relying on subjective observations. Microscopical evaluation is usually conducted both on the nerve and scar tissue. Beyond the type of staining used, the quantitative evaluation of the perineural scar is very important. First of all, a staining solution that is able to visualize collagen fibers specifically (such as Masson's Trichrome, Sirius Red) is preferred to the classical H&E and methylene blue stainings. Quantification of scar tissue is proposed in different ways, but the most adopted is the calculation of the scar tissue formation index (by dividing the mean thickness of the scar tissue by the mean thickness of the nerve tissue) [18,28,[30][31][32][33][34][36][37][38][39][40][41]. Moreover, in these cases, some other authors have used different parameters to quantify (or semiquantify) the scar tissue histologically; sometimes, the structure of scar tissue is only qualitatively described. Some authors have added immunohistochemistry to conventional staining. Very interesting are the attempts to detect scar tissue using antibodies against macrophages and lymphocytes, but it is difficult to obtain a quantification of these findings [21,37,59].
Biomechanical analysis provides another quantitative parameter and consists of measuring the peak force required to pull the nerve from the muscular bed. Different tools have been adopted, and, in general, they consist of applying a continuous force to the nerve until the complete detachment of the nerve from the surrounding tissue. This should be the other main method that is useful to demonstrate the strength of the scar tissue in preclinical analysis. In addition, the structure and function of the nerve must be assessed. Quantification of different parameters (number of fibers, axon and fiber diameter, myelin thickness) are important to assess the degree of nerve regeneration and can be related to electrophysiological parameters and functional evaluations, often assessed in studies dealing with peripheral nerve scarring. In our thoughts, morphological and functional impairments correlate more to a direct nerve injury than an experimental perineural scar. Hence, morphometry and electrophysiology should be associated, especially when nerve injury and repair have been performed or when scar neuropathy has lasted for several months.
Different animal models have been proposed, each one with pros and cons. Mice are easy to house and allow quick and easy surgery. The sciatic nerve allows the performance of all the main analyses described above. Otherwise, in this model of nerve repair, functional and electrophysiological tests are more difficult to perform due to the smallness of the animal itself. Conversely, rats allow the testing of both the sciatic nerve and the median nerve; the size of these nerves is feasible for nerve repair and electrodiagnostic tests. Furthermore, functional evaluation on the median nerve can be carried out, especially when a direct nerve injury is performed.
Due to the limitations of the present models, no definitive conclusion can be derived about the efficacy of antiadhesion devices. In most of the experiments, every treated group showed scar reduction according to the evaluation methods. The most employed antiadhesion devices were polysaccharide-based and collagen-based ones. Their effectiveness was, in most cases, well known in spine surgery, tendon surgery, or abdominal surgery. It is similar for biological barriers: vein wrapping has been described to protect nerve sutures in the past [25,85], with good clinical outcomes. Fat grafts were previously adopted in spine surgery with controversial results; in peripheral nerves, it seems to be effective, and promising results were obtained with Coleman's lipoaspirate [70]. From our review, it has emerged that the application of amniotic membrane can be promising, considering the increased chance of tissue storage [14,39,40]. This requirement can also be considered a limit for this technique. Drugs and other devices can also be promising, but currently, no clinical experience exists.
This preclinical literature review suggests that we reconsider the whole argument of traction neuropathies. This pathology was classified by Millesi [86] in intra-and extraneural scar, but more extensive classification would be effective to better understand the correct treatment. Perineural scarring arises after closed trauma and nerve decompression or can be associated with a repaired nerve. The intraneural involvement is different in each case, and different approaches should be considered. When an intraneural scar is present, it should be treated as a neuroma in continuity. In contrast, when the perineural scar is the main concern, other procedures should be performed. According to our clinical practice and the results obtained in these experimental models, vein wrapping can be a feasible procedure to prevent intraneural and epineural adhesion after nerve suture. Otherwise, in a secondary peripheral nerve decompression, the application of an antiadhesion device, either in the form of gel or film composed of polysaccharide or collagen, could be adequate as well as autologous lipoaspirate, local adipose flap, synovial flap, or amniotic membrane wrapping. The choice, at state of the art, is up to the individual surgeon's experience and availability since no clear evidence exists.
To better understand traction neuropathies, a more extensive classification should be designed by considering the extension of the scar, the amount of fibrous tissue (both preand intraoperative by means of a quantitative scale, as proposed by Petersen [27]), and previous surgery on the nerve (suture, traction injury, nerve decompression). On these, prognostic criteria would be found and a more fitted treatment protocol developed.
In our experience and considering the literature, to completely evaluate scar tissue formation around a nerve, we require a scored macroscopical analysis, a quantitative microscopical analysis conducted with a staining solution that is able to collagen fibers visualize specifically (such as Masson's Trichrome, Sirus Red) in order to clearly measure the thickness and extension of scar tissue, and a computed biomechanical analysis with the appropriate microinstruments. The structure, organization, and function of the nerve should also accompany scar tissue data, but these are not always mandatory since they depend on the type of nerve injury induced.
Finally, this review reports on the different antiadhesion devices that have been experimentally tested so far. Due to the high variability of scar induction and evaluation methods described, it is not possible to compare the results obtained in terms of scar reduction and efficacy of the antiadhesion devices employed.
Author Contributions: A.C., G.R. and P.T. organized the manuscript. A.C., G.R., B.E.F., S.O. and S.R. wrote different sections of the manuscript. All authors contributed to manuscript revision and approved the submitted version of the manuscript.