Reconstructive materials used in surgery: Classification and host response

INTRODUCTION — This topic review will discuss the classification and histologic behavior of reconstructive materials used in surgery. Prosthetic materials in the surgical management of specific conditions are discussed in individual topic reviews.

The indications for and techniques for placing such materials are reviewed separately.

TYPES OF MATERIALS — Surgical reconstructive materials differ by source and include synthetic mesh, autografts, allografts, and xenografts. The advantages and disadvantages of the different material types are found in the table (table 1).

Synthetic materials — Synthetic materials are available as both absorbable (eg, polyglactin 910 [Vicryl], polyglycolic acid [Dexon]) and nonabsorbable mesh (eg, polypropylene [Marlex, Prolene] and expanded polytetrafluoroethylene [ePTFE, Gore-tex]). Compared with biologic grafts, advantages of synthetic materials include greater availability (does not require harvesting) and lower cost of the material. However, infectious and erosion complications, especially with transvaginal surgery, have prompted a search for alternative materials [1]. (See 'Host response' below.)

Autografts — Autograft materials are harvested from the patient undergoing the procedure. Skin autografting is the basis for burn reconstruction, and various tissue flaps can provide coverage for complex soft tissue defects. Tensor fascia lata and rectus fascia, the most commonly used autografts for stress urinary incontinence surgery (eg, bladder neck sling), have been used for decades and yield predictable results [2]. (See "Skin autografting" and "Overview of flaps for soft tissue reconstruction".)

A clear advantage of autografts is that the host response is rarely problematic. However, the use of autografts is limited by morbidity associated with harvesting the tissue (eg, pain, bleeding, infection, hernia formation), as well as varying size, quantity, and quality of tissue [3]. (See 'Host response' below.)

Allografts — Allografts are processed cadaveric fascia lata or acellular dermal matrices (ADM) from human donors (table 2). The material is decellularized and rendered nonimmunogenic by washing processes, which are designed to remove cellular debris without permanently damaging the connective tissue scaffold. (See "Skin substitutes".)

Although tissue banks in North America are accredited through the American Association of Tissue Banks (AATB), evidence-based comparisons among allografts distributed by different companies are difficult because the harvesting, processing, and preservation of these materials are varied and proprietary. (See 'Allograft processing' below.)

Human donors of allografts are screened for blood-borne pathogens; those with risk factors or who test positive for HIV, syphilis, HTLV, or hepatitis B or C are excluded. There have been no reported cases of donor-related infection with these viral infections associated with the use of allografts.

With respect to severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the Physicians Council of the AATB assessed the prospects for mitigating the risk relating to coronavirus disease 2019 (COVID-19) [4]. While testing is available using polymerase chain reaction (PCR), usually from oropharyngeal or nasopharyngeal samples, these diagnostic tests have not been validated for cadaveric samples. Nevertheless, the Physicians Council subgroup has voted to recommend laboratory testing for COVID-19 as part of donor screening, using PCR testing obtained via swab, if available. (See "COVID-19: Issues related to solid organ transplantation", section on 'Donor screening'.)

This class of materials eliminates the morbidity associated with autologous fascia harvest. However, allografts have consistently underperformed compared with autologous and synthetic materials.

Xenografts — Xenografts are acellular extracts of collagen from nonhuman sources, harvested with or without additional extracellular matrix components (table 3). They differ in the source species (bovine or porcine), site of harvest (pericardium, dermis, or small intestine submucosa), and whether or not chemical crosslinking is used in the processing of the material [5].

It is unclear whether architectural differences due to harvest site (eg, dermis with high elastin content versus intestinal submucosa with no elastin content) affect in vivo performance.

Some patients may object to the use of porcine or bovine implants due to religious or cultural issues.

HOST RESPONSE — Histologic behavior to a class of material depends upon the physical and structural properties of the prosthesis. The type of host response is a key determinant of synthetic mesh-related complications. As an example, among patients undergoing a sacrocolpopexy, the risk of mesh exposure was 4.2 times greater for those who had polytetrafluoroethylene (PTFE) compared to those that used non-PTFE mesh [6].

Terminology — Host response to reconstructive materials is described by varying terminology [7], as an example:

●Incorporation: Graft infiltration by host cells, allowing neovascularization and collagen deposition throughout the material.

●Encapsulation: Collagen and connective tissue deposit at the periphery of the material, rather than the infiltration of graft by host cells. (See 'Encapsulation' below.)

●Mixed response: Incorporation occurs at graft pores, and encapsulation occurs around the remaining material.

●Resorption: Material is replaced by host neo-connective tissue.

