Vous êtes sur la page 1sur 10

Hernia (2015) 19:169–178 DOI 10.1007/s10029-014-1307-8

REVIEW

REVIEW

A

review of biocompatibility in hernia repair; considerations

in

vitro and in vivo for selecting the most appropriate repair

material

N.

Bryan C. Battersby

N.

Smart J. Hunt

Received: 14 January 2014 / Accepted: 29 August 2014 / Published online: 13 September 2014 Springer-Verlag France 2014

Abstract Purpose Repair of hernia typically makes use of a pros- thetic material; synthetic or biologic in nature. Any mate- rial which enters the body is subject to interrogation by the inflammation and immune system in addition to numerous other cell families, the outcome of which ultimately determines the success of the repair. In this review, we discuss the fundamental biology which occurs in situ when a biomaterial associates with a tissue, compare and contrast the techniques available to predict this in vitro, and review how features of hernia repair materials specifically may manipulate tissue interrogation and integration. Finally, we conclude our article by examining how biocompatibility impacts surgical practise and how a better understanding of the manner by which materials and tissues interact could benefit hernia repair. Materials and methods A review of the literature was conducted using appropriate scientific search engines in addition to inclusion of findings from the groups’ primary research. Results Using pre-clinical assays to anticipate the bio- compatibility of a medical device is critical; however, to maximise the scientific power of in vitro findings, we must carefully consider the in vivo niche of the cells with which we are working. Excessive in vitro culture or contact to non-self materials can add compounding complexity to

N. Bryan ( &) C. Battersby J. Hunt

Clinical Engineering (UKCTE), Institute of Ageing and Chronic Disease, Duncan Building Ground Floor, Daulby Street, Liverpool L69 3GA, UK e-mail: n.bryan@liv.ac.uk

N. Smart

Exeter Health Science Research Unit, Royal Devon and Exeter Hospital, Barrak Road, Exeter, Devon EX2 5DW, UK

studies involving leucocytes for instance; therefore, we must ensure careful and stringent assay design when developing techniques for assaying pre-clinical biocom- patibility. Furthermore, many of the features associated with hernia repair material design specifically, included to enhance their mechanical or biodegradation characteristics, are inadvertently instructive to cells, and therefore, throughout the prototype stages of a materials develop- ment, regular biocompatibility assessment must be performed. Conclusion The biocompatibility of a material is rate limiting in its ability to function as a medical device. The future of hernia repair materials will rely on close cohesion between the surgical and scientific communities to ensure the most robust biocompatibility assessment techniques, and models are utilised to predict the efficacy of a given material in a particular surgical application.

Keywords Biocompatibility Inflammation Leucocyte Hernia In vitro In vivo

Introduction

Hernia repair often incorporates the use of an exogenous material to provide mechanical support to compromised tissue. Immediately after implanting this material, it is subject to contact with tissue and bodily fluids causing the physiologic processes which allow the body to decide the fate of the material to begin. Ideally, the material integrates with surrounding tissue acting as a scaffold for appropriate cells to colonise and provide permanent mechanical aug- mentation to damaged tissue. However, if this interrogation process is inappropriately stimulated, the material may become encapsulated, degraded and require removal with

123

170

Hernia (2015) 19:169–178

morbidity to previously healthy interfacial tissues. There- fore, understanding the mechanisms at a cellular level with which a material interacts with tissue is paramount in predicting its efficacy and longevity in vivo. This review provides a concise overview of cellular considerations of biocompatibility in addition to considering in detail how specific features of hernia materials may influence this process and the impact that this may have on surgical practise.

Materials and methods

A review of the literature was performed across a range of

scientific citation databases considering surgical and sci- entific research in the area of biocompatibility and hernia repair from over the previous three decades. In addition to this, findings from our research teams own work are also included.

Biocompatibility

Tissue healing involves complementary cellular and path- ophysiologic processes which return compromised tissue to normal function [2, 7]. From the instant, a surgeon places a material in contact with tissue the processes that determine its fate, and ultimately the quality of the patient’s life commence. Material surfaces are coated with proteins from blood, interstitial fluids or exudates which commences

cellular interrogation. A positive outcome is integration with surrounding anatomy in a mechanically stable manner having returned the tissue to normal homoeostasis. How- ever, should a material’s cellular interrogation decide it presents a risk to the function of the organism as a whole it will degraded via enzymes, small molecules and reactive oxygen species [12]. If destruction is not possible, mate- rials are isolated in a capsule of fibrous matrix where risk to surrounding tissue is contained. Broadly, the ability of a material to repress the negative and encourage the positive components of cell interrogation is termed biocompatibility. The interaction between material and tissue is dynamic. Throughout the proceeding hours to years (in the instance

of permanent prostheses), the boundaries of its effects on

the host are pushed as the body attempts to use it as a conduit to tissue regeneration, which may require its remodelling, cell infiltration or removal. We must disregard definitions that portray inflammation as intrinsically negative. A material that stimulates inflammation is a statement which informs us of inevita- bility due to lack of any true biologically inert material. Inflammation is the underlying process which repairs tis- sue. Without inflammation, wounds do not heal, materials

123

do not integrate and infection prevails. A material is required to encourage constructive elements of inflamma- tion to support tissue remodelling through with appropriate debridement and disinfection as opposed to destruction of material, and healthy tissue that occurs during exacerbated inflammation. The cells that decide material fate are a varied with an array of roles unified by a drive to return tissue to normal homoeostasis. Primarily this involves leucocytes, but as time progresses, non-circulatory cells also play a role in material compliance. Within the first hours, a material is investigated by neutrophils; which secrete many destruc- tive proteins and small molecules capable of damaging both material and native tissue. Although historically thought of a cell whose role was purely destruction and decontamination, we now recognise neutrophils as media- tors of inflammation by signalling to attack and recon- naissance cells to orchestrate pro and anti-inflammatory processes [4]. Over the proceeding days, neutrophils are accompanied by macrophages and famed debridement through phago- cytosis, and these cells also signal and are arguably the most rate limiting cell throughout tissue healing [5]. Additionally, these cells also present fragments of digested material using MHC molecules, which drives generation and persistence of long-term and targeted inflammation by lymphocytes. A site of tissue repair will require re-vascularisation meaning that a biomaterial must support migration of blood vessels through its bulk. This provides colonising cells with a mass transport network to prevent necrosis of neo-tissue. As cells enter the material and proliferate their nutrient and waste, import/export requirements require constant remodelling of neo-microvasculature throughout the graft; changing shape, size and direction to ensure that cells within receive support to keep the tissue healing and healthy. The interaction of cells with the periphery of the material must also be considered. Blood cells are rela- tively short lived, and as such, their presence at a material site is constantly turned over. Therefore, we must give thought to cells that remain throughout its existence and provide security and stability beyond the degradable suture used to secure it. A materials edge must permit cell infiltration such that matrix secretion knits the material with surrounding ECM. This is the role of fibroblasts; excessive fibroblast activation or proliferation results in encapsulation; enshrouding fibrous tissue preventing integration, colonisation and vasculari- sation. Therefore, the fibroblast cells must be managed to ensure the correct degree of matrix deposition to provide graft security, without fibrous encapsulation and material isolation.

Hernia (2015) 19:169–178

171

Predicting biocompatibility in vitro and in vivo

Biomaterial development should be guided by clinical

feedback to researchers and companies, to define the ideal requirements for a particular style of repair. Any new material will be subject to direct and indi- rect contact cytotoxicity screening to assess cell vitality in response to a material using biochemical tests. Typi- cally, these report quantitatively (colorimetric or fluori- metric) in response to changes in cell metabolism or membrane integrity. These techniques quantify the pro- portion of live or dead cells after treatment with a material and can be tailored into deduce whether cell death was a consequence of apoptosis or necrosis. Cytotoxicity assays are run over short periods (hours) to deduce acute cytotoxicity or extended over days/weeks to consider modifications to a materials cytotoxicity as it degrades or is remodelled. The cells used for cytotoxicity screening can be tailored to a materials end use. Com- monly, an established cell line is used for initial cyto- toxicity screening which progresses to primary cells appropriate to the anatomy in which the material will be situated, as its development is refined.

In cell specific models, the goal may not solely be one of

confirming cytocompliance, in favour of deducing the influence of a material on the phenotype or function of a relevant cell population. To this end, we must ensure that

in vivo culture platforms are designed with appropriate stringency to mirror the cells native 3D microenvironment as much as possible to obtain translatable data [6].

A second stage in anticipating material biocompatibility

is to subject it to contact with inflammatory cells and using

molecular biology to deduce its influence on cell longevity, phenotype and function. Most commonly, these experi- ments utilise human peripheral blood mononuclear cells (PBMCs), extracted from whole blood via density gradient centrifugation. PBMCs contain the mononuclear fraction of leucocytes; monocytes and lymphocytes; granulocytes are removed during density gradient centrifugation. The reason for utilising mononuclear leucocytes is that granulocytes rapidly become activated by the physical and chemical factors to which they are subjected outside of the cardiovascular system. When these cells are activated, they secrete cytokines and reactive oxygen species (ROS) which cause phenotypic changes across a cell population, making subtle, constructive leucocyte responses difficult to observe. ROS are an incredibly diverse class of signalling molecules capable of both pro- and anti-inflammatory processes, a number of which are detailed in Fig. 1. This highlights the limitations of considering inflam- mation in vitro. Although assays are regularly performed throughout biomaterials science which place leucocytes in contact with a test substance and record their response, we