Materials that undergo encapsulation or mixed response are at increased risk of infection and erosion. (See 'Classification' below.)

The key factor(s) for the host response to each type of material are as follows:

●Synthetic mesh: Fiber diameter, pore size, and weave or knit type.

●Autografts: Autologous tissue, host response is rarely problematic.

●Allografts: Key factor(s) have not been elucidated.

●Xenografts: Removal of host cellular components and chemically crosslinking.

The International Urogynecological Association (IUGA) and the International Continence Society (ICS) have published a Joint Terminology and Classification Report on mesh-related complications that classifies the event based upon category (asymptomatic, symptomatic, infection with or without abscess), location (vagina, urinary tract, bowel, skin or patient compromise), and timing since implantation (within 48 hours of surgery, until 6 months and beyond 6 months from surgery). Clinically, prosthetic material-related complications are manifested as exposures (mesh fibers protruding into the lumen of the vaginal or surrounding organs), mesh contraction or prominence (the wrinkling or shrinkage of the material projecting beyond the surface of the epithelium), infection (including sinus tract formation and abscess), bowel or bladder injury, or bleeding complications and systemic compromise. It is rare to have an infected, unexposed material. In some locations, such as the vagina, it is difficult to differentiate between infection and colonization, and clinical signs of infection are necessary to make the diagnosis. Furthermore, whether these clinical manifestations are due to host response or technical aspects of material use (eg, thickness of vaginal flaps) is unclear.

Synthetic materials — Host response to synthetic prosthetic materials depends on several factors: absorbability, pore size (space between fibrils), weave or knit (mono- or multifilament), and weight (mass per surface area mesh).

Absorbable versus nonabsorbable — Both absorbable and nonabsorbable materials cause initial and chronic inflammatory tissue responses after being implanted [8]. The quantity and quality of local inflammation depend directly on the specific material used.

The implantation of absorbable material (eg, polyglactin 910 [Vicryl] or polyglycolic acid [Dexon]) elicits a chronic foreign body reaction and promotes fibroblast activity. After complete absorption (30 days for polyglactin 910; 90 days for polyglycolic acid), the prosthetic material is replaced by collagen-rich connective tissue [9]. Absorbable materials are less likely to become infected than nonabsorbable materials and are less harmful to viscera [10]. Based on animal studies, one disadvantage to absorbable implants is that the resultant scar tissue weakens after the material is absorbed and may not provide the necessary long-term repair strength [11].

Nonabsorbable prosthetic materials (eg, polypropylene [Prolene, Marlex]) are associated with more connective tissue reaction than absorbable materials. Thus, repair strength is increased due to the continuing presence of the implant and greater scar formation.

Partially absorbable materials have been developed with the aim of decreasing the amount of permanent foreign material while maintaining mechanical resistance. These materials are composites of absorbable and nonabsorbable materials (eg, polypropylene with polyglactin [Vypro] or with poliglecaprone-25 [Ultrapro]) and are lightweight; few reports are available regarding clinical performance [12,13]. (See 'Weight' below.)

Pore size and weave — Pore size influences cellular infiltration, risk of infection, the tendency to form adhesions, and mesh density and flexibility [14,15]. Macropores (>75 microns; 1 mm = 1000 microns) allow for host cell colonization with collagen deposition and angiogenesis, although this has only been reported in animal studies [16,17]. In addition, in vivo evidence suggests that the large pore allows unrestricted immune cell access and minimizes the risk of infection.

Micropores (<10 microns) result in restriction of fibroblast and immune cell colonization to the material surface (eg, PTFE). Therefore, collagen and connective tissue deposition occurs at the periphery rather than by infiltration by host cells. This process is called encapsulation [5].

In addition, microporous materials are at increased risk of becoming infected, as large immune cells (macrophages and natural killer cells, mean diameter 9 to 20 microns) cannot infiltrate the interstices (space between fibers) to phagocytose bacteria (<1 micron) [14].

Weight — Mesh weight, or density, is another descriptor for synthetic materials (figure 1). Generally speaking, the weight is a measure of how much material is present per square meter.

Weight and elasticity are determined by pore size. Meshes with larger pores generally have a lower weight and are more elastic than those with smaller pores. The larger the pores, the less material content and the more flexible the elicited scar [18]. It has been postulated that lightweight materials may be less prone to infection or erosion than heavy-weight materials. Even amongst macroporous materials, those with pore size <1 mm elicit suboptimal tissue in-growth, chronic inflammation, and can undergo fibrotic encapsulation [19,20]. Materials with pore size <1 mm appear to induce stress shielding, a process in which the rigid material impairs force transmission through the tissue leading to suboptimal vaginal smooth muscle response, atrophy, and degeneration. Ideally, surgeons should seek to use materials with pore sizes >1 mm.