must treat this data as a baseline and avoid making direct translation to in vivo inflammatory physiology. This is because the ex vivo environment triggers phenotypic switches similar to invasion of the body with non-self material. One must remember that cells recognise and activate in the presence of materials deemed foreign to healthy tissue, therefore, it should not be simply underes- timated that a tissue culture environment distorts leucocyte function and influences data collection. Factors worthy of consideration include the physical and haemodynamic forces associated with venipuncture and centrifugation, and contact with xenoproteins in foetal calf serum, a component of tissue culture media [11]. Impor- tantly, we must consider the influence of contact with tis- sue culture plastic vessels remembering that leucocytes are constantly suspended in blood. Therefore, a direct surface contact mirrors activation pathways such as contact with damaged tissues or activated endothelium in vascular walls. The compounding activating variables should be accepted as significant when extrapolating to whole organism physiology. Solutions have been developed which address and alle- viate components of this in vitro activating milleau, including non-adherent suspension culture systems using bioreactors, acoustic forces [13], chemically modified surfaces and serum-free media. Typically, the data output from in vitro studies of leu- cocyte activation should include several levels of molec- ular interrogation which support an ultimate assessment of changes in cell function based on transcriptomic, proteomic and secretomic levels. We must take care to understand the implications and limitations of each stage of this workflow. Although PCR and microarray strategies provide insight into cellular behaviour, we must use these genomic/tran- scriptomic level analyses to provide support and robustness for protein and secretion data. Fundamentally, it is the protein level which allows a cell to perform its function, and therefore, whilst changes in gene expression provide a platform level of understanding into cell behaviour, it is pertinent that we do not cease our analyses at a nucleotide level and evaluate changes in proteins responsible for cell function in vivo. A further pitfall to studying material/leucocyte interac- tion in vivo is in using clean and pure populations of PBMCs, we remove the synergy which happens as a result of material, cell and blood interaction. Biomaterial surfaces are rapidly coated with proteins from blood and tissue fluid which include immunomodulatory molecules such as immunoglobulin and compliment. Therefore, whenever possible, we must consider the inclusion of whole blood or human plasma in in vitro biocompatibility workflows. The authors have previously used a whole blood based assay to conclude a materials capacity to activate

123

172

Hernia (2015) 19:169–178

172 Hernia (2015) 19:169–178 Fig. 1 Leucocyte signalling is a vastly intricate process. The important signalling

Fig. 1 Leucocyte signalling is a vastly intricate process. The important signalling aspects of one class of leucocyte produced molecules, ROS, are detailed

leucocytes using chemiluminescent reporting of ROS pro- duction. Using whole blood adds granulocytes into the assay; therefore, we designed experiments to consider the initial phases of material interaction, running assays for 90 min with appropriate controls to gather quantitative data before granulocyte degeneration [9, 10]. After in vitro studies have concluded sufficiently that candidate materials do not possess inherent cytotoxicity nor are they intrinsically pro-inflammatory the next stage in predicting their biocompatibility at a whole organism level are in vivo animal models. Initially, implantation is performed to deduce inflam- matory characteristics of a material in vivo without the compounding variable of a healing or damaged tissue. This is typically a subcutaneous implant with a material placed adjacent to the dorsolumbar musculature in a rodent. This allows free movement of cells through the material, allowing inflammation to proceed and be monitored with- out compromise of the animal should the material fail mechanically. We must remember, however, particularly in device-specific models that animal models often present a material to different mechanical loading (quadrupedal vs. bipedal for instance), biochemistry, and differentiation

123

degradation kinetics based on much shorter lifespan to humans. These factors should all be considered when designing and interpreting data from small animal series. After preliminary animal studies, materials are tested in a more anatomically correct model which describes the defect which it is designed to repair clinically, such as a full thickness abdominal wall excision model in the instance of ventral/incisional hernia. In tissue-specific models, material mechanical properties are also tested to comply with their target tissue. Furthermore, models should subject materials to specific populations of cells with which they would contact clinically, such as muscle cells and tissue-specific populations of leucocytes such as peritoneal macrophages in the instance of abdominal wall repair. Classical analyses are performed by excising tissues after predetermined end points and performing histo- pathology in which the fixed, micron thickness sections are non-specifically stained to colour the cells and ECM to visualise using light microscopy. This enables path- ologically indexing using a numerical scale to semi- quantitatively elucidate material interaction with host physiology.

Hernia (2015) 19:169–178

173

This approach, whilst providing valuable baseline data, is suboptimal for elucidating finer points of cell/material interaction. Fundamentally, the system is subjective, based on the opinion and training of scoring pathologists, and negates of the importance of considering the function of particular populations of cells at an implant in favour of recording their mere presence. Although pathologists undoubtedly poses the skill to discern between lymphocytes and neutrophils for instance based on cell morphology, the presence of these cells provides only basic insight into how a material interacts with tissue. Although greater numbers of neutrophils indicate stronger acute inflammation, we remain devoid of detail concerning cell activation or signalling making true prediction of material fate difficult. Cells with microscopically similar morphology indicate vastly differing tissue consequences which only become apparent with more detailed molecular analyses such as CD4/CD8 T lymphocytes and classically versus alterna- tively activated macrophages. Therefore to quantify and critically, standardise bio- compatibility predictions, we must enhance pathologic interrogation beyond traditional histopathology. This should use validated SOPs to remove variability induced by human scored systems whilst adding value to informa- tion gained purely by tinctorial staining, providing a layer of molecular deduction of the roles and purpose of specific cells at an implant. The clearest alternative is immunohistochemistry, a process which utilises antibodies against antigens associ- ated with particular components of inflammation to visu- alise specific cells and tissue areas by visible light or fluorescent microscopy. Once cells or areas have been specifically allocated a spectral signature, human evalua- tion can be removed in favour of image analysis to count cells or calculate boundaries absolutely [8]. Using absolute quantification inter-sample standardisation is required when cell counting; a material may present with greater numbers of a particular cell simply because the sample is physically larger. Therefore, cells of interest should be normalised to a concurrent nuclear stain to deduce total cell number such as haematoxylin or Hoechst. This technique differentiates morphologically similar yet functionally distinct cells in addition to targeting surface markers associated with defined cell processes such as activation, phagocytosis, migration and apoptosis. Immunohistochemistry also targets secreted molecules which present in intercellular and interstitial spaces allowing signalling substances involved in inflammatory progression and remission to be observed, to real-time monitor inflammatory progression. Immunohistochemistry is, however, a relatively com- plex process made up of several sub-elements which also

have potential to induce user mediated variation in the analytic procedure. There are numerous individual pro- cesses which can induce variation between practitioners including fixation, antigen retrieval, antibody concentra- tions/incubation incubation parameters, visualisation pro- tocol and even mounting and colorimetric/fluorimetric preservation. Therefore, although the technique undoubt- edly has the potential to add value beyond traditional pathologic indexing, we must proceed with caution and ensure operating protocols are generated and followed to produce data which supports the rigour of inter-centre standardisation. A skilled histologist produces sections 4–6-lm thick, ensuring cells are sliced across their axis, presenting their internal cytoplasm. This allows observation of intracellular substances such as precursors to secrete factors and probing the phagolysosome. Depending on the tissue fixation, we may also probe a section at a nucleic acid level using nucleotide hybridisation in which a fluorescently conju- gated nucleotide sequence complementary to an mRNA sequence targets sequences within. Such techniques con- clude the gene expression in cells surrounding an implant, providing a window into their objectives as they decide the material fate. Explants can be dissociated and subject to gene probing using polymerase chain reaction (PCR) analyses. A par- ticularly useful modification to this, reverse transcriptase PCR utilises the enzyme reverse transcriptase to convert mRNA into DNA which enables PCR amplification. This is particularly interesting as elucidates genes which are expressed at a particular moment in time. It is now possible to characterise expression of many genes simultaneously in a single PCR meaning large data can be assembled rapidly. This is useful for deducing novel biomarkers of disease based on high throughput and cost effectiveness of PCR. However, we must ensure during any molecular biology transcripts of clinical interest have their translation to protein confirmed. Histopathology reports as a terminal procedure which does not help to dynamically predict biocompatibility. In vivo, the material will be excised when the animal is killed, clinically a material may only be available patho- logically when it failed and been removed/replaced. Therefore, we were not able to decipher molecular pro- cesses which preceded its failure which would provide a group of dynamic biomarkers to monitor material/tissue interaction throughout its integration/degradation profile. This insight would improve pharmaceutical management or refine surgical intervention to augment the process before clinical failure. Throughout animal studies, we must collection of bio- markers from experimental subjects in addition to histo- pathology. Most obviously, from this blood collection, we

123

174

Hernia (2015) 19:169–178

can monitor populations of cells as they migrate to an implant or quantify substances in serum. We may also consider less invasive fluids specifically urine which can easily be collected from experimental animals. From biologic fluids, there are several techniques to monitor systemic molecules and rapidly screen for sub-

stances of interest. Similarly to immunohistochemistry, this employs antibodies specific to particular molecules to flag them up for luminescence/fluorescence detection. These may be ELISA systems in which antibodies tethered to a substrate bind molecules of interest, which allows quanti- fication with the addition of further antibodies conjugated

to a reporter substance. Recently, these technologies have

been advanced multiplex in which many molecules can be observed in a single assay using spectrally distinct reporter

antibodies with detection apparatus capable of reading many emission spectra simultaneously.

Material factors influencing biocompatibility

Many factors influence biocompatibility of implanted materials. The physical properties of a material, either by design or as by-products of fabrication, inadvertently instruct cells and tissues. When considering biocompati- bility, we must remember material integration is syner- gistic between material and patient tissue; therefore, there are also patient-specific physiological variables which influence material interaction and modify biocompatibility at a personal level. This is highlighted in the Williams definition of biocompatibility ‘the ability of a material to perform with an appropriate host response in a specific application’ we must consider the specificity of an appli- cation to be personal given the heterogeneity of inter- individual tissue dynamics [1]. Therefore, generalisations regarding ubiquitous biocompatibility should be used with extreme caution until sufficient prospective data confirms this throughout a large range of subjects. The skill of the surgeon and appreciation for maintain-

ing tissue vitality also contributes to the biocompatibility of an implant. Excessive tissue manipulation causes damage and extends wound remission. Furthermore, surgical application of the material also influences its biological efficacy, such as material/tissue overlap or suture to material ratio. For this reason, deducing subtleties in material biocompatibility are difficult from retrospective analyses due to the variables introduced by lack of stand- ardised surgical application. Materials modify biocompatibility at the level of chemical and biological composition and physical and mechanical properties. There is no such thing as an inert biomaterial, one which may be applied to a defect purely as