Classification — Nonabsorbable synthetic materials are generally described by type, a classification based on pore size and weave (figure 1) [21]. There are four types (figure 1):

●Type I – Macroporous (>75 microns), monofilament polypropylene (eg, Prolene, Marlex). Type I meshes are further subdivided into heavy-, mid-, and lightweight materials.

Combination absorbable and nonabsorbable mesh material, such as polypropylene with either polyglactin (eg, Vypro) or with poliglecaprone (eg, Ultrapro), are also available. The absorbable component provides temporary wound integrity and maximizes pore size (as the absorbable component undergoes hydrolysis during the intervening two to three months postimplantation) to minimize the risk of mesh exposure.

●Type II – Microporous (<10 microns), eg, expanded polytetrafluoroethylene (eg, Gore-tex).

●Type III – Macroporous (>75 microns) grafts with either microporous components or multifilament fiber structure (eg, polyethylene [Mersilene]). Histologic behavior is similar to type II materials. This category includes some polypropylene materials (eg, ObTape [heat-bonded, heat-welded polypropylene with microporous components] and IVS Tunneler [multifilament polypropylene]). Both materials have been associated with increased rates of infections and erosion [22]; ObTape has been removed from the market, and IVS Tunneler is not being marketed.

●Type IV – Submicroscopic pore size (eg, polypropylene sheet [Cellgard]). These are not commonly used in gynecologic surgery.

Allograft processing — Allograft processing is intended to render a graft nonimmunogenic. The methods companies use to process human tissue to produce allografts vary, and details are not publicly disclosed (table 2), thereby interfering with evidence-based comparisons among allografts.

There are at least four major steps in allograft processing:

Harvesting — Harvesting of fascia lata and dermis is done aseptically from cadavers. For dermis products, the epidermis is mechanically or chemically separated from the underlying dermis; the collagen matrix is then extracted from the dermis.

Cell removal — After harvesting, the material is made acellular by washing it in a solution to extract cellular components. These processes render the allograft nonimmunogenic by eliminating cell surface proteins (eg, human histocompatibility antigen, HLA) and sugar moieties. The solvents must remove cellular debris without permanently denaturing the protein scaffold.

Despite assertions by manufacturers, it appears that the cellular extraction process could be better [23]. In one study, the HLA type of allograft donors could still be identified (using the polymerase chain reaction) in freeze-dried and Tutoplast-processed grafts; Repliform interfered with the assay in the study, and its antigenicity could not be determined. These data suggest that not all contaminating DNA had been extracted. However, this finding does not signify that the grafts were immunogenic. Host rejection is mediated by host lymphocyte detection of cell surface proteins and sugar moieties, not DNA or other intracellular components.

Preservation and sterilization — The collagen and extracellular matrix are preserved via lyophilization (freeze-dried under vacuum) or solvent dehydration. In some tissue banks, the graft is then terminally sterilized (usually with gamma irradiation).

The processes employed for these steps appear to affect the final graft strength. As an example, the rate of lyophilization alters the size of ice crystal formation (the more ice crystals formed, the weaker the final product), and the terminal sterilization temperature influences the degree of free radical formation.

Both lyophilization and sterilization affect the final integrity of the material and, theoretically, graft performance. However, the effect of preimplantation biomechanical characteristics of a graft on in vivo behavior is controversial [7,24-28]. No data on postimplantation biomechanical testing exist for allografts; it appears that preimplantation mechanical testing does not predict in vivo behavior of xenografts [7].

In fact, the few studies that have compared different allografts by preimplantation strength are inconsistent with reports of clinical performance. As an example, one study reported that the preimplantation maximum load to failure (minimum force needed to tear the tissue) and stiffness were significantly higher for solvent dehydrated compared with freeze-dried fascia lata [24]. Nevertheless, solvent-dehydrated allografts have consistently been reported to result in early failure after pubovaginal sling [25,26], sacrocolpopexy [27], and vaginal reinforced repairs [28]. Similarly, the only study that evaluated vaginal repair with a freeze-dried graft reported a high failure rate [29].

Xenograft processing — Similar to allografts, xenografts processing includes harvesting, cell removal, preservation, and sterilization. These processes are generally proprietary, and thus comparisons among different grafts are difficult. In our experience, some patients have had increased tissue reactions with xenografts, raising concerns about how well the processes remove cellular debris. (See 'Allograft processing' above.)