a mechanical bridge without cell interaction, and it is

123

essential to disregard this ideology. Fundamentally, we must consider the way in which biomacromolecules inter- act with a material surface as this gives a cell its first impression of an implant and begins the process of deter- mining acceptance or destruction. Cell surfaces are dynamic and continually adapt to the 3D space to which they are subjected. As cells contact substrate membrane, fluidity allows it to flex and mould to the surface, producing focal adhesions which feed back to the cellular cytoskeleton to change in shape and rigidity. Once intracellular cytoskeletal scaffolding is laid down, the cell produces ECM creating a more favourable environ- ment for colonisation as extracellular receptors recognise this environment is familiar using integrins and other rec- ognition molecules [2]. If a surface environment is unfavourable to cells, the material is perpetually reliant on its sutures to retain its structural capacity as cells will never colonise and produce autogenous matrix to bind the device with surrounding tissues. The converse of this is a surface which signals cells to hypersecrete matrix which results in fibrous encapsula- tion. We must be mindful of the interaction of cells with material surfaces and engineer implants to incorporate known topographic parameters with balanced adhesive and proliferative properties. The physical microenvironment of a biomaterial greatly influences cell response and supports interaction of the material with tissues as a whole. Topography at a micron scale, which is easily overlooked during materials fabri- cation as it may play a minor part in conferring mechanical properties, modifies cytoskeleton remodelling, matrix pro- duction and cells recruitment. Techniques such as atomic force microscopy (AFM) characterise material surfaces to quantify the roughness and randomness of their topography enabling the provision of optimal surface parameters for cell compliance. When a material contacts tissue fluids its surface is immediately coated with protein, so therefore when con- sidering surface topography in vitro, we allow this inevi- table fouling which may mask subtle topographic changes by providing a homogenous protein coat. On this basis, we should also consider a materials protein adsorption capac- ity when performing material evaluation and more specif- ically investigate materials possess preferential adsorption of certain specific proteins. Despite protein coating, it has been demonstrated that nanoscale modifications to material surfaces modify cell fate and function, particularly in the instance of adult stem cells, a pertinent cell in any healing tissue [15]. The hydrophobicity/hydrophilicity surface must also be considered, as this will strongly influence protein adsorp- tion and subsequent cell interaction. Conversely, the field is continually seeking materials in which an anti-adhesive

Hernia (2015) 19:169–178

175

face can be applied to the materials visceral side to inhibit intra-abdominal adhesions. Although a material which is intrinsically repulsive to tissue presents a tantalising sub- stance in herniorrhaphy, we must proceed with caution with anti-adhesive biomaterials and use our ability to repel brings surgical issues caused by compromised integration and poor mechanical stabilisation. There may indeed be more subtle pharmaceutical methods by which to manage intra-abdominal adhesions by targeting and suppressing migration and activation of inflammatory cells [3]. Material integration will also benefit from colonisation, increasing and securing the bond between material and peripheral tissue by providing a meshwork of interfacial cells inside and outside the graft. Therefore, we must ensure that a material has sufficient porosity to allow cell penetration and migration. In the instance of large grafts, this should also allow the penetration of macroscopic tissue architecture such as microvasculature to provide the graft with mass transport capacity and preventing necrosis. Although understanding physical and chemical compo- nents of a surface is critical in anticipating its tissue interaction, we must also consider internal structure and chemistry. As material erodes, degrades or remodel’s its internal structure becomes visible to cells and tissues; therefore, in a material whose bulk is not homogenously distributed resulting in differing internal and external structures or chemistry, we must consider that this continually changing environment modifies tissue behaviour beyond a standard acute inflammatory interaction time frame, as with every new material parameters, as the device degrades comes a fresh wave of inflammatory investigation. In vitro, testing of material degradation is also the subject of debate. Various methods exist which include most basically incubating the material for extensive time courses in simulated body fluids and assessing changes in physical appearance and mechanical properties as time progresses. A further degree of complexity is induced by incubating with single enzymes such as collagenase or enzyme cocktails to mimic the tissue remodelling envi- ronment or the influence of microbial contamination secreting enzymes into the material field. Much like many in vitro tests, however, modelling degradation in vitro is challenging to directly extrapolate to a true in vivo tissue reaction as the actions of continuing waves of cells inter- rogating and potentially digesting the material, in con- junction with the huge and undefined array of metabolically active proteins present in interstitial fluids, is difficult to recreate in vivo. In vitro, it is also difficult to create tests which appreciate the repeated and perhaps unpredictable influence of mechanical loading on a mate- rial and how this may synergise with the tissue environ- ment to modify the materials degradation kinetics.