In addition, xenografts undergo chemical crosslinking, which is one of the key determinants of the host response to xenografts (table 3).

Crosslinking — Noncrosslinked xenografts (eg, Surgisis, Xenform) are purported to serve as biologic scaffolding for fibroblast and angioblast ingrowth; they are replaced by host connective tissue at varying rates, depending on material thickness, method of freezing, and site of implantation [5].

Conversely, cross-linked materials (eg, Pelvicol) are functionally nonporous. As a result, host cellular infiltration does not occur, and the material is surrounded by a connective tissue envelope (ie, encapsulation, similar to synthetic type II materials) [5]. (See 'Classification' above.)

If crosslinked materials are perforated before implantation (eg, Pelvisoft and Fortagen), they undergo a mixed response with incorporation at the perforations and encapsulation in the remainder of the material (similar to what is observed for type III synthetic materials) [5]. (See 'Classification' above.)

Although crosslinking is performed to permanently stabilize the material, crosslinked graft resorption over time is an area of concern. According to the manufacturer report only, the process of resorption would be different for crosslinked and non-crosslinked grafts.

It is unknown whether eventual graft loss affects the structural integrity of the repair. Furthermore, it remains to be elucidated if resorption occurs in all patients implanted with a crosslinked graft or only in a subset of patients rejecting the graft. The following studies illustrate this:

●A small histopathologic study evaluated females who had undergone placement of a midurethral sling using a crosslinked porcine dermal collagen implant (Pelvicol) and who required sling revision due to recurrent stress urinary incontinence or urinary retention [30]. Time from graft placement to biopsy ranged from 15 to 67 months. Although the graft remained encapsulated in the early biopsy specimens, there was a progressive increase in the immunologic response at the graft-host interface at 21 weeks. Histiocyte and multinucleated giant cell infiltration of the graft was detected by 42 weeks [30]. No residual graft was found in the 58- and 67-week samples, suggesting that the materials were resorbed in the interim [5,30].

●In another study, females with a Pelvicol reinforced posterior repair were evaluated upon reoperation one year postoperatively; no residual Pelvicol was found [31].

Encapsulation — Encapsulation is a host response in which collagen and connective tissue deposit at the periphery of the material rather than being infiltrated. Using mesh materials that become encapsulated should be avoided for midurethral sling procedures and vaginal prolapse repairs. Four different kits introduced since the mid-1990s for urinary incontinence (ProteGen, ObTape, and UraTape) and for vaginal prolapse repair (IVS Tunneler) have elicited an encapsulation response (eg, a type II or type III mesh). They were approved by the US Food and Drug Administration but were removed from the market within two to three years of introduction due to an unacceptably high rate of complications (sinus tract formation, urethral and bladder fistula, chronic discharge, elevated rates of exposure) [32]. (See "Surgical management of stress urinary incontinence in females: Retropubic midurethral slings", section on 'Synthetic mesh complications'.)

USE OF RECONSTRUCTIVE MATERIALS

Hernia repair — The use of prosthetic material dramatically reduces the incidence of recurrence associated with ventral, inguinal, and femoral hernia repair. Polypropylene is the most common material used. (See "Overview of treatment for inguinal and femoral hernia in adults" and "Overview of abdominal wall hernias in adults".)

Pelvic organ prolapse — Pelvic organ prolapse (POP) and urinary incontinence are common comorbid disorders in females and can greatly impact the quality of life [33]. (See "Pelvic organ prolapse in females: Epidemiology, risk factors, clinical manifestations, and management" and "Surgical approach to rectal procidentia (rectal prolapse)".)

SUMMARY AND RECOMMENDATIONS

●There are four categories of reconstructive surgical materials: synthetic material, autografts, allografts, and xenografts (table 1). (See 'Types of materials' above.)

●The host response to a class of material depends upon the physical and structural properties of the prosthesis. Synthetic material pore size determines host response. The type of response predisposes to an increased risk of infection and mesh exposure. (See 'Host response' above.)

●Allograft processing is intended to render a material nonimmunogenic. Processing techniques vary by manufacturer, and information is not publicly disclosed, thus limiting evidence-based comparison of allografts. (See 'Allograft processing' above.)

●Chemical crosslinking is a key determinant of the host response to xenografts. (See 'Xenograft processing' above.)

●Macroporous polypropylene mesh is the only material that has consistently been shown to improve outcomes following hernia and abdominal prolapse repairs and midurethral slings for urinary incontinence. Without mesh, these reconstructions are followed by high recurrence rates (eg, hernia, prolapse). (See 'Use of reconstructive materials' above.)