Therefore, although we can broadly predict a materials degradation profile based on in vitro tests, this is only absolutely quantified when a material progresses to an animal model. The rheometry of a device, its stiffness, rigidity and elasticity also influences cell function. Cells reside in specific niches in which they are accustomed to a set of sheer parameters; therefore, subjection to physical forces to which they are unaccustomed has the capacity to modify their response. We must also consider how this component of cell/material interaction may change over time if the mechanics of a device changes due to wear or repeated loading. In the instance of hernia, there are many compounding engineering features which influence the biocompatibility of a prostheses, this is particularly challenging in the instance of biologics [14]. Biologic meshes are subject to proprietary processing steps which are partially or completely undisclosed, this modifies the inflammatory and mechanical characteristics of the tissue to render it suitable as a medical device. A number of these processes result in material changes at the level to instruct cell fate. Starting material may influence biocompatibility by utilising a range of source species and tissues which it would be naive to assume contain identical structural and mechanical features and induce the same cell responses. Proprietary grafts exist which are xenogeneic (porcine or bovine) or allogenic (cadaveric) derived from a range of tissues including dermis, small intestinal submucosa and pericardium. These materials vary in degradation profile, continually unmasking epitopes to inflammatory cells as degeneration modifies material. When materials are decellularised, it is critical that cells and cellular remnants are completely removed or cell pattern recognition receptors will flag the graft as non-self resulting in destruction. The goal therefore is to ensure the cellular component is removed, meaning the remaining matrix con- tains only proteins (largely collagen) that are conserved across individuals and species and provides a 3D environ- ment for cells to colonise, without epitopes for the immune system to differentiate graft from autogenous ECM. In some instances, this core protein structure is modified using cross-linking to join the collagen molecules within the decellularised tissue. Cross-linking increases resistance to bacterial proteases, which increases its persistence in infected fields. It would be reasonable to assume changes induced by cross-linking modify inflammatory character- istics by presenting cells with a protein conformed in a manner that they are not accustomed. However, studies by our own group both in vitro and in vivo have generally demonstrated that cross-linking does not significantly influence cell response in dermis derived biomaterials [30].

123

176

Hernia (2015) 19:169–178

A final variable is sterilisation, with many techniques employed including gamma rays, electron beams and eth- ylene oxide.

The surgical consequences of unsatisfactory biocompatibility

In the light of progress in the scientific understanding of material/tissue interaction, and advances surgical applica- tion, the ideal characteristics of hernia repair materials are perpetually refined. The qualities described by Cumberland and Scales, and Hamer-Hodges and Scott, are as follows:

non-carcinogenic, chemically inert, resistant to mechanical strain, sterile, unresponsive to body and tissue fluids, able to limit foreign-body reaction, modifiable to defect size, and non-allergenic [16]. Required characteristics applied specifically in the instance of biologic meshes include resistance to infection, anti-adhesive, and the capacity to act like native tissue [17]. Additionally, an ideal material should be associated with very little surgical morbidity, such as seroma, easy to handle in open and laparoscopic instances, and cost effective. Due to significant advances in synthetic and composite meshes, a vast array of products for repair of uncomplicated hernias is available [18]. Despite this panel of ideal material properties and the apparent ease at which a material can be engineered to modify its mechanical characteristics, synthetic meshes are not without risk, even in clean, uncontaminated operations, which highlights the need for a more thorough under- standing at a material science level of how tissues and material are interacting. Often the consequences of syn- thetic mesh, failure can be catastrophic. This includes erosion, adhesion, fistulation or encapsulation, which are incredibly debilitating requiring challenging surgeries to correct with potential for life modifying long-term out- comes such as stoma as a consequence of bowel resection during adhesiolysis [31, 32]. Undoubtedly, synthetics improve the quality of life of many people each day, however at present, we can say that these materials are not ideal in every patient, and a more thorough understanding of how the biological and cellular origins of inter-patient differences in response is required as a matter of urgency to take the field forward. Identifying the ideal repair material for complicated hernias remains a challenge. Although classification of her- nia complexity is debated and has been the subject of a number of consensus meetings [19, 20], it is accepted that complex hernias include those in which there is contami- nated or clean–contaminated tissue, exposed bowel or fis- tula, and significant loss of domain such that low- or tension- free repair is not possible. The use of synthetic meshes for complex hernias has been associated with infection,

123

recurrence and significant adhesions to intra-abdominal organs, leading to obstruction, erosion and fistula formation [21]. Infected synthetic meshes are difficult to treat, fre- quently require surgery to explant, and may leave the patient with a colonised defect requiring prolonged treatment [17]. For synthetic mesh repair in the presence of contamination or infection, several authorities recommend a two-stage pro- cedure in which contamination is managed surgically in the first stage, with the hernia repaired 6–12 months later [22]. Biologic meshes are promoted for repair of complex hernias, as these meshes become incorporated into the wound, acting as a scaffold for tissue repair leading to a strong, well-healed, vascularised wound [22]. Due to the nature of biologics, the adhesions associated with synthetic mesh should not occur and vascularisation allows delivery of immune cells and antibiotics [17]. Retrospective con- sideration of biologic mesh efficacy is compromised due to the compounding variables induced throughout their application: the variety of available products, the tech- niques used to repair hernias (onlay, inlay, sublay, com- ponent separation) and the complex nature of hernias being repaired [23, 24]. To date, no data from randomised con- trolled trials specifically designed to investigate biologic meshes is available [21]. Besides the previously mentioned issues surrounding classification of, the classification of ventral hernia repair and subsequent outcome reporting is also the subject of debate [24]. Current literature is based on single centre case series, (and reviews of those series), with several series reporting a variety of hernia complexity, a range of surgical techniques, a variety of mesh types, and use of biologic mesh in both clean and contaminated wounds [2123, 25, 26]. The recently launched European hernia registry is a development designed to bring clarity to reporting of ventral hernia repairs and outcomes, and should provide robust, prospective data [24, 27]. Although overt recur- rence of the hernia might be considered a failure, one author, reporting a series of ventral hernia repairs using a variety of biologic meshes in contaminated or clean–con- taminated tissue, reported a 31.3 % rate of hernia recur- rence at 21.7 months, but described the majority of recurrences as asymptomatic, with only 17.5 % requiring further surgery. The same study showed a recurrence-free survival of 92 % at 1 year, 77 % at 2 years and 51 % at 3 years—a trend that questions the durability of such repairs [22]. Another study, describing repairs of ventral hernias secondary to emergency or trauma laparotomy, reported a 100 % recurrence with AlloDerm at 1 year, and a 31 % recurrence rate with FlexHDAs, all recurrences progressed to further surgery [26]. A recent review of 17 retrospective series showed that recurrence depended upon wound class, with an overall recurrence rate of 13.8 %. The recurrence rates were 2.9 % in clean or clean/contaminated

Hernia (2015) 19:169–178

177

cases, 19.4 % in complicated hernias, and 23.1 % in con- taminated or dirty wounds. The rate of recurrences requiring further surgery is not clear from that review [23]. Associated surgical and wound morbidity is well docu- mented in ventral hernia repairs, with a review of 25 series showing a 46.3 % complication rate, wound infections being most common in 15.9 % of all cases [23]. It is interesting to note that only 4.9 % of infected implants required removal to manage infection; the remainder were salvageable via to non-operative management. Although not all series report seroma formation, that problem is commonly documented following biologic mesh repair. The review of 25 series shows an overall seroma rate of 14.2 % [23]; however, it is likely that the surgical technique, mesh, and wound class all influence seroma risk. A study describing Strattice to repair contaminated or infected ventral hernias documented ser- oma in 28 % of cases [28].

The future of biocompatibility—a surgeon’s perspective

The current literature relating to biologic meshes is not sufficient to support evidence-based decision-making in ventral hernia repair, as it is composed of relatively small series, often with limited follow-up period. Moreover, the series describe significant variations in technique, using a range of biologic meshes. Interpreting data from such series is compounded by the fact that the nature of com- plicated or contaminated ventral hernias is also a signifi- cant source of variability. The only valid conclusion that can be supported from current literature is that it seems safe to use biologics in contaminated fields; however, the recurrence and morbidity rates remain high, especially in challenging cases. As Belyansky and Heniford [29] point out in response to the work of Kissane and Itani [21], large prospective trials would be challenging and expensive, considering the number of variables that have to be con- sidered; however, from a surgical point of view, estab- lishing the role of biologic meshes in this complex field must be driven by evidence from prospective trials, in combination with development of biologic materials.

Conflict of interest NB declares conflict of interest not directly related to the submitted work. CB declares no conflict of interest. NJS declares conflict of interest not directly related to the sub- mitted work. JH declares conflict of interest not directly related to the submitted work.

References

1. Williams DF (2008) On the mechanisms of biocompatibility. Biomaterials 29:2941–2953

2.

Mutsaers SE, Bishop JE, McGrouther G, Laurent GJ (1997) Mechanisms of tissue repair from wound healing to fibrosis. Int J Biochem Cell Biol 29:5–17

3.

ten Raa S, van der Tol P, Sluiter W, Hofland LJ, van Eijck CHJ, Jeekel H (2006) The role of neutrophils and oxygen free radicals in post-operative adhesions. J Surg Res 136:45–52

4.

Butterfield TA, Best TM, Merrick MA (2006) The dual role of neutrophils and macrophages in inflammation: a critical balance between tissue damage and repair. J Athl Train 41:457–465

5.

Anderson JM, Miller KM (1984) Biomaterial biocompatibility and the macrophage. Biomaterials 5:5–10

6.

Kirkpatrick CJ, Krump-Konvalinkova V, Unger RE, Bittinger F, Otto M, Peters K (2002) Tissue response and biomaterial inte- gration: the efficacy of in vitro methods. Biomol Eng 19:211–217

7.

Daubay DA, Franz MG (2003) Acute wound healing: the biology of acute wound failure. Surg Clin North Am 83:463–481

8.

Hunt JA, Vince DG, Williams DF (1993) Image analysis in the evaluation of biomaterials. J Biomed Eng 15:39–45

9.

Bryan N, Ashwin H, Bayon Y, Wohlert S, Smart N, Hunt J (2012) The innate oxygen dependent immune pathway is a sensitive parameter to predict the performance of biological graft materi- als. Biomaterials 33:6380–6392

10.

Bryan N, Ashwin H, Bayon Y, Wohlert S, Smart N, Hunt JA (2012) In vitro activation and degranulation of human acute inflammatory cells in response to direct contact with synthetic hernia repair meshes. Clin Biochem 45:672–676

11.

Bryan N, Andrews KD, Hunt JA (2011) Elucidating the contri- bution of the elemental composition of fetal calf serum towards antigenic expression of primary human umbilical vein endothelial cells in vitro. Biosci Rep 31:199–210

12.

Bryan N, Ashwin H, Bayon Y, Wohlert S, Smart N, Hunt JA (2012) Reactive oxygen species (ROS); a family of fate deciding molecules pivotal in successful wound healing. ECM 24:249–265

13.

Bryan N, Birch P, Bond D, Stanley C, Hunt J (2012) Examination of the effect of acoustic force capture on the ultra purification of leukocyte subpopulations translatable throughout a variety research and diagnostic applications. J Tissue Eng Regen Med.

14.

Smart N, Bryan N, Hunt JA (2012) A scientific evidence for the efficacy of biologic implants for soft tissue reconstruction. Colorectal Dis 14:1–6

15.

Khang D, Choi J, Im YM, Kim YJ, Jang JH, Kang SS, Nam TH, Song J, Park JW (2012) Role of subnano-, nano- and micron- surface features on osteoblast differentiation of bone marrow mesenchymal stem cells. Biomaterials 33:5997–6007

16.

Hamer-Hodges DW, Scott NB (1985) Surgeon’s workshop:

replacement of an abdominal wall defect using expanded PTFE sheet (GORE-TEX). J Royal Coll Surg Edin 30:65–67

17.

Cevasco M, Itani KM (2012) Ventral hernia repair with synthetic, composite and biologic mesh: characteristics, indications and infection profile. Surg Infect 13:209–215

18.

Le D, Deveney CW, Reaven NL, Funk SE, McGaughey KJ, Martindale RG (2012) Mesh choice in ventral hernia: so many choices, so little time. Am J Surg 205:602–607

19.

Muysoms FE, Miserez M, Berrevoet F, Camanelli G, Champault GG, Chelala E, Dietz UA, Eker HH, El Nakadi I, Hauters P, Hidalgo Pascual M, Hoeferlin A, Klinge U, Montgomery A, Sim- mermacher RKJ, Simons MP, Smietanski M, Sommeling C, Tol- lens T, Vierendeels T, Kingsnorth A (2009) Classification of primary and incisional abdominal wall hernias. Hernia 13:407–414

20.

Kanters AE, Krpata DM, Blatnik JA, Novitsky YM, Rosen MJ (2012) Modified hernia grading scale to stratify surgical site occur- rence after open ventral hernia repairs. J Am Coll Surg 215:787–793

21.

Kissane N, Itani KMF (2012) A decade of ventral incisional hernia repairs with biologic acellular dermal matrix: what have we learned? Plast Reconstr Surg 130:194S–202S

123

178

Hernia (2015) 19:169–178

22. Rosen MJ, Krpata DM, Ermlich B, Blatnik J (2013) A 5-year clinical experience with single-staged repairs of infected and contaminated abdominal wall defects utilising biologic mesh. Ann Surg 257:991–996

23. Slater NJ, van der Kolk Hendriks T, van Goor H, Bleichrodt RP (2013) Biologic grafts for ventral hernia repair: a systematic review. Am J Surg 205:220–230

24. Muysoms F, Campanelli G, Champault GG, DeBeaux AC, Dietz UA, Jeekel J, Klinge U, Kockerling F, Mandala V, Montgomery A, Morales Conde S, Puppe F, Simmermacher RKJ, Smietanski M, Miserez M (2012) EuraHS: the development of an interna- tional online platform for registration and outcome measurement of ventral abdominal wall hernia repair. Hernia 16:239–250

25. Hood K, Millikan K, Pittman T, Zelhart M, Secemsky B, Rajan M, Myers J, Luu M (2013) Abdominal wall reconstruction: a case series of ventral hernia repair using the component separation technique with biologic mesh. Am J Surg 205:322–328

26. Bochiccio GV, De Castro GP, Bochiccio KM, Weeks J, Rodri- guez E, Scalea TM (2013) Comparison study of acellular dermal matrices in complicated hernia surgery. J Am Coll Surg

217:606–613

27. Muysoms FE, Deerenberg EB, Peeters E, Agresta F, Berrevoet F, Campanelli G, Ceelen W, Champault GG, Corcione F, Cu- ccurullo D, DeBeaux AC, Dietz UA, Fitzgibbons RJ, Gillon JF,

123

Hilgers RD, Jeekel J, Kyle-Leinhase I, Kockerling F, Mandala V, Montgomery A, Morales-Conde S, Simmermacher RKJ, Schumpelick V, Smietanski M, Walgenbach M, Miserez M (2013) Recommendations for reporting outcomes in abdominal wall repair. Hernia 17:423–433

28. Itani KMF, Rosen M, Vargo D, Awad S, DeNoto G, Butler CE and the RICH study group (2012) Prospective study of single- stage repair of contaminated hernias using a biologic porcine tissue matrix: the RICH study. Surgery 152:498–505

29. Belyansky I, Heniford T (2012) Discussion: a decade of ventral incisional hernia repairs with biologic acellular dermal matrix:

what have we learned? Plast Reconstruct Surg 130:203S–205S

30. Bryan N. Ashwin H, Bayon Y, Wohlert S, Smart N, Hunt JA (2013) Evaluation of the in vivo performance of tissue-based biomaterials in a rat full thickness abdominal wall model. JBMR. doi:10.1002/jbm.b.33050. [Epub ahead of print]

31. Halm JA, de Wall LL, Steyerberg EW, Jeekel J, Lange LW (2007) Intraperitoneal polyproylene mesh repair complicates subsequent abdominal surgery. World J Surg 31:423–431

32. Gray SH, Vick CC, Graham LA, Finan KR, Neumayer LA, Hawn MT (2008) Risk of complications from enterotomy or unplanned bowel resection during elective hernia repair. Arch Surg

143(6):582–586